CN117760063A - Subway air conditioner load prediction method based on air enthalpy value - Google Patents
Subway air conditioner load prediction method based on air enthalpy value Download PDFInfo
- Publication number
- CN117760063A CN117760063A CN202311754298.0A CN202311754298A CN117760063A CN 117760063 A CN117760063 A CN 117760063A CN 202311754298 A CN202311754298 A CN 202311754298A CN 117760063 A CN117760063 A CN 117760063A
- Authority
- CN
- China
- Prior art keywords
- load
- model
- enthalpy value
- outdoor
- air conditioner
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013528 artificial neural network Methods 0.000 claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 23
- 230000006870 function Effects 0.000 claims abstract description 17
- 238000010586 diagram Methods 0.000 claims abstract description 16
- 238000012360 testing method Methods 0.000 claims abstract description 15
- 238000004378 air conditioning Methods 0.000 claims abstract description 14
- 230000017525 heat dissipation Effects 0.000 claims abstract description 12
- 238000004088 simulation Methods 0.000 claims abstract description 12
- 238000012795 verification Methods 0.000 claims abstract description 9
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims abstract description 7
- 238000013329 compounding Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims 2
- 238000013178 mathematical model Methods 0.000 claims 2
- 238000005265 energy consumption Methods 0.000 description 6
- 238000010606 normalization Methods 0.000 description 5
- 238000005286 illumination Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000009499 grossing Methods 0.000 description 3
- 238000012417 linear regression Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001143 conditioned effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- -1 wall thickness Substances 0.000 description 1
Landscapes
- Air Conditioning Control Device (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a subway air conditioner load prediction method based on an air enthalpy value, which comprises the following steps: the time-by-time load of the subway air conditioner in an air conditioning season is simulated through load simulation software, and compared with the historical actual operation load to verify the accuracy of simulation data, and a platform is built in the load simulation software, so that the platform can meet the functions, and meanwhile, the factors with strong correlation such as time-by-time fresh air, outdoor temperature, humidity, people flow and equipment heat dissipation which cause the load are output; dividing the simulated data into a training set, a verification set and a test set, wherein the ratio of the training set to the verification set to the test set is 7:2:1; training the simulated data through a BP neural network algorithm to obtain a first model, and verifying and testing the trained model; the outdoor temperature, outdoor humidity, and load predicted by the first model are extracted. And repeating the process to obtain a second model, changing the outdoor temperature and the outdoor humidity, predicting the load by using the second model, and comparing the load predicted by the first model with the load predicted by the first model to verify the accuracy of the second model. Determining an outdoor enthalpy value by using the outdoor temperature and the outdoor humidity, and constructing a scatter diagram of the enthalpy value and a load predicted by the second model; analyzing the scatter diagram by numerical analysis software, selecting a best fit equation from 10 functions of linearity, logarithm, quadratic, cubic, compounding and the like, and predicting the load by using the enthalpy value by using the best fit equation. The load prediction method improves the load prediction precision, solves the problem that the current load prediction input parameters are difficult to acquire, and is more suitable for application of actual engineering.
Description
Technical Field
The invention relates to an air conditioner load prediction method, in particular to a subway air conditioner load prediction method based on an air enthalpy value.
Background
In recent years, the demands of people on travel modes are higher and higher, subways become the preferred travel modes of people due to less traffic interference, rapidness and convenience and reliable quasi points, the high energy consumption is brought to the subway station air conditioning system while the convenience is high, in the subway operation process, the energy consumption of the station air conditioning system is a main source of energy consumption in the station, and mainly comes from long-time operation of high-power air conditioning equipment of the station, the operation of the equipment is generally directly related to the actual load of the station, and in the actual operation, the energy consumption of the air conditioning system is not matched with the actual demand due to the fact that the load condition of the station cannot be accurately acquired, so that a large amount of energy is wasted. Aiming at the phenomenon, a load prediction technology is proposed by a plurality of people, and the station load at the next moment is predicted to be used for guiding the starting and stopping of the air conditioning equipment, so that the aim of saving energy is fulfilled.
The widely applied load prediction methods at present are an exponential smoothing method, a gray prediction method, a linear regression method and a neural network prediction method, wherein the exponential smoothing method is one of time series prediction technology centers, prediction is carried out according to historical load data, and the linear regression method is an explanatory model based on regression analysis. The grey prediction theory is to accumulate random sequences to weaken the randomness of the random sequences, so as to find the reality law of load development. The air conditioner load prediction model established by the exponential smoothing method has the advantages of simple structure, low prediction cost and good system portability, but cannot effectively utilize data of related factors closely related to building load, so that the prediction precision is difficult to further improve, the processing work difficulty of the gray prediction method on the original data before modeling is high, and the prediction precision is influenced by the processing result of the original data. Multiple linear regression is not desirable in air conditioning predictions because of the nonlinear relationship between most influencing factors and air conditioning load. The more accurate prediction method is to use a neural network to predict the load, the accuracy of the method depends on the selection of input parameters, most of the current methods can select outdoor temperature, humidity, people flow, fresh air quantity, solar radiation and the like as the input parameters, but the data are not easy to acquire in actual engineering, so the method can accurately predict the load, but the acquisition difficulty coefficient of the conditions is large, and the method is unfavorable for the application in the actual engineering.
Disclosure of Invention
The air conditioner load prediction method based on the air enthalpy value is a technical means for predicting future load by analyzing the relation between the historical enthalpy value and the load, and aims to provide a load prediction method suitable for actual engineering. Aiming at the problem that the prior art can not well meet the precision requirement of actual engineering and the predicted input parameters are difficult to acquire. The invention provides an air conditioner load prediction method based on enthalpy values, which improves prediction accuracy and a model R thereof 2 Reaching 0.989. In addition, the variety of input parameters is reduced, so that the acquisition difficulty of the conditions is greatly reduced, and the method is more suitable for application of actual engineering.
The technical scheme of the invention is that the subway air conditioner load prediction method based on the air enthalpy value comprises the following steps:
step one: and simulating the time-by-time load of the subway air conditioner in the air conditioning season through load simulation software, and comparing the load with the historical actual running load to verify the accuracy of simulation data.
Step two: by building the platform in the load simulation software, the platform can meet the functions of outputting the load and outputting factors with strong correlation such as time-by-time fresh air, outdoor temperature, humidity, people flow, equipment heat dissipation and the like which cause the load.
Step three: the simulated data is divided into a training set, a verification set and a test set, and the ratio is 7:2:1.
Step four: the simulated data is trained through the BP neural network algorithm to obtain a first model, and the trained model is verified and tested.
Step five: the outdoor temperature, outdoor humidity, and load predicted by the first model are extracted. And repeating the third step and the fourth step, wherein the process obtains a second model, changes the outdoor temperature and the outdoor humidity, predicts the load by using the second model, and compares the load with the load predicted by the first model to verify the accuracy of the second model.
Step six: and fifthly, determining an outdoor enthalpy value by using the outdoor temperature and the outdoor humidity, and constructing a scatter diagram of the enthalpy value and the load predicted by the second model.
Step seven: and C, analyzing the scatter diagram in the step six by using statistical analysis software, selecting a best fit equation from 10 functions of linearity, logarithm, quadratic, cubic, compounding and the like, and predicting the load by using the enthalpy value by using the best fit equation.
Further, the first step simulates the time-by-time load of the subway air conditioner in the air conditioner season through load simulation software, and compares the time-by-time load with the historical actual operation load to verify the accuracy of simulation data. The method comprises the following specific steps: and building a physical model of the subway in modeling software, importing the built model into energy consumption simulation software for setting subway parameters, preferably, obtaining data with a step length of 1h, then outputting the air conditioner time-by-time load of the subway station, and comparing the output time-by-time load with the actual measurement load of the same year to verify the reliability of the simulation data.
Furthermore, the platform is built in the load simulation software, so that the platform can meet the functions, and simultaneously output the load and output the factors with strong correlation such as the moment-by-moment fresh air quantity, the outdoor temperature, the humidity, the people flow, the equipment heat dissipation and the like which cause the load. The method comprises the following specific steps: firstly, parameters needing to be input, namely, the flow of people, the outdoor temperature, the outdoor humidity, the fresh air quantity, the heat dissipation of equipment and the illumination heat dissipation are selected from the Building module. And then, the selected output parameters are given to an intermediate module for unifying the formats, and finally, the unified formats are transmitted to an output module for visual output of the parameters.
Further, the simulated data are divided into a training set, a verification set and a test set in the ratio of 7:2:1. The method comprises the following specific steps: the simulated data is removed from the remaining 3600 sets of data for the portion of the night that is not air conditioned, wherein the training set is used to train the data set of the neural network model, typically accounting for (70% -80%) of the entire data set. Wherein the test set is used to evaluate the performance of the model on unseen data, typically accounting for (10% -20%) of the entire data set. Where the validation set is used to adjust and select model hyper-parameters typically account for (10% -20%) of the entire dataset. In summary, the training set, the validation set, and the test set are in a ratio of 7:2:1.
Further, training the simulated data through the BP neural network algorithm to obtain a first model, and verifying and testing the trained model. The method comprises the following specific steps: firstly, subway air conditioner load data are extracted from a designated excel file by calling an xlsread function in matlab, then the data are preprocessed, normalization processing is needed to be used for reducing the influence of distribution change of each data, the data distribution is mapped to a determined interval, and the data are normalized to be between [0,1] by using a max-min normalization method. And then determining the structure of the neural network and establishing a model, wherein 6 input parameters of the neural network are determined in the step two, so that the number of input layers of the neural network is 6, the output layers are prediction results of the load of the air conditioning system of the subway station, the number of output layers of the neural network is 1, and the hidden layers of the neural network can be determined by the following formula.
(J-hidden layer node number I-input layer node number K-output layer node number a E [1, 10)]) And determining that the number of hidden layer nodes is 3-13 through calculation, so that the optimal number of hidden layers is 10 through setting different hidden layer nodes and comparing error modes. Defining a modelThe iteration number of (a) is 1000, the learning rate is 0.01, the training minimum error is 1 multiplied by 10 -5 . And then training the neural network, determining the structure and the model of the neural network, and training the neural network by using the normalized data to obtain a first model. And finally, predicting and inversely normalizing, namely predicting the load by using a first model, and inversely normalizing the predicted load and each input parameter to restore each group of data to the original order of magnitude.
Further, the step five extracts the outdoor temperature, the outdoor humidity and the load predicted by the first model. And repeating the third step and the fourth step, wherein the process obtains a second model, changes the outdoor temperature and the outdoor humidity, predicts the load by using the second model, and compares the load with the load predicted by the first model to verify the accuracy of the second model. The method comprises the following specific steps: and (3) repeating the third step and the fourth step by using the load predicted by the first model and the outdoor temperature and the outdoor humidity to obtain a second model. The load is predicted by the second model using the outdoor temperature and the outdoor humidity as input parameters. The load predicted by the first model is compared with the load predicted by the second model to verify the fit of the first model to the second model.
Further, the outdoor temperature and the outdoor humidity in the step five are utilized to determine an outdoor enthalpy value, and a scatter diagram of the enthalpy value and the load predicted by the second model is constructed. Specifically, when the enthalpy value is determined by outdoor temperature and humidity, the program is realized in the programming software by self-modification by using the existing program, and the program can determine the rest air parameters by any two values of the air parameters. Outdoor temperature and outdoor humidity are used to determine the outdoor enthalpy. And constructing a scatter diagram of the outdoor enthalpy value and the load predicted by the second model by using numerical analysis software.
Further, the third step is to analyze the scatter diagram in the fourth step through numerical analysis software, select the best fit equation from 10 functions of linearity, logarithm, quadratic, cubic, compound and the like, and predict the load through the enthalpy value by using the best fit equation. Specifically, the scatter diagram in the step six is fitted by using 10 functions of linearity, logarithm, quadratic, cubic, compounding and the like, the fitting goodness of the 10 functions is compared, and the best fitting equation is selected. And predicting the load through the enthalpy value by utilizing the optimal fitting equation.
Compared with the prior art, the technical scheme of the invention has the following beneficial effects: the load simulation software simulates the time-by-time load of the subway air conditioner in the air conditioner season, the load is compared with the historical actual running load to verify the accuracy of simulation data, a platform is built in the load simulation software, the platform can meet the functions, the influence parameters such as fresh air, outdoor temperature, humidity and human flow which cause the load are output at the same time of outputting the load, the simulated data are trained through the BP neural network algorithm to obtain a first model, and the trained model is verified and tested. Extracting outdoor temperature, outdoor humidity and load. Training the extracted data by using a neural network algorithm to obtain a second model. And constructing a relation between the enthalpy and the load by using the relation between the outdoor temperature, the outdoor humidity and the load, expressing the relation between the enthalpy and the load by using a scatter diagram, fitting the scatter diagram by using a plurality of functions such as linearity, secondary, tertiary, compound and the like, and selecting an equation with the optimal fitting precision. R of the equation 2 When the enthalpy value reaches 0.989, the relation between the enthalpy value and the load can be almost expressed by the equation, and the future load can be directly predicted by the enthalpy value by using the equation. Based on the equation, the future load is predicted, the variety of input parameters is reduced, and the problem that the input parameters of the original prediction scheme are difficult to acquire is solved to a certain extent. The applicability of load prediction in actual engineering is improved.
Drawings
FIG. 1 is a flow chart of a subway air conditioner load prediction method based on an air enthalpy value, which is related to the invention;
fig. 2 is a graph of output platforms of parameters of a subway air conditioner load prediction method based on air enthalpy values;
fig. 3 is a flowchart of a BP neural network algorithm of a subway air conditioner load prediction method based on an air enthalpy value according to the present invention;
fig. 4 is a comparison chart of goodness-of-fit of each model of a subway air conditioner load prediction method based on air enthalpy value;
Detailed Description
The invention is described below in conjunction with the drawings in the specification, it being understood that the preferred embodiments described herein are provided to illustrate and explain the invention and are not intended to limit the invention. Various changes and modifications to the present invention may be made by one skilled in the art after reading the teachings of the present invention, and such equivalents fall within the scope of the present application as defined in the appended claims.
A subway air conditioner load prediction method based on air enthalpy value is shown in figure 1, and the specific implementation method comprises the following steps:
step one: and simulating the time-by-time load of the subway air conditioner in the air conditioning season through load simulation software, and comparing the load with the historical actual running load to verify the accuracy of simulation data. According to the embodiment, the time-by-time load of the subway air conditioner in an air conditioning season is simulated by the load simulation software platform, and the specific steps include building a physical model of a subway station in modeling software, and after the model is built, importing the built model into energy consumption simulation software for setting parameters of the subway station, wherein the parameters comprise the flow rate of people, the outdoor temperature, the outdoor humidity, the fresh air quantity, equipment heat dissipation, illumination heat dissipation, wall materials, wall thickness, soil temperature at the central height position of the subway station and the like. Wherein the fresh air quantity is selected to meet the requirement that the people average is not less than 20m 3 The requirement of/h cannot be lower than 10% of the total indoor air supply. After setting the above parameters indoors, the proper simulation step length and simulation time are finally selected. And the final simulation step length is 1h, the simulation time is 5-11 months, the time for closing the equipment at night is deleted, and the data 3600 groups are finally simulated. The resulting data is compared to the historical actual operating load to verify the accuracy of the simulated data.
Step two: by building the platform in the load simulation software, the platform can meet the functions of outputting the load and outputting factors with strong correlation such as fresh air, outdoor temperature, humidity, people flow, equipment heat dissipation, illumination heat dissipation and the like which cause the load. In this embodiment, a specific platform is shown in fig. 2. In the construction process of the platform, parameters needing to be output, namely, the flow of people, the outdoor temperature, the outdoor humidity, the fresh air quantity, the heat dissipation of equipment and the illumination heat dissipation are selected from the building module. And then, the selected output parameters are given to an intermediate module for unifying the formats, and finally, the unified formats are transmitted to an output module for visual output of the parameters.
Step three: the simulated data is divided into a training set, a verification set and a test set, and the ratio is 7:2:1. In this embodiment, the simulated data is deleted to obtain 3600 groups of data when the air conditioner is not turned on at night, the data volume is moderate, the data set is randomly divided into three parts, namely a training set, a verification set and a test set, the training set is used for training a model, the verification set is used for model selection, and the test set is used for final assessment of a learning method.
Step four: the simulated data is trained through the BP neural network algorithm to obtain a first model, and the trained model is verified and tested. In this embodiment, first, by calling xlsread function in matlab, subway air conditioner load data is extracted from the load file in the first step. Then, the data is preprocessed, normalization is used to map the data distribution to a certain interval in order to reduce the influence of the change of each data distribution, and the data is normalized to [0,1] by using max-min normalization in the embodiment]Between them. Then, determining the neural network structure and establishing a model, and determining 6 input parameters of the neural network in the second step, wherein in the embodiment, the number of input layers of the neural network is 6, and the output layers are the prediction result of the load of the air conditioning system of the subway station, so that the number of output layers of the neural network is 1, and the hidden layers of the neural network can pass through(J-hidden layer node number I-input layer node number K-output layer node number a E [1, 10)]) Determining that the number of hidden layer nodes is 3-13 through calculation, so that the optimal mode is screened out by setting different hidden layer nodes and comparing the error modesThe number of the hidden layers is 10. The iteration number of the specified model is 1000, the learning rate is 0.01, the training minimum error is 1 multiplied by 10 -5 . And then training the neural network, determining the structure and the model of the neural network, and training the neural network by using the normalized data to obtain a first model. And finally, predicting the load by using the first model, and performing inverse normalization processing on the predicted load and each input parameter to restore each data to the original order of magnitude. The specific flow of predicting the load through the neural network is shown in figure 3.
Step five: the outdoor temperature, outdoor humidity, and load predicted by the first model are extracted. And (3) repeating the processes of s3 and s4 to obtain a second model, changing the outdoor temperature and the outdoor humidity, predicting the load by using the second model, and comparing the load with the load predicted by the first model to verify the accuracy of the second model. In this embodiment, the third step and the fourth step are repeated by using the load predicted by the first model and the outdoor temperature and the outdoor humidity, so as to obtain the second model. The load is predicted by the second model using the outdoor temperature and the outdoor humidity as input parameters. In this embodiment, the accuracy of the second model is verified by fitting the load predicted by the first model and the load predicted by the second model by using the regression analysis module of excel.
Step six: and fifthly, determining the outdoor enthalpy value by using the outdoor temperature and the outdoor humidity, and constructing a scatter diagram of the enthalpy value and the load predicted by the second model. In this embodiment, when the enthalpy value is determined by the outdoor temperature and the outdoor humidity, the program is implemented in the programming software by modifying the program by itself, and the program can determine the remaining air parameters by using any two values of the air parameters. Outdoor temperature and outdoor humidity are used to determine the outdoor enthalpy. And constructing a scatter diagram of the outdoor enthalpy value and the load predicted by the second model by using numerical analysis software.
Step seven: analyzing the scatter diagram in the step six by numerical analysis software, selecting the best fitting equation from 10 functions of linearity, logarithm, quadratic, cubic, compounding and the like, and predicting the load directly by the enthalpy value by using the best fitting equation. In this embodiment, the scatter diagram in the sixth step is fitted by using 10 functions of linearity, logarithm, quadratic, cubic, compounding, etc. respectively, the fitting result is shown in fig. 4, and the best fitting equation is selected by comparison. And predicting the load directly through the enthalpy value by utilizing the optimal fitting equation.
The goodness of fit of the predicted load of this example is shown in FIG. 4, by comparison, R of the optimal equation 2 When the enthalpy value reaches 0.989, the relation between the enthalpy value and the load can be almost expressed by the equation, and the future load can be directly predicted by the enthalpy value by using the equation. Based on the equation, the future load is predicted, the variety of input parameters is reduced, and the problem that the input parameters of the original prediction scheme are difficult to acquire is solved to a certain extent. The applicability of load prediction in actual engineering is improved.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the specific embodiments described above, and that the above specific embodiments and descriptions are provided for further illustration of the principles of the present invention, and that various changes and modifications may be made therein without departing from the spirit and scope of the invention as defined in the appended claims. The scope of the invention is defined by the claims and their equivalents.
Claims (8)
1. The subway air conditioner load prediction method based on the air enthalpy value is characterized by comprising the following steps of:
step one: simulating the time-by-time load of the subway air conditioner in an air conditioning season through load simulation software, and comparing the load with the historical actual operation load to verify the accuracy of simulation data;
step two: by building a platform in the load simulation software, the platform can meet the functions of outputting the load and outputting factors with strong correlation such as time-by-time fresh air, outdoor temperature, humidity, people flow, equipment heat dissipation and the like which cause the load;
step three: dividing the simulated data into a training set, a verification set and a test set, wherein the ratio of the training set to the verification set to the test set is 7:2:1;
step four: training the simulated data through a BP neural network algorithm to obtain a first model, and verifying and testing the trained model;
step five: the outdoor temperature, outdoor humidity, and load predicted by the first model are extracted. Repeating the third step and the fourth step, wherein the process obtains a second model, changes the outdoor temperature and the outdoor humidity, predicts the load by using the second model, and compares the load with the load predicted by the first model to verify the accuracy of the second model;
step six: determining an outdoor enthalpy value by using the outdoor temperature and the outdoor humidity in the fifth step, and constructing a scatter diagram of the enthalpy value and the load predicted by the second model;
step seven: and C, analyzing the scatter diagram in the step six by using statistical analysis software, selecting a best fit equation from 10 functions of linearity, logarithm, quadratic, cubic, compounding and the like, and predicting the load by using the enthalpy value by using the best fit equation.
2. The subway air conditioner load prediction method based on the air enthalpy value according to claim 1, wherein:
the load is output, a series of factors causing the load are also output, so that the load is more convenient to observe, a large amount of data required by later machine learning can be obtained, and the problem that a large amount of data required by machine learning is difficult to obtain is solved to a certain extent.
3. The subway air conditioner load prediction method based on the air enthalpy value according to claim 1, wherein:
the training set and the testing set of the first model and the second model are 7:2:1, and the simulation step sizes are the same and are all 1h.
4. The subway air conditioner load prediction method based on the air enthalpy value according to claim 1, wherein:
when the outdoor temperature and outdoor humidity are utilized to determine the outdoor enthalpy value, the existing program is utilized to be modified in programming software, and the program can determine other air parameters through any two values of the air parameters, so that the method is more convenient and more accurate than the traditional method for calculating the enthalpy value.
5. The subway air conditioner load prediction method based on the air enthalpy value according to claim 1, wherein:
and constructing an outdoor temperature, an outdoor humidity and a load relationship on the basis of the first model to obtain a second model.
6. The subway air conditioner load prediction method based on the air enthalpy value according to claim 1, wherein:
the second model can directly utilize the outdoor temperature and the outdoor humidity to predict the load, and input parameters are reduced, so that the calculated amount of air conditioner load prediction is reduced, and the calculation time is shortened.
7. The subway air conditioner load prediction method based on the air enthalpy value according to claim 1, wherein:
in analyzing the relationship between enthalpy and load, multiple mathematical models are created by comparing different mathematical models R 2 A best fit model is selected.
8. The subway air conditioner load prediction method based on the air enthalpy value according to claim 1, wherein:
selected best fit model, R 2 The maximum value reaches 0.989, and the error is small.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311754298.0A CN117760063A (en) | 2023-12-19 | 2023-12-19 | Subway air conditioner load prediction method based on air enthalpy value |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311754298.0A CN117760063A (en) | 2023-12-19 | 2023-12-19 | Subway air conditioner load prediction method based on air enthalpy value |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117760063A true CN117760063A (en) | 2024-03-26 |
Family
ID=90313848
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311754298.0A Pending CN117760063A (en) | 2023-12-19 | 2023-12-19 | Subway air conditioner load prediction method based on air enthalpy value |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117760063A (en) |
-
2023
- 2023-12-19 CN CN202311754298.0A patent/CN117760063A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110046743A (en) | Energy Consumption of Public Buildings prediction technique and system based on GA-ANN | |
| Yalcintas | An energy benchmarking model based on artificial neural network method with a case example for tropical climates | |
| Luo | A novel clustering-enhanced adaptive artificial neural network model for predicting day-ahead building cooling demand | |
| KR20180110940A (en) | Energy preformance profiling system for existing building | |
| CN113983646A (en) | Air conditioner interaction end energy consumption prediction method based on generation countermeasure network and air conditioner | |
| CN111723085B (en) | Method and system for cleaning electric energy consumption data of public building air conditioner | |
| Tian et al. | Calibrating dynamic building energy models using regression model and Bayesian analysis in building retrofit projects | |
| CN114764521A (en) | Garden building shape optimization method and system based on genetic algorithm | |
| CN116205425A (en) | Low-carbon park cold-hot electric load prediction method based on typical database | |
| Chen et al. | AI-Assisted approach for building energy and carbon footprint modeling | |
| CN116796168A (en) | CNN-BiLSTM high-altitude multi-factor power transmission line audible noise prediction method based on multi-head attention mechanism | |
| Shen et al. | Exploring potential for residential energy saving in New York using developed lightweight prototypical building models based on survey data in the past decades | |
| CN111366291A (en) | Water supply pipeline pressure prediction and roughness calibration method based on entropy and artificial neural network | |
| Chen et al. | A hybrid-model forecasting framework for reducing the building energy performance gap | |
| CN117469603B (en) | Multi-water-plant water supply system pressure optimal control method based on big data learning | |
| Liu et al. | Building information modelling-enabled multi-objective optimization for energy consumption parametric analysis in green buildings design using hybrid machine learning algorithms | |
| CN110716998A (en) | Method for spatializing fine-scale population data | |
| Yu et al. | A bottom-up approach for community load prediction based on multi-agent model | |
| Lim et al. | Estimating unknown parameters of a building stock using a stochastic-deterministic-coupled approach | |
| CN108647817A (en) | A kind of prediction technique and forecasting system of energy consumption load | |
| CN117760063A (en) | Subway air conditioner load prediction method based on air enthalpy value | |
| CN115713219A (en) | Method, system, device and medium for optimizing zero-carbon operation building scheme | |
| Wang | Application of deep learning model in building energy consumption prediction | |
| CN114200839A (en) | Office building energy consumption intelligent control model for dynamic monitoring of coupled environmental behaviors | |
| CN112668784A (en) | Regional macro economy prediction model and method based on big data |
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 |