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

CN112524751A - Dynamic air conditioning system energy consumption prediction model construction and prediction method and device - Google Patents

Dynamic air conditioning system energy consumption prediction model construction and prediction method and device Download PDF

Info

Publication number
CN112524751A
CN112524751A CN202011387024.9A CN202011387024A CN112524751A CN 112524751 A CN112524751 A CN 112524751A CN 202011387024 A CN202011387024 A CN 202011387024A CN 112524751 A CN112524751 A CN 112524751A
Authority
CN
China
Prior art keywords
value
energy consumption
conditioning system
predicted
building
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
CN202011387024.9A
Other languages
Chinese (zh)
Other versions
CN112524751B (en
Inventor
白燕
武璐璐
贺引娥
张亚俐
万陶成
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.)
Xian University of Architecture and Technology
Original Assignee
Xian University of Architecture and Technology
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 Xian University of Architecture and Technology filed Critical Xian University of Architecture and Technology
Priority to CN202011387024.9A priority Critical patent/CN112524751B/en
Publication of CN112524751A publication Critical patent/CN112524751A/en
Application granted granted Critical
Publication of CN112524751B publication Critical patent/CN112524751B/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
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
    • F24F11/46—Improving electric energy efficiency or saving
    • F24F11/47—Responding to energy costs
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/64—Electronic processing using pre-stored data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • 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"

Landscapes

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

Abstract

The invention discloses a dynamic air conditioning system energy consumption prediction model building and predicting method and device, which can be used for predicting the dynamic air conditioning system energy consumption time by building an improved particle swarm optimization BP neural network, thereby improving the accuracy of energy consumption prediction and being beneficial to energy saving and optimization control of an air conditioning system; by establishing the dynamic energy consumption control model, the dynamic change trend that the indoor thermal environment effectively follows the outdoor temperature is realized, and the comfort and health unification are realized while the dynamic control of the thermal environment is met.

Description

Dynamic air conditioning system energy consumption prediction model construction and prediction method and device
Technical Field
The invention relates to an energy consumption prediction method and device for an air conditioning system, in particular to a dynamic energy consumption prediction model construction and prediction method and device for the air conditioning system.
Background
The comfortable indoor thermal environment is the premise of physical and mental health and efficient work of people. The health and high energy consumption issues hidden behind the traditional steady-state thermal environment creation model are of great interest to researchers in the relevant field. At present, dynamic thermal environment control provides a favorable way for indoor thermal environment dynamic optimization, can effectively utilize small regular fluctuation of a thermal environment, and realizes comfort and health unification while satisfying the dynamic control of the thermal environment. However, creating a comfortable indoor thermal environment comes at the cost of some system energy consumption. In the total energy consumption of the building, the energy consumption of the heating, ventilating and air conditioning system accounts for more than 50%, and if the energy consumption can be accurately and timely predicted, the energy consumption and the energy consumption of the heating, ventilating and air conditioning system have important significance for energy conservation and optimal control of the air conditioning system.
At present, the main methods for predicting the energy consumption of the air conditioning system comprise linear regression, a support vector machine, a decision tree and an artificial neural network. Linear regression is a statistical analysis method with fast modeling speed, but cannot solve the non-linear problem. The support vector machine is an artificial intelligence method, can map to a high-dimensional space, but is greatly influenced by missing data. The decision tree is a classification method which is easy to understand and implement, but the prediction result of the decision tree is biased to the characteristic of large numerical ratio and cannot well process nonlinear data. Compared with the method, the artificial neural network has strong nonlinear fitting and approximation capabilities and is more suitable for predicting the energy consumption of the air conditioning system. Among them, BPNN is widely used due to its strong nonlinear mapping and fault-tolerant capability, but it also has the defects of slow convergence speed and local minimization. Some researchers use the PSO algorithm to optimize the parameters of the BPNN, and the prediction precision and the convergence speed of the BPNN are improved. However, the PSO algorithm is prone to be locally optimal, so that the accuracy and convergence rate of the prediction result still cannot meet ideal requirements. In addition, the relevance of the input variable and the output variable of the prediction model is effectively mined, and the prediction precision is improved.
Disclosure of Invention
The invention aims to provide a dynamic air conditioning system energy consumption prediction model building and predicting method and device, which are used for solving the problem that the energy consumption predicted by the air conditioning system energy consumption predicting method and device in the prior art is inaccurate.
In order to realize the task, the invention adopts the following technical scheme:
a dynamic air conditioning system energy consumption prediction model building method is used for building an energy consumption prediction model of a building air conditioning system to be predicted, and the method is executed according to the following steps:
step 1, collecting environment data of a building air conditioning system to be predicted at the ith day of a J, wherein the environment data at the ith day of the J comprises an ith day outdoor temperature value, an ith-1 th outdoor relative humidity value, an ith-1 th outdoor wind speed value, an ith-1 th solar direct radiation value, an ith-1 th solar scattered radiation value, an ith day space-time temperature adjustment set value and an ith day space-time adjustment system energy consumption mode, wherein i is 1,2, …, 24, J is 1,2,3, …, J and J are positive integers;
wherein the ith day time-modulation system energy consumption mode comprises zero energy consumption, low energy consumption, medium energy consumption or high energy consumption;
step 2, acquiring an energy consumption value of the j th day i of the building air conditioning system to be predicted;
step 3, setting i to i +1, returning to the step 1 until i to 24, and executing the step 4;
step 4, setting J to be J +1, returning to the step 1 until J is J, and executing the step 5;
step 5, obtaining a plurality of environment data to obtain an environment data set; obtaining a plurality of energy consumption values to obtain a tag set;
step 6, taking the environment data set as input, taking the label set as reference output, and training an improved particle swarm optimization BP neural network;
when the BP neural network is optimized by training the improved particle swarm, the particle speed and the position are updated by adopting a formula II:
Figure BDA0002809978520000031
wherein vτ lDenotes the speed, x, of the τ -th particle at the l-th iterationτ lDenotes the position of the τ -th particle at the l-th iteration, l ═ 0,1,2, …, ∞, r1And r2Are all in [0,1 ]]Random number between, pbestτFor the individual extremum of the τ -th particle, gbestτIs a global extremum of the population, ω represents a weight, ωmaxRepresenting the maximum value of the weight, ωminRepresents the minimum value of the weight, c1Denotes a first acceleration factor, c2Representing a second acceleration factor, itmaxRepresenting the maximum value of the number of iterations, iterRepresenting the current number of iterations, c1fInitial value representing a first acceleration factor, c2fInitial value representing a second acceleration factor, c1FRepresents the final value of the first acceleration factor, c2FRepresents a final value of the second acceleration factor;
and obtaining an energy consumption prediction model.
Further, in step 2, obtaining the energy consumption value of the building air conditioning system to be predicted at the jth day i specifically includes:
step 2.1, obtaining the highest outdoor temperature value of the j th day of the building to be predicted, if the highest outdoor temperature value is greater than T, executing the step 2.2, otherwise, the j th day i energy consumption value of the air conditioning system of the building to be predicted is 0;
step 2.2, obtaining the regulation temperature g (t) of the air conditioning system at the ith day of the j day by adopting the formula Ij,i) The unit is ℃:
Figure BDA0002809978520000041
wherein t isj,iRepresents an outdoor temperature value, t ', at day i of the building to be predicted'max、t'minRespectively, the maximum value and the minimum value of temperature regulation, and the unit is DEG Cj,maxThe maximum value of the j day outdoor temperature of the building to be predicted is given in DEG Cj,minThe minimum value of the outdoor temperature of the j day of the building to be predicted is calculated in units of ℃;
step 2.3, obtaining a building simulation model of the building to be predicted;
and 2.4, obtaining the energy consumption value of the air conditioning system of the building to be predicted at the j th day i according to the building simulation model and the temperature regulated and controlled by the air conditioning system.
Further, the initial value c of the first acceleration factor1f0.5, the final value c of the first acceleration factor1F2.4, the second acceleration factor initial value c2f2.4, the final value c of the second acceleration factor2F0.5, maximum value of weight ωmax0.8, minimum value of weight ωmin＝0.4。
A dynamic air conditioning system energy consumption prediction method is used for predicting the energy consumption of a building air conditioning system to be predicted, and the method is executed according to the following steps:
a, acquiring environmental data of a building air conditioning system to be predicted at a prediction time R, wherein the acquired data at the time R comprises an outdoor temperature value at the time R, an outdoor temperature value at the time R-1, an outdoor relative humidity value at the time R-1, an outdoor wind speed value at the time R-1, a direct solar radiation value at the time R-1, a scattered solar radiation value at the time R-1, a set value of a time-space temperature adjustment degree at the time R and an energy consumption mode of the time-space temperature adjustment system at the time R;
and step B, inputting the predicted environment data of the air conditioning system of the building at the predicted time R into an energy consumption prediction model constructed by the dynamic air conditioning system energy consumption prediction model construction method to obtain the energy consumption value of the air conditioning system of the building at the predicted time R to be predicted.
A dynamic air conditioning system energy consumption prediction model construction device is used for establishing an energy consumption prediction model of a building air conditioning system to be predicted, and comprises an environmental data acquisition module, an energy consumption value acquisition module and a model training module;
the environment data acquisition module is used for acquiring environment data of a building air conditioning system to be predicted at the ith day, wherein the environment data at the ith day comprises an ith day outdoor temperature value, an ith-1 th outdoor relative humidity value, an ith-1 th outdoor wind speed value, an ith-1 th solar direct radiation value, an ith-1 th solar scattered radiation value, an ith day space-time temperature adjustment set value and an ith day space-time modulation system energy consumption mode, wherein i is 1,2, …, 24, J is 1,2,3, …, and J is a positive integer;
wherein the ith time-modulation system energy consumption mode comprises zero energy consumption, low energy consumption, medium energy consumption or high energy consumption;
obtaining a plurality of environment data, and obtaining an environment data set;
the energy consumption value acquisition module is used for acquiring the energy consumption value of the j th day i of the building air conditioning system to be predicted; obtaining a plurality of energy consumption values to obtain a tag set;
the model training module is used for training an improved particle swarm optimization BP neural network by taking the environment data set as input and the label set as reference output;
when the BP neural network is optimized by training the improved particle swarm, the particle speed and the position are updated by adopting a formula II:
Figure BDA0002809978520000061
wherein vτ lDenotes the speed, x, of the τ -th particle at the l-th iterationτ lDenotes the position of the τ -th particle at the l-th iteration, l ═ 0,1,2, …, ∞, r1And r2Are all in [0,1 ]]Random number between, pbestτFor the individual extremum of the τ -th particle, gbestτIs a global extremum of the population, ω represents a weight, ωmaxRepresenting the maximum value of the weight, ωminRepresents the minimum value of the weight, c1Denotes a first acceleration factor, c2Representing a second acceleration factor, itmaxRepresenting the maximum value of the number of iterations, iterRepresenting the current number of iterations, c1fInitial value representing a first acceleration factor, c2fInitial value representing a second acceleration factor, c1FRepresents the final value of the first acceleration factor, c2FRepresents a final value of the second acceleration factor;
and obtaining an energy consumption prediction model.
Furthermore, the energy consumption value acquisition module comprises a judgment submodule, a temperature regulation and control acquisition submodule, a building simulation submodule and an energy consumption value simulation submodule;
the judgment sub-module is used for acquiring the highest outdoor temperature value of the building to be predicted on the jth day, if the highest outdoor temperature value is greater than T, the temperature is adjusted and controlled to obtain the sub-module, and if not, the energy consumption value of the building air conditioning system to be predicted on the jth day is 0;
the regulation and control temperature obtaining submodule is used for obtaining the regulation and control temperature g (t) of the air conditioning system at the ith day of the jth day by adopting a formula Ij,i) The unit is ℃:
Figure BDA0002809978520000071
wherein t isj,iRepresents an outdoor temperature value, t ', at day i of the building to be predicted'max、t'minRespectively, the maximum value and the minimum value of temperature regulation, and the unit is DEG Cj,maxThe maximum value of the j day outdoor temperature of the building to be predicted is given in DEG Cj,minThe minimum value of the outdoor temperature of the j day of the building to be predicted is calculated in units of ℃;
the building simulation submodule is used for obtaining a building simulation model of a building to be predicted;
and the energy consumption value simulation submodule is used for obtaining the energy consumption value of the building air-conditioning system to be predicted at the j th day i according to the building simulation model and the temperature regulated and controlled by the air-conditioning system.
Further, the initial value c of the first acceleration factor1f0.5, the final value c of the first acceleration factor1F2.4, the second acceleration factor initial value c2f2.4, the final value c of the second acceleration factor2F0.5, maximum value of weight ωmax0.8, minimum value of weight ωmin＝0.4。
A dynamic air conditioning system energy consumption prediction device is used for predicting the energy consumption of a building air conditioning system to be predicted, and comprises a data acquisition module and an energy consumption prediction module;
the data acquisition module is used for acquiring environmental data of the building air conditioning system to be predicted at a prediction time R, and the acquired data at the time R comprises an outdoor temperature value at the time R, an outdoor temperature value at the time R-1, an outdoor relative humidity value at the time R-1, an outdoor wind speed value at the time R-1, a direct solar radiation value at the time R-1, a scattered solar radiation value at the time R-1, a set value of a time-space temperature adjustment degree at the time R and an energy consumption mode of the time-space temperature adjustment system at the time R;
the energy consumption prediction module is used for inputting the predicted environment data of the building air conditioning system at the predicted time R into the energy consumption prediction model constructed by the dynamic air conditioning system energy consumption prediction model construction device to obtain the energy consumption value of the building air conditioning system to be predicted at the predicted time R.
Compared with the prior art, the invention has the following technical effects:
1. according to the dynamic air conditioning system energy consumption prediction model building and predicting method and device, the dynamic air conditioning system energy consumption is predicted time by building the improved particle swarm optimization BP neural network, so that the accuracy of energy consumption prediction is improved, and the energy conservation and optimization control of the air conditioning system are facilitated.
2. According to the dynamic energy consumption prediction model building and predicting method and device for the air conditioning system, the dynamic energy consumption control model is built, the dynamic change trend that the indoor thermal environment effectively follows the outdoor temperature is achieved, the dynamic control of the thermal environment is met, and meanwhile the comfort and health are unified.
Drawings
FIG. 1 is a flow chart of a prediction method provided by the present invention;
FIG. 2 is a graph of SSE trend with cluster number provided in an embodiment of the present invention;
FIG. 3 is a graph of clustering results of energy consumption provided in an embodiment of the present invention;
FIG. 4 is a schematic diagram of an energy consumption mode decision tree provided in an embodiment of the present invention;
FIG. 5 is a geometric building simulation model to be predicted provided in an embodiment of the present invention;
FIG. 6 is a graph illustrating a comparison of time-by-time energy consumption of two temperature controlled air conditioners provided in an embodiment of the present invention;
FIG. 7 is a graph of the test results of the PMV-PPD model provided in one embodiment of the present invention;
FIG. 8 is a BPNN structure diagram provided in one embodiment of the invention;
FIG. 9 is a flow chart of the IPSO-BPNN model provided in one embodiment of the invention
FIG. 10 is a comparison graph of the training convergence results provided in one embodiment of the present invention;
fig. 11 is a comparison graph of predicted values and actual values of four models provided in an embodiment of the present invention.
Detailed Description
The following embodiments of the present invention are provided, and it should be noted that the present invention is not limited to the following embodiments, and all equivalent changes based on the technical solutions of the present invention are within the protection scope of the present invention.
The definitions or conceptual connotations involved in the present invention will be explained first:
particle swarm optimization BP neural network model: the weight and the threshold of the BP neural network are used as initial solutions of the particle swarm algorithm, optimization is carried out to obtain an optimal solution, and then the weight and the threshold of the BP neural network are brought in to carry out network training.
Dynamic air conditioning system: the air conditioning system is controlled based on a dynamic temperature regulation strategy.
Direct solar radiation value: refers to the solar radiation that reaches the earth's surface as parallel light without changing the direction of illumination, and has the unit of W/m2。
Solar scattered radiation value: refers to a part of solar radiation which reaches the ground surface from various angles in the sky under the scattering action of gas, dust, aerosol and the like in the atmosphere when passing through the atmosphere, and the unit is W/m2。
Example one
As shown in fig. 1, in this embodiment, a method for constructing an energy consumption prediction model of a dynamic air conditioning system of a building to be predicted is disclosed, where the method is performed according to the following steps:
step 1, collecting environment data of a building air conditioning system to be predicted at the ith day of the jth day, wherein the environment data of the ith day of the jth day comprise an ith day outdoor temperature value, an ith-1 th outdoor relative humidity value, an ith-1 th outdoor wind speed value, an ith day direct solar radiation value, an ith-1 th solar scattered radiation value, an ith day space-time temperature adjustment set value and an ith day space-time adjustment system energy consumption mode, wherein i is 1,2, …, 24, J is 1,2,3, …, J is a positive integer; wherein the energy consumption mode of the ith time-modulation system comprises zero energy consumption, low energy consumption, medium energy consumption or high energy consumption;
in the step, the time-by-time energy consumption data are utilized to determine the environmental data needing to be collected of the building air conditioning system to be predicted, the time-by-time energy consumption data are clustered by adopting a K-Means clustering algorithm, the SSE is calculated to judge the optimal a value, and the energy consumption modes corresponding to different moments are divided;
after energy consumption modes of the air conditioning system are divided, in order to facilitate analysis of distinguishing conditions of the energy consumption modes, whether working days or overtime or not and whether overtime or not are selected as characteristic attributes of the energy consumption modes to classify different energy consumption modes, and a CART decision tree algorithm is adopted to construct an energy consumption mode distinguishing decision tree;
and determining important influence factors of energy consumption as input variables of the prediction model by using grey correlation degree analysis.
In the embodiment, 2208 groups of time-by-time energy consumption data from 6 months to 8 months are analyzed by three methods of clustering, classification and relevance analysis. The data analysis is realized by the following steps:
in the first step, the elbow method is used to determine the optimal a value of 4, and FIG. 2 shows the variation trend of SSE with cluster number. The clustering analysis result is shown in fig. 3, and four pattern labels of zero energy consumption, low energy consumption, medium energy consumption and high energy consumption are respectively added to the four types of data clusters of the clustering result.
And secondly, after the energy consumption modes of the air conditioning system are divided, selecting whether the working day, whether the overtime time and whether the working time are taken as characteristic attributes of the energy consumption modes to classify different energy consumption modes in order to conveniently analyze the judging conditions of the energy consumption modes. And randomly selecting 80% of the data set as a training set, using the rest 20% as a test set, and constructing an energy consumption mode discrimination decision tree by adopting a CART decision tree algorithm. The partial energy consumption data with the characteristic attributes added is shown in table 1.
TABLE 1 partial energy consumption data with added characteristic attributes
Figure BDA0002809978520000111
Whether the working day is used as a root node of the decision tree or not, whether the overtime time and the working time are respectively used as internal nodes or not are known by calculating the kini coefficient, the energy consumption mode is a leaf node, and the established energy consumption mode is used for judging the decision tree as shown in figure 4. As can be seen from the decision tree, the four energy consumption modes can be classified according to different time category attributes.
And thirdly, selecting the outdoor temperature at the moment i, the outdoor temperature at the moment i-1, the outdoor relative humidity at the moment i-1, the outdoor wind speed at the moment i-1, the direct solar radiation at the moment i-1, the solar scattered radiation at the moment i-1, the set value of the air-conditioning temperature at the moment i-1 and the energy consumption mode of the air-conditioning system at the moment i-1 as correlation factors, and analyzing and judging important influence factors of the energy consumption at the moment i by using gray correlation degrees. Based on the above result of classifying the energy consumption at the time i, since the zero energy consumption mode can be directly estimated according to the classification rule, the corresponding sample sequence is removed from the 2208 group of data set, and the 741 group of samples after removal are selected to calculate the association degree of the energy consumption data at the time i and the association factor thereof.
The calculation results of the correlation degree are shown in table 2, and the invention determines 8 influencing factors with the correlation factor higher than 0.6 as the input items of the prediction model.
TABLE 2 correlation degree of each influence factor and air conditioner energy consumption at t moment
Figure BDA0002809978520000121
In the embodiment, the unit of the outdoor wind speed at the moment i-1 is m/s, the unit of the outdoor temperature at the moment i-1 and the unit of the air-conditioning temperature set value at the moment i are all C, the unit of the outdoor relative humidity at the moment i-1 is%, and the unit of the direct solar radiation at the moment i-1 is W/m2The unit of the solar scattered radiation at the moment i-1 is W/m2And the zero energy consumption, the low energy consumption, the medium energy consumption and the high energy consumption in the energy consumption mode of the air conditioning system at the moment i are respectively expressed by 0,1,2 and 3 and are dimensionless parameters.
Step 2, acquiring an energy consumption value of the j th day i of the building air conditioning system to be predicted;
in this step, the method for obtaining the energy consumption value of the j th day i of the building air conditioning system to be predicted may be obtained through a simulation experiment.
Optionally, the obtaining of the energy consumption value of the j th day i of the building air conditioning system to be predicted specifically includes:
step 2.1, obtaining the highest outdoor temperature value of the j th day of the building to be predicted, if the highest outdoor temperature value is greater than T, executing the step 2.2, otherwise, the j th day i energy consumption value of the air conditioning system of the building to be predicted is 0;
in the present embodiment, T is set to 27 ℃, and only the time required to turn on the air conditioner, that is, the air conditioner operation time, among the operation times is considered. Therefore, the dynamic temperature regulation can be carried out when the outdoor maximum temperature exceeds 27 ℃ in the working time.
Step 2.2, obtaining the air conditioner regulating temperature g (t) at the ith time of the jth day by adopting a formula Ij,i) The unit is ℃:
Figure BDA0002809978520000131
wherein t isj,iRepresents an outdoor temperature value, t ', at day i of the building to be predicted'max、t'minRespectively, the maximum value and the minimum value of temperature regulation, and the unit is DEG Cj,maxFor buildings to be predictedThe maximum outdoor temperature on day j of the subject is given in degrees Celsiusj,minThe minimum value of the outdoor temperature of the j day of the building to be predicted is calculated in units of ℃;
in this example, the maximum temperature control value was 28 ℃ and the minimum temperature control value was 26 ℃.
The regulation and control strategy provided in the embodiment realizes that the indoor thermal environment effectively follows the dynamic change trend of the outdoor temperature, and realizes the unity of comfort and health while satisfying the dynamic control of the thermal environment.
Step 2.3, obtaining a building simulation model of the building to be predicted;
in this example, the energy plus software was used to obtain a geometric building simulation model to be predicted, as shown in fig. 5.
And 2.4, obtaining the energy consumption value of the air conditioning system of the building to be predicted at the j th day i according to the building simulation model and the temperature regulated and controlled by the air conditioning system.
In this embodiment, the energy consumption of the variable air volume air conditioning system based on the dynamic temperature regulation strategy is simulated by applying energy yPlus software.
Regulating and controlling the temperature g (t) of the building simulation model and the air conditionerj,i) And inputting the energy consumption value into EnergyPlus software to obtain the energy consumption value of the ith day of the building to be predicted.
In this embodiment, the human thermal comfort of the dynamic thermal environment control is examined. The thermal comfort was evaluated using the PMV-PPD index proposed by the Danish student's professor Fanger. Since most of the personnel in the office environment sit still or do only slight activities, with reference to the relevant parameter settings of the ASHRAE standard, the following assumptions are made: the external work of the human body is 0W/m2The metabolic rate is 62.5W/m2The indoor wind speed is 0.3m/s, the clothing thermal resistance is 0.5clo, the average radiation temperature is the indoor air temperature, and the test result is shown in figure 7.
In this embodiment, a model of an air-conditioning office building in the region of western security is built, the building has 12 floors in total, the window-wall ratio is 0.4, and the total area is 19296.12m2Table 3 shows a basic outline of the building. Density of people in indoor load is 4m2The power of lighting is 11W/m2And the power of the device is20W/m2. All the building design parameters refer to DBJ T6161-60-2011 public building energy-saving design standard. The energy consumption of the variable air volume air conditioning system under two temperature controls is simulated by using energy plus, and the comparison result is shown in fig. 6.
TABLE 3 basic overview of the construction
Figure BDA0002809978520000151
Step 3, setting i to i +1, returning to the step 1 until i to 24, and executing the step 4;
step 4, setting J to be J +1, returning to the step 1 until J is J, and executing the step 5;
step 5, obtaining a plurality of environment data to obtain an environment data set;
obtaining a plurality of energy consumption values to obtain a tag set;
in this embodiment, the environmental data and the corresponding energy consumption value in a period of time are finally obtained through multiple times of collection, for example, the environmental data from 8 o 'clock to 18 o' clock every 6 to 8 months in 2020 and the corresponding energy consumption value at each moment are collected.
Step 6, taking the environment data set as input, taking the label set as reference output, and training an improved particle swarm optimization BP neural network;
when the BP neural network is optimized by training the improved particle swarm, the particle speed and the position are updated by adopting a formula II:
Figure BDA0002809978520000152
wherein vτ lDenotes the speed, x, of the τ -th particle at the l-th iterationτ lDenotes the position of the τ -th particle at the l-th iteration, l ═ 0,1,2, …, ∞, r1And r2Are all in [0,1 ]]Random number between, pbestτFor the individual extremum of the τ -th particle, gbestτIs a global extremum of the population, ω represents a weight, ωmaxRepresenting the maximum value of the weight, ωminRepresents the minimum value of the weight, c1Denotes a first acceleration factor, c2Representing a second acceleration factor, itmaxRepresenting the maximum value of the number of iterations, iterRepresenting the current number of iterations, c1fInitial value representing a first acceleration factor, c2fInitial value representing a second acceleration factor, c1FRepresents the final value of the first acceleration factor, c2FRepresents a final value of the second acceleration factor;
and obtaining an energy consumption prediction model.
In this embodiment, as shown in fig. 9, the training of the improved particle swarm BP neural network specifically includes:
firstly, establishing a BPNN model, as shown in FIG. 8, taking the environmental data collected in step 1 as an input item of the model, taking the energy consumption value obtained in step 2 as an output item of the model, and determining the number of hidden layer nodes according to empirical formula III and a trial-and-error method.
Figure BDA0002809978520000161
In the formula, gamma is a constant between 0 and 10, and alpha, beta and lambda are the node numbers of the input layer, the output layer and the hidden layer respectively.
Randomly generating a population of particles TμAs an initial solution to the neural network.
Tμ＝(wμ1,wμ2,…,wμd)T,μ＝1,2,…,z
d＝α·γ+γ·β+γ+β
Wherein d is the search space dimension and z is the population number.
In the second step, the particles T in the first step are treatedμCarrying out network training on the weight and the threshold assigned to the BPNN, and calculating the individual T in the population T according to the formula IVμIs adapted to
Figure BDA0002809978520000162
In the formula, Co(z) and Yo(z) are each independentlyAnd g is the number of samples in the training set.
And thirdly, updating the individual extremum and the global extremum of the particle group.
And step four, updating the particle speed and position by adopting a formula II:
Figure BDA0002809978520000171
wherein vτ lDenotes the speed, x, of the τ -th particle at the l-th iterationτ lDenotes the position of the τ -th particle at the l-th iteration, l ═ 0,1,2, …, ∞, r1And r2Are all in [0,1 ]]Random number between, pbestτFor the individual extremum of the τ -th particle, gbestτIs a global extremum of the population, ω represents a weight, ωmaxRepresenting the maximum value of the weight, ωminRepresents the minimum value of the weight, c1Denotes a first acceleration factor, c2Representing a second acceleration factor, itmaxRepresenting the maximum value of the number of iterations, iterRepresenting the current number of iterations, c1fInitial value representing a first acceleration factor, c2fInitial value representing a second acceleration factor, c1FRepresents the final value of the first acceleration factor, c2FRepresents a final value of the second acceleration factor;
calculating the fitness value, the individual extreme value and the global extreme value of the new particle group again, updating the particle group according to the second step, and continuously and circularly iterating until the iteration is 200 times, namely iter＝200。
And fifthly, assigning the optimal population particles to the weight and the threshold of the BPNN for training until the iteration times are met, and outputting a result.
Preferably, the first acceleration factor initial value c1f0.5, the final value c of the first acceleration factor1F2.4, second acceleration factor initial value c2f2.4, second acceleration factor final value c2F0.5, maximum value of weight ωmax0.8, minimum value of weight ωmin＝0.4。
In this embodiment, 80% of the time-by-time energy consumption data sets of the 741 air conditioning systems are randomly selected as training sets of the network, and the remaining 20% are used as test sets. And taking the ith outdoor temperature value, the ith-1 outdoor relative humidity value, the ith-1 outdoor wind speed value, the ith-1 solar direct radiation value, the ith-1 solar scattered radiation value, the ith time-space temperature regulation setting value and the ith time-space temperature regulation system energy consumption mode as input items of the model, taking the ith time energy consumption data as output items of the model, and finally determining that the number of nodes of the hidden layer is 13, so that the network structure is 8-13-1, as shown in figure 8.
And respectively training a traditional single BPNN model and a BPNN model, a PSO-BPNN model and an IPSO-BPNN model which are established on the basis of analyzing energy consumption by adopting an integration method by using the same data set. The maximum iteration number of the BPNN model is 100, the target error is 0.001, and the learning rate is 0.01; the evolution algebra of the PSO-BPNN model and the IPSO-BPNN model is 200, the population size is 100, and the velocity range of the particles is [ -1,1]In the range of [ -5,5 ]]. Acceleration factor c in PSO-BPNN model1And c2Both are 2, the inertial weight is 0.7; acceleration factor c in IPSO-BPNN model1And c2Respectively in a dynamic adjustment range of [0.5,2.4 ]]And [2.4,0.5 ]]Dynamic adjustment range of inertial weight [0.4,0.8 ]]. The iterative convergence conditions of the IPSO-BPNN model and the PSO-BPNN model are shown in FIG. 10, and it can be seen that the PSO-BPNN model requires 92 steps of iteration to achieve convergence, while the IPSO-BPNN model only requires 27 steps of iteration to achieve convergence, and the convergence speed is high. FIG. 11 is a comparison of predicted values and actual values for four models.
Example two
In this embodiment, a dynamic prediction method for energy consumption of an air conditioning system of a building to be predicted is disclosed, and the method is performed according to the following steps:
a, acquiring environmental data of a building air conditioning system to be predicted at a prediction time R, wherein the acquired data at the time R comprises an outdoor temperature value at the time R, an outdoor temperature value at the time R-1, an outdoor relative humidity value at the time R-1, an outdoor wind speed value at the time R-1, a direct solar radiation value at the time R-1, a solar scattering radiation value at the time R-1, a time R time temperature setting value and a time R time conditioning system energy consumption mode;
and step B, inputting the predicted environment data of the air conditioning system of the building at the predicted time R into an energy consumption prediction model constructed by the dynamic air conditioning system energy consumption prediction model construction method to obtain the energy consumption value of the air conditioning system of the building at the predicted time R to be predicted.
In this embodiment, the accuracy of the prediction method is tested, and as shown in table 4, the accuracy of the prediction model is evaluated by combining three performance evaluation indexes, namely, MAPE, MAE and RMSE, and the smaller the value, the higher the accuracy. It can be seen that the BPNN model, the PSO-BPNN model and the IPSO-BPNN model which are established on the basis of analyzing the energy consumption by adopting the integration method have higher prediction precision than the traditional single BPNN model, and the MAPE is reduced by 3.4%.
TABLE 4 error index analysis
Figure BDA0002809978520000191
The result shows that MAPE corresponding to the BPNN model, the PSO-BPNN model, the IPSO-BPNN model and the traditional single BPNN model which are established on the basis of analyzing the energy consumption by adopting the integration method are respectively 6.13%, 5.59%, 4.87% and 8.24%. It can be seen that the neural network models established on the basis of analyzing energy consumption by adopting an integration method are higher in precision than the traditional single BPNN model, and the IPSO algorithm can better make up for the defects of local minimum traps, low convergence speed and the like existing in the BPNN and the PSO-BPNN.
EXAMPLE III
The embodiment discloses a dynamic air conditioning system energy consumption prediction model construction device, which is used for establishing an energy consumption prediction model of a building air conditioning system to be predicted, and comprises an environmental data acquisition module, an energy consumption value acquisition module and a model training module;
the environment data acquisition module is used for acquiring environment data of a building air conditioning system to be predicted at the ith day, wherein the environment data at the ith day comprises an ith day outdoor temperature value, an ith-1 th outdoor relative humidity value, an ith-1 th outdoor wind speed value, an ith-1 th solar direct radiation value, an ith-1 th solar scattered radiation value, an ith day space-time temperature adjustment set value and an ith day space-time adjustment system energy consumption mode, wherein i is 1,2, …, 24, J is 1,2,3, …, J and J are positive integers;
wherein the energy consumption mode of the ith time-modulation system comprises zero energy consumption, low energy consumption, medium energy consumption or high energy consumption;
obtaining a plurality of environment data, and obtaining an environment data set;
the energy consumption value acquisition module is used for acquiring the energy consumption value of the j th day i of the building air conditioning system to be predicted; obtaining a plurality of energy consumption values to obtain a tag set;
the model training module is used for training and improving the particle swarm optimization BP neural network by taking the environment data set as input and the label set as reference output;
when the BP neural network is optimized by training the improved particle swarm, the particle speed and the position are updated by adopting a formula II:
Figure BDA0002809978520000211
wherein vτ lDenotes the speed, x, of the τ -th particle at the l-th iterationτ lDenotes the position of the τ -th particle at the l-th iteration, l ═ 0,1,2, …, ∞, r1And r2Are all in [0,1 ]]Random number between, pbestτFor the individual extremum of the τ -th particle, gbestτIs a global extremum of the population, ω represents a weight, ωmaxRepresenting the maximum value of the weight, ωminRepresents the minimum value of the weight, c1Denotes a first acceleration factor, c2Representing a second acceleration factor, itmaxRepresenting the maximum value of the number of iterations, iterRepresenting the current number of iterations, c1fInitial value representing a first acceleration factor, c2fInitial value representing a second acceleration factor, c1FRepresents the final value of the first acceleration factor, c2FRepresents a final value of the second acceleration factor;
and obtaining an energy consumption prediction model.
Optionally, the energy consumption value obtaining module comprises a judging submodule, a temperature regulating and controlling obtaining submodule, a building simulation submodule and an energy consumption value simulation submodule;
the judgment submodule is used for acquiring the highest outdoor temperature value of the building to be predicted on the jth day, if the highest outdoor temperature value is greater than T, the temperature is adjusted and controlled to obtain the submodule, and if not, the energy consumption value of the building air conditioning system to be predicted on the jth day is 0;
the regulation and control temperature obtaining submodule is used for obtaining the regulation and control temperature g (t) of the air conditioning system at the ith day of the j day by adopting a formula Ij,i) The unit is ℃:
Figure BDA0002809978520000221
wherein t isj,iRepresents an outdoor temperature value, t ', at day i of the building to be predicted'max、t'minRespectively, the maximum value and the minimum value of temperature regulation, and the unit is DEG Cj,maxThe maximum value of the j day outdoor temperature of the building to be predicted is given in DEG Cj,minThe minimum value of the outdoor temperature of the j day of the building to be predicted is calculated in units of ℃;
the building simulation submodule is used for obtaining a building simulation model of a building to be predicted;
and the energy consumption value simulation submodule is used for obtaining the energy consumption value of the building air conditioning system to be predicted at the jth day i according to the building simulation model and the temperature regulated and controlled by the air conditioning system.
Optionally, a first acceleration factor initial value c1f0.5, the final value c of the first acceleration factor1F2.4, second acceleration factor initial value c2f2.4, second acceleration factor final value c2F0.5, maximum value of weight ωmax0.8, minimum value of weight ωmin＝0.4。
Example four
The embodiment discloses a dynamic air conditioning system energy consumption prediction device, which is used for predicting the energy consumption of a building air conditioning system to be predicted and comprises a data acquisition module and an energy consumption prediction module;
the data acquisition module is used for acquiring environmental data of the building air conditioning system to be predicted at a prediction time R, and the acquired data at the time R comprises an outdoor temperature value at the time R, an outdoor temperature value at the time R-1, an outdoor relative humidity value at the time R-1, an outdoor wind speed value at the time R-1, a direct solar radiation value at the time R-1, a scattered solar radiation value at the time R-1, a set value of a time-space temperature adjustment degree at the time R and an energy consumption mode of the time-space temperature adjustment system at the time R;
the energy consumption prediction module is used for inputting the predicted environment data of the building air conditioning system at the prediction time R into the energy consumption prediction model constructed by the dynamic air conditioning system energy consumption prediction model construction device in the third embodiment, and obtaining the energy consumption value of the building air conditioning system to be predicted at the prediction time R.
Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention may be implemented by software plus necessary general hardware, and certainly may also be implemented by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present invention may be substantially implemented or a part of the technical solutions contributing to the prior art may be embodied in the form of a software product, where the computer software product is stored in a readable storage medium, such as a floppy disk, a hard disk, or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

Claims (8)

1. A dynamic air conditioning system energy consumption prediction model construction method is used for establishing an energy consumption prediction model of a building air conditioning system to be predicted, and is characterized by being executed according to the following steps:
step 1, collecting environment data of a building air conditioning system to be predicted at the ith day of a J, wherein the environment data at the ith day of the J comprises an ith day outdoor temperature value, an ith-1 th outdoor relative humidity value, an ith-1 th outdoor wind speed value, an ith-1 th solar direct radiation value, an ith-1 th solar scattered radiation value, an ith day space-time temperature adjustment set value and an ith day space-time adjustment system energy consumption mode, wherein i is 1,2, …, 24, J is 1,2,3, …, J and J are positive integers;
wherein the ith day time-modulation system energy consumption mode comprises zero energy consumption, low energy consumption, medium energy consumption or high energy consumption;
step 2, acquiring an energy consumption value of the j th day i of the building air conditioning system to be predicted;
step 3, setting i to i +1, returning to the step 1 until i to 24, and executing the step 4;
step 4, setting J to be J +1, returning to the step 1 until J is J, and executing the step 5;
step 5, obtaining a plurality of environment data to obtain an environment data set; obtaining a plurality of energy consumption values to obtain a tag set;
step 6, taking the environment data set as input, taking the label set as reference output, and training an improved particle swarm optimization BP neural network;
when the BP neural network is optimized by training the improved particle swarm, the particle speed and the position are updated by adopting a formula II:
Figure FDA0002809978510000021
wherein vτ lDenotes the speed, x, of the τ -th particle at the l-th iterationτ lDenotes the position of the τ -th particle at the l-th iteration, l ═ 0,1,2, …, ∞, r1And r2Are all in [0,1 ]]Random number between, pbestτFor the individual extremum of the τ -th particle, gbestτIs a global extremum of the population, ω represents a weight, ωmaxRepresenting the maximum value of the weight, ωminRepresents the minimum value of the weight, c1Denotes a first acceleration factor, c2Representing a second acceleration factor, itmaxRepresenting the maximum value of the number of iterations, iterIndicates the currentNumber of iterations, c1fInitial value representing a first acceleration factor, c2fInitial value representing a second acceleration factor, c1FRepresents the final value of the first acceleration factor, c2FRepresents a final value of the second acceleration factor;
and obtaining an energy consumption prediction model.
2. The method for constructing the dynamic energy consumption prediction model of the air conditioning system as claimed in claim 1, wherein in the step 2, the obtaining of the j-th and i-th energy consumption values of the building air conditioning system to be predicted specifically comprises:
step 2.1, obtaining the highest outdoor temperature value of the j th day of the building to be predicted, if the highest outdoor temperature value is greater than T, executing the step 2.2, otherwise, the j th day i energy consumption value of the air conditioning system of the building to be predicted is 0;
step 2.2, obtaining the regulation temperature g (t) of the air conditioning system at the ith day of the j day by adopting the formula Ij,i) The unit is ℃:
Figure FDA0002809978510000031
wherein t isj,iRepresents an outdoor temperature value, t ', at day i of the building to be predicted'max、t'minRespectively, the maximum value and the minimum value of temperature regulation, and the unit is DEG Cj,maxThe maximum value of the j day outdoor temperature of the building to be predicted is given in DEG Cj,minThe minimum value of the outdoor temperature of the j day of the building to be predicted is calculated in units of ℃;
step 2.3, obtaining a building simulation model of the building to be predicted;
and 2.4, obtaining the energy consumption value of the air conditioning system of the building to be predicted at the j th day i according to the building simulation model and the temperature regulated and controlled by the air conditioning system.
3. The method as claimed in claim 1, wherein the initial value c of the first acceleration factor is set as the initial value1f0.5, the final value c of the first acceleration factor1F2.4, the second acceleration factor initial value c2f2.4, the final value c of the second acceleration factor2F0.5, maximum value of weight ωmax0.8, minimum value of weight ωmin＝0.4。
4. A dynamic air conditioning system energy consumption prediction method is used for predicting the energy consumption of a building air conditioning system to be predicted, and is characterized in that the method is executed according to the following steps:
a, acquiring environmental data of a building air conditioning system to be predicted at a prediction time R, wherein the acquired data at the time R comprises an outdoor temperature value at the time R, an outdoor temperature value at the time R-1, an outdoor relative humidity value at the time R-1, an outdoor wind speed value at the time R-1, a direct solar radiation value at the time R-1, a scattered solar radiation value at the time R-1, a set value of a time-space temperature adjustment degree at the time R and an energy consumption mode of the time-space temperature adjustment system at the time R;
and step B, inputting the predicted environment data of the air conditioning system of the building at the predicted time R into the energy consumption prediction model constructed by the dynamic air conditioning system energy consumption prediction model construction method according to any one of claims 1 to 3, and obtaining the energy consumption value of the air conditioning system of the building at the predicted time R to be predicted.
5. A dynamic air conditioning system energy consumption prediction model construction device is used for establishing an energy consumption prediction model of a building air conditioning system to be predicted and is characterized by comprising an environmental data acquisition module, an energy consumption value acquisition module and a model training module;
the environment data acquisition module is used for acquiring environment data of a building air conditioning system to be predicted at the ith day, wherein the environment data at the ith day comprises an ith day outdoor temperature value, an ith-1 th outdoor relative humidity value, an ith-1 th outdoor wind speed value, an ith-1 th solar direct radiation value, an ith-1 th solar scattered radiation value, an ith day space-time temperature adjustment set value and an ith day space-time modulation system energy consumption mode, wherein i is 1,2, …, 24, J is 1,2,3, …, and J is a positive integer;
wherein the ith time-modulation system energy consumption mode comprises zero energy consumption, low energy consumption, medium energy consumption or high energy consumption;
obtaining a plurality of environment data, and obtaining an environment data set;
the energy consumption value acquisition module is used for acquiring the energy consumption value of the j th day i of the building air conditioning system to be predicted; obtaining a plurality of energy consumption values to obtain a tag set;
the model training module is used for training an improved particle swarm optimization BP neural network by taking the environment data set as input and the label set as reference output;
when the BP neural network is optimized by training the improved particle swarm, the particle speed and the position are updated by adopting a formula II:
Figure FDA0002809978510000051
wherein vτ lDenotes the speed, x, of the τ -th particle at the l-th iterationτ lDenotes the position of the τ -th particle at the l-th iteration, l ═ 0,1,2, …, ∞, r1And r2Are all in [0,1 ]]Random number between, pbestτFor the individual extremum of the τ -th particle, gbestτIs a global extremum of the population, ω represents a weight, ωmaxRepresenting the maximum value of the weight, ωminRepresents the minimum value of the weight, c1Denotes a first acceleration factor, c2Representing a second acceleration factor, itmaxRepresenting the maximum value of the number of iterations, iterRepresenting the current number of iterations, c1fInitial value representing a first acceleration factor, c2fInitial value representing a second acceleration factor, c1FRepresents the final value of the first acceleration factor, c2FRepresents a final value of the second acceleration factor;
and obtaining an energy consumption prediction model.
6. The dynamic air conditioning system energy consumption prediction model construction device of claim 5, wherein the energy consumption value acquisition module comprises a judgment sub-module, a regulation and control temperature acquisition sub-module, a building simulation sub-module, and an energy consumption value simulation sub-module;
the judgment sub-module is used for acquiring the highest outdoor temperature value of the building to be predicted on the jth day, if the highest outdoor temperature value is greater than T, the temperature is adjusted and controlled to obtain the sub-module, and if not, the energy consumption value of the building air conditioning system to be predicted on the jth day is 0;
the regulation and control temperature obtaining submodule is used for obtaining the regulation and control temperature g (t) of the air conditioning system at the ith day of the jth day by adopting a formula Ij,i) The unit is ℃:
Figure FDA0002809978510000061
wherein t isj,iRepresents an outdoor temperature value, t ', at day i of the building to be predicted'max、t'minRespectively, the maximum value and the minimum value of temperature regulation, and the unit is DEG Cj,maxThe maximum value of the j day outdoor temperature of the building to be predicted is given in DEG Cj,minThe minimum value of the outdoor temperature of the j day of the building to be predicted is calculated in units of ℃;
the building simulation submodule is used for obtaining a building simulation model of a building to be predicted;
and the energy consumption value simulation submodule is used for obtaining the energy consumption value of the building air-conditioning system to be predicted at the j th day i according to the building simulation model and the temperature regulated and controlled by the air-conditioning system.
7. The dynamic air conditioning system energy consumption prediction model building apparatus of claim 5, wherein the initial value c of the first acceleration factor1f0.5, the final value c of the first acceleration factor1F2.4, the second acceleration factor initial value c2f2.4, the final value c of the second acceleration factor2F0.5, maximum value of weight ωmax0.8, minimum weightValue omegamin＝0.4。
8. A dynamic air conditioning system energy consumption prediction device is used for predicting the energy consumption of a building air conditioning system to be predicted and is characterized by comprising a data acquisition module and an energy consumption prediction module;
the data acquisition module is used for acquiring environmental data of the building air conditioning system to be predicted at a prediction time R, and the acquired data at the time R comprises an outdoor temperature value at the time R, an outdoor temperature value at the time R-1, an outdoor relative humidity value at the time R-1, an outdoor wind speed value at the time R-1, a direct solar radiation value at the time R-1, a scattered solar radiation value at the time R-1, a set value of a time-space temperature adjustment degree at the time R and an energy consumption mode of the time-space temperature adjustment system at the time R;
the energy consumption prediction module is used for inputting the predicted environment data of the building air conditioning system at the predicted time R into the energy consumption prediction model constructed by the dynamic air conditioning system energy consumption prediction model construction device according to any one of claims 5 to 7, and obtaining the energy consumption value of the building air conditioning system at the predicted time R to be predicted.
CN202011387024.9A 2020-12-01 2020-12-01 Dynamic air conditioning system energy consumption prediction model construction and prediction method and device Active CN112524751B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011387024.9A CN112524751B (en) 2020-12-01 2020-12-01 Dynamic air conditioning system energy consumption prediction model construction and prediction method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011387024.9A CN112524751B (en) 2020-12-01 2020-12-01 Dynamic air conditioning system energy consumption prediction model construction and prediction method and device

Publications (2)

Publication Number Publication Date
CN112524751A true CN112524751A (en) 2021-03-19
CN112524751B CN112524751B (en) 2022-04-19

Family

ID=74996001

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011387024.9A Active CN112524751B (en) 2020-12-01 2020-12-01 Dynamic air conditioning system energy consumption prediction model construction and prediction method and device

Country Status (1)

Country Link
CN (1) CN112524751B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115513940A (en) * 2022-10-13 2022-12-23 深圳鸿富城建设集团有限公司 Heating and ventilation power supply system of data center and power supply method thereof
CN115618462A (en) * 2022-10-10 2023-01-17 哈尔滨工业大学 Urban block local wind-heat environment coupling prediction method based on reduced scale model physical simulation
CN115907191A (en) * 2022-12-08 2023-04-04 山东建筑大学 Adaptive building photovoltaic skin model prediction control method
CN117075566A (en) * 2023-10-13 2023-11-17 深圳市明源云链互联网科技有限公司 Energy consumption optimization method, device, equipment and computer readable storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976221A (en) * 2010-10-14 2011-02-16 北京航空航天大学 Particle swarm taboo combination-based parallel test task dispatching method and platform
CN104484715A (en) * 2014-11-28 2015-04-01 江苏大学 Neural network and particle swarm optimization algorithm-based building energy consumption predicting method
CN107037728A (en) * 2017-03-22 2017-08-11 安徽农业大学 Greenhouse optimal control method based on multiple objective gray particle cluster algorithm
CN111649457A (en) * 2020-05-13 2020-09-11 中国科学院广州能源研究所 Dynamic predictive machine learning type air conditioner energy-saving control method
CN112884012A (en) * 2021-01-26 2021-06-01 山东历控能源有限公司 Building energy consumption prediction method based on support vector machine principle

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976221A (en) * 2010-10-14 2011-02-16 北京航空航天大学 Particle swarm taboo combination-based parallel test task dispatching method and platform
CN104484715A (en) * 2014-11-28 2015-04-01 江苏大学 Neural network and particle swarm optimization algorithm-based building energy consumption predicting method
CN107037728A (en) * 2017-03-22 2017-08-11 安徽农业大学 Greenhouse optimal control method based on multiple objective gray particle cluster algorithm
CN111649457A (en) * 2020-05-13 2020-09-11 中国科学院广州能源研究所 Dynamic predictive machine learning type air conditioner energy-saving control method
CN112884012A (en) * 2021-01-26 2021-06-01 山东历控能源有限公司 Building energy consumption prediction method based on support vector machine principle

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115618462A (en) * 2022-10-10 2023-01-17 哈尔滨工业大学 Urban block local wind-heat environment coupling prediction method based on reduced scale model physical simulation
CN115513940A (en) * 2022-10-13 2022-12-23 深圳鸿富城建设集团有限公司 Heating and ventilation power supply system of data center and power supply method thereof
CN115907191A (en) * 2022-12-08 2023-04-04 山东建筑大学 Adaptive building photovoltaic skin model prediction control method
CN117075566A (en) * 2023-10-13 2023-11-17 深圳市明源云链互联网科技有限公司 Energy consumption optimization method, device, equipment and computer readable storage medium
CN117075566B (en) * 2023-10-13 2024-03-12 深圳市明源云链互联网科技有限公司 Energy consumption optimization method, device, equipment and computer readable storage medium

Also Published As

Publication number Publication date
CN112524751B (en) 2022-04-19

Similar Documents

Publication Publication Date Title
CN112524751B (en) Dynamic air conditioning system energy consumption prediction model construction and prediction method and device
Cai et al. Predicting the energy consumption of residential buildings for regional electricity supply-side and demand-side management
CN106920006B (en) Subway station air conditioning system energy consumption prediction method based on ISOA-LSSVM
Tian et al. Predictive model of energy consumption for office building by using improved GWO-BP
CN103853106B (en) A kind of energy consumption Prediction Parameters optimization method of building energy supplied equipment
Han et al. Energy saving of buildings for reducing carbon dioxide emissions using novel dendrite net integrated adaptive mean square gradient
Xuemei et al. Building cooling load forecasting model based on LS-SVM
Liu et al. Identifying the most significant input parameters for predicting district heating load using an association rule algorithm
CN112415924A (en) Energy-saving optimization method and system for air conditioning system
CN110276393A (en) A kind of compound prediction technique of green building energy consumption
CN109858700A (en) BP neural network heating system energy consumption prediction technique based on similar screening sample
CN105320184A (en) Intelligent monitoring system of indoor environment of building
Zhao et al. Heating load prediction of residential district using hybrid model based on CNN
CN113762387B (en) Multi-element load prediction method for data center station based on hybrid model prediction
CN112418495A (en) Building energy consumption prediction method based on longicorn stigma optimization algorithm and neural network
CN114046593A (en) Dynamic predictive machine learning type air conditioner energy-saving control method and system
Sözer et al. Predicting the indoor thermal data for heating season based on short-term measurements to calibrate the simulation set-points
CN113947261A (en) Optimization decision support method for building energy conservation transformation
CN109919374A (en) Prediction of Stock Price method based on APSO-BP neural network
CN114240687A (en) Energy hosting efficiency analysis method suitable for comprehensive energy system
Zhang et al. Comparing the linear and logarithm normalized artificial neural networks in inverse design of aircraft cabin environment
Zhang et al. Enhancing multi-scenario data-driven energy consumption prediction in campus buildings by selecting appropriate inputs and improving algorithms with attention mechanisms
Wang et al. Prediction of heating load fluctuation based on fuzzy information granulation and support vector machine
Liao et al. Building energy efficiency assessment base on predict-center criterion under diversified conditions
CN111401638B (en) Spatial load prediction method based on extreme learning machine and load density index method

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