Air conditioner power prediction and control method based on comprehensive intelligent zero-carbon power plant cloud platform

Abstract

The invention relates to an air conditioner power prediction and control method based on a comprehensive intelligent zero-carbon power plant cloud platform. The method comprises the steps that S1, historical operation data and real-time operation data of an air conditioner and a temperature and humidity sensor are collected; s2, on the basis of the actual layout of the air conditioner and the structure of the room, a multi-level neural network model is constructed and used for predicting the power of the air conditioner; s3, training the neural network model by adopting historical operation data; s4, establishing an optimization algorithm objective function of the optimal air conditioner control strategy; s5, determining an optimal air conditioner control strategy through an optimization algorithm; s6, deploying the neural network model and an air conditioner control strategy optimization algorithm on a comprehensive intelligent zero-carbon power plant cloud platform; and S7, after the comprehensive intelligent zero-carbon power plant cloud platform receives a demand response instruction issued by the power grid, an optimal control instruction is obtained through an embedded algorithm, and the instruction is sent to each air conditioner device. Compared with the prior art, the method has the advantages of low calculation cost, more accurate prediction effect and the like.

Description

TECHNICAL FIELD

[0001] The present application relates to an air conditioner power prediction and control method, in particular to an air conditioner power prediction and control method based on a comprehensive intelligent zero-carbon power plant cloud platform. BACKGROUND

[0002] Power supply and demand balance is very important for power grid safety. At present, virtual power plants participating in power grid demand response has become a way to cut peak and fill valley for power grid. Air conditioners, as one of the main power consumption equipment in shopping malls, office buildings and other scenarios, are a very potential virtual power plant regulation resource. Connecting air conditioner resources to the comprehensive intelligent zero-carbon power plant platform, and adjusting the operation parameters of the air conditioner through the platform to predict and control the power of the air conditioner, is an important research topic.

[0003] In the prior art, air conditioner power prediction and control models are generally divided into neural network models and models derived through heat transfer mechanisms. Through retrieval, Chinese patent publication No. CN120777687A discloses an air conditioner control method, device, equipment and medium, and specifically proposes a prediction control method that uses a neural network model, combines human comfort indicators, and uses a genetic algorithm for optimization. At the same time, Chinese patent publication No. CN120292677A discloses an air conditioner cooling load prediction method and system, and specifically proposes a mechanism for constructing a prediction feature matrix to predict air conditioner power, based on building microclimate characteristics, user behavior patterns, and wall dynamic thermal properties, using dynamic thermal resistance calculation to construct a wall thermal resistance model and calculate building instantaneous heat flow data.

[0004] For the method of the prior art neural network model, the problem is that the structure of the neural network model itself is not combined with air conditioner layout, room structure and other information. At the same time, the optimization algorithm does not consider the weights of different air conditioners and rooms, which reduces the rationality of the control strategy. For the model method of the mechanism, the current problem is that the heat transfer model is complex and the modeling cost is high.

[0005] Therefore, how to reduce the calculation cost while improving the prediction accuracy has become a technical problem to be solved. SUMMARY

[0006] The purpose of the present application is to overcome the defects of the prior art and provide an air conditioner power prediction and control method based on a comprehensive intelligent zero-carbon power plant cloud platform, which has low calculation cost and more accurate prediction effect.

[0007] The purpose of the present application can be achieved by the following technical solutions: According to one aspect of the present application, an air conditioner power prediction and control method based on a comprehensive intelligent zero-carbon power plant cloud platform is provided, which specifically comprises: Step S1: Collect historical and real-time operating data of air conditioning and temperature and humidity sensors through the integrated smart zero-carbon power plant cloud platform; Step S2: Based on the actual layout of the air conditioner and the structure of the room, construct a multi-level neural network model to predict the air conditioner power. Step S3: Train the neural network model using the historical running data from step S1; Step S4: Establish the objective function of the optimization algorithm for the optimal air conditioning control strategy; Step S5: Determine the optimal air conditioning control strategy using the optimization algorithm established in step S4. Step S6: Deploy the neural network model and air conditioning control strategy optimization algorithm on the integrated smart zero-carbon power plant cloud platform; Step S7: After receiving the demand response command issued by the power grid, the integrated smart zero-carbon power plant cloud platform obtains the optimal control command through the embedded algorithm and sends the command to each air conditioning device.

[0008] As a preferred technical solution, step S1 specifically includes: Step S101: Install and configure the wireless communication gateway on site; Step S102: Collect historical and real-time operating data of air conditioner set temperature, set operating mode, set fan speed, indoor temperature, indoor humidity and air conditioner operating power, and send them to the integrated smart zero-carbon power plant cloud platform through the wireless communication gateway. Step S103: Collect historical and real-time operating data of indoor temperature, indoor humidity, outdoor temperature and outdoor humidity from the temperature and humidity sensor, and upload them to the integrated smart zero-carbon power plant cloud platform via a wireless communication gateway.

[0009] As a preferred technical solution, in step S101, the wireless communication gateway is configured with IP and port as the IP and port of the target integrated smart zero-carbon power plant cloud server; and the on-site air conditioning equipment is connected to the wireless communication gateway through shielded twisted pair cable or network cable.

[0010] As a preferred technical solution, step S2 specifically includes: In step S201, each independent air conditioner constitutes a first-level neural network, and the parameters of each air conditioner and temperature and humidity sensor are used as input parameters of the input layer of their respective neural network models. In step S202, the air conditioners in the same room constitute the second-level neural network. For each second-level neural network model, the output of the output layer of the neural network model corresponding to the air conditioner in step S201 is used as the input parameter of the input layer of the second-level neural network model. In step S203, the interconnected rooms constitute a third-level neural network. For each third-level neural network model, the output of the neural network model output layer of the corresponding room in step S202 is used as the input parameter of the input layer of the third-level neural network model. In step S204, if the room structure has multiple levels, the neural network model is constructed in the manner described in step S203.

[0011] As a preferred technical solution, the neural network model at each level contains at least three hidden layers, each hidden layer contains several neurons, each layer is fully connected to the previous layer, a bias is added, and ReLU is used as the activation function.

[0012] As a preferred technical solution, step S3 specifically includes: Step S301: The air conditioner set temperature, set operating mode, set fan speed, indoor temperature and indoor humidity processed in step S1, as well as the historical data of indoor temperature, indoor humidity, outdoor temperature and outdoor humidity from the temperature and humidity sensor, are used as input parameters in the training sample, and the historical data of the air conditioner operating power curve within one hour are used as output parameters in the training sample. Step S302, let the first... The weight matrix of the hidden layer is , bias is The input vector is The output vector is Then the forward propagation function of this layer is The input dimension of each hidden layer is... The output dimension is ; Step S303: The training loss function uses mean squared error; Step S304: Use the optimization function to perform backpropagation and update the loss value to the weight parameters of each hidden layer; Step S305: Iterate through the above steps to train the model until the model loss is less than the preset value or the number of iterations reaches the preset upper limit.

[0013] As a preferred technical solution, the objective function of the optimization algorithm in step S4 includes the deviation between the predicted power and the target power under different air conditioning control commands, human comfort, and the weight of each room.

[0014] As a preferred technical solution, the scoring of the deviation between the predicted power and the target power under different air conditioning control commands is specifically as follows: Let the rating coefficient be... Let the air conditioning control command parameters and temperature and humidity parameters be used as inputs to the neural network model in step 3), to obtain the predicted power curve data for the next hour. Let the target value of the demand response be... Then the score for this item is... for The specific human comfort score is as follows: Let the rating coefficient be... Let the number of people in each room be... Let the air conditioning control command parameters for each room be... Let the optimal control command parameters for each room be... Then the score for this item is... for The specific weighted scores for each room are as follows: Let the rating coefficient be... Let the importance weight of each room be . Let the air conditioning control command parameters for each room be... Let the optimal control command parameters for each room be... Then the score for this item is... for The objective function of the optimization algorithm is

[0015] As a preferred technical solution, step S5 specifically includes: Step S501: Establish the initial population using the DE differential evolution algorithm. It contains NP individuals , The dimension is the air conditioning control command parameter. To generate randomly within their respective parameter domains; Step S502: Perform mutation operations on the individual using the DE differential evolution algorithm. The variant individuals are , in , , Three distinct individuals are randomly selected from the population. It is a variable factor; Step S503: Using the DE differential evolution algorithm, crossover operations are performed on the mutated individuals to generate experimental individuals. in Crossover probability factor; Step S504: Using a greedy algorithm, the original individual and the experimental individual are respectively substituted into the objective function for calculation, and the individual with the better objective function result is selected as the next generation; Step S505: Repeat the above steps until the number of algorithm iterations reaches the predetermined maximum number, or the population optimal solution reaches the predetermined accuracy, to obtain the optimal air conditioning control strategy parameters.

[0016] As a preferred technical solution, step S7 specifically includes: Step S701: Obtain the demand response instructions of the virtual power plant from the integrated smart zero-carbon power plant cloud platform; Step S702: The real-time operating status of the air conditioner and the real-time data of the temperature and humidity sensors collected by the integrated smart zero-carbon power plant cloud platform, as well as the target response quantity of the demand response, are substituted into the air conditioner control strategy optimization algorithm to obtain the optimal air conditioner control strategy. Step S703: Send the optimal control command to each air conditioning unit.

[0017] Compared with the prior art, the present invention has the following advantages: 1) This invention uses the actual location of the air conditioner and the room layout to construct a neural network model. Compared with directly using the parameters of all air conditioners as the parameters of the same input layer of the neural network, it can remove the mutual interference between air conditioners or rooms that are far apart, making the training results more realistic and more accurate. 2) This invention uses the deviation between the predicted power and the target power after air conditioning control, human comfort, and the weight of each room as the objective function for optimizing the air conditioning control strategy. This allows the final air conditioning control command to simultaneously satisfy the effects of accuracy when the air conditioning participates in grid demand response, user comfort during air conditioning adjustment, and the rationality of adopting different strategies for rooms of different importance. 3) This invention utilizes a smart zero-carbon power plant cloud platform to deploy air conditioning power prediction and control algorithms for real-time calculation and control, which can reduce the cost of purchasing, installing and deploying on-site equipment, and enable air conditioning to participate in grid demand response as an aggregated resource at a lower cost. Attached Figure Description

[0018] Figure 1 This is a flowchart of the model construction process of the present invention; Figure 2 This is a flowchart illustrating the specific control process of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] This invention provides an air conditioning power prediction model and control method based on a multi-level neural network deployed on a comprehensive smart zero-carbon power plant cloud platform. The method collects historical and real-time data from the air conditioning system through the cloud platform. Simultaneously, based on the actual layout of the air conditioning units and the room structure, the neural network model is divided into multiple layers. The air conditioning operating status and variables such as temperature and humidity are used as input parameters for the model's perception layer, and the power curve is used as the output parameter for the output layer. Historical data is used for model training. Furthermore, this method scores the effectiveness of different control commands based on indicators such as the deviation between predicted and target power under different air conditioning control commands, human comfort, and the weight of each room. The scoring result serves as the objective function of an optimization algorithm. Through optimization, the most reasonable air conditioning control scheme that meets the target power for demand response adjustment can be obtained. After receiving the grid demand response command, the comprehensive smart zero-carbon power plant cloud platform distributes the optimal control command obtained by this control algorithm to each air conditioning unit, realizing the function of air conditioning as an aggregated resource participating in grid demand response.

[0021] The neural network model of this invention combines the actual layout of air conditioners and rooms, resulting in simple modeling, low computational cost, and more accurate prediction. Furthermore, the optimization algorithm for control commands employs comfort indices and weighted indices for each room, making the control commands more rational. In addition, deploying air conditioning power prediction and control algorithms on a comprehensive smart zero-carbon power plant cloud platform for real-time calculation and control can reduce the costs of purchasing, installing, and deploying on-site equipment.

[0022] like Figure 1 and Figure 2 As shown, the specific implementation process of the present invention is as follows: 1) Collect historical and real-time operating data of air conditioners and temperature and humidity sensors through the integrated smart zero-carbon power plant cloud platform.

[0023] ① Install and configure the wireless communication gateway on-site. Configure the IP address and port of the wireless communication gateway to match the IP address and port of the target integrated smart zero-carbon power plant cloud server. Connect the on-site air conditioning equipment to the wireless communication gateway using shielded twisted-pair cables or network cables; the specific connection method depends on the air conditioning model.

[0024] ② Collect historical and real-time operating data of the air conditioner's set temperature, operating mode, fan speed, indoor temperature, indoor humidity, and operating power. The collected data is then transmitted to the integrated smart zero-carbon power plant cloud platform via a wireless communication gateway.

[0025] ③ Collect historical and real-time operating data of indoor temperature, indoor humidity, outdoor temperature, and outdoor humidity from temperature and humidity sensors. The collected data is then transmitted to the integrated smart zero-carbon power plant cloud platform via a wireless communication gateway.

[0026] At the same time, all data are normalized, and erroneous data are corrected and noise is removed by interpolation and other methods.

[0027] 2) Based on the actual layout of the air conditioner and the room structure, construct a neural network model.

[0028] ① Each independent air conditioner constitutes its own first-level neural network, and the parameters of each air conditioner and temperature and humidity sensor are used as the input parameters of the input layer of their respective neural network models.

[0029] ② The air conditioners in the same room constitute the second-level neural network. For each second-level neural network model, the output of the corresponding air conditioner neural network model output layer in ① is used as the input parameter of the input layer of the second-level neural network model.

[0030] ③ The interconnected rooms constitute a third-level neural network. For each third-level neural network model, the output of the neural network model output layer of the corresponding room in ② is used as the input parameter of the input layer of the third-level neural network model.

[0031] ④ If the room structure has multiple levels, continue building the neural network model in the manner described in ③, and so on.

[0032] Each of the above neural network models should contain at least three hidden layers, each containing several neurons. Each layer should be fully connected to the previous layer, with a bias applied, and ReLU should be used as the activation function.

[0033] 3) Train a multi-level neural network model based on historical data.

[0034] The historical data of the air conditioner's set temperature, set operating mode, set fan speed, indoor temperature, indoor humidity, and outdoor temperature and humidity from the temperature and humidity sensor (processed in step 1) are used as input parameters in the training samples, and the historical data of the air conditioner's operating power curve over one hour are used as output parameters in the training samples.

[0035] Let the input dimension of each hidden layer be . The output dimension is The input dimension of the first hidden layer is determined by the dimension of the model input parameters, the input dimension of the remaining hidden layers is the number of neurons in the previous hidden layer, the output dimension of the last hidden layer is determined by the dimension of the model output parameters, and the output dimension of the remaining hidden layers is the number of neurons in the next hidden layer.

[0036] Let the first The weight matrix of the hidden layer is , bias is The input vector is The output vector is Then the forward propagation function of this layer is The training loss function uses mean squared error, which is calculated by comparing the predicted power curve for the next hour with the actual historical power curve for the next hour.

[0037] Backpropagation is performed using an optimization function to update the loss values ​​to the weight parameters of each hidden layer.

[0038] Continue iterating through the above steps to train the model until the model loss is less than a preset value or the number of iterations reaches a preset limit.

[0039] The trained multi-level neural network model needs to be saved for subsequent steps to calculate the predicted power curve of the air conditioning strategy.

[0040] 4) Establish the objective function of the optimization algorithm for the optimal air conditioning control strategy.

[0041] The objective function of the optimization algorithm consists of the following three parts: the deviation between the predicted power and the target power under different air conditioning control commands, human comfort, and the weight of each room.

[0042] For scoring the deviation between predicted power and target power under different air conditioning control commands, let the scoring coefficient be... Let the air conditioning control command parameters and temperature and humidity parameters be used as inputs to the neural network model in step 3), to obtain the predicted power curve data for the next hour. Let the target value of the demand response be... Then the score for this item is... for For human comfort rating, let the rating coefficient be... Let the number of people in each room be... Let the air conditioning control command parameters for each room be... Let the optimal control command parameters for each room be... Then the score for this item is... for For the weighted scoring of each room, let the scoring coefficient be... Let the importance weight of each room be . Let the air conditioning control command parameters for each room be... Let the optimal control command parameters for each room be... Then the score for this item is... for The objective function of the optimization algorithm is then: 5) Determine the optimal air conditioning control strategy through an optimization algorithm.

[0043] ① Establish the initial population using the DE differential evolution algorithm. It contains NP individuals , The dimension is the air conditioning control command parameter. To generate randomly within their respective parameter domains.

[0044] ② The individual is mutated using the DE differential evolution algorithm. The variant individuals are , In the formula: , , Three distinct individuals are randomly selected from the population. It is a variable factor.

[0045] ③ Using the DE differential evolution algorithm, crossover operations are performed on the mutated individuals to generate experimental individuals. In the formula: This is the crossover probability factor.

[0046] ④ Using a greedy algorithm, the original individual and the experimental individual are respectively substituted into the objective function for calculation, and the individual with the better objective function result is selected as the next generation.

[0047] ⑤ Repeat the above steps until the algorithm reaches the predetermined maximum number of iterations, or the population optimal solution reaches the predetermined accuracy, to obtain the optimal air conditioning control strategy parameters.

[0048] 6) Deploy the air conditioning power prediction neural network model and air conditioning control strategy optimization algorithm on the integrated smart zero-carbon power plant cloud platform.

[0049] 7) After receiving the demand response command issued by the power grid, the integrated smart zero-carbon power plant cloud platform issues the optimal control command obtained by the control algorithm to each air conditioning device, so as to realize the function of air conditioning as an aggregated resource to participate in the power grid demand response.

[0050] ①The integrated smart zero-carbon power plant cloud platform obtains demand response instructions from the virtual power plant through the interface with the power grid operation and management platform.

[0051] ② The real-time operating status of the air conditioner, the real-time data of the temperature and humidity sensors, and the target response quantity of the demand response collected by the integrated smart zero-carbon power plant cloud platform are substituted into the air conditioner control strategy optimization algorithm to obtain the optimal air conditioner control strategy.

[0052] ③ Based on the air conditioning equipment communication protocol, the optimal control command is sent to each air conditioning unit.

[0053] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform, characterized in that, The method specifically includes: Step S1: Collect historical and real-time operating data of air conditioning and temperature and humidity sensors through the integrated smart zero-carbon power plant cloud platform; Step S2: Based on the actual layout of the air conditioner and the structure of the room, construct a multi-level neural network model to predict the air conditioner power. Step S3: Train the neural network model using the historical running data from step S1; Step S4: Establish the objective function of the optimization algorithm for the optimal air conditioning control strategy; Step S5: Determine the optimal air conditioning control strategy using the optimization algorithm established in step S4. Step S6: Deploy the neural network model and air conditioning control strategy optimization algorithm on the integrated smart zero-carbon power plant cloud platform; Step S7: After receiving the demand response command issued by the power grid, the integrated smart zero-carbon power plant cloud platform obtains the optimal control command through the embedded algorithm and sends the command to each air conditioning device.

2. The method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform according to claim 1, characterized in that, Step S1 specifically involves: Step S101: Install and configure the wireless communication gateway on site; Step S102: Collect historical and real-time operating data of air conditioner set temperature, set operating mode, set fan speed, indoor temperature, indoor humidity and air conditioner operating power, and send them to the integrated smart zero-carbon power plant cloud platform through the wireless communication gateway. Step S103: Collect historical and real-time operating data of indoor temperature, indoor humidity, outdoor temperature and outdoor humidity from the temperature and humidity sensor, and upload them to the integrated smart zero-carbon power plant cloud platform via a wireless communication gateway.

3. The method for predicting and controlling air conditioning power based on a comprehensive intelligent zero-carbon power plant cloud platform according to claim 2, characterized in that, In step S101, the wireless communication gateway is configured with IP and port as the IP and port of the target integrated smart zero-carbon power plant cloud server; and the on-site air conditioning equipment is connected to the wireless communication gateway through shielded twisted pair cable or network cable.

4. The method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform according to claim 1, characterized in that, Step S2 specifically involves: In step S201, each independent air conditioner constitutes a first-level neural network, and the parameters of each air conditioner and temperature and humidity sensor are used as input parameters of the input layer of their respective neural network models. In step S202, the air conditioners in the same room constitute the second-level neural network. For each second-level neural network model, the output of the output layer of the neural network model corresponding to the air conditioner in step S201 is used as the input parameter of the input layer of the second-level neural network model. In step S203, the interconnected rooms form a third-level neural network. For each third-level neural network model, the output of the neural network model output layer of the corresponding room in step S202 is used as the input parameter of the input layer of the third-level neural network model. In step S204, if the room structure has multiple levels, the neural network model is constructed in the manner described in step S203.

5. The method for predicting and controlling air conditioning power based on a comprehensive intelligent zero-carbon power plant cloud platform according to claim 4, characterized in that, The neural network model at each level contains at least three hidden layers, each containing several neurons. Each layer is fully connected to the layer above it, biased, and uses ReLU as the activation function.

6. The method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform according to claim 1, characterized in that, Step S3 specifically involves: Step S301: The air conditioner set temperature, set operating mode, set fan speed, indoor temperature and indoor humidity processed in step S1, as well as the historical data of indoor temperature, indoor humidity, outdoor temperature and outdoor humidity from the temperature and humidity sensor, are used as input parameters in the training sample, and the historical data of the air conditioner operating power curve within one hour are used as output parameters in the training sample. Step S302, let the first... The weight matrix of the hidden layer is , bias is The input vector is The output vector is Then the forward propagation function of this layer is The input dimension of each hidden layer is... The output dimension is ; Step S303: The training loss function uses mean squared error; Step S304: Use the optimization function to perform backpropagation and update the loss value to the weight parameters of each hidden layer; Step S305: Iterate through the above steps to train the model until the model loss is less than the preset value or the number of iterations reaches the preset upper limit.

7. The method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform according to claim 1, characterized in that, The objective function of the optimization algorithm in step S4 includes the deviation between the predicted power and the target power under different air conditioning control commands, human comfort, and the weight of each room.

8. The method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform according to claim 7, characterized in that, The specific scoring of the deviation between predicted power and target power under different air conditioning control commands is as follows: Let the rating coefficient be... Let the air conditioning control command parameters and temperature and humidity parameters be used as inputs to the neural network model in step 3), to obtain the predicted power curve data for the next hour. Let the target value of the demand response be... Then the score for this item is... for The specific human comfort score is as follows: Let the rating coefficient be... Let the number of people in each room be... Let the air conditioning control command parameters for each room be... Let the optimal control command parameters for each room be... Then the score for this item is... for The specific weighted scores for each room are as follows: Let the rating coefficient be... Let the importance weight of each room be . Let the air conditioning control command parameters for each room be... Let the optimal control command parameters for each room be... Then the score for this item is... for The objective function of the optimization algorithm is 9. The method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform according to claim 1, characterized in that, Step S5 specifically involves: Step S501: Establish the initial population using the DE differential evolution algorithm. It contains NP individuals , The dimension is the air conditioning control command parameter. To generate randomly within their respective parameter domains; Step S502: Perform mutation operations on the individual using the DE differential evolution algorithm. The variant individuals are ， in , , Three distinct individuals are randomly selected from the population. It is a variable factor; Step S503: Using the DE differential evolution algorithm, crossover operations are performed on the mutated individuals to generate experimental individuals. in Crossover probability factor; Step S504: Using a greedy algorithm, the original individual and the experimental individual are respectively substituted into the objective function for calculation, and the individual with the better objective function result is selected as the next generation; Step S505: Repeat the above steps until the number of algorithm iterations reaches the predetermined maximum number, or the population optimal solution reaches the predetermined accuracy, to obtain the optimal air conditioning control strategy parameters.

10. The method for predicting and controlling air conditioning power based on a comprehensive smart zero-carbon power plant cloud platform according to claim 1, characterized in that, Step S7 specifically involves: Step S701: Obtain the demand response instructions of the virtual power plant from the integrated smart zero-carbon power plant cloud platform; Step S702: The real-time operating status of the air conditioner and the real-time data of the temperature and humidity sensors collected by the integrated smart zero-carbon power plant cloud platform, as well as the target response quantity of the demand response, are substituted into the air conditioner control strategy optimization algorithm to obtain the optimal air conditioner control strategy. Step S703: Send the optimal control command to each air conditioning unit.

Images