CN118246699B

CN118246699B - Urban water affair scientific scheduling method and device, storage medium and electronic equipment

Info

Publication number: CN118246699B
Application number: CN202410658230.0A
Authority: CN
Inventors: 童麒源; 李旭; 林峰; 陈颂华; 赵治成; 邱泽; 刘文彬; 邱雅旭; 何锦; 陈颖聪; 肖霖; 苏洋; 林鑫
Original assignee: Shenzhen Water Group Co ltd
Current assignee: Shenzhen Water Group Co ltd
Priority date: 2024-05-27
Filing date: 2024-05-27
Publication date: 2024-09-20
Anticipated expiration: 2044-05-27
Also published as: CN118246699A

Abstract

The application relates to a scientific scheduling method, a device, a storage medium and electronic equipment for urban water affairs, and relates to the technical field of water affair scheduling, wherein the method comprises the following steps: predicting a target delivery pressure and a target delivery flow corresponding to a scheduling short-term evaluation period through a first prediction model based on a target water pump state and a corresponding total water demand; and finally determining the target clean water tank liquid level in the scheduled short-term evaluation period through a second prediction model corresponding to the urban water plant based on the target outgoing flow and the target incoming flow. Predicting the target water pump energy consumption of the urban water plant through a third prediction model corresponding to the urban water plant based on the target water pump state and the target delivery pressure; and determining the total score of the corresponding single scheduling scheme through a preset scoring rule, and determining the single scheduling scheme with the maximum total score as the optimal scheduling scheme. The application has the effect of reducing the difficulty of determining a reasonable scheduling scheme.

Description

Urban water affair scientific scheduling method and device, storage medium and electronic equipment

Technical Field

The application relates to the technical field of water affair scheduling, in particular to a scientific scheduling method and device for urban water affairs, a storage medium and electronic equipment.

Background

The urban water service scheduling refers to scientific scheduling and management of water resources to ensure the safety, high efficiency and sustainability of urban water supply, and relates to unified planning, coordination and management of a plurality of links such as water sources, water plants, water supply networks, water users and the like so as to realize the optimal configuration and reasonable utilization of the water resources. The main tasks of urban water affair scheduling include: reasonably arranging a water taking plan of a water source area according to urban water demand; and the production and water supply plans of each water plant are coordinated, so that the stability and sufficiency of the water supply amount are ensured.

At present, with the acceleration of the urban process and the continuous increase of the complexity of water demand, so that urban water service scheduling faces a plurality of challenges, and mainly how to realize the rationality and superiority of a scheduling scheme of urban water service is generally as follows: based on the water supply network hydraulic model, the effect of the scheduling scheme of the urban water affairs is evaluated to determine a better scheduling scheme, but in the urban water affair scheduling process, the involved water supply network is more complex, and based on the water supply network hydraulic model, the process of simulating and evaluating the scheduling scheme is more complex and tedious, so that the difficulty of determining a reasonable scheduling scheme is higher.

Disclosure of Invention

In order to reduce the difficulty of determining a reasonable scheduling scheme, the application provides a scientific scheduling method, device, storage medium and electronic equipment for urban water affairs.

In a first aspect of the present application, there is provided a method for scientific scheduling of urban water affairs, comprising:

Acquiring a first prediction model corresponding to a municipal water plant, wherein the first prediction model is used for predicting the delivery pressure and delivery flow of the municipal water plant;

Determining the total water demand of urban users in a scheduled short-term evaluation period, acquiring a target water pump state of the urban water plant in a single scheduling scheme at a scheduling time, and predicting a target delivery pressure and a target delivery flow corresponding to the scheduled short-term evaluation period through the first prediction model based on the target water pump state and the total water demand, wherein the scheduled short-term evaluation period is a short-term period from the scheduling time to a short-term preset time;

Determining a target factory entering flow corresponding to the urban water plant under the single scheduling scheme, predicting a target clean water tank liquid level change corresponding to the urban water plant under the scheduling short-term evaluation period through a second prediction model corresponding to the urban water plant based on the target factory leaving flow and the target factory entering flow, and determining a target clean water tank liquid level under the scheduling short-term evaluation period according to the target clean water tank liquid level change, wherein the second prediction model is used for predicting the clean water tank liquid level change of the urban water plant;

Predicting the target water pump energy consumption of the urban water plant through a third prediction model corresponding to the urban water plant based on the target water pump state and the target delivery pressure, wherein the third prediction model is used for predicting the water pump energy consumption of the urban water plant;

based on the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme, determining the total score of the corresponding single scheduling scheme through a preset scoring rule, and determining the single scheduling scheme with the maximum total score as an optimal scheduling scheme.

By adopting the technical scheme, according to the total water demand of the urban users in the scheduling short-term evaluation period and the target water pump state of each water pump of the urban water plant in the single scheduling scheme to be executed at the scheduling moment, factors influencing the delivery pressure and the target delivery flow generated by the single scheduling scheme after the single scheduling scheme is executed at the scheduling moment, namely the scheduling short-term evaluation period are determined, and further, the target delivery pressure and the target delivery flow in the scheduling short-term evaluation period are predicted and obtained based on the target water pump state and the total water demand through a first prediction model. Further, determining the target factory inflow of the urban water plant after the single scheduling scheme is executed, inputting the target factory inflow and the target factory outflow in the scheduled short-term evaluation period into a second prediction model, and finally determining the target clean water tank liquid level in the scheduled short-term evaluation period after the single scheduling scheme is executed. And inputting the state of the target water pump and the target delivery pressure into a third prediction model to obtain the target water pump energy consumption in a scheduling short-term evaluation period, and finally scoring the corresponding single scheduling scheme based on the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times, so that the rationality and superiority of the single scheduling scheme are better evaluated, and finally, the single scheduling scheme with the maximum total score is determined as the optimal scheduling scheme, thereby rapidly and accurately evaluating the scheduling scheme and reducing the difficulty of determining the reasonable scheduling scheme.

Optionally, the determining, based on the target factory pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme, the total score of the corresponding single scheduling scheme according to a preset scoring rule specifically includes:

Respectively scoring the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme through a preset scoring function to obtain corresponding index scores;

Determining a corresponding first weight according to the scheduling short-term evaluation period, and determining a second weight corresponding to the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times through a hierarchical analysis method;

and carrying out weighted average on the index scores according to the first weight and the second weight to obtain the total score of the corresponding single scheduling scheme.

By adopting the technical scheme, after the index scores of the four indexes of the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times of each single scheduling scheme are determined, the performance of each index is then evaluated from the time dimension, namely, the first weight is determined according to each moment in the scheduling short-term evaluation period. Further, the second weight of each index is independently determined from the index dimension, so that the influence degree of the second weight on the rationality of the corresponding single scheduling scheme is evaluated, and finally, the corresponding index is weighted and averaged according to the first weight and the second weight of the index, so that the total score of the corresponding single scheduling scheme is obtained, and the rationality and the superiority of the single scheduling scheme are objectively and comprehensively evaluated.

Optionally, the scoring, by a preset scoring function, the target factory pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme respectively to obtain corresponding index scores specifically includes:

Scoring the target delivery pressure corresponding to each single scheduling scheme through a preset first scoring function to obtain a corresponding first index score, wherein the larger the target delivery pressure is, the lower the first index score is, the larger the target delivery pressure is when the target delivery pressure exceeds a pressure upper limit value, and the first index score is 1 when the target delivery pressure is not above a pressure lower limit value, the larger the target delivery pressure is, and the first index score is between a pressure lower limit value and a pressure upper limit value;

scoring the target clean water tank liquid level corresponding to each single scheduling scheme through a preset second scoring function to obtain a corresponding second index score, wherein when the target clean water tank liquid level exceeds a liquid level upper limit value, the larger the target clean water tank liquid level is, the lower the second index score is, when the target clean water tank liquid level does not exceed a liquid level lower limit value, the larger the target clean water tank liquid level is, the larger the second index score is, and when the target clean water tank liquid level is between a liquid level lower limit value and a liquid level upper limit value, the second index score is 1;

Scoring the energy consumption of the target water pump corresponding to each single scheduling scheme through a preset third scoring function to obtain a corresponding third index score, wherein the third scoring function is semi-normal distribution, and the larger the energy consumption of the target water pump is, the smaller the third index score is;

And scoring the preset scheduling times corresponding to each single scheduling scheme through a preset fourth scoring function to obtain a corresponding fourth index score, wherein the fourth scoring function is an indication function.

By adopting the technical scheme, the first index score of the target delivery pressure of the single scheduling scheme is determined through the first scoring function, so that the reasonable degree of the delivery pressure after the single scheduling scheme is executed at the scheduling time is evaluated to a certain extent. And similarly, sequentially evaluating the clean water tank liquid level, the water pump energy consumption and the reasonable degree of the preset scheduling times after the single scheduling scheme through the second scoring function, the third scoring function and the fourth scoring function, so that the rationality and the superiority of the single scheduling scheme executed at the scheduling moment are conveniently evaluated subsequently.

Optionally, the obtaining a first prediction model corresponding to the urban water plant specifically includes:

Constructing an equation expression of the factory pressure or the factory flow of the urban water factory;

determining final values of all parameters in the equation expression, substituting each final value into the equation expression to obtain a first prediction model, storing the first prediction model into a database, and obtaining the first prediction model from the database, wherein the equation expression is as follows:

In the method, in the process of the invention, The delivery pressure or delivery flow of the urban water plant at the moment t; The total water demand at time t; in order to be the state of the water pump i at time t, Is a dummy variable of the time period in the day, if the time t belongs to the time period j in the day; Otherwise take，The value range is (0, + -infinity), parameter c is the intercept term, parameterTo change the total water demandConversion of delivery pressure or delivery flowEffects, parameters ofFor the effect of the water pump on the change of the delivery pressure or delivery flow,For the threshold value of the state of the water pump i at the moment t, the parameterFor the influence of dummy variable in daily period on delivery pressure or delivery flow conversion, parametersIs an error term.

By adopting the technical proposal, the utility model has the advantages that,And (3) withThe product of the two water pumps can better reflect the influence degree of the opening of each water pump on the delivery pressure or delivery flow, and the influence degree of each water pump is accumulated to obtain. Since the time period in the day also has an influence on the delivery pressure or delivery flow, the methodAnd (3) withReflecting the influence of each time period on the delivery pressure or delivery flow, and summing the time periodsAnd (3) withAnd thus better reflects the overall impact of all time-of-day periods on the factory pressure or the factory flow. And finally, considering the influence of the total water demand on the delivery pressure or the delivery flow, finally determining an equation expression of the delivery pressure or the delivery flow, further determining the final value of each parameter in the equation expression, substituting the final value into the equation expression, and determining a first prediction model capable of accurately predicting the delivery pressure or the delivery flow.

Optionally, the method further comprises:

If the scheduling operation in the optimal scheduling scheme is the start-stop of the urban water plant water pump, when a future long-term optimal scheduling scheme from the current moment to a long-term preset moment is required to be determined, determining a scheduling trigger moment corresponding to each single scheduling scheme in a period from the scheduling moment to the long-term preset moment, wherein the scheduling trigger moment is a future moment when the delivery pressure is lower than a pressure lower limit value or exceeds a pressure upper limit value, and the scheduling moment is earlier than the long-term preset moment;

Based on the scoring rule, selecting an optimal scheduling scheme to be triggered from all scheduling schemes corresponding to each scheduling triggering time of the same single scheduling scheme;

Combining each single scheduling scheme with each optimal scheduling scheme to be triggered to obtain a scheme chain, and determining the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain;

and determining the total score of the corresponding scheme chain based on the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain through the scoring rule, and determining a starting point scheme in the scheme chain with the maximum total score as the future long-term optimal scheduling scheme, wherein the starting point scheme is the scheduling scheme with the earliest scheduling time node in the corresponding scheme chain.

By adopting the technical scheme, the optimal scheduling scheme to be triggered at each scheduling triggering time of the single scheduling scheme is evaluated through the preset scoring rule, then the single scheduling scheme is combined with the corresponding optimal scheduling scheme to be triggered to obtain a scheme chain, each scheme chain is regarded as an integral scheduling scheme, the total score of the scheme chain is determined continuously through the preset scoring rule, and the starting point scheme in the scheme chain with the maximum total score is selected as a future long-term optimal scheduling scheme, so that the long-term optimal scheduling scheme at the scheduling time is accurately determined.

Optionally, the method includes:

Determining the total score of each scheme chain as a target dependent variable, and determining the starting point scheme in each scheme chain as a target independent variable;

and fitting a function relation between the target independent variable and the target dependent variable through a preset neural network model to obtain a value function.

By adopting the technical scheme, the function relation between the target dependent variable and the target independent variable is fitted through the preset neural network model to obtain the value function, so that the effect of the whole scheme chain after execution is not required to be simulated in the subsequent determination of the long-term optimal scheduling scheme, the value function is determined only by the starting point scheme in the scheme chain, and the starting point scheme in the scheme chain with the maximum value function is determined as the long-term optimal scheduling scheme, thereby obtaining the superior scheduling scheme at the scheduling moment in a faster way under the long-term consideration.

Optionally, the determining the target incoming flow corresponding to the urban water plant under the single scheduling scheme specifically includes:

determining a raw water pump state of a raw water pump room in the single scheduling scheme;

and predicting the target incoming flow corresponding to the urban water plant through a preset fourth prediction model based on the raw water pump state, wherein the fourth prediction model is used for predicting the incoming flow of the urban water plant.

By adopting the technical scheme, the water pump state of the raw water pump room can influence the water outlet flow of the raw water pump room and further influence the factory inlet flow of the urban water factory, so that the raw water pump state in a single scheduling scheme at the scheduling time is determined and is input into a fourth prediction model, the target factory inlet flow of a short-term evaluation period after the single scheduling scheme is obtained can be predicted, and the subsequent determination of an optimal scheduling scheme based on the target factory inlet flow is facilitated.

In a second aspect of the present application, there is provided a municipal water science scheduling device, comprising in particular:

The model acquisition module is used for acquiring a first prediction model corresponding to the urban water plant, wherein the first prediction model is used for predicting the delivery pressure and delivery flow of the urban water plant;

the dispatching system comprises a dispatching determination module, a dispatching control module and a dispatching control module, wherein the dispatching determination module is used for determining the total water demand of urban users in a dispatching short-term evaluation period, acquiring the target water pump state of the urban water plant in a single dispatching scheme at a dispatching moment, and predicting the target dispatching pressure and the target dispatching flow corresponding to the dispatching short-term evaluation period through the first prediction model based on the target water pump state and the total water demand, wherein the dispatching short-term evaluation period is a short-term period from the dispatching moment to a short-term preset moment;

The liquid level determining module is used for determining target factory entering flow corresponding to the urban water plant under the single scheduling scheme, predicting target clean water tank liquid level change corresponding to the urban water plant in the scheduling period of the scheduling short-term evaluation period through a second prediction model corresponding to the urban water plant based on the target factory leaving flow and the target factory entering flow, and determining target clean water tank liquid level in the scheduling period of the scheduling short-term evaluation period according to the target clean water tank liquid level change, wherein the second prediction model is used for predicting clean water tank liquid level change of the urban water plant;

the energy consumption determining module is used for predicting the target water pump energy consumption of the urban water plant through a third prediction model corresponding to the urban water plant based on the target water pump state and the target delivery pressure, and the third prediction model is used for predicting the water pump energy consumption of the urban water plant;

The scheme determining module is used for determining the total score of the corresponding single scheduling scheme through a preset scoring rule based on the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme, and determining the single scheduling scheme with the maximum total score as the optimal scheduling scheme.

By adopting the technical scheme, the model acquisition module acquires the first prediction model, the delivery determination module predicts the target delivery pressure and the target delivery flow of the urban water plant in the dispatching short-term evaluation period through the first prediction model based on the total water demand and the target water pump state, and then the liquid level determination module predicts the target clean water tank liquid level change through the second prediction model, so that the corresponding target clean water tank liquid level in the dispatching short-term evaluation period is determined. And then, the energy consumption determining module predicts the energy consumption of the target water pump through a third prediction model based on the state of the target water pump and the target delivery pressure, and finally, the scheme determining module scores each single scheduling scheme and determines the single scheduling scheme with the maximum total score as the optimal scheduling scheme.

In a third aspect of the application there is provided a computer readable storage medium having a computer program stored therein, which when loaded and executed by a processor performs the method steps of any of the first aspects.

In a fourth aspect of the present application, there is provided an electronic device, comprising:

a processor, a memory and a computer program stored in the memory and capable of running on the processor, the processor being configured to load and execute the computer program stored in the memory to cause the electronic device to perform the method according to any one of the first aspects.

In summary, the present application includes at least one of the following beneficial technical effects: and predicting the target delivery pressure and the target delivery flow based on the target water pump state and the total water demand through a first prediction model. Further, determining the target factory entering flow of the urban water factory after the single scheduling scheme is executed, inputting the target factory entering flow and the target factory leaving flow at the scheduling time into a second prediction model, and finally determining the target clean water tank liquid level after the single scheduling scheme is executed. And inputting the state of the target water pump and the target delivery pressure into a third prediction model to obtain the target water pump energy consumption in a scheduling short-term evaluation period, and finally scoring the corresponding single scheduling scheme on four indexes of the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times, so that the rationality and superiority of the single scheduling scheme are better evaluated, and finally, the single scheduling scheme with the maximum total score is determined as the optimal scheduling scheme, thereby rapidly and accurately evaluating the scheduling scheme and reducing the difficulty of determining the reasonable scheduling scheme.

Drawings

FIG. 1 is a schematic flow chart of a scientific scheduling method for urban water affairs, which is provided by the embodiment of the application;

FIG. 2 is a schematic diagram of a scoring function provided by an embodiment of the present application;

FIG. 3 is a schematic flow chart of another method for scientific scheduling of urban water affairs according to an embodiment of the application;

Fig. 4 is a schematic structural diagram of a scientific scheduling device for urban water affairs according to an embodiment of the present application;

Fig. 5 is a schematic structural diagram of another urban water science scheduling device according to an embodiment of the present application.

Reference numerals illustrate: 11. a model acquisition module; 12. a factory determining module; 13. a liquid level determination module; 14. an energy consumption determining module; 15. a scheme determination module; 16. a long-term optimization module; 17. and (5) a rapid optimization module.

Detailed Description

In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present application, not all embodiments.

In describing embodiments of the present application, words such as "exemplary," "such as" or "for example" are used to mean serving as examples, illustrations or explanations. Any embodiment or design described herein as "illustrative," "such as" or "for example" is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "illustratively," "such as" or "for example," etc., is intended to present related concepts in a concrete fashion.

In the description of the embodiments of the present application, the term "and/or" is merely an association relationship describing an association object, and indicates that three relationships may exist, for example, a and/or B may indicate: a alone, B alone, and both A and B. In addition, unless otherwise indicated, the term "plurality" means two or more. For example, a plurality of systems means two or more systems, and a plurality of screen terminals means two or more screen terminals. Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating an indicated technical feature. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.

Referring to fig. 1, an embodiment of the application discloses a flow diagram of a municipal water science scheduling method, which can be realized by a computer program or can be operated on a municipal water science scheduling device based on von neumann system. The computer program can be integrated in an application or can be run as a stand-alone tool class application, and specifically comprises:

S101: and acquiring a first prediction model corresponding to the urban water plant, wherein the first prediction model is used for predicting the factory pressure and the factory flow of the urban water plant.

Specifically, the urban water plant is an important component of the urban water supply system, and tap water meeting the standard of drinking water is supplied to users of the city through a water pump. In embodiments of the application, the municipal water plant may be a single water plant, and in other embodiments, the municipal water plant may represent multiple water plants. The urban water plant comprises a clean water tank, and the clean water tank is a regulation structure for uniformly supplying water to the urban water plant and meeting the uneven water consumption of users. The specific process is as follows: when water is supplied to the urban water plant, water at a water source area is firstly conveyed into the urban water plant through a raw water pump room and finally into a clean water tank, so that the liquid level of the clean water tank is controlled within a reasonable range. Further, a first prediction model corresponding to the urban water plant is obtained, and as the delivery pressure and delivery flow of the urban water plant are influenced by the total water demand of urban users and the water pump state of the urban water plant, namely the water pump is turned on and off, for example, the water pump state is unchanged, the larger the total water demand is, the smaller the delivery pressure is generally; when the water pump is started by the urban water plant or other urban water plants under the condition that the total water demand is unchanged, the delivery pressure of the urban water plant is increased, and a first preset prediction model is constructed based on the relation, so that the first preset prediction model can predict the delivery pressure and delivery flow of the urban water plant at the future time according to the total water demand and the water pump state. The delivery pressure refers to the pressure when the urban water flows out from the water outlet of the urban water plant, and is used for ensuring that the water can be smoothly conveyed to various areas of the city. The factory flow refers to the water quantity from the water outlet of the urban water plant in unit time, and reflects the water supply capacity and efficiency of the urban water plant.

In one implementation manner, a first prediction model corresponding to the urban water plant is obtained, specifically:

The equation expression for constructing the factory pressure or the factory flow of the urban water factory is as follows:

In the method, in the process of the invention, The delivery pressure or delivery flow of the urban water plant at the moment t; The total water demand at time t; in order to be the state of the water pump i at time t, Is a dummy variable of the time period in the day, if the time t belongs to the time period j in the day; Otherwise take，The value range is (0, + -infinity), parameter c is the intercept term, parameterTo change the total water demandConversion of delivery pressure or delivery flowIs used for the effect of (a) and (b),Conversion parameters, parameters for delivery pressure or delivery flowFor the effect of the water pump on the change of the delivery pressure or delivery flow,For the threshold value of the state of the water pump i at the moment t, the parameterFor the influence of dummy variable in daily period on delivery pressure or delivery flow conversion, parametersIs an error term. M represents a water pump M, and K represents a time period K in the day.

When (when)Less thanIn the time-course of which the first and second contact surfaces,Taking 0, the change of the water pump state does not affect the delivery pressure or delivery flow. When the time t does not belong to the intra-day period j,That is, the daily period does not affect the factory pressure or the factory flow.

The water pump i may be a water pump of a city water plant, or may be a water pump of a city water plant and/or a water pump of another city water plant. In addition, since the distribution ratio of the total water demand among urban users varies, the delivery pressure or the delivery flow is affected, and such distribution ratio is closely related to the time of day. Thus, in the embodiment of the present application, the intra-day period dummy variable is also taken as an independent variable. The daily period can be understood as a period affecting the delivery pressure or the delivery flow, and the dummy variable of the daily period is a virtual variable representing the influence of the daily period on the delivery pressure or the delivery flow, and is usually 0 or 1. In other embodiments, if the water pump i is a variable frequency pump, thenThe ratio of the i frequency of the water pump at the moment t to the rated frequency of the water pump at the moment t can be used for reflecting the state of the water pump at the moment t, and parameters are similar to the stateThe frequency of the water pump i can be increased by 1 unit to have an effect on the delivery pressure or delivery flow conversion.

Further, determining final values of parameters in the equation expression, substituting each final value into the equation expression to obtain a first prediction model, storing the first prediction model into a database, and acquiring the first prediction model from the database, wherein the specific process is as follows: acquiring historical data of delivery pressure or delivery flow, total water demand and water pump state in a first preset time before the current moment, wherein the first preset time can be the past 1 year or 3 years, and then determining parameters in an equation expression by using a nonlinear least square method (NonlinearLeastSquare, NLS)、、、AndAnd constructing prior distribution of the corresponding parameters based on the corresponding estimated values, thereby limiting the value range of the parameters. Further, based on historical data in a second preset time before the current moment and prior distribution of each parameter, a mean value of posterior distribution corresponding to each parameter is obtained by using Markov chain Monte Carlo (Markov Chain Monte Carlo, MCMC), the mean value of posterior distribution is determined to be a final value of the corresponding parameter, each final value is substituted into an equation expression, a first prediction model is obtained and stored in a database, and when future time factory pressure or factory flow prediction is needed, the first preset model is directly obtained from the database. It should be noted that the second preset time may be 30 days in the past. In addition, the MCMC algorithm is an algorithm that uses markov chains to sample from a random distribution. The specific process of determining the mean value of posterior distribution corresponding to each parameter by using the MCMC algorithm is as follows: and constructing a likelihood function according to the historical data in the second preset time, namely comparing the historical data with the predicted data of the first predicted model in training, calculating a likelihood value, and finally constructing the likelihood function. And then calculating the product of the likelihood function and the prior distribution to obtain posterior distribution, finally generating a sample of the parameter through an MCMC algorithm, constructing a corresponding Markov chain to enable the corresponding Markov chain to be continuously approximate to the posterior distribution, and finally calculating statistics (such as a mean value and the like) of the posterior distribution through the sample which approximates to the posterior distribution. This is because it is difficult to directly calculate the statistics of complex posterior distributions. In other embodiments, a Hamiltonian Monte Carlo (Hybrid Monte Carlo, HMC) algorithm may also be used to determine the mean of the posterior distribution corresponding to each parameter, where the specific process is: 1. hamiltonian system modeling: 1) For a given probability distribution p (θ1), its corresponding negative log likelihood is constructed as a potential energy function U (θ1). 2) A set of momentum variables q of the same dimensions as the parameter vector are introduced, giving a uniform distribution as a kinetic energy function K (q). 2. Kinetic simulation: 1) And (3) position updating: using the position update law of hamilton dynamics (based on newton's second law), the instantaneous velocity of the parameter vector at the current momentum is calculated and the parameter position is updated by integration (typically using Leapfrog integrators). Here, the time step ϵ and the gradient ∂ U/∂ θ1 of the potential energy function are key parameters. 2) Momentum update: after each position update, the potential energy function U (θ1) and the kinetic energy function K (q) are recalculated according to the new position, and the hamilton function H (θ1, q) is updated. Then, it is decided whether to accept the new parameter position by Metropolis-Hastings acceptance criterion. The probability of acceptance is typically calculated based on the difference between the old and new Hamiltonian. 3. Iteration and sampling: repeating the dynamic simulation process to generate a series of samples. In each iteration, the system evolves according to the current Hamiltonian and acceptance criteria and generates a new sample. These samples will progressively follow the target profile p (θ1). 4. And (3) convergence judgment: convergence criteria are set to determine when to stop iteration. And finally calculating the average value of posterior distribution based on the generated samples.

S102: determining the total water demand of urban users in a scheduling short-term evaluation period, acquiring the target water pump state of the urban water plant in a single scheduling scheme at the scheduling moment, and predicting the target delivery pressure and the target delivery flow corresponding to the scheduling short-term evaluation period through a first prediction model based on the target water pump state and the total water demand, wherein the scheduling short-term evaluation period is a short-term period from the scheduling moment to a short-term preset moment.

Specifically, if the water service scheduling related personnel needs to determine the optimal scheduling scheme of the scheduling time after the current time, the total water demand of urban users at each time under the scheduling short-term evaluation period is first determined, and one feasible determination mode is as follows: training a Time-series dense encoder (Time-SERIES DENSE Encoder, tiDE) through historical data of total water demand of urban users and corresponding covariate characteristics to obtain a trained future water demand prediction model, wherein the Time-series dense encoder is suitable for long-term Time-series prediction. Covariates are characterized by time covariates (e.g., what minute is within an hour, what hour is within a day, what day is within a week, etc.), days from holidays (e.g., days from mid autumn, days from labor, etc.), weather variables (e.g., maximum temperature, minimum temperature, whether rainfall, etc.), and holiday information (e.g., workday 0, saturday 1, holiday 2, etc.), as these covariates all have an impact on the total demand of urban users. And finally, predicting the total water demand of urban users at each time in the short-term evaluation period of the scheduling through the future water demand prediction model. Wherein the scheduling time is a time node for executing a single scheduling scheme in the future.

Further, a single scheduling scheme is selected from a preset scheme set, wherein the scheme set comprises at least one scheduling scheme which can be executed at a future scheduling moment, and the single scheduling scheme is one scheduling scheme of the scheme set. The state of each water pump in the urban water plant in the selected single scheduling scheme is determined as a target water pump state, for example, the target water pump state can be that 1 water pump in the urban water plant is started or 1 water pump is closed. Further, the target water pump status and the total water demand of this single scheduling scheme are input into the first predictive model, i.e. the target water pump status is taken asThe total water demand is taken asSubstituting the target delivery pressure and the target delivery flow of the urban water plant into an equation expression of the final value of the determined parameter to obtain the target delivery pressure and the target delivery flow of the urban water plant at the corresponding time in the scheduled short-term evaluation period. In other embodiments, if there is also a status of the city other water plant water pump in the single scheduling scheme, the target water pump status, the water pump status at the time of the city other water plant scheduling, and the total water demand for each of the scheduled short-term assessment periods are input into the first predictive model for prediction.

It should be noted that, for example, when the scheduling time is 10 points, after the single scheduling scheme is executed by 10 points, the conditions of the delivery pressure and the delivery flow of the urban water plant in the latter half hour are determined, and then the scheduling short-term evaluation period is 10 points-10 points for 30 minutes, so that the effect of the single scheduling scheme after being executed is evaluated to a certain extent, and the subsequent determination of the optimal scheduling scheme is facilitated.

S103: and determining the target factory entering flow corresponding to the urban water plant under a single scheduling scheme, predicting the target clean water tank liquid level change corresponding to the urban water plant under a short-term evaluation period of scheduling by using a second prediction model corresponding to the urban water plant based on the target factory leaving flow and the target factory entering flow, and determining the target clean water tank liquid level under the short-term evaluation period of scheduling according to the target clean water tank liquid level change, wherein the second prediction model is used for predicting the clean water tank liquid level change of the urban water plant.

Specifically, after the execution of the single scheduling scheme, after the determination of the target delivery pressure and the target delivery flow of the urban water plant at each moment in the scheduling short-term evaluation period, the target flow of the urban water plant corresponding to the urban water plant in the scheduling short-term evaluation period after the execution of the scheduling moment needs to be further determined, wherein the target flow of the urban water plant refers to the raw water flow entering the urban water plant at the scheduling moment. One possible way of determining is: the state of the raw water pump house is extracted from the single scheduling scheme, i.e., the raw water pump in the raw water pump house is turned on or off. Further, since the water outlet flow of the raw water pump room is increased after the new 1 water pump is started, the factory inlet flow of the city water factory entering the downstream is synchronously increased under the condition that no valve adjustment exists and no other city water factory exists, therefore, the historical water pump states of each group of the raw water pump room and the corresponding historical factory inlet flow of the city water factory in the past 30 days are obtained and used as training samples, then the training samples are sequentially divided into training sets, verification sets and test sets according to the dividing ratio of 8:1:1, then the training samples in the training sets are input into the existing machine learning model, the training is performed on the training sets, the model performance is optimized, then the training process is monitored by using the training samples in the verification sets, overfitting is prevented, finally the test sets are input into the model, the performance of the model is evaluated, and finally the model capable of predicting the training completion model of the corresponding factory inlet flow based on the water pump state of the raw water pump room, namely the factory inlet flow prediction model is obtained. The machine learning model may be a convolutional neural network model. Further, the raw water pump state is input into a factory entering flow prediction model, so that the target factory entering flow corresponding to the urban water factory is predicted.

Further, the target factory flow and the target factory entering flow are input into a second prediction model, so that the target clean water tank liquid level change corresponding to the urban water plant at each moment in a scheduled short-term evaluation period is obtained, wherein the second prediction model is constructed based on the function relation of clean water tank liquid level change= (factory entering flow-factory flow)/clean water tank bottom area, the prior art is omitted, and finally the clean water tank liquid level at the current moment in the urban water plant and the target clean water tank liquid level change in the scheduled short-term evaluation period are accumulated and summed to obtain the target clean water tank liquid level of the urban water plant in the scheduled short-term evaluation period.

S104: and predicting the target water pump energy consumption of the urban water plant through a third prediction model corresponding to the urban water plant based on the target water pump state and the target delivery pressure, wherein the third prediction model is used for predicting the water pump energy consumption of the urban water plant.

Specifically, the target water pump state and the target delivery pressure of the urban water plant in the short-term evaluation period are input into a third prediction model, and the target water pump energy consumption of the urban water plant in the short-term evaluation period is predicted, wherein the target water pump energy consumption refers to the total energy consumption of all water pumps started in the urban water plant. The energy consumption of the water pump is taken as a dependent variable when the delivery pressure is kept unchanged, and the water pump is in an on state, so that the energy consumption can be increased; when the state of the water pump is kept unchanged, the delivery pressure rises, and the energy consumption of the water pump generally decreases. The method comprises the steps of obtaining historical data of historical water pump states, historical delivery pressures and historical water pump energy consumption in the past 30 days, taking the historical data as training samples, sequentially dividing the training samples into a training set, a verification set and a test set according to a dividing ratio of 6:2:2, inputting the training samples in the training set into an existing machine learning model, training the training samples, optimizing model performance, monitoring a training process by using the training samples in the verification set, preventing overfitting, inputting the test set into the model, evaluating the performance of the model, and finally obtaining a training completed model capable of predicting corresponding water pump energy consumption based on the water pump states and the delivery pressures, namely a third prediction model. Further, the target water pump state and the target delivery pressure at each moment in the scheduled short-term evaluation period are input into a third prediction model, and the corresponding target water pump energy consumption is predicted. And by analogy, the target delivery pressure, the target clean water tank liquid level and the target water pump energy consumption corresponding to other single scheduling schemes in the scheme set can be determined. The machine learning model can adopt a cyclic neural network model or a long-term and short-term memory network model.

S105: based on the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme, determining the total score of the corresponding single scheduling scheme through a preset scoring rule, and determining the single scheduling scheme with the maximum total score as the optimal scheduling scheme.

In one implementation mode, the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme are respectively scored through a preset scoring function, so that corresponding index scores are obtained;

And carrying out weighted scoring on the scores of the indexes according to the first weight and the second weight to obtain the total score of the corresponding single scheduling scheme.

Specifically, the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme in the scheme set are respectively scored one by one through a preset scoring function, so that corresponding index scores are obtained, wherein the maximum value of the index scores is 1, and the minimum value of the index scores is 0. The method comprises the following steps: for the same single scheduling scheme, the corresponding target factory pressure is scored through a preset first scoring function, so that a corresponding first index score is obtained, in the embodiment of the application, the first scoring function is a trapezoid scoring function, as shown in (a) in fig. 2, when the target factory pressure exceeds the upper pressure limit value, the higher the target factory pressure is, the lower the first index score is, when the target factory pressure does not exceed the lower pressure limit value, the higher the target factory pressure is, the higher the first index score is, and when the target factory pressure is between the lower pressure limit value and the upper pressure limit value, the first index score is 1.

Further, the corresponding target clean water tank liquid level is scored through a preset second scoring function, so that a corresponding second index score is obtained, the second scoring function is also a trapezoid scoring function, as shown in (b) in fig. 2, when the target clean water tank liquid level exceeds the liquid level upper limit value, the larger the target clean water tank liquid level is, the lower the second index score is, when the target clean water tank liquid level does not exceed the liquid level lower limit value, the larger the target clean water tank liquid level is, and when the target clean water tank liquid level is between the liquid level lower limit value and the liquid level upper limit value, the second index score is 1.

Further, the corresponding target water pump energy consumption is scored through a preset third scoring function, so that a corresponding third index score is obtained, the third scoring function is semi-normal distribution, and as shown in (c) in fig. 2, the larger the target water pump energy consumption is, the smaller the third index score is.

Further, scoring the corresponding preset scheduling times through a preset fourth scoring function to obtain a corresponding fourth index score, wherein the fourth scoring function is an indication function, the indication function is a function with a value of 1 when a specific condition is met, and otherwise, the indication function is a function with a value of 0. The single scheduling scheme at the scheduling time does not have any scheduling action, the preset scheduling times are 0, and the fourth index score is 1; there is a scheduling action, the preset scheduling times are not 0, and the fourth index score is 0. The feasible mode of determining the preset scheduling times is as follows: the water service dispatching related personnel are input through a terminal, and the terminal can be a smart phone or a personal computer. In addition, when the preset scheduling times are 0, the corresponding single scheduling scheme has the highest rationality and superiority. In other embodiments, the fourth index score corresponding to the preset scheduling times may also be matched through a fourth index score matching table preset in the database, where the smaller the preset scheduling times, the higher the rationality and superiority of the corresponding single scheduling scheme, and the higher the corresponding fourth index score, the fourth index score matching table includes different preset scheduling times and corresponding fourth index scores, which are all set based on manual experience, and one possible setting manner is that: the preset scheduling times are 0, and the corresponding fourth index score is 1; the preset scheduling times are 1, and the corresponding fourth index score is 0.9; the preset scheduling times are 2, and the corresponding fourth index score is 0.8; the preset scheduling times are 3, the corresponding fourth index score is 0.7, and the preset scheduling times are increased by 1, and the corresponding fourth index score is decreased by 0.1. In still another embodiment, the fourth scoring function may be a function that is semi-normally distributed, where the smaller the number of preset schedules, the higher the corresponding fourth index score.

After each index score of a single scheduling scheme is determined, corresponding first weights are determined according to each moment in a scheduling short-term evaluation period and a preset decreasing exponential function in a database, the first weights decrease in an exponential function mode along with the increase of the moment, the smaller the moment is, the closer to the current moment, and the more objective the predicted delivery pressure, clean water tank liquid level or water pump energy consumption is after the single scheduling scheme is executed at the scheduling moment, the larger the first weights are. Wherein the decreasing exponential function is set by a person based on experience, one possible setting way is: the decreasing exponential function is set to f (x) =0.9 ^x, f (x) is the first weight, x is the time instant. And then, respectively determining a target delivery pressure, a target clean water tank liquid level, target water pump energy consumption and a second weight corresponding to each index in preset scheduling times by an analytic hierarchy process, wherein the indexes are factors influencing the rationality and superiority of a single scheduling scheme at scheduling time. Among these, analytic Hierarchy Process (AHP) is a decision analysis method that helps a decision maker to choose between multiple criteria or targets by building a multi-level structural model. The specific process of determining the second weight corresponding to each index through the analytic hierarchy process is as follows: assigning the relative importance of each index to the remaining index based on a 1-9 scale, wherein, for example, the relative importance of index a to index B represents: the importance of index a for scheduling scheme evaluation is compared to the relative importance of index B for scheduling scheme evaluation. Further, based on the respective assignments, a judgment matrix is constructed, the elements in each column of the judgment matrix representing the relative importance of the individual index and each of the remaining indexes. And finally, calculating the product of each column in the judgment matrix, taking the square root of each product, and carrying out normalization processing on each square root to finally obtain the second weight of each index.

And finally, carrying out weighted average on each index score, namely a first index score, a second index score, a third index score and a fourth index score, of the same single scheduling scheme at each moment in the scheduling short-term evaluation period according to the first weight and the second weight to obtain the total score of the corresponding single scheduling scheme, wherein the higher the total score is, the more reasonable and superior the single scheduling scheme is. Further, the single scheduling scheme with the largest total score in a plurality of single scheduling schemes at the scheduling time is determined to be the optimal scheduling scheme at the scheduling time.

Referring to fig. 3, an embodiment of the present application discloses a schematic flow chart of another method for scheduling urban water affairs science, which can be implemented by a computer program or can be run on an urban water affairs science scheduling device based on von neumann system. The computer program can be integrated in an application or can be run as a stand-alone tool class application, and specifically comprises:

S201: and acquiring a first prediction model corresponding to the urban water plant, wherein the first prediction model is used for predicting the factory pressure and the factory flow of the urban water plant.

S202: determining the total water demand of urban users in a scheduling short-term evaluation period, acquiring the target water pump state of the urban water plant in a single scheduling scheme at the scheduling moment, and predicting the target delivery pressure and the target delivery flow corresponding to the scheduling short-term evaluation period through a first prediction model based on the target water pump state and the total water demand, wherein the scheduling short-term evaluation period is a short-term period from the scheduling moment to a short-term preset moment.

S203: and determining the target factory entering flow corresponding to the urban water plant under a single scheduling scheme, predicting the target clean water tank liquid level change corresponding to the urban water plant under a short-term evaluation period of scheduling by using a second prediction model corresponding to the urban water plant based on the target factory leaving flow and the target factory entering flow, and determining the target clean water tank liquid level under the short-term evaluation period of scheduling according to the target clean water tank liquid level change, wherein the second prediction model is used for predicting the clean water tank liquid level change of the urban water plant.

S204: and predicting the target water pump energy consumption of the urban water plant through a third prediction model corresponding to the urban water plant based on the target water pump state and the target delivery pressure, wherein the third prediction model is used for predicting the water pump energy consumption of the urban water plant.

S205: based on the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme, determining the total score of the corresponding single scheduling scheme through a preset scoring rule, and determining the single scheduling scheme with the maximum total score as the optimal scheduling scheme.

Specifically, reference may be made to steps S101-S105, which are not described herein.

S206: if the scheduling operation in the optimal scheduling scheme is the on-off of the water pump of the urban water plant, when the future long-term optimal scheduling scheme from the current moment to the long-term preset moment needs to be determined, determining the scheduling trigger moment corresponding to each single scheduling scheme in the period from the scheduling moment to the long-term preset moment.

Specifically, if the dispatching operation is the start and stop of the water pump of the urban water plant in the optimal dispatching scheme at the dispatching time, for example, the dispatching time is 7 points 00 minutes, and the water plant A starts 1 water pump. The long-term preset time is preset in advance, the scheduling time is earlier than the long-term preset time, the short-term preset time is earlier than the long-term preset time, the currently determined optimal scheduling scheme is short-term, and is not optimal in a long-term period from the current time to the long-term preset time, so that the further optimal scheduling scheme relates to discrete variable optimization, and the on-off quantity of the water pumps of the urban water plant is a discrete value. The optimization of discrete variables refers to the optimization of the water pump on-off, and comprises the moment and the number of the water pump on-off. Further, determining a scheduling trigger time corresponding to each single scheduling scheme in a period from the scheduling time to a long-term preset time, wherein the scheduling trigger time is a future time when the factory pressure of the urban water plant is lower than a pressure lower limit value or exceeds a pressure upper limit value, when the factory pressure is lower than a pressure lower limit value or exceeds the pressure upper limit value, the state of the existing water pump is maintained continuously, so that water of partial users cannot be effectively supplied, at the moment, a water pump needs to be increased or decreased, and then the time when the condition for maintaining the safety of the water supply process is no longer met is called the scheduling trigger time. One possible way to determine the trigger time of the schedule is: and predicting the factory pressure of the urban water plant at each moment after the scheduling moment in real time through the first prediction model, and determining the corresponding moment as the scheduling triggering moment when the factory pressure exceeds the upper pressure limit value or is lower than the lower pressure limit value. It should be noted that, each individual scheduling scheme is a scheduling scheme corresponding to a scheduling time.

S207: and selecting an optimal waiting triggering scheduling scheme from the scheduling schemes corresponding to each scheduling triggering moment of the same single scheduling scheme based on the scoring rule.

S208: and combining each single scheduling scheme with each corresponding optimal scheduling scheme to be triggered to obtain a scheme chain, and determining the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain.

S209: and determining the total score of the corresponding scheme chain based on the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain through a scoring rule, and determining a starting point scheme in the scheme chain with the maximum total score as a future long-term optimal scheduling scheme.

Specifically, after the scheduling trigger time of each single scheduling scheme in a period from the scheduling time to a long-term preset time is determined, determining the factory pressure, the clean water tank liquid level, the water pump energy consumption and index scores corresponding to the preset scheduling times of each scheduling scheme corresponding to each scheduling trigger time of the same single scheduling scheme through a preset scoring function, finally determining a corresponding total score, and further selecting the scheduling scheme with the maximum total score as the optimal scheduling scheme to be triggered at the corresponding scheduling trigger time. Thus, each scheduling trigger time of the same single scheduling scheme determines a corresponding optimal scheduling scheme to be triggered. See step S105, and will not be described in detail herein.

Further, the single scheduling scheme is combined with each corresponding optimal scheduling scheme to be triggered to obtain a scheme chain, and the scheme chain corresponding to all the single scheduling schemes is obtained by analogy. Next, each scheme chain is used as an integral scheduling scheme, and the corresponding final delivery pressure, final clean water tank liquid level, final water pump energy consumption and final scheduling times are determined, and specific reference may be made to steps S101-S104, which are not described herein. And then sequentially grading the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times of each scheme chain through a preset grading function, finally determining the total score of the corresponding scheme chain, and finally selecting the starting point scheme in the scheme chain with the maximum total score to determine the future long-term optimal scheduling scheme, namely determining the scheduling scheme with more reasonable scheduling time under a long-term angle. The fewer the final scheduling times corresponding to a single scheme chain, namely the fewer the existing scheduling actions, the better the overall scheduling scheme formed by the scheme chain. In addition, the starting point scheme is the earliest scheduling scheme of the scheduling time node in the corresponding scheme chain.

In other embodiments, the total score of each scheme chain may be determined as a target dependent variable, the starting point scheme in each scheme chain is determined as a target independent variable, and further, a functional relationship between the target dependent variable and the target independent variable is fitted through a preset neural network model to obtain a value function, so that when the long-term optimal scheduling scheme is determined later, the effect after the whole scheme chain is executed is not required to be simulated, only the starting point scheme in the scheme chain is required to determine the value function, and the starting point scheme with the maximum value function is determined as the optimal long-term scheduling scheme at the scheduling moment. The neural network model may be a convolutional neural network model, and in other embodiments, may be a BP neural network model.

The implementation principle of the urban water affair scientific scheduling method of the embodiment of the application is as follows: and predicting the target delivery pressure and the target delivery flow based on the target water pump state and the total water demand through a first prediction model. Further, determining the target factory entering flow of the urban water factory after the single scheduling scheme is executed, inputting the target factory entering flow and the target factory leaving flow at the scheduling time into a second prediction model, and finally determining the target clean water tank liquid level after the single scheduling scheme is executed. And inputting the state of the target water pump and the target delivery pressure into a third prediction model to obtain the target water pump energy consumption in a scheduling short-term evaluation period, and finally scoring the corresponding single scheduling scheme on four indexes of the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times, so that the rationality and superiority of the single scheduling scheme are better evaluated, and finally, the single scheduling scheme with the maximum total score is determined as the optimal scheduling scheme, thereby rapidly and accurately evaluating the scheduling scheme and reducing the difficulty of determining the reasonable scheduling scheme.

The following are examples of the apparatus of the present application that may be used to perform the method embodiments of the present application. For details not disclosed in the embodiments of the apparatus of the present application, please refer to the embodiments of the method of the present application.

Referring to fig. 4, a schematic structural diagram of a scientific scheduling device for urban water affairs according to an embodiment of the application is shown. The device for scheduling the urban water science can be realized into all or part of the device through software, hardware or the combination of the software and the hardware. The device comprises a model acquisition module 11, a factory determination module 12, a liquid level determination module 13, an energy consumption determination module 14 and a scheme determination module 15.

The model obtaining module 11 is configured to obtain a first prediction model corresponding to the urban water plant, where the first prediction model is used to predict a factory pressure and a factory flow of the urban water plant;

The factory determining module 12 is configured to determine a total water demand of a city user in a scheduled short-term evaluation period, and obtain a target water pump state of the city water plant in a single scheduling scheme at a scheduled time, and predict, based on the target water pump state and the total water demand, a target factory pressure and a target factory flow corresponding to the scheduled short-term evaluation period through the first prediction model, where the scheduled short-term evaluation period is a short-term period from the scheduled time to a short-term preset time;

the liquid level determining module 13 is configured to determine a target factory inflow corresponding to the urban water plant under the single scheduling scheme, predict, based on the target factory inflow and the target factory inflow, a target clean water tank liquid level change corresponding to the urban water plant under the scheduling period of the scheduling short-term evaluation period according to a second prediction model corresponding to the urban water plant, and determine a target clean water tank liquid level under the scheduling period of the scheduling short-term evaluation period according to the target clean water tank liquid level change, where the second prediction model is used for predicting the clean water tank liquid level change of the urban water plant;

The energy consumption determining module 14 is configured to predict the target energy consumption of the water pump of the urban water plant through a third prediction model corresponding to the urban water plant based on the target water pump state and the target delivery pressure, where the third prediction model is used to predict the energy consumption of the water pump of the urban water plant;

the solution determining module 15 is configured to determine, according to a preset scoring rule, a total score of a corresponding single scheduling solution based on a target factory pressure, a target clean water tank level, a target water pump energy consumption and a preset scheduling number corresponding to each single scheduling solution, and determine a single scheduling solution with a maximum total score as an optimal scheduling solution.

Optionally, the scheme determining module 15 is specifically configured to:

And carrying out weighted average on the scores of the indexes according to the first weight and the second weight to obtain the total score of the corresponding single scheduling scheme.

Optionally, the scheme determining module 15 is specifically further configured to:

scoring the target delivery pressure corresponding to each single scheduling scheme through a preset first scoring function to obtain a corresponding first index score, wherein the larger the target delivery pressure is, the lower the first index score is, the larger the target delivery pressure is when the target delivery pressure exceeds the upper pressure limit value, the larger the target delivery pressure is when the target delivery pressure does not exceed the lower pressure limit value, the larger the first index score is, and the first index score is 1 when the target delivery pressure is between the lower pressure limit value and the upper pressure limit value;

Scoring the target clean water tank liquid level corresponding to each single scheduling scheme through a preset second scoring function to obtain a corresponding second index score, wherein when the target clean water tank liquid level exceeds a liquid level upper limit value, the target clean water tank liquid level is larger, the second index score is lower, when the target clean water tank liquid level does not exceed a liquid level lower limit value, the target clean water tank liquid level is larger, the second index score is larger, and when the target clean water tank liquid level is between a liquid level lower limit value and a liquid level upper limit value, the second index score is 1;

Scoring the energy consumption of the target water pump corresponding to each single scheduling scheme through a preset third scoring function to obtain a corresponding third index score, wherein the third scoring function is semi-normally distributed, and the larger the energy consumption of the target water pump is, the smaller the third index score is;

Optionally, the model acquisition module 11 is specifically configured to:

and determining the final value of each parameter in the equation expression, substituting each final value into the equation expression, obtaining a first prediction model, storing the first prediction model into a database, and obtaining the first prediction model from the database.

Optionally, as shown in fig. 5, the apparatus further includes a long-term optimization module 16, specifically configured to:

If the dispatching operation in the optimal dispatching scheme is the start-stop of the urban water plant water pump, when a future long-term optimal dispatching scheme from the current moment to the long-term preset moment is required to be determined, a dispatching trigger moment corresponding to each single dispatching scheme in a period from the dispatching moment to the long-term preset moment is determined, wherein the dispatching trigger moment is a future moment when the factory pressure is lower than the pressure lower limit value or exceeds the pressure upper limit value, and the dispatching moment is earlier than the long-term preset moment;

based on a scoring rule, selecting an optimal waiting triggering scheduling scheme from all scheduling schemes corresponding to each scheduling triggering moment of the same single scheduling scheme;

Combining each single scheduling scheme with each corresponding optimal scheduling scheme to be triggered to obtain a scheme chain, and determining the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain;

And determining the total score of the corresponding scheme chain based on the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain through a scoring rule, and determining a starting point scheme in the scheme chain with the maximum total score as a future long-term optimal scheduling scheme.

Optionally, the apparatus further comprises a fast optimization module 17, specifically configured to:

Optionally, the liquid level determining module 13 is specifically configured to:

determining a raw water pump state of a raw water pump room in a single scheduling scheme;

And predicting the target incoming flow corresponding to the urban water plant through a preset fourth prediction model based on the state of the raw water pump, wherein the fourth prediction model is used for predicting the incoming flow of the urban water plant.

It should be noted that, when executing the urban water science scheduling method, the urban water science scheduling device provided in the foregoing embodiment is only exemplified by the division of the foregoing functional modules, and in practical application, the foregoing functional allocation may be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the embodiment of the present invention provides a device for scientific scheduling of urban water affairs and an embodiment of a method for scientific scheduling of urban water affairs, which are the same conception, and detailed implementation processes are shown in the embodiment of the method, and are not described herein.

The embodiment of the application also discloses a computer readable storage medium, and the computer readable storage medium stores a computer program, wherein the computer program adopts the urban water science scheduling method of the embodiment when being executed by a processor.

The computer program may be stored in a computer readable medium, where the computer program includes computer program code, where the computer program code may be in a source code form, an object code form, an executable file form, or some middleware form, etc., and the computer readable medium includes any entity or device capable of carrying the computer program code, a recording medium, a usb disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a Random Access Memory (RAM), an electrical carrier signal, a telecommunication signal, a software distribution medium, etc., where the computer readable medium includes, but is not limited to, the above components.

The urban water science scheduling method of the embodiment is stored in the computer readable storage medium through the computer readable storage medium, and is loaded and executed on a processor, so that the storage and application of the method are convenient.

The embodiment of the application also discloses electronic equipment, wherein a computer program is stored in a computer readable storage medium, and when the computer program is loaded and executed by a processor, the urban water affair scientific scheduling method is adopted.

The electronic device may be an electronic device such as a desktop computer, a notebook computer, or a cloud server, and the electronic device includes, but is not limited to, a processor and a memory, for example, the electronic device may further include an input/output device, a network access device, a bus, and the like.

The processor may be a Central Processing Unit (CPU), or of course, according to actual use, other general purpose processors, digital Signal Processors (DSP), application Specific Integrated Circuits (ASIC), ready-made programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., and the general purpose processor may be a microprocessor or any conventional processor, etc., which is not limited in this respect.

The memory may be an internal storage unit of the electronic device, for example, a hard disk or a memory of the electronic device, or may be an external storage device of the electronic device, for example, a plug-in hard disk, a Smart Memory Card (SMC), a secure digital card (SD), or a flash memory card (FC) provided on the electronic device, or the like, and may be a combination of the internal storage unit of the electronic device and the external storage device, where the memory is used to store a computer program and other programs and data required by the electronic device, and the memory may be used to temporarily store data that has been output or is to be output, which is not limited by the present application.

The urban water science scheduling method of the embodiment is stored in the memory of the electronic device through the electronic device, and is loaded and executed on the processor of the electronic device, so that the method is convenient to use.

The foregoing is merely exemplary embodiments of the present disclosure and is not intended to limit the scope of the present disclosure. That is, equivalent changes and modifications are contemplated by the teachings of this disclosure, which fall within the scope of the present disclosure. Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a scope and spirit of the disclosure being indicated by the claims.

Claims

1. A method for scientific scheduling of urban water affairs, the method comprising:

determining the total score of the corresponding single scheduling scheme through a preset scoring rule based on the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme, and determining the single scheduling scheme with the maximum total score as an optimal scheduling scheme;

Determining the total score of the corresponding scheme chain based on the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain through the scoring rule, and determining a starting point scheme in the scheme chain with the maximum total score as the future long-term optimal scheduling scheme, wherein the starting point scheme is the scheduling scheme with the earliest scheduling time node in the corresponding scheme chain; determining the total score of each scheme chain as a target dependent variable, and determining the starting point scheme in each scheme chain as a target independent variable;

2. The urban water science scheduling method according to claim 1, wherein the determining the total score of the corresponding single scheduling scheme by a preset scoring rule based on the target delivery pressure, the target clean water tank level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme specifically comprises:

3. The urban water science scheduling method according to claim 2, wherein the scoring, by a preset scoring function, the target delivery pressure, the target clean water tank level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme respectively to obtain corresponding index scores, comprises:

4. The method for scientific scheduling of urban water works according to claim 1, wherein said obtaining a first prediction model corresponding to an urban water works comprises:

；

In the method, in the process of the invention, The delivery pressure or delivery flow of the urban water plant at the moment t; The total water demand at time t; in order to be the state of the water pump i at time t, Is a dummy variable of the time period in the day, if the time t belongs to the time period j in the day; Otherwise take，The value range is (0, + -infinity), parameter c is the intercept term, parameterTo change the total water demandConversion of delivery pressure or delivery flowEffects, parameters ofFor the effect of the water pump on the change of the delivery pressure or delivery flow,For the threshold value of the state of the water pump i at the moment t, the parameterFor the influence of dummy variable in daily period on delivery pressure or delivery flow conversion, parametersAs an error term, M represents the total number of water pumps, and K represents the total number of time periods in the day.

5. The method for scientific scheduling of municipal water works according to claim 1, wherein the determining the target incoming flow of the municipal water works under the single scheduling scheme specifically comprises:

6. A municipal water science scheduling device, comprising:

the model acquisition module (11) is used for acquiring a first prediction model corresponding to the urban water plant, wherein the first prediction model is used for predicting the factory pressure and the factory flow of the urban water plant;

the factory determining module (12) is used for determining the total water demand of urban users in a dispatching short-term evaluation period, acquiring a target water pump state of the urban water plant in a single dispatching scheme at a dispatching moment, and predicting target factory pressure and target factory flow corresponding to the dispatching short-term evaluation period through the first prediction model based on the target water pump state and the total water demand, wherein the dispatching short-term evaluation period is a short-term period from the dispatching moment to a short-term preset moment;

The liquid level determining module (13) is used for determining target factory entering flow corresponding to the urban water plant under the single scheduling scheme, predicting target clean water tank liquid level change corresponding to the urban water plant under the scheduling period of the scheduling short-term evaluation period through a second prediction model corresponding to the urban water plant based on the target factory leaving flow and the target factory entering flow, and determining target clean water tank liquid level under the scheduling period of the scheduling short-term evaluation period according to the target clean water tank liquid level change, wherein the second prediction model is used for predicting the clean water tank liquid level change of the urban water plant;

The energy consumption determining module (14) is used for predicting the target water pump energy consumption of the urban water plant through a third prediction model corresponding to the urban water plant based on the target water pump state and the target delivery pressure, and the third prediction model is used for predicting the water pump energy consumption of the urban water plant;

The scheme determining module (15) is used for determining the total score of the corresponding single scheduling scheme through a preset scoring rule based on the target delivery pressure, the target clean water tank liquid level, the target water pump energy consumption and the preset scheduling times corresponding to each single scheduling scheme, and determining the single scheduling scheme with the maximum total score as an optimal scheduling scheme;

A long-term optimizing module (16) configured to determine, if a scheduling operation in the optimal scheduling scheme is on or off of the water pump of the urban water plant, a scheduling trigger time corresponding to each individual scheduling scheme in a period from a current time to a long-term preset time when a future long-term optimal scheduling scheme from the current time to the long-term preset time needs to be determined, where the scheduling trigger time is a future time when a factory pressure is lower than a pressure lower limit value or the factory pressure exceeds a pressure upper limit value, and the scheduling time is earlier than the long-term preset time;

determining the total score of the corresponding scheme chain based on the final delivery pressure, the final clean water tank liquid level, the final water pump energy consumption and the final scheduling times corresponding to each scheme chain through the scoring rule, and determining a starting point scheme in the scheme chain with the maximum total score as the future long-term optimal scheduling scheme, wherein the starting point scheme is the scheduling scheme with the earliest scheduling time node in the corresponding scheme chain;

a fast optimization module (17) for determining the total score of each of the solution chains as a target dependent variable and the starting point solution in each of the solution chains as a target independent variable;

7. A computer readable storage medium having a computer program stored therein, characterized in that the method according to any of claims 1-5 is employed when the computer program is loaded and executed by a processor.

8. An electronic device comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, characterized in that the method according to any of claims 1-5 is used when the computer program is loaded and executed by the processor.