[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN114909707A - Heat supply secondary network regulation and control method based on intelligent balancing device and reinforcement learning - Google Patents

Heat supply secondary network regulation and control method based on intelligent balancing device and reinforcement learning Download PDF

Info

Publication number
CN114909707A
CN114909707A CN202210432789.2A CN202210432789A CN114909707A CN 114909707 A CN114909707 A CN 114909707A CN 202210432789 A CN202210432789 A CN 202210432789A CN 114909707 A CN114909707 A CN 114909707A
Authority
CN
China
Prior art keywords
data
unit building
heat
model
training
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210432789.2A
Other languages
Chinese (zh)
Other versions
CN114909707B (en
Inventor
刘定杰
谢金芳
穆佩红
赵琼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Yingji Power Technology Co ltd
Original Assignee
Zhejiang Yingji Power Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang Yingji Power Technology Co ltd filed Critical Zhejiang Yingji Power Technology Co ltd
Priority to CN202210432789.2A priority Critical patent/CN114909707B/en
Publication of CN114909707A publication Critical patent/CN114909707A/en
Application granted granted Critical
Publication of CN114909707B publication Critical patent/CN114909707B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24D—DOMESTIC- OR SPACE-HEATING SYSTEMS, e.g. CENTRAL HEATING SYSTEMS; DOMESTIC HOT-WATER SUPPLY SYSTEMS; ELEMENTS OR COMPONENTS THEREFOR
    • F24D19/00—Details
    • F24D19/10—Arrangement or mounting of control or safety devices
    • F24D19/1006—Arrangement or mounting of control or safety devices for water heating systems
    • F24D19/1009—Arrangement or mounting of control or safety devices for water heating systems for central heating
    • F24D19/1015—Arrangement or mounting of control or safety devices for water heating systems for central heating using a valve or valves
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/042—Knowledge-based neural networks; Logical representations of neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G06N5/022—Knowledge engineering; Knowledge acquisition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P80/00—Climate change mitigation technologies for sector-wide applications
    • Y02P80/10—Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Economics (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Human Resources & Organizations (AREA)
  • Molecular Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Mechanical Engineering (AREA)
  • Development Economics (AREA)
  • Public Health (AREA)
  • Thermal Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Air Conditioning Control Device (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a heat supply secondary network regulation and control method based on an intelligent balancing device and reinforcement learning, which comprises the following steps: establishing a digital twin model of the heat supply secondary network unit building by adopting a mechanism modeling and data identification method; an intelligent balancing device, an electric regulating valve and a heat meter for improving the circulation flow in a unit building are connected in front of the unit building with unfavorable circulation flow characteristics; acquiring historical heat supply operation data, weather data and indoor temperature based on a digital twin model of a heat supply secondary network unit building, and calculating to obtain the heat required in the unit building through a hybrid prediction model and a model weight value; when the required heat in the unit building is unchanged and the circulation flow is insufficient, calculating by adopting a reinforcement learning algorithm to obtain a regulation and control strategy of the intelligent balancing device and the electric regulating valve, and improving the circulation flow in the unit building; and based on the digital twin model of the heat supply secondary network unit building, issuing and executing the regulation and control strategy after the regulation and control strategy is verified and subjected to energy-saving analysis.

Description

Heat supply secondary network regulation and control method based on intelligent balancing device and reinforcement learning
Technical Field
The invention belongs to the technical field of intelligent heat supply, and particularly relates to a heat supply secondary network regulation and control method based on an intelligent balance device and reinforcement learning.
Background
In current central heating system operation work, unit building energy consumption is reducing gradually, heat supply second grade pipe network design construction is progressively standardized, and the heat supply enterprise pays close attention to user's heat quality more and reduces the energy consumption for the equipment annex level of regulation constantly improves, and terminal heat dissipation equipment form is more various, and electrical apparatus automatic control equipment drops into in a large number, and the mode of heat supply operation regulation also constantly changes, and the comprehensiveness is stronger.
For the unit building with poor circulation characteristics of the system in the building, due to the fact that no effective adjusting means exists, a user at the tail end of the system in the building changes a radiator privately because the room temperature does not reach the standard, the heat dissipation area is increased, the flow is increased, the water supply temperature of the user at the back is lower, the difference of the room temperature among the users becomes larger, the required flow of the whole unit building far exceeds the design flow, the heating station can only increase the consumption of electric energy by replacing a larger water pump, and therefore the flow of a secondary network is further increased.
Aiming at a unit building with poor building internal circulation characteristics, how to improve the circulation flow in the unit building and the temperature of supply and return water and realize heat supply and energy conservation is a problem which is urgently needed to be solved at present.
Based on the technical problems, a new heating secondary network regulation and control method based on an intelligent balance device and reinforcement learning needs to be designed.
Disclosure of Invention
The invention aims to solve the technical problem of overcoming the defects of the prior art and provides a heat supply secondary network regulation and control method based on an intelligent balance device and reinforcement learning.
In order to solve the technical problems, the technical scheme of the invention is as follows:
the invention provides a heat supply secondary network regulation and control method based on an intelligent balancing device and reinforcement learning, which comprises the following steps:
s1, establishing a digital twin model of the heat supply secondary network unit building by adopting a mechanism modeling and data identification method;
step S2, connecting an intelligent balancing device, an electric regulating valve and a heat meter for improving the circulation flow in the unit building in front of the unit building with unfavorable circulation flow characteristics;
s3, acquiring historical heat supply operation data, weather data and indoor temperature based on the digital twin model of the heat supply secondary network unit building, and calculating to obtain the heat required in the unit building through a hybrid prediction model and a model weight value;
step S4, when the required heat in the unit building is not changed and the circulation flow is insufficient, calculating by adopting a reinforcement learning algorithm to obtain a regulation and control strategy of the intelligent balancing device and the electric control valve, and improving the circulation flow in the unit building;
and S5, based on the digital twin model of the heat supply secondary network unit building, issuing and executing the regulation strategy after the regulation strategy is verified and subjected to energy-saving analysis.
Further, in step S1, establishing a digital twin model of the heating secondary network unit building by using a mechanism modeling and data identification method specifically includes:
establishing a digital twin model comprising a physical entity, a virtual entity, a twin data service and connecting elements among all components of a second-level network unit building;
the physical entity is the basis of the digital twin model and is a data source driven by the whole digital twin model; the virtual entities and the physical entities are mapped one by one and interacted in real time, elements of a physical space are depicted from multiple dimensions and multiple scales, the actual process of the physical entities is simulated, and element data is analyzed, evaluated, predicted and controlled; the twin data service integrates physical space information and virtual space information, ensures real-time performance of data transmission, provides knowledge base data including an intelligent algorithm, a model, rule standards and expert experiences, and forms a twin database by fusing the physical information, multi-temporal-spatial correlation information and the knowledge base data; the connection among the components is to realize the interconnection of the components, and the real-time acquisition and feedback of data are realized between the physical entity and the twin data service through a sensor and a protocol transmission specification; data transmission is carried out between the physical entity and the virtual entity through a protocol, physical information is transmitted to the virtual space in real time to update the correction model, and the virtual entity controls the physical entity in real time through an actuator; the information transmission between the virtual entity and the twin data service is realized through a database interface;
and identifying the digital twin model, accessing the multi-working-condition real-time operation data of the second-level network unit building into the established digital twin model, and performing self-adaptive identification correction on the simulation result of the digital twin model by adopting a reverse identification method to obtain the identified and corrected digital twin model of the heat supply second-level network unit building.
Further, in step S3, obtaining historical heat supply operation data, weather data and indoor temperature based on the digital twin model of the heat supply secondary network unit building, and obtaining the heat required by the unit building through calculation of a hybrid prediction model and a model weight value, specifically including:
based on a digital twin model of a heat supply secondary network unit building, acquiring historical multidimensional heat supply data of the unit building and corresponding historical heat demand real data, preprocessing the data to obtain a sample set of a heat demand prediction model in the unit building, wherein the sample set at least comprises historical indoor temperature, weather data, unit building water supply and return temperature, unit building water supply flow and unit building heat demand, and dividing the sample set into a training data set and a testing data set;
selecting two training prediction models, and sequentially training the two training prediction models through a training data set to obtain corresponding prediction models of the heat required in the unit building;
sequentially inputting the test data sets into the demand heat prediction models in the two unit buildings to obtain corresponding demand heat test data;
calculating the weight values of the two training prediction models by adopting an optimal weight value combination strategy according to the demand heat test data and the historical demand heat real data;
obtaining the multidimensional heat supply data of the unit building at the next moment, and inputting the data into the demand heat prediction models in the two unit buildings to obtain corresponding demand heat prediction values;
and calculating according to the predicted values of the required heat of the two training prediction models and the corresponding weight values to obtain the final required heat in the unit building.
Further, the calculating the weight values of the two training prediction models by adopting an optimal weight value combination strategy according to the demand heat test data and the historical demand heat real data specifically comprises:
setting the predicted value of the required heat of the first training prediction model as f 1 The predicted value of the required heat of the second training prediction model is f 2 Predicted value f of required heat 1 Corresponding weight value is w 1 Predicted value f of required heat 2 Corresponding weight value is w 2 ；
According to the required heat test data and the historical required heat real data, calculating to obtain prediction deviation values of two training prediction models, wherein the prediction deviation value of the first training prediction model is e 1 The prediction deviation value of the second training prediction model is e 2 ；
Calculating the square sum of the deviation of the two training prediction models as S according to the weight value and the prediction deviation value;
setting an objective function of the optimal weight combination strategy as deviation square sum S minimization, wherein the constraint condition is as follows: w is a 1 +w 2 ＝1；w 1 ＞0，w 2 ＞0；
And solving the objective function to obtain the optimal weight values of the two training prediction models.
Further, in step S4, when the required heat in the unit building is not changed and the circulation flow is insufficient, a reinforcement learning algorithm is used to calculate and obtain a regulation and control strategy of the intelligent balancing device and the electric control valve, so as to increase the circulation flow in the unit building, which specifically includes:
when the heat required in the unit building is unchanged and the circulation flow is detected to be insufficient, the fact that the water supply circulation flow cannot meet the heat required in the unit building is indicated, based on the obtained heat supply operation data, weather data and indoor temperature as states, when the Q value is maximum is obtained through the combination of rule constraint analysis and dulling-DQN reasoning, the final intelligent balance device and the regulation and control strategy of the electric regulating valve output action are issued to actual execution equipment for execution, the circulation flow in the unit building is improved, and after the execution, energy-saving reward feedback and data states before and after the execution action are obtained; storing the data in a database, and performing training update on the Dueling-DQN model by regularly extracting the data to replace an old model;
under the condition that input and output of the Dueling-DQN model are not changed, the Dueling-DQN neural network algorithm splits a network structure into two branches before an output layer: and the value estimation branch and the advantage estimation branch are used for respectively estimating the value of the state and the values of different actions in the state, and then the two network branches are linearly combined into an output layer to realize accurate estimation of the Q value.
Furthermore, the Dueling-DQN neural network algorithm is enabled to operate in a safe range by setting rule constraint, and parameters of the Dueling-DQN neural network are continuously optimized by setting a reward function;
determining a plurality of safe action ranges according to different rule constraints, meeting the heat required in a unit building, calculating action output of a set value of the temperature of the supply return water through a Dueling-DQN neural network algorithm, selecting the action with the maximum Q value, judging whether the set value is in the rule constraint range, if so, executing the action according to the Dueling-DQN neural network algorithm output action, otherwise, executing the action according to the rule constraint output action, and simultaneously executing a punishment reward for the action.
Further, the training update of the Dueling-DQN model includes: taking heat supply operation data, weather data and indoor temperature data stored in a database as a current state s, outputting an intelligent balancing device and an electric control valve action a through a dulling-DQN neural network algorithm, and obtaining an instant reward r after execution; when the environment changes and reaches a new state s ', storing (s, a, r, s') as training sample data in a database, and introducing an experience playback mechanism for offline training of the model;
the dulling-DQN neural network algorithm is trained and updated at fixed times per day: and loading the current dulling-DQN neural network algorithm model parameters, then randomly extracting data samples from a storage library each time for training to obtain new model parameters and replace the old model.
Further, the reward function settings of the dulling-DQN neural network algorithm include energy savings and security constraint rewards.
Further, the Dueling-DQN neural network algorithm comprises two common hidden layers, the value estimation branch comprises one hidden layer, and the dominance estimation branch comprises one hidden layer.
The invention has the beneficial effects that:
on one hand, the required heat in the unit building is obtained through calculation of a hybrid prediction model and a model weight value, two sub-prediction models are established, the hybrid prediction model is established through weighting two model weights, and the model has high prediction precision; on the other hand, in order to save training time and improve training efficiency, the Dueling DQN algorithm is provided for training a neural network, a better action is output according to input heat supply operation data, weather data, indoor temperature and the like to drive the intelligent balancing device and the electric control valve to change the flow rate of the circulation in the building, and the algorithm further improves the training stability while ensuring the performance.
Additional features and advantages will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a heat supply secondary network regulation and control method based on an intelligent balancing device and reinforcement learning according to the present invention;
FIG. 2 is a diagram of the Dueling-DQN neural network structure of the present invention;
FIG. 3 is a schematic block diagram of a secondary grid regulation and control strategy obtained based on the Dueling-DQN neural network algorithm.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flow chart of a heating secondary network regulation and control method based on an intelligent balance device and reinforcement learning according to the invention.
As shown in fig. 1, the present embodiment provides a heating secondary network regulation and control method based on an intelligent balancing device and reinforcement learning, which includes:
s1, establishing a digital twin model of the heat supply secondary network unit building by adopting a mechanism modeling and data identification method;
step S2, connecting an intelligent balancing device, an electric regulating valve and a heat meter for improving the circulation flow in the unit building in front of the unit building with unfavorable circulation flow characteristics;
s3, acquiring historical heat supply operation data, weather data and indoor temperature based on the digital twin model of the heat supply secondary network unit building, and calculating to obtain the heat required in the unit building through a hybrid prediction model and a model weight value;
step S4, when the required heat in the unit building is not changed and the circulation flow is insufficient, calculating by adopting a reinforcement learning algorithm to obtain a regulation and control strategy of the intelligent balancing device and the electric regulating valve, and improving the circulation flow in the unit building;
and S5, based on the digital twin model of the heat supply secondary network unit building, issuing and executing the regulation strategy after the regulation strategy is verified and subjected to energy-saving analysis.
In the practical application, can unify intelligent balancing unit and the electrical control valve of inserting before the unit building, the calorimeter is balancing unit before the building, include outdoor temperature sensor at least, supply water temperature sensor, a water pump, manual balance valve, electrical control valve, differential pressure controller, return water temperature sensor, can separate the operating mode of second grade pipe network and the operating mode of system in the building, solve the not good problem of building inner loop characteristic with intelligent balancing unit and electrical control valve before the building, second grade net and the interior operating mode of building are mutual noninterference, promote building inner loop flow and energy saving, solve the interior water imbalance problem of building. After the balance device in front of the building is connected, the main pipe network adopts a large-temperature-difference and small-flow operation mode, and the users of the tail end unit building adopt a large-flow and small-temperature-difference operation mode, so that the resistance of the main pipe network is relatively small, and the resistance of the tail end users is relatively large. According to the hydraulic stability, the operation mode has good balance of the whole system, is not easy to generate hydraulic imbalance problems, and can be adjusted by adjusting the circulating flow rate even if the hydraulic imbalance problems occur locally.
In this embodiment, in step S1, establishing a digital twin model of the heating secondary network unit building by using a mechanism modeling and data identification method specifically includes:
establishing a digital twin model comprising a physical entity, a virtual entity, a twin data service and connecting elements among all components of a second-level network unit building;
the physical entity is the basis of the digital twin model and is a data source driven by the whole digital twin model; the virtual entities and the physical entities are mapped one by one and interacted in real time, elements of a physical space are depicted from multiple dimensions and multiple scales, the actual process of the physical entities is simulated, and element data is analyzed, evaluated, predicted and controlled; the twin data service integrates physical space information and virtual space information, ensures real-time performance of data transmission, provides knowledge base data comprising an intelligent algorithm, a model, rule standards and expert experience, and forms a twin database by fusing the physical information, multi-space-time correlation information and the knowledge base data; the connection among the components is to realize the interconnection of the components, and the real-time acquisition and feedback of data are realized between the physical entity and the twin data service through a sensor and a protocol transmission specification; data transmission is carried out between the physical entity and the virtual entity through a protocol, physical information is transmitted to the virtual space in real time to update the correction model, and the virtual entity controls the physical entity in real time through an actuator; the information transmission between the virtual entity and the twin data service is realized through a database interface;
and identifying the digital twin model, accessing the multi-working-condition real-time operation data of the second-level network unit building into the established digital twin model, and performing self-adaptive identification correction on the simulation result of the digital twin model by adopting a reverse identification method to obtain the identified and corrected digital twin model of the heat supply second-level network unit building.
In this embodiment, in step S3, obtaining historical heat supply operation data, weather data, and indoor temperature based on the digital twin model of the unit building of the second-level heat supply network, and calculating to obtain the heat required by the unit building through a hybrid prediction model and a model weight value specifically includes:
based on a digital twin model of a heat supply secondary network unit building, acquiring historical multidimensional heat supply data and corresponding historical heat demand real data of the unit building, preprocessing the data to obtain a sample set of a heat demand prediction model in the unit building, wherein the sample set at least comprises historical indoor temperature, weather data, unit building water supply and return temperature, unit building water supply flow and unit building heat demand, and dividing the sample set into a training data set and a testing data set;
selecting two training prediction models, and sequentially training the two training prediction models through a training data set to obtain corresponding prediction models of the heat required in the unit building;
sequentially inputting the test data sets into the demand heat prediction models in the two unit buildings to obtain corresponding demand heat test data;
calculating the weight values of the two training prediction models by adopting an optimal weight value combination strategy according to the demand heat test data and the historical demand heat real data;
obtaining the multidimensional heat supply data of the unit building at the next moment, and inputting the data into the demand heat prediction models in the two unit buildings to obtain corresponding demand heat prediction values;
and calculating to obtain the final required heat in the unit building according to the predicted values of the required heat of the two training prediction models and the corresponding weight values.
In this embodiment, the calculating the weight values of the two training prediction models by using an optimal weight combination strategy according to the demand heat test data and the historical demand heat real data specifically includes:
setting the predicted value of the required heat of the first training prediction model as f 1 The predicted value of the required heat of the second training prediction model is f 2 Predicted value f of required heat 1 Corresponding weight value is w 1 Predicted value f of required heat 2 Corresponding weight value is w 2 ；
Calculating to obtain the prediction deviation values of the two training prediction models according to the heat demand test data and the historical heat demand real data, wherein the prediction deviation value of the first training prediction model is e 1 The prediction deviation value of the second training prediction model is e 2 ；
Calculating the square sum of the deviation of the two training prediction models as S according to the weight value and the prediction deviation value;
setting an objective function of the optimal weight combination strategy as deviation square sum S minimization, wherein the constraint condition is as follows: w is a 1 +w 2 ＝1；w 1 ＞0，w 2 ＞0；
And solving the objective function to obtain the optimal weight values of the two training prediction models.
FIG. 2 is a diagram of the Dueling-DQN neural network structure according to the present invention.
FIG. 3 is a schematic block diagram of a two-level net regulation strategy obtained based on a Dueling-DQN neural network algorithm according to the present invention.
As shown in fig. 2 and 3, in this embodiment, in step S4, when the required heat in the unit building is not changed and the circulation flow is insufficient, a reinforcement learning algorithm is used to calculate and obtain a regulation and control strategy of the intelligent balancing device and the electric control valve, so as to increase the circulation flow in the unit building, specifically including:
when the heat required in the unit building is unchanged and the circulation flow is detected to be insufficient, the fact that the water supply circulation flow cannot meet the heat required in the unit building is indicated, based on the obtained heat supply operation data, weather data and indoor temperature as states, when the Q value is maximum is obtained through the combination of rule constraint analysis and dulling-DQN reasoning, the final intelligent balance device and the regulation and control strategy of the electric regulating valve output action are issued to actual execution equipment for execution, the circulation flow in the unit building is improved, and after the execution, energy-saving reward feedback and data states before and after the execution action are obtained; storing the data in a database, and performing training update on the Dueling-DQN model by regularly extracting the data to replace an old model;
under the condition that input and output of the Dueling-DQN model are not changed, the Dueling-DQN neural network algorithm splits a network structure into two branches before an output layer: and the value estimation branch and the advantage estimation branch are used for respectively estimating the value of the state and the values of different actions in the state, and then the two network branches are linearly combined into an output layer to realize accurate estimation of the Q value.
In practical application, the Dueling-DQN neural network algorithm divides a Q network into a value function and an advantage function, wherein the value function is only related to a state s and is not related to a specifically adopted action a, and the value function is marked as V (s, w, beta); the merit function is then determined by state s and action a, denoted as a (s, a, w, α), and the final Q value is expressed as:
Q(s,a,w,α,β)＝V(s,w,β)+A(s,a,w,α)；
w is a network parameter of the common part; alpha is a network parameter of the individual merit function; beta is a network parameter of the individual cost function.
In the embodiment, the Dueling-DQN neural network algorithm is operated in a safe range by setting rule constraint, and the parameters of the Dueling-DQN neural network are continuously optimized by setting a reward function;
determining a plurality of safe action ranges according to different rule constraints, meeting the heat required in a unit building, calculating action output of a set value of the temperature of the supply return water through a Dueling-DQN neural network algorithm, selecting the action with the maximum Q value, judging whether the set value is in the rule constraint range, if so, executing the action according to the Dueling-DQN neural network algorithm output action, otherwise, executing the action according to the rule constraint output action, and simultaneously executing a punishment reward for the action.
In this embodiment, the training update of the Dueling-DQN model includes: taking heat supply operation data, weather data and indoor temperature data stored in a database as a current state s, outputting an intelligent balancing device and an electric control valve action a through a Dueling-DQN neural network algorithm, and obtaining an instant reward r after execution; when the environment changes and reaches a new state s ', storing (s, a, r, s') as training sample data in a database, and introducing an experience playback mechanism for offline training of the model;
the dulling-DQN neural network algorithm is trained and updated at fixed times per day: and loading the current dulling-DQN neural network algorithm model parameters, then randomly extracting data samples from a storage library each time for training to obtain new model parameters and replace the old model.
In this embodiment, the reward function settings of the dulling-DQN neural network algorithm include energy savings and security constraint rewards.
In this embodiment, the Dueling-DQN neural network algorithm includes two common hidden layers, the value estimation branch includes one hidden layer, and the dominance estimation branch includes one hidden layer.
It should be noted that the Dueling-DQN neural network algorithm includes, in addition to three neural networks of an input layer, a hidden layer, and an output layer, two sub-network structures respectively corresponding to the cost function and the dominance function network portions, and the output layer of the Q network is obtained by linearly combining the outputs of the cost function and the dominance function network.
The Dueling-DQN neural network algorithm flow comprises the following steps:
inputting an algorithm: iteration round number T, state characteristic dimension n, action set A, exploration rate epsilon, batch gradient descent sample number m, attenuation factor gamma, current Q network Q, target Q network Q', and target Q network parameter updating frequency P.
And (3) outputting an algorithm: q network parameters.
(1) The Q network parameter ω and the parameter ω 'of the target Q network Q' are initialized, and the value Q corresponding to all the states and actions are initialized. The empirical playback unit D is initialized.
(2) And performing iteration.
1) The first state S in the state sequence is initialized, with a feature vector of Φ (S).
2) And taking phi (S) as the input in the Q network to obtain Q values of outputs corresponding to all actions, and selecting the corresponding action A by an epsilon-greedy method.
3) And selecting and executing the current action A in the state S to obtain the feature vector phi (S') and the reward value R of the next state, and judging whether the state is a termination state.
4) Storing { Φ (S), a, R, Φ (S') } in the empirical playback unit D.
5) Let S be S'
6) Collecting m samples from an empirical playback unit D, and calculating a current target Q value y j Where j is 1, 2.. m, then:
Figure BDA0003611588430000091
7) loss function according to mean square error
Figure BDA0003611588430000092
And updating all parameters omega of the Q network by adopting back propagation.
8) When i% P is 1, the target Q network parameter ω' is updated.
9) And judging whether the S' is in a termination state, if so, ending the current iteration, and if not, turning to the step 2).
In the several embodiments provided in the present application, it should be understood that the disclosed system and method may be implemented in other ways. The system embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In light of the foregoing description of the preferred embodiment of the present invention, many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The technical scope of the present invention is not limited to the content of the specification, and must be determined according to the scope of the claims.

Claims (9)

1. A heat supply secondary network regulation and control method based on an intelligent balance device and reinforcement learning is characterized by comprising the following steps:
s1, establishing a digital twin model of the heat supply secondary network unit building by adopting a mechanism modeling and data identification method;
step S2, connecting an intelligent balancing device, an electric regulating valve and a heat meter for improving the circulation flow in the unit building in front of the unit building with unfavorable circulation flow characteristics;
s3, acquiring historical heat supply operation data, weather data and indoor temperature based on the digital twin model of the heat supply secondary network unit building, and calculating to obtain the heat required in the unit building through a hybrid prediction model and a model weight value;
step S4, when the required heat in the unit building is not changed and the circulation flow is insufficient, calculating by adopting a reinforcement learning algorithm to obtain a regulation and control strategy of the intelligent balancing device and the electric control valve, and improving the circulation flow in the unit building;
and S5, based on the digital twin model of the heat supply secondary network unit building, issuing and executing the regulation strategy after the regulation strategy is verified and subjected to energy-saving analysis.
2. The method as claimed in claim 1, wherein in step S1, the building digital twin model of the heating secondary network unit building by using mechanism modeling and data identification method specifically includes:
establishing a digital twin model comprising a physical entity, a virtual entity, a twin data service and connecting elements among all components of a second-level network unit building;
the physical entity is a data source of the whole digital twin model;
the virtual entity carries out simulation on the actual process of the physical entity and carries out data analysis, evaluation, prediction and control on the element data;
the twin data service integrates physical space information and virtual space information, ensures real-time performance of data transmission, provides knowledge base data comprising an intelligent algorithm, a model, rule standards and expert experience, and forms a twin database by fusing the physical information, multi-space-time correlation information and the knowledge base data;
the connection among the components is used for realizing the interconnection and intercommunication of the components, and the real-time acquisition and feedback of data are realized between the physical entity and the twin data service through a sensor and a protocol transmission specification;
data transmission is carried out between the physical entity and the virtual entity through a protocol, physical information is transmitted to a virtual space in real time to update a correction model, and the virtual entity controls the physical entity in real time through an actuator;
the virtual entity and the twin data service are in information transmission through a database interface;
and identifying the digital twin model, accessing the multi-working-condition real-time operation data of the second-level network unit building into the established digital twin model, and performing self-adaptive identification correction on the simulation result of the digital twin model by adopting a reverse identification method to obtain the identified and corrected digital twin model of the heat supply second-level network unit building.
3. The method according to claim 1, wherein in step S3, based on a digital twin model of a unit building of the secondary heating network, historical heating operation data, weather data and indoor temperature are obtained, and a required heat in the unit building is obtained by calculating a hybrid prediction model and a model weight value, and specifically includes:
based on a digital twin model of a heat supply secondary network unit building, acquiring historical multidimensional heat supply data of the unit building and corresponding historical heat demand real data, preprocessing the data to obtain a sample set of a heat demand prediction model in the unit building, wherein the sample set at least comprises historical indoor temperature, weather data, unit building water supply and return temperature, unit building water supply flow and unit building heat demand, and dividing the sample set into a training data set and a testing data set;
selecting two training prediction models, and sequentially training the two training prediction models through a training data set to obtain a corresponding prediction model of the heat required in the unit building;
sequentially inputting the test data sets into the demand heat prediction models in the two unit buildings to obtain corresponding demand heat test data;
calculating the weight values of the two training prediction models by adopting an optimal weight value combination strategy according to the demand heat test data and the historical demand heat real data;
obtaining the multidimensional heat supply data of the unit building at the next moment, and inputting the data into the demand heat prediction models in the two unit buildings to obtain corresponding demand heat prediction values;
and calculating according to the predicted values of the required heat of the two training prediction models and the corresponding weight values to obtain the final required heat in the unit building.
4. The method according to claim 3, wherein the calculating the weight values of the two training prediction models by using an optimal weight combination strategy according to the demand heat test data and the historical demand heat real data specifically comprises:
setting the predicted value of the required heat of the first training prediction model as f 1 The predicted value of the required heat of the second training prediction model is f 2 Predicted value f of required heat 1 Corresponding weight value is w 1 Predicted value f of required heat 2 Corresponding weight value is w 2 ；
Calculating to obtain the prediction deviation values of the two training prediction models according to the heat demand test data and the historical heat demand real data, wherein the prediction deviation value of the first training prediction model is e 1 The prediction deviation value of the second training prediction model is e 2 ；
Calculating the square sum of the deviation of the two training prediction models as S according to the weight value and the prediction deviation value;
the objective function of the optimal weight combination strategy is set to be the deviation square sum S minimum, and the constraint condition is as follows:
w 1 +w 2 ＝1；w 1 ＞0，w 2 ＞0；
and solving the objective function to obtain the optimal weight values of the two training prediction models.
5. The method as claimed in claim 1, wherein in step S4, when the heat demand in the unit building is not changed and the circulation flow is insufficient, the method adopts a reinforcement learning algorithm to calculate the regulation strategy of the intelligent balancing device and the electric control valve, and the circulation flow in the unit building is increased, specifically comprising:
when the heat required in the unit building is unchanged and the circulation flow is detected to be insufficient, the fact that the water supply circulation flow cannot meet the heat required in the unit building is indicated, based on the obtained heat supply operation data, weather data and indoor temperature as states, when the Q value is maximum is obtained through the combination of rule constraint analysis and dulling-DQN reasoning, the final intelligent balance device and the regulation and control strategy of the electric regulating valve output action are issued to actual execution equipment for execution, the circulation flow in the unit building is improved, and after the execution, energy-saving reward feedback and data states before and after the execution action are obtained; storing the data in a database, and performing training update on the Dueling-DQN model by regularly extracting the data to replace an old model;
under the condition that input and output of the Dueling-DQN model are not changed, the Dueling-DQN neural network algorithm splits a network structure into two branches before an output layer: and the value estimation branch and the advantage estimation branch are used for respectively estimating the value of the state and the values of different actions in the state, and then the two network branches are linearly combined into an output layer to realize accurate estimation of the Q value.
6. A heating secondary network regulation method according to claim 5, characterized in that: the method also comprises the steps of enabling a Dueling-DQN neural network algorithm to operate in a safe range by setting rule constraint, and continuously optimizing the Dueling-DQN neural network parameters by setting a reward function;
determining a plurality of safe action ranges according to different rule constraints, meeting the heat required in a unit building, calculating action output of a set value of the temperature of the supply return water through a Dueling-DQN neural network algorithm, selecting the action with the maximum Q value, judging whether the set value is in the rule constraint range, if so, executing the action according to the Dueling-DQN neural network algorithm output action, otherwise, executing the action according to the rule constraint output action, and simultaneously executing a punishment reward for the action.
7. A heating secondary network regulation method according to claim 5, characterized in that: the method further comprises the step of training and updating the Dueling-DQN model, and specifically comprises the following steps:
taking heat supply operation data, weather data and indoor temperature data stored in a database as a current state s, outputting an intelligent balancing device and an electric control valve action a through a dulling-DQN neural network algorithm, and obtaining an instant reward r after execution; when the environment changes and reaches a new state s ', storing (s, a, r, s') as training sample data in a database, and introducing an experience playback mechanism for offline training of the model;
the dulling-DQN neural network algorithm is trained and updated at fixed time every day: and loading the current dulling-DQN neural network algorithm model parameters, then randomly extracting data samples from a storage library each time for training to obtain new model parameters and replace the old model.
8. A heating secondary grid regulation method according to claim 7, characterized in that the reward function settings of the dulling-DQN neural network algorithm include energy saving and safety constraint rewards.
9. The heating secondary grid regulation and control method of claim 8, wherein the Dueling-DQN neural network algorithm comprises two common hidden layers, the value estimation branch comprises one hidden layer, and the dominance estimation branch comprises one hidden layer.
CN202210432789.2A 2022-04-24 2022-04-24 Heat supply secondary network regulation and control method based on intelligent balance device and reinforcement learning Active CN114909707B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210432789.2A CN114909707B (en) 2022-04-24 2022-04-24 Heat supply secondary network regulation and control method based on intelligent balance device and reinforcement learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210432789.2A CN114909707B (en) 2022-04-24 2022-04-24 Heat supply secondary network regulation and control method based on intelligent balance device and reinforcement learning

Publications (2)

Publication Number Publication Date
CN114909707A true CN114909707A (en) 2022-08-16
CN114909707B CN114909707B (en) 2023-10-10

Family

ID=82764264

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210432789.2A Active CN114909707B (en) 2022-04-24 2022-04-24 Heat supply secondary network regulation and control method based on intelligent balance device and reinforcement learning

Country Status (1)

Country Link
CN (1) CN114909707B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116757095A (en) * 2023-08-14 2023-09-15 国网浙江省电力有限公司宁波供电公司 Electric power system operation method, device and medium based on cloud edge end cooperation

Citations (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002022278A (en) * 2001-04-26 2002-01-23 Noritz Corp Hot water feeding device
CN104390253A (en) * 2014-10-27 2015-03-04 朱杰 Centralized heating system based on flow independent type heat radiator tail ends and control method
CN108679683A (en) * 2018-05-24 2018-10-19 中联西北工程设计研究院有限公司 A kind of inlet device and hot and cold water flow allocation method of the unpowered injector of band
US20190360711A1 (en) * 2018-05-22 2019-11-28 Seokyoung Systems Method and device for controlling power supply to heating, ventilating, and air-conditioning (hvac) system for building based on target temperature
CN111023223A (en) * 2019-12-30 2020-04-17 南京国之鑫科技有限公司 Heating heat supply network intelligent hydraulic balance system based on cloud and return water temperature
CN111306609A (en) * 2020-02-27 2020-06-19 中国第一汽车股份有限公司 Building time-sharing control heating temperature energy-saving system
CN111561732A (en) * 2020-05-18 2020-08-21 瑞纳智能设备股份有限公司 Heat exchange station heat supply adjusting method and system based on artificial intelligence
CN111580382A (en) * 2020-05-18 2020-08-25 瑞纳智能设备股份有限公司 Unit-level heat supply adjusting method and system based on artificial intelligence
CN111578371A (en) * 2020-05-22 2020-08-25 浙江大学 Data-driven accurate regulation and control method for urban centralized heating system
US20200355391A1 (en) * 2017-04-25 2020-11-12 Johnson Controls Technology Company Predictive building control system with neural network based comfort prediction
CN112460741A (en) * 2020-11-23 2021-03-09 香港中文大学（深圳） Control method of building heating, ventilation and air conditioning system
CN113028494A (en) * 2021-03-18 2021-06-25 山东琅卡博能源科技股份有限公司 Intelligent heat supply dynamic hydraulic balance control method
CN113091123A (en) * 2021-05-11 2021-07-09 杭州英集动力科技有限公司 Building unit heat supply system regulation and control method based on digital twin model
CA3177372A1 (en) * 2020-04-28 2021-11-04 Strong Force Tp Portfolio 2022, Llc Digital twin systems and methods for transportation systems
CN113606650A (en) * 2021-07-23 2021-11-05 淄博热力有限公司 Intelligent heat supply room temperature regulation and control system based on machine learning algorithm
CN113657031A (en) * 2021-08-12 2021-11-16 杭州英集动力科技有限公司 Digital twin-based heat supply scheduling automation realization method, system and platform
CN113719887A (en) * 2021-08-10 2021-11-30 华能山东发电有限公司烟台发电厂 Intelligent balance heat supply system
WO2021259474A1 (en) * 2020-06-24 2021-12-30 Ecosync Ltd. Heating control system
GB202116859D0 (en) * 2021-06-29 2022-01-05 Univ Jiangsu Intelligent parallel pumping system and optimal regulating method thereof

Patent Citations (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002022278A (en) * 2001-04-26 2002-01-23 Noritz Corp Hot water feeding device
CN104390253A (en) * 2014-10-27 2015-03-04 朱杰 Centralized heating system based on flow independent type heat radiator tail ends and control method
US20200355391A1 (en) * 2017-04-25 2020-11-12 Johnson Controls Technology Company Predictive building control system with neural network based comfort prediction
US20190360711A1 (en) * 2018-05-22 2019-11-28 Seokyoung Systems Method and device for controlling power supply to heating, ventilating, and air-conditioning (hvac) system for building based on target temperature
CN108679683A (en) * 2018-05-24 2018-10-19 中联西北工程设计研究院有限公司 A kind of inlet device and hot and cold water flow allocation method of the unpowered injector of band
CN111023223A (en) * 2019-12-30 2020-04-17 南京国之鑫科技有限公司 Heating heat supply network intelligent hydraulic balance system based on cloud and return water temperature
CN111306609A (en) * 2020-02-27 2020-06-19 中国第一汽车股份有限公司 Building time-sharing control heating temperature energy-saving system
CA3177372A1 (en) * 2020-04-28 2021-11-04 Strong Force Tp Portfolio 2022, Llc Digital twin systems and methods for transportation systems
CN111561732A (en) * 2020-05-18 2020-08-21 瑞纳智能设备股份有限公司 Heat exchange station heat supply adjusting method and system based on artificial intelligence
CN111580382A (en) * 2020-05-18 2020-08-25 瑞纳智能设备股份有限公司 Unit-level heat supply adjusting method and system based on artificial intelligence
CN111578371A (en) * 2020-05-22 2020-08-25 浙江大学 Data-driven accurate regulation and control method for urban centralized heating system
WO2021259474A1 (en) * 2020-06-24 2021-12-30 Ecosync Ltd. Heating control system
CN112460741A (en) * 2020-11-23 2021-03-09 香港中文大学（深圳） Control method of building heating, ventilation and air conditioning system
CN113028494A (en) * 2021-03-18 2021-06-25 山东琅卡博能源科技股份有限公司 Intelligent heat supply dynamic hydraulic balance control method
CN113091123A (en) * 2021-05-11 2021-07-09 杭州英集动力科技有限公司 Building unit heat supply system regulation and control method based on digital twin model
GB202116859D0 (en) * 2021-06-29 2022-01-05 Univ Jiangsu Intelligent parallel pumping system and optimal regulating method thereof
CN113606650A (en) * 2021-07-23 2021-11-05 淄博热力有限公司 Intelligent heat supply room temperature regulation and control system based on machine learning algorithm
CN113719887A (en) * 2021-08-10 2021-11-30 华能山东发电有限公司烟台发电厂 Intelligent balance heat supply system
CN113657031A (en) * 2021-08-12 2021-11-16 杭州英集动力科技有限公司 Digital twin-based heat supply scheduling automation realization method, system and platform

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116757095A (en) * 2023-08-14 2023-09-15 国网浙江省电力有限公司宁波供电公司 Electric power system operation method, device and medium based on cloud edge end cooperation
CN116757095B (en) * 2023-08-14 2023-11-07 国网浙江省电力有限公司宁波供电公司 Electric power system operation method, device and medium based on cloud edge end cooperation

Also Published As

Publication number Publication date
CN114909707B (en) 2023-10-10

Similar Documents

Publication Publication Date Title
CN103912966B (en) A kind of earth source heat pump refrigeration system optimal control method
CN111795484A (en) Intelligent air conditioner control method and system
CN111561734B (en) Heating household valve adjusting method, system and equipment based on artificial intelligence
CN113657660B (en) Heat source load prediction method based on substation load and heat supply network hysteresis model
CN114777192B (en) Secondary network heat supply autonomous optimization regulation and control method based on data association and deep learning
CN110610275B (en) Variable air volume air conditioner load prediction method and system based on ACQPSO-ELM
CN111461466B (en) Heating valve adjusting method, system and equipment based on LSTM time sequence
CN114200839B (en) Intelligent office building energy consumption control model for dynamic monitoring of coupling environment behaviors
CN114811713B (en) Two-level network inter-user balanced heat supply regulation and control method based on mixed deep learning
CN109919356A (en) One kind being based on BP neural network section water demand prediction method
CN110097929A (en) A kind of blast furnace molten iron silicon content on-line prediction method
CN114909706B (en) Two-level network balance regulation and control method based on reinforcement learning algorithm and differential pressure control
CN114909707B (en) Heat supply secondary network regulation and control method based on intelligent balance device and reinforcement learning
CN115751441A (en) Heat supply system heating station heat regulation method and system based on secondary side flow
CN114777193A (en) Secondary network household regulation and control mode switching method and system for heat supply system
CN115013863B (en) Autonomous optimization regulation and control method for heat supply system of jet pump based on digital twin model
CN110705756A (en) Electric power energy consumption optimization control method based on input convex neural network
CN113701232B (en) Heat supply system building-level regulation and control method and system based on temperature diversity analysis
CN117454168A (en) Building energy consumption prediction method and system
CN116227704A (en) Edge calculation and cloud calculation-based heat supply energy consumption diagnosis and energy-saving regulation and control method
CN115013862B (en) Autonomous optimal operation method of heating system based on jet pump and auxiliary circulating pump
Grzenda et al. Heat consumption prediction with multiple hybrid models
CN118134227B (en) Short-term building cold-hot load prediction method and system based on hybrid model
CN116839091B (en) Heat exchange station automatic control parameter setting method based on deep learning
CN113657670B (en) Ocean humidity prediction method, system, device and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant