Energy efficiency evaluation method for station-city integrated three-dimensional network space

Abstract

The application discloses a station-city integrated three-dimensional network space energy efficiency evaluation method, and belongs to the technical field of energy efficiency evaluation based on dynamic modeling. The application solves the problem of low accuracy of evaluation results caused by the traditional energy efficiency evaluation method based on air conditioner operation mechanism in the prior art, which does not consider the influence of different space energy efficiency influencing factors on air conditioner body energy efficiency; the application establishes a station-city integrated three-dimensional network space information model to obtain space information; dynamic monitoring data related to air conditioner energy efficiency is acquired; the space information and the dynamic monitoring data are registered to obtain registered air conditioner energy efficiency influencing factor data and air conditioner energy consumption calculation index data; iterative training is performed by using a data set and a loss function to obtain a trained station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on a KAN model; and air conditioner energy efficiency evaluation based on energy efficiency grades is realized through hierarchical regulations. The application reduces the time consumption of energy efficiency calculation and can be applied to energy efficiency evaluation under a station-city integrated space.

Description

TECHNICAL FIELD

[0001] The present application relates to a station-city integration three-dimensional network space energy efficiency evaluation method, belonging to the technical field of energy efficiency evaluation based on dynamic modeling. BACKGROUND

[0002] Station-city integration is also called "station-city integration". Macroscopically, it refers to the relationship between railway and urban rail transit stations and the surrounding city in terms of functional space, traffic mechanism and other factors. On the project level, station-city integration refers to taking measures to realize the coordinated development of the hub and its surrounding area by combining with the city's own conditions and relying on the spillover effect of railway and rail development. Station-city integration advocates the close integration of important transportation sites and urban space, that is, arranging daily office, residence and city service functions within the walking distance of the station to provide convenient living conditions and economic activity conditions for citizens.

[0003] The station-city integration three-dimensional network space covers the traditional comprehensive passenger hub space structure and various types of space areas connected with the city. As the development direction of future comprehensive transportation hubs, the performance evaluation of its green operation is increasingly prominent. How to quickly and accurately evaluate the energy efficiency based on the operation mechanism of air conditioning has also put forward higher requirements. In the prior art one, the patent document with publication (announcement) number CN113175733B discloses a method for calculating the energy efficiency of an air conditioner, an air conditioner and a storage medium, but it does not consider the influence of vehicle flow, pedestrian flow, space structure and environmental changes in the actual engineering application process on the energy efficiency of the air conditioner. In the prior art two, the patent document with publication (announcement) number CN110059801B discloses an air conditioner energy efficiency control method based on a neural network, but it does not consider the influence of different space energy efficiency influencing factors on the air conditioning system energy efficiency during the operation of the central air conditioner in different spaces, and cannot determine the weight of different influencing factors.

[0004] In view of the above, there is a need for a station-city integration three-dimensional network space energy efficiency evaluation method. SUMMARY

[0005] In the following, a brief summary of the present application is given in order to provide a basic understanding of some aspects of the present application. It should be understood that this summary is not an exhaustive overview of the present application. It is not intended to identify key or important parts of the present application nor is it intended to limit the scope of the present application. Its purpose is merely to present some concepts in a simplified form as a prelude to a more detailed description to be discussed later.

[0006] In view of this, in order to solve the problem of low accuracy of the evaluation result caused by the traditional energy efficiency evaluation method based on the operation mechanism of the air conditioner in the prior art, which does not consider the influence of different space energy efficiency factors on the energy efficiency of the air conditioning body, the present application provides an energy efficiency evaluation method for a station-city integrated three-dimensional network space.

[0007] The technical scheme is as follows: an energy efficiency evaluation method for a station-city integrated three-dimensional network space, comprising the following steps:

[0008] S1. Based on the energy space area of the air conditioner, a station-city integrated three-dimensional network space information model is established to obtain space information.

[0009] S2. According to the station-city integrated three-dimensional network space information model and the collection system arranged by the energy space area of the air conditioner, dynamic monitoring data related to the energy efficiency of the air conditioner is obtained.

[0010] S3. The space information and the dynamic monitoring data are registered to determine the relationship between the space information and the dynamic monitoring data, and the registered air conditioner energy efficiency influencing factor data and air conditioner energy consumption calculation index data are obtained.

[0011] S4. A data set is constructed through the registered air conditioner energy efficiency influencing factor data and the air conditioner energy consumption calculation index data, and the data set and the established loss function are used to iteratively train the deep learning neural network structure based on the KAN model to obtain a trained station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model.

[0012] S5. The trained station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model is used to output the air conditioner energy consumption, the refrigeration capacity and the space influence weight, and the central air conditioning body energy efficiency is determined through hierarchical regulation to realize the air conditioner energy efficiency evaluation based on the energy efficiency grade.

[0013] Further, in S1, the air conditioning energy space area of the station-city integrated three-dimensional network core is determined, which includes the waiting room, the subway platform, the transfer channel, the transfer functional space and the centralized commercial space.

[0014] The station-city integrated three-dimensional network space information model is established according to the structure and power design construction drawing of the air conditioning energy space area, which includes the space information of the space size, the space ventilation port, the air conditioner inlet and outlet port and the air conditioner inlet and outlet position in the air conditioning energy space area of the station-city integrated three-dimensional network core.

[0015] Further, in S2, the following steps are included:

[0016] S21. Taking the air conditioner energy efficiency influencing factor data as the input parameter of the energy efficiency dynamic evaluation, arranging the front-end perception sensor in the air conditioner energy space area of the station-city integrated three-dimensional network core according to the station-city integrated three-dimensional network space information model, combining the front-end perception sensor, the collection instrument and the industrial computer to establish the energy efficiency influencing factor data collection system, and collecting the dynamic air conditioner energy efficiency influencing factor data;

[0017] S22. Taking the air conditioner energy consumption calculation index data as the air conditioner system energy consumption parameter and the refrigerating capacity parameter of the energy efficiency dynamic evaluation, monitoring and obtaining the air conditioner energy consumption data of different spaces of the station-city according to a reasonable point arrangement scheme by using the sensor in the air conditioner room corresponding to the air conditioner energy space area of the station-city integrated three-dimensional network core, combining the sensor, the collection instrument and the industrial computer to establish the energy consumption calculation index data collection system, and collecting the dynamic air conditioner energy consumption calculation index data.

[0018] Further, in the S3, the following steps are specifically included:

[0019] S31. Matching the space information with the air conditioner energy efficiency influencing factor data and the air conditioner energy consumption calculation index data respectively to obtain matching information;

[0020] In the S31, the air conditioner energy efficiency influencing factor data is matched with the information label in the station-city integrated three-dimensional network space information model according to the space area where the sensor corresponding to the data collection process is located;

[0021] The air conditioner energy consumption calculation index data is matched with the space information label in the station-city integrated three-dimensional network space information model according to the air conditioner action space area;

[0022] S32. Labeling the dynamic monitoring data according to the matching information, adding a space information data field to all data in the process of collecting all dynamic monitoring data according to the matched space information, and obtaining the registered air conditioner energy efficiency influencing factor data and the air conditioner energy consumption calculation index data.

[0023] Further, in the S4, the following steps are included:

[0024] S41. For the spatio-temporal heterogeneity problem of multi-source data, a hierarchical spatio-temporal encoder based on a reference grid is constructed, time axis normalization and spatial interpolation methods are adopted to unify the detection data of different sampling frequencies and spatial resolutions to a common spatio-temporal coordinate system, and an input data set and an output data set are constructed, which respectively contain a training data set and a test data set;

[0025] In the S41, the registered air conditioner energy efficiency influencing factor data is used to construct the input data set;

[0026] The registered air conditioner energy consumption calculation index data is used to construct the output data set;

[0027] S42. Initialize the model parameters, that is, define the deep learning neural network structure based on the KAN model according to the number of input variables, the number of output variables, including the number of input neurons, the number of output neurons, the number of neural network layers, and the number of neurons in the intermediate layer;

[0028] S43. Establish a loss function of the deep learning neural network structure based on the KAN model according to the energy consumption and the refrigerating capacity;

[0029] Loss function is expressed as:

[0030] ;

[0031] wherein, is the energy consumption weight, is the energy consumption loss value, is the refrigerating capacity weight, is the refrigerating capacity loss value;

[0032] S44. According to the loss function , using the quasi-Newton method in the second-order optimization method, define an iterative optimization function;

[0033] S45. Input the input data set into the deep learning neural network structure based on the KAN model for iterative training, and obtain the trained station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model through evaluation;

[0034] In the S45, the input data set is input into the deep learning neural network structure based on the KAN model, and the deep learning neural network structure based on the KAN model is adaptively iteratively trained according to the iterative optimization function, so that the loss value reaches below the allowable value, and the allowable value is set to 0.01. After reaching the allowable value, the training is automatically stopped, and the station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model is output;

[0035] The test set data is input into the station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model trained to evaluate the accuracy, and when the evaluation progress reaches 99%, it is confirmed that the station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model is successfully trained.

[0036] Further, in the S5, the following steps are included:

[0037] S51. Obtain real-time station-city integrated three-dimensional network different space flow, environment, and equipment monitoring data;

[0038] S52. Input the real-time monitoring data obtained in step S51 into the KAN model-based station-city fusion air conditioner energy efficiency operation dynamic evaluation model trained, and output air conditioner energy consumption, refrigerating capacity and space influence weight;

[0039] S53. Determine the air conditioner energy efficiency based on the air conditioner energy efficiency calculation formula, refrigerating capacity and air conditioner energy consumption through hierarchical provisions;

[0040] In the S53, the air conditioner energy efficiency calculation formula is represented as:

[0041] ;

[0042] Among them, is the air conditioner energy efficiency ratio of the jth space, is the refrigerating capacity of the jth space, is the air conditioner energy consumption of the jth space; the energy efficiency is judged by the hierarchical provisions in the public material of Shenzhen Public Building Central Air Conditioning System Energy Efficiency Evaluation Standard, when ≥6, the space air conditioner energy efficiency is first class, when 6.0> ≥5.5, the space air conditioner energy efficiency is second class, when 5.5> ≥5.0, the space air conditioner energy efficiency is third class, when 5.0> ≥4.0, the space air conditioner energy efficiency meets the limit value, when 4.0> , the space air conditioner energy efficiency does not meet the limit value;

[0043] S54. Determine the central air conditioning system energy efficiency based on the overall air conditioner energy efficiency calculation formula and real-time station-city fusion three-dimensional network different space people flow, environment, equipment monitoring data through hierarchical provisions;

[0044] In the S54, the overall air conditioner energy efficiency calculation formula is represented as:

[0045] ;

[0046] ;

[0047] Among them, is the central air conditioning system energy efficiency ratio, is the influence weight of the jth space, is the influence weight of the people flow under the jth space, is the people flow parameter data, is the influence weight of the environment under the jth space, is the environment parameter data, is the influence weight of the space under the jth space, is the space parameter data, is the influence weight of the device performance in the jth space, D is the device performance parameter data, the energy efficiency is classified according to the public building central air conditioning system energy efficiency evaluation standard in the public material, when ≥6, the central air conditioning body energy efficiency is first class, when 6.0 ≥5.5, the central air conditioning body energy efficiency is second class, when 5.5 ≥5.0, the central air conditioning body energy efficiency is third class, when 5.0 ≥4.0, the central air conditioning body energy efficiency meets the limit value, when 4.0 , the central air conditioning body energy efficiency does not meet the limit value.

[0048] The beneficial effects of the present application are as follows: the present application provides a station-city integrated three-dimensional network space air conditioner energy efficiency operation state dynamic evaluation method, compared with the traditional method, the present application solves the air conditioner energy efficiency evaluation under the action of different energy efficiency influencing factors in different functional spaces in the station-city integrated complex three-dimensional network space environment, and the weight of different energy efficiency influencing factors can be determined; through the construction of the station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model, the present application realizes the local and overall precise evaluation of the air conditioner energy efficiency under the operation of the station-city integrated three-dimensional network space, and the station-city integrated air conditioner energy efficiency operation dynamic evaluation model based on the KAN model realizes the local and overall dynamic real-time evaluation of the air conditioner energy efficiency under the operation of the station-city integrated three-dimensional network space, and the present application can solve the problems of time-consuming and low simulation efficiency in general air conditioner energy efficiency calculation, and realizes real-time evaluation. BRIEF DESCRIPTION OF DRAWINGS

[0049] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application, the schematic embodiments of the present application and the description thereof are used to explain the present application, and do not constitute improper limitation on the present application. In the drawings:

[0050] Figure 1 It is the flow chart of the energy efficiency evaluation method of the station-city integrated three-dimensional network space. DETAILED DESCRIPTION

[0051] In order to make the technical solutions and advantages in the embodiments of the present application more clear and explicit, the exemplary embodiments of the present application are further described in detail below with reference to the drawings, obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments of the present application. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0052] Reference Figure 1 The energy efficiency evaluation method of the station-city integrated three-dimensional network space is described in detail, which specifically includes the following steps:

[0053] S1. Based on the air-conditioning energy space area, a station-city integrated three-dimensional network space information model is established, and space information is obtained;

[0054] S2. According to the station-city integrated three-dimensional network space information model and the collection system arranged according to the air-conditioning energy space area, dynamic monitoring data related to air-conditioning energy efficiency is obtained;

[0055] S3. The space information and the dynamic monitoring data are registered, the relationship between the space information and the dynamic monitoring data is determined, and air-conditioning energy efficiency influencing factor data and air-conditioning energy consumption calculation index data after registration are obtained;

[0056] S4. A data set is constructed through the air-conditioning energy efficiency influencing factor data and the air-conditioning energy consumption calculation index data after registration, the data set and the established loss function are used to iteratively train a deep learning neural network structure based on a KAN model, and a trained station-city integrated air-conditioning energy efficiency operation dynamic evaluation model based on the KAN model is obtained;

[0057] S5. Through the trained station-city integrated air-conditioning energy efficiency operation dynamic evaluation model based on the KAN model, air-conditioning energy consumption, refrigerating capacity and space influence weight are output, central air conditioning system energy efficiency is determined through hierarchical regulations, and air-conditioning energy efficiency evaluation based on energy efficiency grade is realized.

[0058] Further, in S1, the air-conditioning energy space area of the station-city integrated three-dimensional network core is determined, which includes waiting rooms, subway platforms, transfer channels, transfer functional spaces and concentrated commercial spaces, etc.

[0059] The station-city integrated three-dimensional network space information model is established according to the structure of the air-conditioning energy space area and the power design construction drawing, which includes space information such as space size, space ventilation port, air-conditioning inlet and outlet port, air-conditioning inlet and outlet position in the air-conditioning energy space area of the station-city integrated three-dimensional network core.

[0060] Further, in S2, the following steps are included:

[0061] S21. The air-conditioning energy efficiency influencing factor data is used as an input parameter for energy efficiency dynamic evaluation, front-end perception sensors including cameras, thermometers, etc. are arranged in the air-conditioning energy space area of the station-city integrated three-dimensional network core according to the station-city integrated three-dimensional network space information model, an energy efficiency influencing factor data collection system is established by combining the front-end perception sensors, a collection instrument and an industrial computer, and dynamic air-conditioning energy efficiency influencing factor data is collected;

[0062] S22. The air conditioning energy consumption calculation index data is used as the air conditioning system energy consumption parameter and the refrigerating capacity parameter of the energy efficiency dynamic evaluation. According to the station-city integrated three-dimensional network space information model, the air conditioning energy consumption data of different spaces of the station is monitored and obtained by using temperature and humidity, flow meter, electric meter and other sensors according to a reasonable distribution scheme. The energy consumption calculation index data acquisition system is established by combining the sensors, the acquisition instrument and the industrial computer to acquire the dynamic air conditioning energy consumption calculation index data.

[0063] Specifically, with reference to Table 1, different influencing factors are determined by using different acquisition methods, sampling frequencies and data thresholds to determine the key influencing factors of the air conditioning energy efficiency in the station-city integrated space.

[0064] Table 1

[0065]

[0066] With reference to Table 2, the key calculation indexes of the air conditioning energy consumption in the station-city integrated space are determined.

[0067] Table 2

[0068]

[0069] Further, the S3 specifically includes the following steps:

[0070] S31. The space information, the air conditioning energy efficiency influencing factor data and the air conditioning energy consumption calculation index data are matched respectively to obtain matching information.

[0071] In the S31, the air conditioning energy efficiency influencing factor data is matched with the information label in the station-city integrated three-dimensional network space information model according to the space region where the sensor corresponding to the data acquisition process is located.

[0072] The air conditioning energy consumption calculation index data is matched with the space information label in the station-city integrated three-dimensional network space information model according to the space region where the air conditioner is located.

[0073] S32. The dynamic monitoring data is labeled according to the matching information. A space information data field is added to all the data according to the matched space information in the process of collecting all the dynamic monitoring data to obtain the registered air conditioning energy efficiency influencing factor data and the air conditioning energy consumption calculation index data.

[0074] Further, the S4 includes the following steps:

[0075] S41. In view of the spatio-temporal heterogeneity of multi-source data, a hierarchical spatio-temporal encoder based on a benchmark grid is constructed, time axis normalization and spatial interpolation methods are used to unify detection data of different sampling frequencies and spatial resolutions to a common spatio-temporal coordinate system, and input data set and output data set are constructed, which respectively contain training data set and test data set;

[0076] In the S41, the input data set is constructed by using the registered air conditioner energy efficiency influencing factor data;

[0077] The output data set is constructed by using the registered air conditioner energy consumption calculation index data;

[0078] S42. Initialize model parameters, that is, define the deep learning neural network structure based on KAN model according to the number of input variables and the number of output variables, which includes the number of input neurons (which depends on the number of collected air conditioner energy efficiency influencing factors, generally 2-10), the number of output neurons (which depends on the number of collected air conditioner energy efficiency influencing factors, generally 2-10), the number of neural network layers (generally 3-5 layers, including input and output layers), and the number of neurons in the middle layer (generally less than 50);

[0079] S43. Establish the loss function of the deep learning neural network structure based on KAN model according to the energy consumption and the refrigerating capacity;

[0080] Loss function is expressed as:

[0081] ;

[0082] Wherein, is the energy consumption weight, is the energy consumption loss value, is the refrigerating capacity weight, is the refrigerating capacity loss value;

[0083] The energy consumption loss value and the refrigerating capacity loss value are obtained by formula calculation based on the data loss function ;

[0084] The data loss function is expressed as:

[0085] ;

[0086] Wherein, is the true value, is the predicted value, is the sample number;

[0087] S44. According to the loss function , using a quasi-Newton method in a second-order optimization method, defining an iterative optimization function;

[0088] S45. Input the input data set into the KAN model-based deep learning neural network structure for iterative training, and obtain the trained KAN model-based station-city integrated air conditioner energy efficiency operation dynamic evaluation model through evaluation;

[0089] In the S45, the input data set is input into the KAN model-based deep learning neural network structure, and the KAN model-based deep learning neural network structure is adaptively iteratively trained according to the iterative optimization function, so that the loss value reaches below the allowable value, and the allowable value is set to 0.01. After reaching the allowable value, the training is automatically stopped, and the KAN model-based station-city integrated air conditioner energy efficiency operation dynamic evaluation model is output;

[0090] The test set data is input into the KAN model-based station-city integrated air conditioner energy efficiency operation dynamic evaluation model trained, and the accuracy is evaluated. When the evaluation progress reaches 99%, it is confirmed that the KAN model-based station-city integrated air conditioner energy efficiency operation dynamic evaluation model is successfully trained.

[0091] Further, in the S5, the following steps are included:

[0092] S51. Obtain real-time station-city integrated three-dimensional network different space people flow, environment, and equipment monitoring data;

[0093] S52. Input the real-time monitoring data obtained in step S51 into the KAN model-based station-city integrated air conditioner energy efficiency operation dynamic evaluation model trained, and output air conditioner energy consumption, refrigerating capacity, and space influence weight data;

[0094] S53. Based on the air conditioner energy efficiency calculation formula, the refrigerating capacity, and the air conditioner energy consumption, the air conditioner energy efficiency is obtained through hierarchical regulation;

[0095] In the S53, the air conditioner energy efficiency calculation formula is represented as:

[0096] ;

[0097] Wherein, is the air conditioner energy efficiency ratio of the jth space, is the refrigerating capacity of the jth space, is the air conditioner energy consumption of the jth space; the hierarchical regulation of the energy efficiency is judged by using the public material of Shenzhen public building central air conditioning system energy efficiency evaluation standard, when ≥6, the space air conditioner energy efficiency is first class, when 6.0> ≥5.5, the space air conditioner energy efficiency is second class, when 5.5> When the energy efficiency ratio is ≥5.0, the space air conditioner is rated as Level 3; when 5.0 > When the energy efficiency ratio is ≥4.0, the space air conditioning meets the limit; when 4.0 > At that time, the energy efficiency of the space air conditioner did not meet the limit;

[0098] S54. Based on the overall air conditioning energy efficiency calculation formula and real-time monitoring data of people flow, environment and equipment in different spaces of the station-city integrated three-dimensional network, the overall energy efficiency of the central air conditioning is determined through classification regulations;

[0099] In S54, the overall air conditioning energy efficiency calculation formula is expressed as follows:

[0100] ;

[0101] ;

[0102] in, The overall energy efficiency ratio of the central air conditioning system. The influence weight of the j-th space, Let the influence weight of the flow of people in the j-th space be denoted as . For human flow parameter data, The influence weight of the environment in the j-th space is... For environmental parameter data, The influence weight of the space under the j-th space, For spatial parameter data, Let be the influence weight of equipment performance in the j-th space, and D be the equipment performance parameter data. The energy efficiency classification is adopted according to the Shenzhen Municipal Public Building Central Air Conditioning System Energy Efficiency Evaluation Standard published in public materials. When... When the energy efficiency rating is ≥6, the overall energy efficiency of the central air conditioning system is Level 1; when 6.0 > When the energy efficiency ratio is ≥5.5, the overall energy efficiency rating of the central air conditioning system is Level 2; when 5.5 > When the energy efficiency ratio is ≥5.0, the overall energy efficiency rating of the central air conditioning system is Level 3; when 5.0 > When the energy efficiency ratio is ≥4.0, the overall energy efficiency of the central air conditioning system meets the limit; when 4.0 > At that time, the overall energy efficiency of the central air conditioning system did not meet the limit.

[0103] While the application has been described in accordance with the various embodiments shown and described, it is to be understood that the application is not limited to those precise embodiments, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the present application. It is intended that the scope of the application should only be limited as recited in the appended claims.

Claims

1. An energy efficiency assessment method for a station-city integrated three-dimensional network space, characterized in that, Includes the following steps: S1. Based on the spatial area of ​​air conditioning energy consumption, establish a three-dimensional network spatial information model of station-city integration to obtain spatial information; S2. Based on the station-city integrated three-dimensional network spatial information model and the data acquisition system deployed in the spatial area of ​​air conditioning energy consumption, obtain dynamic monitoring data related to air conditioning energy efficiency; S3. Register spatial information and dynamic monitoring data, determine the relationship between spatial information and dynamic monitoring data, and obtain the registered data on air conditioning energy efficiency influencing factors and air conditioning energy consumption calculation index data; S4. A dataset is constructed using the registered data of factors affecting air conditioning energy efficiency and the data of air conditioning energy consumption calculation indicators. The dataset and the established loss function are used to iteratively train the deep learning neural network structure based on the KAN model to obtain the trained dynamic evaluation model of station-city integrated air conditioning energy efficiency operation based on the KAN model. S5. Through the trained KAN-based station-city integrated air conditioning energy efficiency operation dynamic evaluation model, the air conditioning energy consumption, cooling capacity and space impact weights are output. Through the classification regulations, the overall energy efficiency of the central air conditioning is judged, and the air conditioning energy efficiency evaluation based on the energy efficiency level is realized. S4 includes the following steps: S41. To address the spatiotemporal heterogeneity of multi-source data, a hierarchical spatiotemporal encoder based on a reference grid is constructed. By employing time axis normalization and spatial interpolation methods, detection data with different sampling frequencies and spatial resolutions are unified into a common spatiotemporal coordinate system. Input and output datasets are constructed, which respectively contain training and test datasets. In step S41, the input dataset is constructed using the registered data on factors affecting air conditioning energy efficiency. The output dataset is constructed using the registered air conditioning energy consumption calculation index data; S42. Initialize model parameters, that is, define the deep learning neural network structure based on the KAN model according to the requirements of the number of input variables and the number of output variables, including the number of input neurons, the number of output neurons, the number of neural network layers, and the number of neurons in the intermediate layers. S43. Establish a loss function for a deep learning neural network structure based on the KAN model, according to energy consumption and cooling capacity; loss function Represented as: ； in, As energy consumption weight, This represents the energy loss value. Weighted by cooling capacity, This represents the cooling capacity loss value. S44. Based on the loss function Using the quasi-Newton method in second-order optimization, an iterative optimization function is defined; S45. Input the input dataset into a deep learning neural network structure based on the KAN model for iterative training, and obtain the trained dynamic evaluation model of station-city integrated air conditioning energy efficiency operation based on the KAN model after evaluation. In step S45, the input dataset is input into the deep learning neural network structure based on the KAN model. The deep learning neural network structure based on the KAN model is adaptively iteratively trained according to the iterative optimization function until the loss value reaches below the allowable value. The allowable value is set to 0.

01. After the allowable value is reached, the training is automatically stopped, and the dynamic evaluation model of station-city integrated air conditioning energy efficiency operation based on the KAN model is output. The test set data is input into the trained dynamic evaluation model of station-city integrated air conditioning energy efficiency operation based on the KAN model, and the accuracy is evaluated. When the evaluation progress reaches 99%, the training of the dynamic evaluation model of station-city integrated air conditioning energy efficiency operation based on the KAN model is confirmed to be successful.

2. The energy efficiency assessment method for the integrated station-city three-dimensional network space according to claim 1, characterized in that, In S1, the air-conditioning energy-consuming space area of ​​the core of the station-city integrated three-dimensional network is determined, which includes waiting room, subway platform, transfer passage, transfer function space and centralized commercial space. Based on the structure and power design and construction drawings of the air conditioning energy consumption space area, a spatial information model of the station-city integrated three-dimensional network is established. It includes spatial information such as the size of the space, the ventilation openings, the air conditioning inlet and outlet, and the location of the air conditioning inlet and outlet in the core air conditioning energy consumption space area of ​​the station-city integrated three-dimensional network.

3. The energy efficiency assessment method for the integrated station-city three-dimensional network space according to claim 2, characterized in that, S2 includes the following steps: S21. Using the data on factors affecting air conditioning energy efficiency as input parameters for dynamic energy efficiency assessment, and based on the spatial information model of the station-city integrated three-dimensional network, front-end sensing sensors are deployed in the air conditioning energy consumption space area of ​​the core of the station-city integrated three-dimensional network. A data acquisition system for energy efficiency factors is established by combining the front-end sensing sensors, data acquisition instruments, and industrial control computers to collect dynamic data on factors affecting air conditioning energy efficiency. S22. Using the air conditioning energy consumption calculation index data as the energy consumption parameters and cooling capacity parameters of the air conditioning system for dynamic energy efficiency evaluation, based on the station-city integrated three-dimensional network spatial information model, in the air conditioning energy consumption space corresponding to the core air conditioning energy consumption space area of ​​the station-city integrated three-dimensional network, sensors are used to monitor and obtain air conditioning energy consumption data of different spaces in the station-city according to a reasonable deployment plan. An energy consumption calculation index data acquisition system is established by combining sensors, data acquisition instruments, and industrial control computers to collect dynamic air conditioning energy consumption calculation index data.

4. The energy efficiency assessment method for the integrated station-city three-dimensional network space according to claim 3, characterized in that, S3 specifically includes the following steps: S31. Match spatial information with data on factors affecting air conditioning energy efficiency and data on air conditioning energy consumption calculation indicators to obtain matching information; In step S31, the data on air conditioning energy efficiency influencing factors are matched with the information labels in the station-city integrated three-dimensional network spatial information model based on the spatial area where the corresponding sensor is located during the data acquisition process. Based on the spatial area where air conditioning operates, the calculated air conditioning energy consumption index data is matched with the spatial information labels in the station-city integrated three-dimensional network spatial information model; S32. Label the dynamic monitoring data according to the matching information. During the collection of all dynamic monitoring data, add a spatial information data field to all data according to the matched spatial information to obtain the registered air conditioning energy efficiency influencing factor data and air conditioning energy consumption calculation index data.

5. The energy efficiency assessment method for the integrated station-city three-dimensional network space according to claim 4, characterized in that, S5 includes the following steps: S51. Obtain real-time monitoring data on pedestrian flow, environment, and equipment in different spaces of the integrated station-city network; S52. Input the real-time monitoring data obtained in step S51 into the trained KAN-based station-city integrated air conditioning energy efficiency operation dynamic evaluation model, and output the air conditioning energy consumption, cooling capacity and spatial impact weights. S53. Based on the air conditioner energy efficiency calculation formula, cooling capacity, and air conditioner energy consumption, the air conditioner energy efficiency is determined through classification regulations; In S53, the formula for calculating the energy efficiency of an air conditioner is expressed as follows: ； in, Let J be the air conditioning energy efficiency ratio of the j-th space. Let J be the cooling capacity of the j-th space. Let j be the air conditioning energy consumption of the j-th space; the energy efficiency is judged according to the Shenzhen Municipal Public Building Central Air Conditioning System Energy Efficiency Evaluation Standard in the publicly available materials. When the energy efficiency rating is ≥6, the space air conditioner is rated as Level 1. When 6.0 > When the energy efficiency ratio is ≥5.5, the space air conditioner is rated as Level 2. When 5.5 > When the energy efficiency ratio is ≥5.0, the space air conditioner is rated as Level 3; when 5.0 > When the energy efficiency ratio is ≥4.0, the space air conditioning meets the limit; when 4.0 > At that time, the energy efficiency of the space air conditioner did not meet the limit; S54. Based on the overall air conditioning energy efficiency calculation formula and real-time monitoring data of people flow, environment and equipment in different spaces of the station-city integrated three-dimensional network, the overall energy efficiency of the central air conditioning is determined through classification regulations; In S54, the overall air conditioning energy efficiency calculation formula is expressed as follows: ； ； in, The overall energy efficiency ratio of the central air conditioning system. The influence weight of the j-th space, Let the influence weight of the flow of people in the j-th space be denoted as . For human flow parameter data, The influence weight of the environment in the j-th space is... For environmental parameter data, The influence weight of the space under the j-th space, For spatial parameter data, Let be the influence weight of equipment performance in the j-th space, and D be the equipment performance parameter data. The energy efficiency classification is adopted according to the Shenzhen Municipal Public Building Central Air Conditioning System Energy Efficiency Evaluation Standard published in public materials. When... When the energy efficiency rating is ≥6, the overall energy efficiency of the central air conditioning system is Level 1; when 6.0 > When the energy efficiency ratio is ≥5.5, the overall energy efficiency rating of the central air conditioning system is Level 2; when 5.5 > When the energy efficiency ratio is ≥5.0, the overall energy efficiency rating of the central air conditioning system is Level 3; when 5.0 > When the energy efficiency ratio is ≥4.0, the overall energy efficiency of the central air conditioning system meets the limit; when 4.0 > At that time, the overall energy efficiency of the central air conditioning system did not meet the limit.

Images