Ventilation air conditioner control method based on load prediction

Abstract

The invention relates to the technical field of energy and energy conservation, in particular to a ventilation air conditioner control method based on load prediction. The technical problems that in the prior art, a ventilation air conditioner control method has certain hysteresis quality, advanced adjustment cannot be achieved, and autonomous adjustment and control cannot be conducted on the occasion with large flow are solved. According to the technical scheme, the ventilation air conditioner control method based on load prediction comprises the steps of analyzing the relation between the operation parameters of each component device of an air conditioner system and the building cooling load according to a model prediction result, fitting the change of the relation into a curve, and compiling an operation control scheme under typical working conditions according to the fitted relation curve. The building cooling load prediction value within a period of time can be obtained according to the algorithm, the system formulates the air conditioner system control strategy according to the prediction value, the problems of unreasonable building ventilation air conditioner operation mode and control lag can be solved, the operation energy efficiency of the air conditioner system is improved, on-demand cooling is effectively achieved, and the energy consumption is reduced. And intelligent building operation management is powerfully supported.

Description

Technical Field

[0001] The present invention relates to the field of energy and energy-saving technology, and in particular to a ventilation and air-conditioning control method based on load prediction. Background Art

[0002] In the energy management system of public buildings, ventilation and air conditioning are key energy-consuming equipment; in areas with cold winters and hot summers, ventilation and air conditioning can account for about 50% of the total energy consumption of buildings; therefore, the control technology of ventilation and air conditioning has always been an important task in the field of building intelligence; at present, the operation of ventilation and air conditioning in public buildings has problems such as unreasonable selection and design, low degree of automation, and low system operation efficiency; because the equipment selection and design capacity are generally large, many public buildings have not reached the standard operating conditions for a long time. It is common in subway stations, large commercial complexes, etc., where air-conditioning systems designed for long-term operating conditions fail to meet expectations for a long time. The phenomenon of "a big horse pulling a small cart" is more prominent, and ventilation The air-conditioning system consumes too much energy. Most of the air-conditioning systems are under manual management and cannot adjust the operating mode according to actual demand. The operating modes such as fresh air volume, temperature and operating time at the ventilation and air-conditioning terminals are unreasonable, resulting in many unreasonable system operations. Even in air-conditioning systems with certain automatic control capabilities, such as the classic PID control system, the control is based on the data changes of various sensors, which is a closed-loop control. The system as a whole has a certain lag and cannot achieve advance adjustment. It cannot be autonomously adjusted according to environmental and passenger flow factors in places with large passenger flow such as stations and airports. It can neither provide a better environmental comfort level nor cause a lot of energy waste. Summary of the invention

[0003] In order to overcome the certain lag of the ventilation and air-conditioning control method under the existing technology, it is impossible to achieve advance adjustment. In places with large passenger flow such as stations and airports, it is impossible to perform autonomous adjustment and control according to environmental and passenger flow factors; it can neither provide a better environmental comfort level nor cause a lot of energy waste.

[0004] The technical solution of the present invention is: a ventilation and air conditioning control method based on load prediction:

[0005] S1: Collect historical operating data from air-conditioning systems in places such as subways and airports, including but not limited to ambient temperature, humidity, passenger flow, and operating parameters of each component of the air-conditioning system;

[0006] S2: Clean the collected data, remove outliers and missing values, etc., to ensure the accuracy and completeness of the data; organize the cleaned data in chronological order and divide it into training sets and test sets;

[0007] S3: Multiple linear regression analysis is used as the modeling method. According to historical data, the key factors affecting the building cooling load are selected as independent variables, and the building cooling load is used as the dependent variable. Using the training set data, the mathematical model between the independent variable and the dependent variable is established through regression analysis to determine the coefficients between the variables.

[0008] S4: According to the model prediction results, analyze the relationship between the operating parameters of each component of the air-conditioning system and the building cooling load, and fit their changes into a curve;

[0009] S5: According to the fitted relationship curve, compile the operation control plan under typical working conditions; sequence the various parameters in the control strategy into a parameter sequence group as the input variable of the ventilation and air-conditioning control system;

[0010] S6: deploy the control system and set the initial value of the control variable of a single device according to the parameter sequence group of the control model;

[0011] S7: Monitor the changes in the building's cooling load in real time, select appropriate parameters from the parameter sequence group for adjustment according to the actual cooling load changes, and continuously compare the on-site operation curve and the simulation curve until the two are consistent, ensuring that the control system can accurately respond to changes in the building's cooling load;

[0012] S8: Continuously collect operating data and user feedback from the control system, analyze system performance, and regularly update the building cooling load model based on new data and feedback to improve prediction accuracy.

[0013] Collect historical operating data from air-conditioning systems in places such as subways and airports, including but not limited to ambient temperature, humidity, passenger flow, and operating parameters of each component of the air-conditioning system; the operating parameters of each component of the air-conditioning system include compressor power and fan speed; the specific steps are as follows:

[0014] S201: Identify the data source target, i.e., the ventilation and air conditioning systems of large public buildings such as subways and airports;

[0015] S202: Capturing equipment parameters and environmental data that affect cooling load in the air-conditioning system through sensors and data acquisition equipment;

[0016] S203: extracting operation data for a period of time in the past from historical data, including but not limited to ambient temperature, humidity, passenger flow, and various operation parameters of the air conditioning system;

[0017] S204: Create a detailed data log for the collected data, recording timestamps, data values, and any abnormal conditions;

[0018] S205: Unify the data format and output it.

[0019] Clean the collected data to remove outliers and missing values ​​to ensure the accuracy and completeness of the data: organize the cleaned data in chronological order and divide it into training sets and test sets for subsequent modeling and verification; the specific steps are as follows:

[0020] S301: Select a filling method or directly delete records containing missing values ​​according to the data type and missing conditions;

[0021] S302: For numerical data, if the missing ratio is not high, the mean or median can be used for filling; if the missing ratio is high, consider whether the column data has a significant impact on the model prediction. If so, the column can be deleted;

[0022] S303: For categorical data, fill using mode or fill based on context information;

[0023] S304: Record processing results;

[0024] S305: Calculate the mean, standard deviation, median, IQR and other statistics of each column of data;

[0025] S306: When using the standard deviation method, set a threshold, such as mean ± 3 times the standard deviation, and data outside this range are considered outliers;

[0026] S307: When using the IQR method, calculate the first quartile of Q1 and the third quartile of Q3, then calculate IQR = Q3-Q1, and consider data less than Q1-1.5IQR or greater than Q3+1.5IQR as outliers;

[0027] S308: Selectively delete the identified outliers;

[0028] S309: Check and convert date and time columns to a unified format; check numeric columns to ensure that there are no non-numeric characters, and convert data types as needed; unify the encoding format for categorical data;

[0029] S310: Check the time column in the data set to ensure that the time format is uniform and correct;

[0030] S311: sorting the data according to the time column;

[0031] S312: Analyze historical data to identify features that are highly relevant to the prediction target; select the feature set that is ultimately used for model training based on constructed derived features, such as temperature difference, humidity difference, passenger flow change rate, etc.;

[0032] S313: Divide the data set into a training set and a test set, and determine the division ratio; divide the data set using a time series segmentation method to ensure that the training set and the test set are consistent in data distribution;

[0033] S314: Normalize or standardize the numerical features.

[0034] The multivariate linear regression analysis method is used as the modeling method. According to historical data, the key factors affecting the building cooling load are selected as independent variables, and the building cooling load is used as the dependent variable. Using the training set data, the mathematical model between the independent variable and the dependent variable is established through regression analysis to determine the coefficients between the variables. The specific steps are as follows:

[0035] S401: Based on the data analysis results, select variables that have a significant impact on the building cooling load;

[0036] S402: Clarify the calculation method of building cooling load or directly use the existing cooling load data as the dependent variable;

[0037] S403: Setting a multiple linear regression model and determining the basic form of the model;

[0038] That is, Y = β0 + β1X1 + β2X2 + ... + βnXn + ∈; where Y is the building cooling load, Xi is the influencing factor, i.e., the independent variable, βi is the coefficient to be determined, and ∈ is the error term;

[0039] S404: Calculate the average value of ventilation and air conditioning power during the building operation period based on historical data. The calculation formula is as follows:

[0040] Steady-state building air conditioning basic cooling load Qd = average ventilation and air conditioning power during building operation period × air conditioning energy efficiency ratio COP;

[0041] Qc=P×COP+Qn1+Qm1;

[0042] Among them, Qc represents the cooling load of ventilation and air conditioning, P represents the electric power of ventilation and air conditioning, i.e. the load value monitored in real time by the energy management system, COP represents the energy efficiency ratio of air conditioning, Qn1 represents the cooling load caused by the change of fresh air volume demand, and Qm1 represents the cooling load caused by the change of passenger flow;

[0043] S405: selecting dates with similar operating conditions to the forecast date from historical data, and using the electric load curves of these dates as the basis of the load curve of the forecast date;

[0044] S406: Calculate the fresh air volume Fi according to the change in the number of people in the building N and the fresh air supply per person E. The calculation formula is as follows:

[0045] Fi = N × E;

[0046] S407: Calculate the cooling load Qn1 caused by the change in fresh air demand using the fresh air volume Fi, air density K, air specific heat Ra, fresh air temperature Ti and outflow air temperature To. The calculation formula is as follows:

[0047] Qn1 = Fi × K × Ra × (Ti-To);

[0048] S408: regularly obtain the latest passenger flow data from the passenger flow system, access control system, etc., and update the predicted values ​​of fresh air volume and fresh air cooling load;

[0049] S409: Calculate the cooling load Qm1 caused by the change in the number of people in the building per unit time according to the change in the number of people N, the cluster coefficient A, the heat dissipation of human bodies Qr and the cooling load coefficient s of human body heat dissipation. The calculation formula is as follows:

[0050] Qm1=N×A×Qr×s;

[0051] S410: According to different building types, query the parameter values ​​such as A, Qr, s, etc. from relevant design specifications or theories;

[0052] S411: Add the basic cooling load Qd of the steady-state building air conditioning, the cooling load Qn1 caused by the change in fresh air volume demand, and the cooling load Qm1 caused by the change in the flow of people, and obtain the total ventilation and air conditioning cooling load prediction value calculation formula Qc=Qd+Qn1+Qm1;

[0053] S412: dynamically adjusting the predicted value based on real-time data, such as current crowd flow changes and fresh air temperature;

[0054] S413: Use the test set data to verify the prediction model and evaluate the prediction accuracy and generalization ability;

[0055] S414: Adjustment and optimization of the model based on the verification results

[0056] According to the model prediction results, the relationship between the operating parameters of each component of the air-conditioning system and the building cooling load is analyzed, and its changes are fitted into a curve. The specific steps are as follows:

[0057] S601: Perform a preliminary fit on the selected model using the collected data to obtain initial model parameters;

[0058] S602: Evaluate the fitting effect of the model by calculating statistical indicators of the fitting curve, such as R^2 value, mean square error (MSE), and root mean square error (RMSE);

[0059] S603: Optimizing the model, adjusting model parameters through an iterative optimization algorithm to improve the fitting effect of the model;

[0060] S604: Observe the shape and trend of the relationship curve obtained by fitting, and analyze the internal relationship between the independent variable and the dependent variable;

[0061] S605: Analyze the shape and trend of the curve; for example, the curve may indicate that increasing the compressor power within a certain range can significantly increase the supply of cooling load, but the effect gradually decreases beyond a certain range.

[0062] According to the fitted relationship curve, the operation control plan under typical working conditions is compiled; the parameters in the control strategy are serialized into a parameter sequence group as the input variable of the ventilation and air-conditioning control system; specifically, the following steps are included:

[0063] S701: Identify typical air-conditioning operating conditions, such as high-load conditions, low-load conditions, and transitional season conditions, based on historical data and actual operating conditions;

[0064] S702: For each typical working condition, determine the optimal number of devices to be started under different working conditions based on the fitted relationship curve and the cooling load prediction results; consider the energy efficiency ratio of the equipment and the load distribution balance to minimize energy consumption and extend the life of the equipment;

[0065] S703: According to the equipment characteristics and load change trend, set a reasonable load increase and decrease rate to smoothly adjust the equipment operation status and avoid frequent start and stop impact on the equipment; the load increase and decrease rate should comprehensively consider factors such as equipment response time, system stability and energy efficiency ratio;

[0066] S704: Setting control target values ​​of key parameters such as outlet water temperature and outlet air temperature according to the building cooling load demand and air conditioning system performance;

[0067] S705: converting the above control strategy into specific control logic, including equipment start and stop conditions, load addition and reduction logic, and parameter adjustment rules;

[0068] S706: sort out various parameters in the control strategy, including the number of equipment turned on, load increase and decrease rate, and target values ​​of key parameters;

[0069] S707: Serializing the parameters into a parameter sequence group in a certain format and order, and ensuring that the format of the parameter sequence group matches the input interface of the ventilation and air conditioning control system so that it can be smoothly transmitted to the control system;

[0070] S708: Verify and test the serialized parameter sequence group to ensure that it can be correctly parsed and executed in the control system;

[0071] S709: Using the serialized parameter sequence group as input variables of the ventilation and air-conditioning control system.

[0072] Deploy the control system and set the initial value of the control variable of a single device according to the parameter sequence group of the control model. The specific steps are as follows:

[0073] S801: Reading a serialized parameter sequence group from a storage medium;

[0074] S802: Parse the parameter sequence group and convert it into a format and data structure that can be recognized by the control system;

[0075] S803: Load the parsed parameter sequence group into the memory of the control system, ready for subsequent control logic processing;

[0076] S804: traverse each parameter item in the parameter sequence group to identify parameters related to the control variables of a single device;

[0077] S805: For each single device control variable, according to its corresponding parameter value, set the initial value of the variable in the control system; for example, if the initial operating frequency of a compressor is specified in the parameter sequence group, then set the frequency value as the initial operating frequency of the compressor;

[0078] S806: Check whether the initial values ​​of all single device control variables have been set correctly to ensure that there are no omissions or errors.

[0079] Monitor the changes in the building's cooling load in real time, select appropriate parameters from the parameter sequence group for adjustment according to the actual cooling load changes, and continuously compare the on-site operation curve and the simulation curve until the two are consistent, ensuring that the control system can accurately respond to changes in the building's cooling load; the specific steps are as follows:

[0080] S901: Start the real-time monitoring system of the ventilation and air conditioning control system to continuously collect real-time cooling load data inside the building;

[0081] S902: Compare the real-time cooling load data with the cooling load data output by the prediction model to evaluate the deviation between the actual current load and the predicted value;

[0082] S903: Selecting relevant control parameters from the parameter sequence group according to the change trend of the real-time cooling load and the output of the prediction model;

[0083] S904: Calculate the parameter value that needs to be adjusted based on the deviation between the real-time load and the predicted load and the current operating state of the system;

[0084] S905: Send the calculated parameter value to the actuator of the ventilation and air conditioning control system to adjust the operating status of related equipment;

[0085] S906: Record the actual operation data of the ventilation and air conditioning system after adjusting the parameters, including the operation parameters of each device and the change of the cooling load inside the building;

[0086] S907: Use these data to draw curves of on-site operation, including a cooling load change curve and an equipment operation parameter change curve;

[0087] S908: Compare the field operation curve with the simulation curve generated based on the prediction model and the parameter sequence group, and analyze the difference and trend between the two;

[0088] S909: If there is a large difference between the on-site operation curve and the simulation curve, it means that the current control strategy or parameter setting is not accurate enough and needs to be adjusted;

[0089] S910: reselect or adjust parameter values ​​in the parameter sequence group, or optimize the control strategy according to the comparison result;

[0090] S911: repeating steps S904 to S910, and continuously iteratively adjusting parameters and control strategies until the field operation curve and the simulation curve are consistent or meet the preset accuracy requirements;

[0091] S912: After parameter adjustment and optimization are completed, system stability verification is performed for a period of time;

[0092] S913: Monitor the system's operating status and various parameter indicators to ensure that the system can respond stably and accurately when the load changes;

[0093] S914: If the system performance is stable and meets the requirements, the adjustment process is completed; if there are still problems, return to step S909 for further adjustment and optimization.

[0094] Continuously collect operating data and user feedback from the control system, analyze system performance, and regularly update the building cooling load model based on new data and feedback to improve prediction accuracy. The specific steps are as follows:

[0095] S1001: Deploy a data collection system to ensure that the operation data of the ventilation and air conditioning control system can be continuously and accurately collected, including equipment operation status, energy consumption data, and environmental parameters;

[0096] S1002: Organize, classify and store the collected operation data and user feedback to ensure the integrity and accessibility of the data;

[0097] S1003: Conduct in-depth analysis on the collected operating data, identify trends, anomalies and correlations in the data, and evaluate the prediction accuracy of the current cooling load model;

[0098] S1004: Based on the data analysis results, evaluate the performance of the current cooling load model, including prediction error, stability, and adaptability;

[0099] S1005: Update the cooling load model based on the evaluation results and new data input, including adjusting model parameters, improving algorithms, introducing new influencing factors, etc., to improve the prediction accuracy and robustness of the model;

[0100] S1006: Use the new data set to verify the updated model to ensure that the prediction performance of the updated model is improved;

[0101] S1007: Based on the model update results and actual needs, evaluate the current air conditioning system control strategy, analyze the performance of the strategy in terms of energy saving, comfort, equipment life, etc., and identify potential areas for improvement;

[0102] S1008: Develop an optimization plan based on the evaluation results, including adjusting the number of devices to be started, optimizing the load-on / load-off ratio, adjusting the setting values ​​of key parameters, etc., to improve the overall operating efficiency of the system;

[0103] S1009: Test the optimized control strategy in a simulation environment to evaluate its effectiveness in actual operation;

[0104] S1010: Deploy the optimized control strategy to the actual system and continuously monitor the system operation status and performance indicators.

[0105] Beneficial effects of the present invention:

[0106] 1. Compared with the existing ventilation and air conditioning control system, which has a certain hysteresis as a whole and cannot realize advance adjustment, it cannot perform autonomous adjustment and control according to environmental and human flow factors in places with large passenger flow such as stations and airports, which can neither provide a good environmental comfort nor cause a lot of energy waste. This method predicts the air conditioning cooling load by considering factors such as human flow and environment based on the prediction algorithm, and guides the operation of the ventilation and air conditioning system according to the prediction results, so as to first obtain the predicted value of the building cooling load within a period of time through the algorithm, and then use the predicted value of the building cooling load as a parameter combined with the traditional control method to realize the advance control of the system, so as to solve the problems of unreasonable ventilation and air conditioning operation mode and system hysteresis.

[0107] 2. By establishing a building cooling load model, automatic control is performed according to the load prediction algorithm, and the building cooling load is simulated by using the multivariate linear regression analysis method. This method is to establish a model based on historical data, analyze and summarize the historical data through regression analysis, so as to determine the relationship between variables, predict future values ​​according to this relationship, analyze the relationship between the operating parameters of each component equipment of the air-conditioning system and the building cooling load, and fit its changes into a curve; according to the fitted relationship curve, compile an operation control plan under typical working conditions to form a set of air-conditioning operation setting parameter sequences, and use them as input variables of the ventilation and air-conditioning control system, and give them through the air-conditioning system parameter sequence group of the control model. According to different working conditions, different parameters of the sequence group are used, and according to the changes in the actual building cooling load, the above set values ​​are adjusted until the curve of the on-site operation and the simulation curve are consistent, so as to first obtain the predicted value of the building cooling load within a period of time through the algorithm, and then use the predicted value of the building cooling load as a parameter combined with the traditional control method to realize the advanced control of the system, and solve the problems of unreasonable ventilation and air-conditioning operation mode and system lag

[0108] 3. Analyze system performance by continuously collecting operating data and user feedback from the control system. Update the cooling load model based on the evaluation results and new data input, including adjusting model parameters, improving algorithms, introducing new influencing factors, etc., to improve the prediction accuracy and robustness of the model. Regularly update the building cooling load model based on new data and feedback to improve prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0109] Figure 1 What is shown is a schematic flow chart of the steps of the ventilation and air conditioning control method based on load prediction of the present invention;

[0110] Figure 2 What is presented is a schematic flow chart of an execution scheme of the ventilation and air-conditioning control method based on load forecasting of the present invention. DETAILED DESCRIPTION

[0111] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0112] See also Figure 1-2 The present invention provides an embodiment: a ventilation and air conditioning control method based on load prediction:

[0113] S1: Collect historical operating data from air-conditioning systems in places such as subways and airports, including but not limited to ambient temperature, humidity, passenger flow, and operating parameters of each component of the air-conditioning system;

[0114] S2: Clean the collected data, remove outliers and missing values, etc., to ensure the accuracy and completeness of the data; organize the cleaned data in chronological order and divide it into training sets and test sets;

[0115] S3: Multiple linear regression analysis is used as the modeling method. According to historical data, the key factors affecting the building cooling load are selected as independent variables, and the building cooling load is used as the dependent variable. Using the training set data, the mathematical model between the independent variable and the dependent variable is established through regression analysis to determine the coefficients between the variables.

[0116] S4: According to the model prediction results, analyze the relationship between the operating parameters of each component of the air-conditioning system and the building cooling load, and fit their changes into a curve;

[0117] S5: According to the fitted relationship curve, compile the operation control plan under typical working conditions; sequence the various parameters in the control strategy into a parameter sequence group as the input variable of the ventilation and air-conditioning control system;

[0118] S6: deploy the control system and set the initial value of the control variable of a single device according to the parameter sequence group of the control model;

[0119] S7: Monitor the changes in the building's cooling load in real time, select appropriate parameters from the parameter sequence group for adjustment according to the actual cooling load changes, and continuously compare the on-site operation curve and the simulation curve until the two are consistent, ensuring that the control system can accurately respond to changes in the building's cooling load;

[0120] S8: Continuously collect operating data and user feedback from the control system, analyze system performance, and regularly update the building cooling load model based on new data and feedback to improve prediction accuracy.

[0121] Collect historical operating data from air-conditioning systems in places such as subways and airports, including but not limited to ambient temperature, humidity, passenger flow, and operating parameters of each component of the air-conditioning system; the operating parameters of each component of the air-conditioning system include compressor power and fan speed; the specific steps are as follows:

[0122] S201: Identify the data source target, i.e., the ventilation and air conditioning systems of large public buildings such as subways and airports;

[0123] S202: Capturing equipment parameters and environmental data that affect cooling load in the air-conditioning system through sensors and data acquisition equipment;

[0124] S203: extracting operation data for a period of time in the past from historical data, including but not limited to ambient temperature, humidity, passenger flow, and various operation parameters of the air conditioning system;

[0125] S204: Create a detailed data log for the collected data, recording timestamps, data values, and any abnormal conditions;

[0126] S205: Unify the data format and output it.

[0127] Clean the collected data to remove outliers and missing values ​​to ensure the accuracy and completeness of the data: organize the cleaned data in chronological order and divide it into training sets and test sets for subsequent modeling and verification; the specific steps are as follows:

[0128] S301: Select a filling method or directly delete records containing missing values ​​according to the data type and missing conditions;

[0129] S302: For numerical data, if the missing ratio is not high, the mean or median can be used for filling; if the missing ratio is high, consider whether the column data has a significant impact on the model prediction. If so, the column can be deleted;

[0130] S303: For categorical data, fill using mode or fill based on context information;

[0131] S304: Record processing results;

[0132] S305: Calculate the mean, standard deviation, median, IQR and other statistics of each column of data;

[0133] S306: When using the standard deviation method, set a threshold, such as mean ± 3 times the standard deviation, and data outside this range are considered outliers;

[0134] S307: When using the IQR method, calculate the first quartile of Q1 and the third quartile of Q3, then calculate IQR = Q3-Q1, and consider data less than Q1-1.5IQR or greater than Q3+1.5IQR as outliers;

[0135] S308: Selectively delete the identified outliers;

[0136] S309: Check and convert date and time columns to a unified format; check numeric columns to ensure that there are no non-numeric characters, and convert data types as needed; unify the encoding format for categorical data;

[0137] S310: Check the time column in the data set to ensure that the time format is uniform and correct;

[0138] S311: sorting the data according to the time column;

[0139] S312: Analyze historical data to identify features that are highly relevant to the prediction target; select the feature set that is ultimately used for model training based on constructed derived features, such as temperature difference, humidity difference, passenger flow change rate, etc.;

[0140] S313: Divide the data set into a training set and a test set, and determine the division ratio; divide the data set using a time series segmentation method to ensure that the training set and the test set are consistent in data distribution;

[0141] S314: Normalize or standardize the numerical features.

[0142] The multivariate linear regression analysis method is used as the modeling method. According to historical data, the key factors affecting the building cooling load are selected as independent variables, and the building cooling load is used as the dependent variable. Using the training set data, the mathematical model between the independent variable and the dependent variable is established through regression analysis to determine the coefficients between the variables. The specific steps are as follows:

[0143] S401: Based on the data analysis results, select variables that have a significant impact on the building cooling load;

[0144] S402: Clarify the calculation method of building cooling load or directly use the existing cooling load data as the dependent variable;

[0145] S403: Setting a multiple linear regression model and determining the basic form of the model;

[0146] That is, Y = β0 + β1X1 + β2X2 + ... + βnXn + ∈; where Y is the building cooling load, Xi is the influencing factor, i.e., the independent variable, βi is the coefficient to be determined, and ∈ is the error term;

[0147] S404: Calculate the average value of ventilation and air conditioning power during the building operation period based on historical data. The calculation formula is as follows:

[0148] Steady-state building air conditioning basic cooling load Qd = average ventilation and air conditioning power during building operation period × air conditioning energy efficiency ratio COP;

[0149] Qc=P×COP+Qn1+Qm1;

[0150] Among them, Qc represents the cooling load of ventilation and air conditioning, P represents the electric power of ventilation and air conditioning, i.e. the load value monitored in real time by the energy management system, COP represents the energy efficiency ratio of air conditioning, Qn1 represents the cooling load caused by the change of fresh air volume demand, and Qm1 represents the cooling load caused by the change of passenger flow;

[0151] S405: selecting dates with similar operating conditions to the forecast date from historical data, and using the electric load curves of these dates as the basis of the load curve of the forecast date;

[0152] S406: Calculate the fresh air volume Fi according to the change in the number of people in the building N and the fresh air supply per person E. The calculation formula is as follows:

[0153] Fi = N × E;

[0154] S407: Calculate the cooling load Qn1 caused by the change in fresh air demand using the fresh air volume Fi, air density K, air specific heat Ra, fresh air temperature Ti and outflow air temperature To. The calculation formula is as follows:

[0155] Qn1 = Fi × K × Ra × (Ti-To);

[0156] S408: regularly obtain the latest passenger flow data from the passenger flow system, access control system, etc., and update the predicted values ​​of fresh air volume and fresh air cooling load;

[0157] S409: Calculate the cooling load Qm1 caused by the change in the number of people in the building per unit time according to the change in the number of people N, the cluster coefficient A, the heat dissipation of human bodies Qr and the cooling load coefficient s of human body heat dissipation. The calculation formula is as follows:

[0158] Qm1=N×A×Qr×s;

[0159] S410: According to different building types, query the parameter values ​​such as A, Qr, s, etc. from relevant design specifications or theories;

[0160] S411: Add the basic cooling load Qd of the steady-state building air conditioning, the cooling load Qn1 caused by the change in fresh air volume demand, and the cooling load Qm1 caused by the change in the flow of people, and obtain the total ventilation and air conditioning cooling load prediction value calculation formula Qc=Qd+Qn1+Qm1;

[0161] S412: dynamically adjusting the predicted value based on real-time data, such as current crowd flow changes and fresh air temperature;

[0162] S413: Use the test set data to verify the prediction model and evaluate the prediction accuracy and generalization ability;

[0163] S414: Adjustment and optimization of the model based on the verification results

[0164] According to the model prediction results, the relationship between the operating parameters of each component of the air-conditioning system and the building cooling load is analyzed, and its changes are fitted into a curve. The specific steps are as follows:

[0165] S601: Perform a preliminary fit on the selected model using the collected data to obtain initial model parameters;

[0166] S602: Evaluate the fitting effect of the model by calculating statistical indicators of the fitting curve, such as R^2 value, mean square error (MSE), and root mean square error (RMSE);

[0167] S603: Optimizing the model, adjusting model parameters through an iterative optimization algorithm to improve the fitting effect of the model;

[0168] S604: Observe the shape and trend of the relationship curve obtained by fitting, and analyze the internal relationship between the independent variable and the dependent variable;

[0169] S605: Analyze the shape and trend of the curve; for example, the curve may indicate that increasing the compressor power within a certain range can significantly increase the supply of cooling load, but the effect gradually decreases beyond a certain range.

[0170] According to the fitted relationship curve, the operation control plan under typical working conditions is compiled; the parameters in the control strategy are serialized into a parameter sequence group as the input variable of the ventilation and air-conditioning control system; specifically, the following steps are included:

[0171] S701: Identify typical air-conditioning operating conditions, such as high-load conditions, low-load conditions, and transitional season conditions, based on historical data and actual operating conditions;

[0172] S702: For each typical working condition, determine the optimal number of devices to be started under different working conditions based on the fitted relationship curve and the cooling load prediction results; consider the energy efficiency ratio of the equipment and the load distribution balance to minimize energy consumption and extend the life of the equipment;

[0173] S703: According to the equipment characteristics and load change trend, set a reasonable load increase and decrease rate to smoothly adjust the equipment operation status and avoid frequent start and stop impact on the equipment; the load increase and decrease rate should comprehensively consider factors such as equipment response time, system stability and energy efficiency ratio;

[0174] S704: Setting control target values ​​of key parameters such as outlet water temperature and outlet air temperature according to the building cooling load demand and air conditioning system performance;

[0175] S705: converting the above control strategy into specific control logic, including equipment start and stop conditions, load addition and reduction logic, and parameter adjustment rules;

[0176] S706: sort out various parameters in the control strategy, including the number of equipment turned on, load increase and decrease rate, and target values ​​of key parameters;

[0177] S707: Serializing the parameters into a parameter sequence group in a certain format and order, and ensuring that the format of the parameter sequence group matches the input interface of the ventilation and air conditioning control system so that it can be smoothly transmitted to the control system;

[0178] S708: Verify and test the serialized parameter sequence group to ensure that it can be correctly parsed and executed in the control system;

[0179] S709: Using the serialized parameter sequence group as input variables of the ventilation and air-conditioning control system.

[0180] Deploy the control system and set the initial value of the control variable of a single device according to the parameter sequence group of the control model. The specific steps are as follows:

[0181] S801: Reading a serialized parameter sequence group from a storage medium;

[0182] S802: Parse the parameter sequence group and convert it into a format and data structure that can be recognized by the control system;

[0183] S803: Load the parsed parameter sequence group into the memory of the control system, ready for subsequent control logic processing;

[0184] S804: traverse each parameter item in the parameter sequence group to identify parameters related to the control variables of a single device;

[0185] S805: For each single device control variable, according to its corresponding parameter value, set the initial value of the variable in the control system; for example, if the initial operating frequency of a compressor is specified in the parameter sequence group, then set the frequency value as the initial operating frequency of the compressor;

[0186] S806: Check whether the initial values ​​of all single device control variables have been set correctly to ensure that there are no omissions or errors.

[0187] Monitor the changes in the building's cooling load in real time, select appropriate parameters from the parameter sequence group for adjustment according to the actual cooling load changes, and continuously compare the on-site operation curve and the simulation curve until the two are consistent, to ensure that the control system can accurately respond to changes in the building's cooling load; the specific steps are as follows:

[0188] S901: Start the real-time monitoring system of the ventilation and air conditioning control system to continuously collect real-time cooling load data inside the building;

[0189] S902: Compare the real-time cooling load data with the cooling load data output by the prediction model to evaluate the deviation between the actual current load and the predicted value;

[0190] S903: Selecting relevant control parameters from the parameter sequence group according to the change trend of the real-time cooling load and the output of the prediction model;

[0191] S904: Calculate the parameter value that needs to be adjusted based on the deviation between the real-time load and the predicted load and the current operating state of the system;

[0192] S905: Send the calculated parameter value to the actuator of the ventilation and air conditioning control system to adjust the operating status of related equipment;

[0193] S906: Record the actual operation data of the ventilation and air conditioning system after adjusting the parameters, including the operation parameters of each device and the change of the cooling load inside the building;

[0194] S907: Use these data to draw curves of on-site operation, including a cooling load change curve and an equipment operation parameter change curve;

[0195] S908: Compare the field operation curve with the simulation curve generated based on the prediction model and the parameter sequence group, and analyze the difference and trend between the two;

[0196] S909: If there is a large difference between the on-site operation curve and the simulation curve, it means that the current control strategy or parameter setting is not accurate enough and needs to be adjusted;

[0197] S910: reselect or adjust parameter values ​​in the parameter sequence group, or optimize the control strategy according to the comparison result;

[0198] S911: repeating steps S904 to S910, and continuously iteratively adjusting parameters and control strategies until the field operation curve and the simulation curve are consistent or meet the preset accuracy requirements;

[0199] S912: After parameter adjustment and optimization are completed, system stability verification is performed for a period of time;

[0200] S913: Monitor the system's operating status and various parameter indicators to ensure that the system can respond stably and accurately when the load changes;

[0201] S914: If the system performance is stable and meets the requirements, the adjustment process is completed; if there are still problems, return to step S909 for further adjustment and optimization.

[0202] Continuously collect operating data and user feedback from the control system, analyze system performance, and regularly update the building cooling load model based on new data and feedback to improve prediction accuracy. The specific steps are as follows:

[0203] S1001: Deploy a data collection system to ensure that the operation data of the ventilation and air conditioning control system can be continuously and accurately collected, including equipment operation status, energy consumption data, and environmental parameters;

[0204] S1002: Organize, classify and store the collected operation data and user feedback to ensure the integrity and accessibility of the data;

[0205] S1003: Conduct in-depth analysis on the collected operating data, identify trends, anomalies and correlations in the data, and evaluate the prediction accuracy of the current cooling load model;

[0206] S1004: Based on the data analysis results, evaluate the performance of the current cooling load model, including prediction error, stability, and adaptability;

[0207] S1005: Update the cooling load model based on the evaluation results and new data input, including adjusting model parameters, improving algorithms, introducing new influencing factors, etc., to improve the prediction accuracy and robustness of the model;

[0208] S1006: Use the new data set to verify the updated model to ensure that the prediction performance of the updated model is improved;

[0209] S1007: Based on the model update results and actual needs, evaluate the current air conditioning system control strategy, analyze the performance of the strategy in terms of energy saving, comfort, equipment life, etc., and identify potential areas for improvement;

[0210] S1008: Develop an optimization plan based on the evaluation results, including adjusting the number of devices to be started, optimizing the load-on / load-off ratio, adjusting the setting values ​​of key parameters, etc., to improve the overall operating efficiency of the system;

[0211] S1009: Test the optimized control strategy in a simulation environment to evaluate its effectiveness in actual operation;

[0212] S1010: Deploy the optimized control strategy to the actual system and continuously monitor the system operation status and performance indicators.

[0213] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of those skilled in the art without departing from the spirit of the present invention.

Claims

1. A ventilation and air conditioning control method based on load forecasting; characterized in that: S1: Collect historical operating data from air-conditioning systems in places such as subways and airports, including but not limited to ambient temperature, humidity, passenger flow, and operating parameters of each component of the air-conditioning system; S2: Clean the collected data, remove outliers and missing values, etc., to ensure the accuracy and completeness of the data; organize the cleaned data in chronological order and divide it into training sets and test sets; S3: Multiple linear regression analysis is used as the modeling method. According to historical data, the key factors affecting the building cooling load are selected as independent variables, and the building cooling load is used as the dependent variable. Using the training set data, the mathematical model between the independent variable and the dependent variable is established through regression analysis to determine the coefficients between the variables. S4: According to the model prediction results, analyze the relationship between the operating parameters of each component of the air-conditioning system and the building cooling load, and fit their changes into a curve; S5: According to the fitted relationship curve, compile the operation control plan under typical working conditions; sequence the various parameters in the control strategy into a parameter sequence group as the input variable of the ventilation and air-conditioning control system; S6: deploy the control system and set the initial value of the control variable of a single device according to the parameter sequence group of the control model; S7: Monitor the changes in the building's cooling load in real time, select appropriate parameters from the parameter sequence group for adjustment according to the actual cooling load changes, and continuously compare the on-site operation curve and the simulation curve until the two are consistent, ensuring that the control system can accurately respond to changes in the building's cooling load; S8: Continuously collect operating data and user feedback from the control system, analyze system performance, and regularly update the building cooling load model based on new data and feedback to improve prediction accuracy.

2. A ventilation and air conditioning control method based on load forecasting according to claim 1, characterized in that: Collect historical operating data from air-conditioning systems in places such as subways and airports, including but not limited to ambient temperature, humidity, passenger flow, and operating parameters of each component of the air-conditioning system; the operating parameters of each component of the air-conditioning system include compressor power and fan speed; the specific steps are as follows: S201: Identify the data source target, i.e., the ventilation and air conditioning systems of large public buildings such as subways and airports; S202: Capturing equipment parameters and environmental data that affect cooling load in the air-conditioning system through sensors and data acquisition equipment; S203: extracting operation data for a period of time in the past from historical data, including but not limited to ambient temperature, humidity, passenger flow, and various operation parameters of the air conditioning system; S204: Create a detailed data log for the collected data, recording timestamps, data values, and any abnormal conditions; S205: Unify the data format and output it.

3. A ventilation and air conditioning control method based on load forecasting according to claim 2, characterized in that: Clean the collected data to remove outliers and missing values ​​to ensure the accuracy and completeness of the data: organize the cleaned data in chronological order and divide it into training sets and test sets for subsequent modeling and verification; the specific steps are as follows: S301: Select a filling method or directly delete records containing missing values ​​according to the data type and missing conditions; S302: For numerical data, if the missing ratio is not high, the mean or median can be used for filling; if the missing ratio is high, consider whether the column data has a significant impact on the model prediction. If so, the column can be deleted; S303: For categorical data, fill using mode or fill based on context information; S304: Record processing results; S305: Calculate the mean, standard deviation, median, IQR and other statistics of each column of data; S306: When using the standard deviation method, set a threshold, such as mean ± 3 times the standard deviation, and data outside this range are considered outliers; S307: When using the IQR method, calculate the first quartile of Q1 and the third quartile of Q3, then calculate IQR = Q3-Q1, and consider data less than Q1-1.5IQR or greater than Q3+1.5IQR as outliers; S308: Selectively delete the identified outliers; S309: Check and convert the date and time columns into a unified format; Check numeric columns to ensure there are no non-numeric characters and convert data types as needed; for categorical data, unify the encoding format; S310: Check the time column in the data set to ensure that the time format is uniform and correct; S311: sorting the data according to the time column; S312: Analyze historical data to identify features that are highly relevant to the prediction target; select the feature set that is ultimately used for model training based on constructed derived features, such as temperature difference, humidity difference, passenger flow change rate, etc.; S313: Divide the data set into a training set and a test set, and determine the division ratio; divide the data set using a time series segmentation method to ensure that the training set and the test set are consistent in data distribution; S314: Normalize or standardize the numerical features.

4. A ventilation and air conditioning control method based on load forecasting according to claim 3, characterized in that: The multivariate linear regression analysis method is used as the modeling method. According to historical data, the key factors affecting the building cooling load are selected as independent variables, and the building cooling load is used as the dependent variable. Using the training set data, the mathematical model between the independent variable and the dependent variable is established through regression analysis to determine the coefficients between the variables. The specific steps are as follows: S401: Based on the data analysis results, select variables that have a significant impact on the building cooling load; S402: Clarify the calculation method of building cooling load or directly use the existing cooling load data as the dependent variable; S403: Setting a multiple linear regression model and determining the basic form of the model; That is, Y=β0+β1X1+β2X2+…+βnXn+∈; Where Y is the building cooling load, Xi is the influencing factors, i.e., the independent variables, βi is the coefficient to be determined, and ∈ is the error term; S404: Calculate the average value of ventilation and air conditioning power during the building operation period based on historical data. The calculation formula is as follows: Steady-state building air conditioning basic cooling load Qd = average ventilation and air conditioning power during building operation period × air conditioning energy efficiency ratio COP; Qc=P×COP+Qn1+Qm1; Among them, Qc represents the cooling load of ventilation and air conditioning, P represents the electric power of ventilation and air conditioning, i.e. the load value monitored in real time by the energy management system, COP represents the energy efficiency ratio of air conditioning, Qn1 represents the cooling load caused by the change of fresh air volume demand, and Qm1 represents the cooling load caused by the change of passenger flow; S405: selecting dates with similar operating conditions to the forecast date from historical data, and using the electric load curves of these dates as the basis of the load curve of the forecast date; S406: Calculate the fresh air volume Fi according to the change in the number of people in the building N and the fresh air supply per person E. The calculation formula is as follows: Fi = N × E; S407: Calculate the cooling load Qn1 caused by the change in fresh air demand using the fresh air volume Fi, air density K, air specific heat Ra, fresh air temperature Ti and outflow air temperature To. The calculation formula is as follows: Qn1 = Fi × K × Ra × (Ti-To); S408: regularly obtain the latest passenger flow data from the passenger flow system, access control system, etc., and update the predicted values ​​of fresh air volume and fresh air cooling load; S409: Calculate the cooling load Qm1 caused by the change in the number of people in the building per unit time according to the change in the number of people N, the cluster coefficient A, the heat dissipation of human bodies Qr and the cooling load coefficient s of human body heat dissipation. The calculation formula is as follows: Qm1=N×A×Qr×s; S410: According to different building types, query the parameter values ​​such as A, Qr, s, etc. from relevant design specifications or theories.

5. A ventilation and air conditioning control method based on load forecasting according to claim 4, characterized in that: The multivariate linear regression analysis method is used as the modeling method. According to historical data, the key factors affecting the building cooling load are selected as independent variables, and the building cooling load is used as the dependent variable. The training set data is used to establish a mathematical model between the independent variable and the dependent variable through regression analysis, and the coefficients between the variables are determined; the following steps are also included: S411: Add the basic cooling load Qd of the steady-state building air conditioning, the cooling load Qn1 caused by the change in fresh air volume demand, and the cooling load Qm1 caused by the change in the flow of people, and obtain the total ventilation and air conditioning cooling load prediction value calculation formula Qc=Qd+Qn1+Qm1; S412: dynamically adjusting the predicted value based on real-time data, such as current crowd flow changes and fresh air temperature; S413: Use the test set data to verify the prediction model and evaluate the prediction accuracy and generalization ability; S414: Adjust and optimize the model according to the verification results.

6. A ventilation and air conditioning control method based on load forecasting according to claim 5, characterized in that: According to the model prediction results, the relationship between the operating parameters of each component of the air-conditioning system and the building cooling load is analyzed, and its changes are fitted into a curve. The specific steps are as follows: S601: Perform a preliminary fit on the selected model using the collected data to obtain initial model parameters; S602: Evaluate the fitting effect of the model by calculating statistical indicators of the fitting curve, such as R^2 value, mean square error (MSE), and root mean square error (RMSE); S603: Optimizing the model, adjusting model parameters through an iterative optimization algorithm to improve the fitting effect of the model; S604: Observe the shape and trend of the relationship curve obtained by fitting, and analyze the internal relationship between the independent variable and the dependent variable; S605: Analyze the shape and trend of the curve; for example, the curve may indicate that increasing the compressor power within a certain range can significantly increase the supply of cooling load, but the effect gradually decreases beyond a certain range.

7. A ventilation and air conditioning control method based on load forecasting according to claim 6, characterized in that: According to the fitted relationship curve, the operation control plan under typical working conditions is compiled; the various parameters in the control strategy are serialized into a parameter sequence group as the input variables of the ventilation and air-conditioning control system; The specific steps include: S701: Identify typical air-conditioning operating conditions, such as high-load conditions, low-load conditions, and transitional season conditions, based on historical data and actual operating conditions; S702: For each typical working condition, determine the optimal number of devices to be started under different working conditions based on the fitted relationship curve and the cooling load prediction results; consider the energy efficiency ratio of the equipment and the load distribution balance to minimize energy consumption and extend the life of the equipment; S703: According to the equipment characteristics and load change trend, set a reasonable load increase and decrease rate to smoothly adjust the equipment operation status and avoid frequent start and stop impact on the equipment; the load increase and decrease rate should comprehensively consider factors such as equipment response time, system stability and energy efficiency ratio; S704: Setting control target values ​​of key parameters such as outlet water temperature and outlet air temperature according to the building cooling load demand and air conditioning system performance; S705: converting the above control strategy into specific control logic, including equipment start and stop conditions, load addition and reduction logic, and parameter adjustment rules; S706: sort out various parameters in the control strategy, including the number of equipment turned on, load increase and decrease rate, and target values ​​of key parameters; S707: Serializing the parameters into a parameter sequence group in a certain format and order, and ensuring that the format of the parameter sequence group matches the input interface of the ventilation and air conditioning control system so that it can be smoothly transmitted to the control system; S708: Verify and test the serialized parameter sequence group to ensure that it can be correctly parsed and executed in the control system; S709: Using the serialized parameter sequence group as input variables of the ventilation and air-conditioning control system.

8. A ventilation and air conditioning control method based on load forecasting according to claim 7, characterized in that: Deploy the control system and set the initial value of the control variable of a single device according to the parameter sequence group of the control model. The specific steps are as follows: S801: Reading a serialized parameter sequence group from a storage medium; S802: Parse the parameter sequence group and convert it into a format and data structure that can be recognized by the control system; S803: Load the parsed parameter sequence group into the memory of the control system, ready for subsequent control logic processing; S804: traverse each parameter item in the parameter sequence group to identify parameters related to the control variables of a single device; S805: For each single device control variable, according to its corresponding parameter value, set the initial value of the variable in the control system; for example, if the initial operating frequency of a compressor is specified in the parameter sequence group, then set the frequency value as the initial operating frequency of the compressor; S806: Check whether the initial values ​​of all single device control variables have been set correctly to ensure that there are no omissions or errors.

9. A ventilation and air conditioning control method based on load forecasting according to claim 8, characterized in that: Monitor the changes in the building's cooling load in real time, select appropriate parameters from the parameter sequence group for adjustment according to the actual cooling load changes, and continuously compare the on-site operation curve and the simulation curve until the two are consistent, ensuring that the control system can accurately respond to changes in the building's cooling load; the specific steps are as follows: S901: Start the real-time monitoring system of the ventilation and air conditioning control system to continuously collect real-time cooling load data inside the building; S902: Compare the real-time cooling load data with the cooling load data output by the prediction model to evaluate the deviation between the actual current load and the predicted value; S903: Selecting relevant control parameters from the parameter sequence group according to the change trend of the real-time cooling load and the output of the prediction model; S904: Calculate the parameter value that needs to be adjusted based on the deviation between the real-time load and the predicted load and the current operating state of the system; S905: Send the calculated parameter value to the actuator of the ventilation and air conditioning control system to adjust the operating status of related equipment; S906: Record the actual operation data of the ventilation and air conditioning system after adjusting the parameters, including the operation parameters of each device and the change of the cooling load inside the building; S907: Use these data to draw curves of on-site operation, including a cooling load change curve and an equipment operation parameter change curve; S908: Compare the field operation curve with the simulation curve generated based on the prediction model and the parameter sequence group, and analyze the difference and trend between the two; S909: If there is a large difference between the on-site operation curve and the simulation curve, it means that the current control strategy or parameter setting is not accurate enough and needs to be adjusted; S910: reselect or adjust parameter values ​​in the parameter sequence group, or optimize the control strategy according to the comparison result; S911: repeating steps S904 to S910, and continuously iteratively adjusting parameters and control strategies until the field operation curve and the simulation curve are consistent or meet the preset accuracy requirements; S912: After parameter adjustment and optimization are completed, system stability verification is performed for a period of time; S913: Monitor the system's operating status and various parameter indicators to ensure that the system can respond stably and accurately when the load changes; S914: If the system performance is stable and meets the requirements, the adjustment process is completed; if there are still problems, return to step S909 for further adjustment and optimization.

10. A ventilation and air conditioning control method based on load forecasting according to claim 9, characterized in that: Continuously collect operating data and user feedback from the control system, analyze system performance, and regularly update the building cooling load model based on new data and feedback to improve prediction accuracy. The specific steps are as follows: S1001: Deploy a data collection system to ensure that the operation data of the ventilation and air conditioning control system can be continuously and accurately collected, including equipment operation status, energy consumption data, and environmental parameters; S1002: Organize, classify and store the collected operation data and user feedback to ensure the integrity and accessibility of the data; S1003: Conduct in-depth analysis on the collected operating data, identify trends, anomalies and correlations in the data, and evaluate the prediction accuracy of the current cooling load model; S1004: Based on the data analysis results, evaluate the performance of the current cooling load model, including prediction error, stability, and adaptability; S1005: Update the cooling load model based on the evaluation results and new data input, including adjusting model parameters, improving algorithms, introducing new influencing factors, etc., to improve the prediction accuracy and robustness of the model; S1006: Use the new data set to verify the updated model to ensure that the prediction performance of the updated model is improved; S1007: Based on the model update results and actual needs, evaluate the current air conditioning system control strategy, analyze the performance of the strategy in terms of energy saving, comfort, equipment life, etc., and identify potential areas for improvement; S1008: Develop an optimization plan based on the evaluation results, including adjusting the number of devices to be started, optimizing the load-on / load-off ratio, adjusting the setting values ​​of key parameters, etc., to improve the overall operating efficiency of the system; S1009: Test the optimized control strategy in a simulation environment to evaluate its effectiveness in actual operation; S1010: Deploy the optimized control strategy to the actual system and continuously monitor the system operation status and performance indicators.

Images