Self-correcting subway station air conditioner energy consumption simulation system and method

Abstract

The invention relates to the field of metro station air conditioning system energy consumption prediction, and discloses a self-correcting metro station air conditioning energy consumption simulation system and method.The simulation system comprises a main cause analysis module and a self-correcting module, and the self-correcting module establishes a corresponding energy consumption simulation model and optimizes the energy consumption simulation model; the main cause analysis module comprises a data acquisition module and a data analysis module; the self-correction module comprises a model building module and an optimization module. The invention further discloses a simulation method based on the system, principal component analysis is performed on measured data of the subway station air conditioner system, an optimization algorithm is utilized to automatically correct an energy consumption simulation model, and the corrected energy consumption simulation model is applied to the subway station air conditioner control system. Therefore, the energy efficiency of the air-conditioning system of the subway station is quickly and accurately predicted, and efficient and low-consumption operation of the air-conditioning system is improved.

Description

A self-calibrating energy consumption simulation system and method for subway station air conditioning Technical Field

[0001] This invention relates to the field of energy consumption prediction for subway station air conditioning systems, and specifically to a self-calibrating subway station air conditioning energy consumption simulation system and method. Background Technology

[0002] With the rapid development of urban rail transit in my country, the energy consumption of subway stations is also increasing, with air conditioning systems accounting for as much as 30-50% of the energy consumption. Due to the large building span (>200m), special design characteristics (large differences between near-term and long-term design parameters), and special disturbance characteristics (large fluctuations in passenger flow and infiltration air volume) of subway stations, the feedback control system is ineffective and energy consumption is high. Currently, the air conditioning feedback control system is also limited by large simulation errors in air conditioning energy consumption, resulting in low overall energy efficiency of the current subway station air conditioning system. The problem of high energy consumption urgently needs to be solved.

[0003] The energy consumption of subway station air conditioning systems is affected by numerous factors, and many parameters cannot be accurately measured, or the design values ​​differ significantly from the actual values. Parameter calibration is necessary to improve the accuracy of energy consumption simulation models. Traditional manual calibration methods suffer from high requirements for personnel expertise, high calibration difficulty, and long calibration times, making it impossible to quickly and accurately predict the energy efficiency of subway station air conditioning systems and limiting their efficient and low-consumption operation. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a self-calibrating subway station air conditioning energy consumption simulation system and method. By performing principal component analysis on measured data of the subway station air conditioning system and using optimization algorithms to automatically calibrate the energy consumption simulation model, the calibrated energy consumption simulation model is applied to the subway station air conditioning control system, thereby quickly and accurately predicting the energy efficiency of the subway station air conditioning system and achieving efficient and low-consumption operation of the air conditioning system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a self-calibrating subway station air conditioning energy consumption simulation system, comprising: a principal component analysis module for performing principal component analysis on measured data of the subway station air conditioning system and a self-calibration module; the self-calibration module establishes a corresponding energy consumption simulation model and optimizes the energy consumption simulation model by collecting subway station CAD drawings and design and operating parameter information; the principal component analysis module includes a data acquisition module for collecting measured data of the air conditioning system affecting the energy consumption of the subway station and a data analysis module for performing principal component analysis on the measured data of the subway station air conditioning system collected by the data acquisition module and determining the principal components that have the main impact on the energy consumption of the air conditioning system; the self-calibration module includes a model building module for establishing a corresponding energy consumption simulation model based on subway station CAD drawings and design and operating parameter information and an optimization module for optimizing the energy consumption simulation model using a Bayesian optimization algorithm.

[0006] Furthermore, the data acquisition module is mainly used to collect measured data on outdoor meteorological parameters, passenger flow, equipment heat generation, infiltration air volume, lighting density, building envelope, building area, and design temperature that affect the air conditioning system of subway stations.

[0007] Furthermore, the data analysis module uses principal component analysis to determine the principal component data affecting the energy consumption of the air conditioning system.

[0008] Furthermore, the energy consumption simulation model established by the model building module is the EnergyPlus model, in which the calculation formula for air conditioning load is as follows:

[0009] in: For the load of the air conditioning system, It is the sum of the heat gained from the internal heat source. Heat is obtained from the internal heat source. For the number of internal heat sources, It is the sum of convective heat transfer on the building surface. For the number of building surfaces, The convective heat transfer coefficient between the building surface and the air. The area of ​​the surface. The surface temperature, The room's air temperature. For heat exchange between different internal zones This refers to the number of internal areas. This refers to the air exchange volume and mass flow rate between different areas. The specific heat capacity of air. For air temperature in different areas, For the heat gain of infiltration air, The mass flow rate of the infiltration air. The temperature of the infiltration air.

[0010] Furthermore, the optimization module utilizes a Bayesian optimization algorithm to optimize the energy consumption simulation model, specifically including: generating initial sample points in the parameter space using Latin hypercube sampling, then running the energy consumption simulation model to obtain an initial dataset; and using the obtained initial dataset to obtain sample point data {(x1, y1), (x2, y2),…(x n , y n The process involves: establishing a probabilistic surrogate model for the objective function using sample point data; predicting the function value at unknown points using the probabilistic surrogate model; finding evaluation points using the data acquisition module to effectively utilize limited resources to find the optimal solution; running the energy consumption simulation model to obtain the corresponding load results using the parameter values ​​corresponding to the evaluation points collected by the data acquisition module; updating the dataset; and repeating the iteration until the error index meets the preset error value.

[0011] A self-calibrating method for simulating the energy consumption of subway station air conditioning includes the following steps: 1) Using a data acquisition module to collect measured data of the air conditioning system affecting its energy consumption; 2) Using a data analysis module to analyze the measured data and determine the main data influencing the system's energy consumption; 3) Using a model building module to establish an energy consumption simulation model and inputting the collected measured data into the model; 4) Comparing the simulated energy consumption data output by the model with the actual measured energy consumption data to determine the calibration target; 5) Determining the parameters to be calibrated in the model based on the main data influencing the system's energy consumption determined by the data analysis module; 6) Optimizing the parameters based on the calibration target using an optimization algorithm; 7) Re-inputting the optimized parameters into the model and comparing the simulated energy consumption data again with the actual measured energy consumption data to determine the calibration target, iterating multiple times; 8) Stopping the iteration when the mean mean deviation error (NMBE) is less than 1% and the root mean square error coefficient of variation (CV) is less than 3%.

[0012] Furthermore, the measured data of the air conditioning system include outdoor meteorological parameters, passenger flow, equipment heat generation, infiltration air volume, lighting density, building envelope, building area, and design temperature.

[0013] Furthermore, the energy consumption simulation model is an EnergyPlus model established based on subway station CAD drawings and design, and operating parameter information. The calculation formula for air conditioning load is as follows:

[0014] in: For the load of the air conditioning system, It is the sum of the heat gained from the internal heat source. Heat is obtained from the internal heat source. For the number of internal heat sources, It is the sum of convective heat transfer on the building surface. For the number of building surfaces, The convective heat transfer coefficient between the building surface and the air. The area of ​​the surface. The surface temperature, The room's air temperature. For heat exchange between different internal zones This refers to the number of internal areas. This refers to the air exchange volume and mass flow rate between different areas. The specific heat capacity of air. For air temperature in different areas, For the heat gain of infiltration air, The mass flow rate of the infiltration air. The temperature of the infiltration air.

[0015] Furthermore, the step of using the data analysis module to analyze the measured data of the air conditioning system collected by the data acquisition module and determine the main parameters affecting the energy consumption of the air conditioning system involves the following steps: preprocessing the measured data of the air conditioning system collected by the data acquisition module by converting the format, merging files, and performing time scale conversion; processing the preprocessed air conditioning system data by deleting outliers and filling in missing values; and using feature selection to perform feature selection on the processed measured data of the air conditioning system to determine the most important parameters affecting the energy consumption of the air conditioning system.

[0016] Furthermore, the optimization algorithm is used to optimize the parameters to be corrected in the energy consumption simulation model according to the correction target of the energy consumption simulation model. The optimization algorithm is a Bayesian optimization algorithm, and the specific optimization steps are as follows: initial sample points are generated in the parameter space using Latin hypercube sampling, and then the energy consumption simulation model is run to obtain the initial dataset; the sample point data {(x1, y1), (x2, y2),…(x... n , y nThe process involves: establishing a probabilistic surrogate model for the objective function using sample point data; predicting the function value at unknown points using the probabilistic surrogate model; finding evaluation points using the data acquisition module to effectively utilize limited resources to find the optimal solution; running the energy consumption simulation model to obtain the corresponding load results using the parameter values ​​corresponding to the evaluation points collected by the data acquisition module; updating the dataset; and repeating the iteration until the error index meets the preset error value.

[0017] The beneficial effects of this invention are as follows: The self-calibrating subway station air conditioning energy consumption simulation system and method proposed in this application, through principal component analysis of the measured data of the subway station air conditioning system, and by using optimization algorithms to automatically calibrate the energy consumption simulation model, applies the calibrated energy consumption simulation model to the air conditioning control system of the subway station, thereby quickly and accurately predicting the energy efficiency of the subway station air conditioning system, realizing the efficient and low-consumption operation of the air conditioning system, and solving the problems of difficult, time-consuming and inaccurate calibration of current energy consumption models.

[0018] The principal component analysis method proposed in this application uses principal component analysis on actual operating data of subway stations to identify the most significant influencing factors on the energy consumption of subway station air conditioning systems, which can significantly improve the efficiency of self-calibration of energy consumption simulation models. The proposed method for improving the accuracy of self-calibration using a simplified EnergyPlus model can reduce the difficulty and time required to establish energy consumption simulation models. It can also use artificial intelligence algorithms to automatically calibrate energy consumption simulation models, significantly reducing the time required for calibration and greatly improving the accuracy of energy consumption simulation models. This provides a basis for accurately controlling the operation of subway station air conditioning systems based on energy consumption simulation models. Attached Figure Description

[0019] Figure 1 is a system block diagram of a self-calibrating subway station air conditioning energy consumption simulation system according to the present invention; Figure 2 is a flowchart of a self-calibrating subway station air conditioning energy consumption simulation method according to the present invention; Figure 3 is a schematic diagram of a self-calibrating subway station air conditioning energy consumption simulation method according to the present invention; Figure 4 is a graph showing the changes in parameters of the EnergyPlus energy consumption simulation model according to the present invention after self-calibration; Figure 5 is a comparison graph of the EnergyPlus energy consumption simulation model before and after calibration according to the present invention.

[0020] Attached diagram labels: 1. Main cause analysis module; 2. Data acquisition module; 3. Data analysis module; 4. Self-calibration module; 5. Model building module; 6. Optimization module. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and specific embodiments: Embodiment 1 As shown in Figure 1, a self-calibrating subway station air conditioning energy consumption simulation system includes: a principal component analysis module 1 for performing principal component analysis on measured data of the subway station air conditioning system and a self-calibration module 4. The self-calibration module 4 establishes a corresponding energy consumption simulation model and optimizes the energy consumption simulation model by collecting CAD drawings and design and operation parameter information of the subway station. The principal component analysis module 1 includes a data acquisition module 2 for collecting measured data of the air conditioning system that affects the energy consumption of the air conditioning system in the subway station and a data analysis module 3 for performing principal component analysis on the measured data of the subway station air conditioning system collected by the data acquisition module 2 and determining the principal components that have the main impact on the energy consumption of the air conditioning system.

[0022] The self-calibration module 4 includes a model building module 5 that uses subway station CAD drawings and designs, operating parameter information, and a corresponding energy consumption simulation model, as well as an optimization module 6 that uses a Bayesian optimization algorithm to optimize the energy consumption simulation model.

[0023] Data acquisition module 2 is mainly used to collect measured data on outdoor meteorological parameters, passenger flow, equipment heat generation, infiltration air volume, lighting density, building envelope, building area, and design temperature that affect the air conditioning system of subway stations. Data analysis module 3 uses principal component analysis to determine the principal component data that affect the energy consumption of the air conditioning system.

[0024] The energy consumption simulation model established by module 5 is the EnergyPlus model, in which the calculation formula for air conditioning load is as follows:

[0025] in: For the load of the air conditioning system, It is the sum of the heat gained from the internal heat source. Heat is obtained from the internal heat source. For the number of internal heat sources, It is the sum of convective heat transfer on the building surface. For the number of building surfaces, The convective heat transfer coefficient between the building surface and the air. The area of ​​the surface. The surface temperature, The room's air temperature. For heat exchange between different internal zones This refers to the number of internal areas. This refers to the air exchange volume and mass flow rate between different areas. The specific heat capacity of air. For air temperature in different areas, For the heat gain of infiltration air, The mass flow rate of the infiltration air. The temperature of the infiltration air.

[0026] Optimization module 6 uses the Bayesian optimization algorithm to optimize the energy consumption simulation model. Specifically, it includes: determining the input parameters of the energy consumption simulation model that need correction, such as outdoor meteorological parameters, indoor design temperature, passenger flow, equipment power, and their value ranges. The defined input parameter value range is θ = {θ = (θ1, θ2,…θ…}. n} | θ i ∈[θ i_min , θ i_max [i=1, 2, …n]; Latin Hypercube Sampling (LHS) is used to divide the dimension of each input parameter into m equally probable spaces; for each parameter θ i Generate m equally spaced points within the interval [0,1] and randomly arrange these points. The corresponding value of each point is LHS_value. j Actual parameter θ i The value of θ i,j =θ i,j_min + (θ i,j_max -θ i,j_min ) × LHS_value j The relationship between the input and output spaces is established, approximating a Gaussian distribution f(θ) ~ GP(m(θ), k(θ,θ')), where m(θ) is the mean function, describing the average value of the function under different input values; k(θ, θ')) is the covariance function, describing the uncertainty of the function under different input values. Using parameter space data and a surrogate model, the variable parameter values ​​that minimize the acquisition function are calculated, which become the next acquisition point. The acquisition function is in the form θ. i,jnew = argminm(θ)+βk(θ, θ'), where β is a regularization parameter used to balance exploration and utilization by selecting the best known or predicted point, and exploring more areas by trying to move away from the known point; using the EPPY library in Python to call the EnergyPlus model, modifying the parameter θ in the input file and performing energy consumption simulation, outputting the energy consumption simulation results, and storing the energy consumption simulation results in the Gaussian process model, repeating the above process iteratively until the error exponents NMBE and CV(RMSE) between the energy consumption simulation results and the measured energy consumption results meet the preset error requirements.

[0027] Example 2, as shown in Figure 2, describes a self-calibrating method for simulating the energy consumption of subway station air conditioning. The specific steps are as follows: 1) Using a data acquisition module to collect measured data of the air conditioning system affecting its energy consumption; 2) Using a data analysis module to analyze the measured data and determine the main data influencing the system's energy consumption; 3) Using a model building module to establish an energy consumption simulation model and inputting the collected measured data into the model; 4) Comparing the simulated energy consumption data output from the model with the actual measured energy consumption data to determine the calibration target; 5) Based on... The data analysis module determines the main data affecting the energy consumption of the air conditioning system to identify the parameters to be corrected in the energy consumption simulation model. Based on the correction target of the energy consumption simulation model, the parameters to be corrected are optimized using an optimization algorithm. The optimized correction parameters are then input back into the energy consumption simulation model, and the simulated energy consumption data output by the model is compared with the actual measured energy consumption data to determine the correction target. This process is repeated multiple times. Iteration continues until the mean mean deviation error (NMBE) is less than 1% and the root mean square error coefficient of variation (CV) (RMSE) is less than 3%, at which point iteration stops. The calculation method is as follows:

[0028] In the formula, It is the normalized average deviation error of the index. These are actual measured data. It is simulated data. It is the number of data points. It sums all the data;

[0029] In the formula, It is the root mean square error coefficient of variation. These are actual measured data. It is simulated data. It is the number of data points. It sums all the data.

[0030] Currently, domestic and international standards require that the average simulation data be NMBE < 10% and CV (RMSE) < 30%. However, in practical applications, it has been found that the simulation error specified in the standards is too large and the effect on the control of the air conditioning system is very poor. The self-correcting subway station air conditioning energy consumption simulation method proposed in this invention can achieve a simulation error of NMBE < 1% and CV (RMSE) < 3%.

[0031] The measured data for the air conditioning system include outdoor meteorological parameters, passenger flow, equipment heat generation, infiltration air volume, lighting density, building envelope, building area, and design temperature. The energy consumption simulation model is an EnergyPlus model built based on subway station CAD drawings, design, and operating parameter information. The calculation formula for air conditioning load is as follows:

[0032] in: For the load of the air conditioning system, It is the sum of the heat gained from the internal heat source. Heat is obtained from the internal heat source. For the number of internal heat sources, It is the sum of convective heat transfer on the building surface. For the number of building surfaces, The convective heat transfer coefficient between the building surface and the air. The area of ​​the surface. The surface temperature, The room's air temperature. For heat exchange between different internal zones This refers to the number of internal areas. This refers to the air exchange volume and mass flow rate between different areas. The specific heat capacity of air. For air temperature in different areas, For the heat gain of infiltration air, The mass flow rate of the infiltration air. The temperature of the infiltration air.

[0033] The data analysis module analyzes the measured data of the air conditioning system collected by the data acquisition module and determines the main data affecting the energy consumption of the air conditioning system. The specific steps are as follows: converting the format of the collected measured data of the air conditioning system, merging files and converting the time scale; preprocessing the measured data of the air conditioning system, deleting outliers and filling missing values; and using feature selection to select features from the processed measured data of the air conditioning system to determine the most important parameters affecting the energy consumption of the air conditioning system.

[0034] An optimization algorithm is used to optimize the parameters of the energy consumption simulation model to be corrected based on the correction target. The optimization algorithm is a Bayesian optimization algorithm. The specific optimization steps are as follows: Determine the input parameters of the energy consumption simulation model to be corrected, such as outdoor meteorological parameters, indoor design temperature, passenger flow, equipment power and their value ranges. The defined input parameter value range is θ = {θ = (θ1, θ2,…θ...} n} | θ i ∈[θ i_min , θi_max [i=1, 2, …n]; Latin Hypercube Sampling (LHS) is used to divide the dimension of each input parameter into m equally probable spaces; for each parameter θ i Generate m equally spaced points within the interval [0,1] and randomly arrange these points. The corresponding value of each point is LHS_value. j Actual parameter θ i The value of θ i,j =θ i,j_min + (θ i,j_max -θ i,j_min ) × LHS_value j The relationship between the input and output spaces is established, approximating a Gaussian distribution f(θ) ~ GP(m(θ), k(θ,θ')), where m(θ) is the mean function, describing the average value of the function under different input values; k(θ, θ')) is the covariance function, describing the uncertainty of the function under different input values. Using parameter space data and a surrogate model, the variable parameter values ​​that minimize the acquisition function are calculated, which become the next acquisition point. The acquisition function is in the form θ. i,jnew = argminm(θ)+βk(θ, θ'), where β is a regularization parameter used to balance exploration and utilization by selecting the best known or predicted point, and exploring more areas by trying to move away from the known point; using the EPPY library in Python to call the EnergyPlus model, modifying the parameter θ in the input file and performing energy consumption simulation, outputting the energy consumption simulation results, and storing the energy consumption simulation results in the Gaussian process model, repeating the above process iteratively until the error exponents NMBE and CV(RMSE) between the energy consumption simulation results and the measured energy consumption results meet the preset error requirements.

[0035] As shown in Figure 3, we take a subway station as an example for analysis, collect its operating parameters and energy consumption data for a whole year (it is recommended to be more than one month), and then use principal component analysis to determine its main influencing factors, such as outdoor meteorological parameters, indoor design temperature, passenger flow, and equipment power.

[0036] Based on the selected subway station CAD drawings and design specifications, an EnergyPlus energy consumption simulation model for the corresponding subway station is established. The calculation formula for air conditioning load is as follows:

[0037] in: For the load of the air conditioning system, It is the sum of the heat gained from the internal heat source. Heat is obtained from the internal heat source. For the number of internal heat sources, It is the sum of convective heat transfer on the building surface. For the number of building surfaces, The convective heat transfer coefficient between the building surface and the air. The area of ​​the surface. The surface temperature, The room's air temperature. For heat exchange between different internal zones This refers to the number of internal areas. This refers to the air exchange volume and mass flow rate between different areas. The specific heat capacity of air. For air temperature in different areas, For the heat gain of infiltration air, The mass flow rate of the infiltration air. The temperature of the infiltration air.

[0038] Since this station has relatively complete data and is equipped with a weather station to monitor outdoor meteorological parameters, the main factors affecting the air conditioning system, namely passenger flow and infiltration air volume, are difficult to obtain accurately. Therefore, this example uses passenger flow as the parameter to be corrected for analysis. To evaluate the accuracy of the corrected energy consumption simulation model, this example assumes a passenger flow of 10,000 people, and uses the energy consumption simulation results based on this parameter as validation data. The formula for calculating the heat generated by people in the EnergyPlus energy consumption simulation model is as follows:

[0039] in For personnel to obtain heat, The number of areas within the building, For the number of personnel, For the heat gain of a single person, For personnel timetable.

[0040] The metabolic rate was estimated based on the rate of movement and standing. The estimation is based on the passenger flow distribution coefficients given in the specifications for the station hall and platform. The main parameter that is difficult to obtain is the number of people, i.e., passenger flow. Since accurate passenger flow cannot be obtained under the initial operating conditions, only a reasonable range of passenger flow variation of 0-50,000 people can be estimated based on data analysis. At this time, the self-calibration module is run, using Python software to run optimization algorithms and the eppy library to realize the automatic calibration, simulation and data processing of EnergyPlus software. In this example, the calibration target is set to NMBE < 1%. After the self-calibration module completes the calibration target, it will automatically output the energy consumption model, error results and parameter values.

[0041] Referring to Figure 4, this is a graph showing the changes in the self-calibration parameters of the EnergyPlus energy consumption simulation model in the embodiments of this application.

[0042] Under initial conditions, the self-calibration module iteratively optimizes based on the given passenger flow range of 0-50,000 people. The first simulation is conducted using the user-set initial passenger flow of 1,000 people. The simulation results show a large error compared to the verification data. Therefore, the self-calibration module performs initial optimization and undergoes 15 iterations. It simulates the results of different parameters between the minimum and maximum values ​​in the passenger flow range and automatically processes the data to obtain the error between the simulation and the research data.

[0043] In the initial iteration phase, the error was found to be minimal near the parameters set in step 10. Therefore, a finer iteration phase was initiated for further simulation optimization. In step 29, the EnergyPlus energy consumption simulation model achieved a simulation error (NMBE) of less than 1%, meeting the optimization requirements. The iteration was then stopped, and the corresponding model and data were output. The corrected passenger flow was 10,054 person-times, very close to the 10,000 person-times set in this embodiment.

[0044] Referring to Figure 5, this is a comparison of the EnergyPlus energy consumption simulation model before and after correction in the embodiments of this application.

[0045] This embodiment selects the air conditioning load results for a whole week for comparison, where "□" represents the results of the verification data, and the other two lines represent the results predicted by the EnergyPlus software before and after the iteration, respectively. It can be seen that using the self-calibration method provided by this invention, the deviation between the simulation results and the verification data is very small, significantly reducing the time required for energy consumption correction while maintaining high accuracy.

[0046] For those skilled in the art, various other corresponding changes and modifications can be made based on the technical solutions and concepts described above, and all such changes and modifications should fall within the protection scope of the claims of this invention.

Claims

1. A self-calibrating subway station air conditioning energy consumption simulation system, characterized in that, include: This module is used for principal component analysis of measured data from subway station air conditioning systems. It includes a main cause analysis module and a self-calibration module. The self-calibration module uses collected subway station CAD drawings, design parameters, and operational parameters to establish and optimize a corresponding energy consumption simulation model. The main cause analysis module includes a data acquisition module for collecting measured data on air conditioning systems affecting energy consumption in subway stations, and a data analysis module for performing principal component analysis on the collected data to determine the principal components that have the main impact on air conditioning system energy consumption. The self-calibration module includes a model building module for establishing a corresponding energy consumption simulation model using subway station CAD drawings, design parameters, and operational parameters, and an optimization module for optimizing the energy consumption simulation model using a Bayesian optimization algorithm.

2. The self-calibrating subway station air conditioning energy consumption simulation system as described in claim 1, characterized in that, The data acquisition module is mainly used to collect measured data on outdoor meteorological parameters, passenger flow, equipment heat generation, infiltration air volume, lighting density, building envelope, building area, and design temperature that affect the air conditioning system of subway stations.

3. The self-calibrating subway station air conditioning energy consumption simulation system as described in claim 1, characterized in that, The data analysis module uses principal component analysis to determine the principal component data affecting the energy consumption of the air conditioning system.

4. The self-calibrating subway station air conditioning energy consumption simulation system as described in claim 1, characterized in that, The energy consumption simulation model established by the model building module is the EnergyPlus model, in which the calculation formula for air conditioning load is as follows: in: For the load of the air conditioning system, It is the sum of the heat gained from the internal heat source. Heat is obtained from the internal heat source. For the number of internal heat sources, It is the sum of convective heat transfer on the building surface. For the number of building surfaces, The convective heat transfer coefficient between the building surface and the air. The area of ​​the surface. The surface temperature, The room's air temperature. For heat exchange between different internal zones This refers to the number of internal areas. This refers to the air exchange volume and mass flow rate between different areas. The specific heat capacity of air. For air temperature in different areas, For the heat gain of infiltration air, The mass flow rate of the infiltration air. The temperature of the infiltration air.

5. The self-calibrating subway station air conditioning energy consumption simulation system as described in claim 1, characterized in that, The optimization module utilizes a Bayesian optimization algorithm to optimize the energy consumption simulation model. Specifically, it includes: generating initial sample points in the parameter space using Latin hypercube sampling, then running the energy consumption simulation model to obtain an initial dataset; and using the obtained initial dataset to obtain sample point data {(x1, y1), (x2, y2),…(x n , y n The process involves: establishing a probabilistic surrogate model for the objective function using sample point data; predicting the function value at unknown points using the probabilistic surrogate model; finding evaluation points using the data acquisition module to effectively utilize limited resources to find the optimal solution; running the energy consumption simulation model to obtain the corresponding load results using the parameter values ​​corresponding to the evaluation points collected by the data acquisition module; updating the dataset; and repeating the iteration until the error index meets the preset error value.

6. A self-calibrating method for simulating the energy consumption of air conditioning systems in subway stations, comprising the following steps: using a data acquisition module to collect measured data of the air conditioning system affecting the energy consumption of the subway station air conditioning system; using a data analysis module to analyze the measured data of the air conditioning system collected by the data acquisition module and determine the main data affecting the energy consumption of the air conditioning system; using a model building module to build an energy consumption simulation model and inputting the collected measured data of the air conditioning system into the energy consumption simulation model; comparing the energy consumption simulation data output by the energy consumption simulation model with the actual measured energy consumption data to determine the calibration target; determining the parameters to be calibrated of the energy consumption simulation model based on the main data affecting the energy consumption of the air conditioning system determined by the data analysis module; optimizing the parameters to be calibrated of the energy consumption simulation model using an optimization algorithm based on the calibration target of the energy consumption simulation model; inputting the optimized calibration parameters back into the energy consumption simulation model, comparing the energy consumption simulation data output by the energy consumption simulation model with the actual measured energy consumption data again to determine the calibration target, and iterating multiple times; stopping the iteration when the mean deviation error (NMBE) is less than 1% and the root mean square error coefficient of variation (CV) (RMSE) is less than 3%.

7. The self-calibrating subway station air conditioning energy consumption simulation method as described in claim 6, characterized in that, The measured data of the air conditioning system include outdoor meteorological parameters, passenger flow, equipment heat generation, infiltration air volume, lighting density, building envelope, building area, and design temperature.

8. The self-calibrating subway station air conditioning energy consumption simulation method as described in claim 6, characterized in that, The energy consumption simulation model is an EnergyPlus model established based on subway station CAD drawings and design, and operating parameter information. The calculation formula for air conditioning load is as follows: in: For the load of the air conditioning system, It is the sum of the heat gained from the internal heat source. Heat is obtained from the internal heat source. For the number of internal heat sources, It is the sum of convective heat transfer on the building surface. For the number of building surfaces, The convective heat transfer coefficient between the building surface and the air. The area of ​​the surface. The surface temperature, The room's air temperature. For heat exchange between different internal zones This refers to the number of internal areas. This refers to the air exchange volume and mass flow rate between different areas. The specific heat capacity of air. For air temperature in different areas, For the heat gain of infiltration air, The mass flow rate of the infiltration air. The temperature of the infiltration air.

9. The self-calibrating method for simulating the energy consumption of subway station air conditioning as described in claim 6, characterized in that, The process of using a data analysis module to analyze the measured data of the air conditioning system collected by the data acquisition module and determine the main data affecting the energy consumption of the air conditioning system involves the following steps: preprocessing the measured data of the air conditioning system collected by the data acquisition module by converting the format, merging files, and converting the time scale; processing the preprocessed air conditioning system data by deleting outliers and filling in missing values; and using feature selection to select features from the processed measured data of the air conditioning system to determine the most important parameters affecting the energy consumption of the air conditioning system.

10. The self-calibrating subway station air conditioning energy consumption simulation method as described in claim 6, characterized in that, The optimization algorithm is used to optimize the parameters to be corrected in the energy consumption simulation model according to the correction target. The optimization algorithm is a Bayesian optimization algorithm, and the specific optimization steps are as follows: initial sample points are generated in the parameter space using Latin hypercube sampling, and then the energy consumption simulation model is run to obtain the initial dataset; the sample point data {(x1, y1), (x2, y2),…(x... n , y n The process involves: establishing a probabilistic surrogate model for the objective function using sample point data; predicting the function value at unknown points using the probabilistic surrogate model; finding evaluation points using the data acquisition module to effectively utilize limited resources to find the optimal solution; running the energy consumption simulation model to obtain the corresponding load results using the parameter values ​​corresponding to the evaluation points collected by the data acquisition module; updating the dataset; and repeating the iteration until the error index meets the preset error value.

Images