Cold station power control method and device and electronic equipment

Abstract

The invention relates to the technical field of cold station regulation and control, and discloses a cold station power control method and device and electronic equipment, and the method comprises the steps: obtaining the target control power of a cold station; predicting future overall control parameters of the cold station based on the environmental parameters, the target control power and the operation data of the cold station; based on the overall control parameters of the cold station, power prediction data and comfort prediction data of the cold station are obtained; obtaining control parameter basic data of the cold station according to the power prediction data, the comfort degree prediction data and the overall control parameters of the cold station; and controlling the cold station to operate based on the control parameter basic data. Through the method, the operation power of the cold station can be effectively controlled while the comfort level is considered, and the utilization efficiency of electric power resources is improved.

Description

Technical Field

[0001] The present application relates to the technical field of cooling station control, and in particular to a cooling station power control method, device and electronic equipment. Background Art

[0002] With the rapid development of power systems in recent years, demand-side response and flexible load regulation have become a focus of attention. During peak load periods, appropriately reducing the power consumption of some loads to achieve target power response has significant application value. This regulation method not only effectively alleviates grid pressure but also improves the efficiency of power resource utilization, offering broad market prospects and enormous development potential.

[0003] Central air-conditioning cooling stations, as systems commonly operated in the summer, play an important role in load-side power regulation. Aggregating a large number of adjustable load resources, such as cooling stations, can generate a large amount of load adjustment space, effectively addressing the contradiction between electricity supply and demand. However, related power control technologies have many limitations: on the one hand, related measures often prioritize the overall building performance, resulting in a large degree of sacrifice in human comfort, making it difficult to achieve optimal comfort during the adjustment process; on the other hand, existing technologies have low efficiency and poor effect in regulating cooling station power, making it difficult to accurately balance the relationship between power control and comfort, and often resulting in problems such as lag in regulation and insensitive response. In addition, the lack of a systematic control strategy results in insufficient flexibility and adaptability in cooling station power regulation, making it impossible to effectively respond to complex and changing power demands and environmental conditions.

[0004] Therefore, achieving efficient and precise regulation of cooling station power while minimizing the impact on human comfort remains a critical issue. A breakthrough in this area will significantly improve the overall efficiency of cooling station power regulation, providing important support for the stable operation of the power system and the optimal allocation of resources. Summary of the Invention

[0005] The embodiments of the present application aim to solve at least one of the technical problems in the related art to a certain extent. To this end, the embodiments of the present application provide a cooling station power control method, device, electronic device, computer product, and medium.

[0006] An embodiment of the present application provides a cooling station power control method, the method comprising: obtaining a target control power of the cooling station; predicting future overall control parameters of the cooling station based on environmental parameters, the target control power, and cooling station operation data; obtaining power prediction data and comfort prediction data of the cooling station based on the overall control parameters of the cooling station; obtaining basic control parameter data of the cooling station based on the power prediction data, the comfort prediction data, and the overall control parameters of the cooling station; and controlling the operation of the cooling station based on the basic control parameter data.

[0007] In some embodiments, basic data of control parameters of the cooling station are obtained based on power prediction data, comfort prediction data and overall control parameters of the cooling station, including: when the power prediction data is within the range of target control power and the comfort prediction data is within a reasonable comfort range, the overall control parameters of the cooling station are determined as the basic data of control parameters of the cooling station; when the power prediction data is not within the range of target control power, or the comfort prediction data is not within a reasonable comfort range, the basic data of control parameters of the cooling station are re-predicted based on the power prediction data and the comfort prediction data.

[0008] In some embodiments, the control parameter basic data includes multiple control parameter basic values corresponding to multiple time periods; controlling the operation of the cold station based on the control parameter basic data includes: obtaining the overall control parameters corresponding to the current time period based on the control parameter basic values corresponding to the current time period; when controlling the operation of the cold station based on the overall control parameters corresponding to the current time period, obtaining the overall control parameters corresponding to the next time period according to the actual power of operation and the current environmental parameters; when the next time period arrives, controlling the operation of the cold station based on the overall control parameters corresponding to the next time period.

[0009] In some embodiments, when the operation of the cooling station is controlled based on the overall control parameters corresponding to the current time period, the overall control parameters corresponding to the next time period are obtained according to the actual operating power and the current environmental parameters, including: when the actual power is within the range of the target control power and the current environmental parameters are within the preset environmental threshold range, the basic value of the control parameter corresponding to the next time period is determined as the overall control parameter corresponding to the next time period; when the actual power is not within the range of the target control power, or the current environmental parameters are not within the preset environmental threshold range, the overall control parameters corresponding to the next time period are re-predicted based on the target data from the base time period to the current time period, wherein the target data includes at least one of the basic control parameter data, the operation data of the cooling station, the environmental data, and the comfort data.

[0010] In some embodiments, the power prediction data of the cold station is predicted by a power prediction model, which includes multiple long and short-term memory network layers and fully connected layers. The power prediction model is trained in the following manner: obtaining power sample data and power labels corresponding to the power sample data, wherein the power sample data includes outdoor temperature, outdoor humidity, and overall control parameters; using multiple long and short-term memory network layers and fully connected layers to process the power sample data to obtain a power prediction result; determining the error between the power prediction result and the power label by means of a root mean square error, and adjusting the model parameters of the power prediction model based on the error to obtain a trained power prediction model.

[0011] In some embodiments, the comfort prediction data of the cooling station is obtained through a comfort prediction model, the power prediction model includes multiple long short-term memory network layers and fully connected layers, and the comfort prediction model is trained in the following manner: obtaining comfort sample data and comfort labels corresponding to the comfort sample data, wherein the comfort sample data includes at least one of outdoor temperature, outdoor humidity, overall control parameters, and weighted number of manual operations, wherein the weighted number of manual operations includes the weighted average number of heating operations, cooling operations, and wind speed adjustment operations; using multiple long short-term memory network layers and fully connected layers to process the comfort sample data to obtain a comfort prediction result; determining the error between the comfort prediction result and the comfort label by means of a root mean square error method, and adjusting the model parameters of the comfort prediction model based on the error to obtain a trained comfort prediction model.

[0012] In some embodiments, the basic data of the control parameters of the cold station is obtained through a reinforcement learning model, and the reinforcement learning model is trained in the following manner: obtaining historical data of the cold station, wherein the historical data of the cold station includes at least one of historical basic data of control parameters, historical operation data, historical outdoor temperature, historical outdoor humidity and historical comfort data; training the reinforcement learning model based on the historical data of the cold station.

[0013] In some embodiments, a reinforcement learning model is trained based on historical data of the cold station, including: training a reinforcement learning model according to a reward function based on the historical data of the cold station, wherein the reward function is associated with at least one of the cold station efficiency reward function, the comfort sample data penalty function, the constraint penalty function, and the final reward function; wherein the cold station efficiency reward function is associated with the real-time cooling capacity, real-time power, and reward coefficient of the cold station; wherein the comfort sample data penalty function is associated with the weighted number of manual operations and the weighted number of operations penalty coefficient; wherein the constraint penalty function is associated with the demand power, real-time power, and penalty coefficient of the cold station.

[0014] An embodiment of the present application provides a power control device for a cooling station, which includes: an acquisition module for acquiring the target control power of the cooling station; a prediction module for predicting future overall control parameters of the cooling station based on environmental parameters, target control power and cooling station operation data, and also for obtaining power prediction data and comfort prediction data of the cooling station based on the overall control parameters of the cooling station; an inference module for obtaining basic control parameter data of the cooling station based on the power prediction data, comfort prediction data and overall control parameters of the cooling station; and a control module for controlling the operation of the cooling station based on the basic control parameter data of the cooling station.

[0015] An embodiment of the present application provides an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions that can be executed by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method of any of the above embodiments.

[0016] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method of any of the above embodiments are implemented.

[0017] An embodiment of the present application provides a computer program product, including a computer program. When the computer program is executed by a processor, the steps of the method according to any one of the above embodiments are implemented.

[0018] The cooling station power control method provided in the embodiments of the present application can achieve high-precision adjustment of cooling station power by obtaining the power control curve in advance, ensuring that the power output meets the target control power range. At the same time, based on environmental parameters and cooling station operation data, it dynamically predicts the future overall control parameters of the cooling station, significantly improving the system response speed and prediction accuracy. Through the comfort prediction model, this method effectively maintains the indoor environment within a reasonable range of human comfort while meeting the target control power range. When the power or comfort prediction data does not meet the requirements, the cooling station overall control parameters are optimized through multiple reasoning to achieve a dynamic balance between power and comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic flow chart of a cooling station power control method provided in an embodiment of the present application;

[0020] Figure 2 A flowchart illustrating the implementation of a cooling station power control method according to an embodiment of the present application;

[0021] Figure 3 A schematic diagram of a power control device for a cooling station according to an embodiment of the present application;

[0022] Figure 4 A block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0024] With the rapid development of power systems in recent years, demand-side response and flexible load regulation have become a focus of attention. During peak load periods, appropriately reducing the power consumption of some loads to achieve target power response has significant application value. This regulation method not only effectively alleviates grid pressure but also improves the efficiency of power resource utilization, offering broad market prospects and enormous development potential.

[0025] Central air conditioning cooling stations, as systems commonly operated in the summer, play an important role in load-side power regulation. Aggregating a large number of adjustable load resources, such as cooling stations, can generate a large amount of load adjustment space, effectively addressing the contradiction between electricity supply and demand. However, the relevant power flexible regulation technology has many limitations: on the one hand, the relevant means often prioritize the overall building performance, resulting in a large degree of sacrifice in human comfort, making it difficult to achieve optimal comfort during the adjustment process; on the other hand, the existing technology has low efficiency and poor effect in regulating cooling station power, making it difficult to accurately balance the relationship between power control and comfort, and often resulting in problems such as lag in regulation and insensitive response. In addition, the lack of a systematic control strategy leads to insufficient flexibility and adaptability in cooling station power regulation, making it impossible to effectively respond to complex and changing power demands and environmental conditions.

[0026] In summary, many technical challenges remain in the current field of flexible cooling plant power regulation, including the accuracy of power prediction, the balance between comfort and cooling plant power, and the optimization of cooling plant time response characteristics. Therefore, the study of a systematic and efficient method for flexible cooling plant power regulation has important practical significance and application value.

[0027] Therefore, the purpose of the present invention is to propose a cooling station power control method that can systematically and flexibly control cooling station power to address the problems of significant comfort sacrifice, low regulation efficiency, and delayed response during the power regulation process in existing technologies. This method constructs a precise power regulation model based on multi-dimensional data such as building energy load, cooling station power forecasts, and comfort forecasts, enabling rapid and accurate response to power limits. Furthermore, by optimizing the control strategy, the impact of power regulation on human comfort is minimized while ensuring overall building performance, significantly improving the accuracy and adaptability of cooling station power regulation.

[0028] Figure 1 A flow chart of a cooling station power control method provided in an embodiment of the present application.

[0029] like Figure 1 As shown, an embodiment of the present application provides a cooling station power control method 100, which includes:

[0030] Step 110: Obtain the target control power of the cooling station.

[0031] For example, the target control power of the cooling station can be read in advance through the demand reading module. For example, the target control power can be data from the power control curve issued by the power grid dispatcher and received and parsed by this module. For example, data from the cooling station power control curve for the next 24 hours (00:00-24:00) can be obtained and parsed within a time window of 22:00 ± 30 minutes per day. The data in the power control curve can be 96 target control powers and other cooling station operation data at intervals of 15 minutes. The power control curve can also include an upper threshold value for the cooling station power.

[0032] Step 120 : predicting future overall control parameters of the cooling station based on the environmental parameters, the target control power, and the cooling station operation data.

[0033] For example, the environmental parameters could be the outdoor temperature and humidity for the next 24 hours as predicted by the weather forecast. Based on the 96 target control powers and cooling plant operating data in the power control curve, a reinforcement learning algorithm is used to predict the 96 cooling plant overall control parameters for the next 24 hours within a time window of 22:00 ±30 minutes daily. The time node of each cooling plant control parameter corresponds to the time node of the 96 target control powers.

[0034] Step 130 : obtaining power prediction data and comfort prediction data of the cooling station based on the overall control parameters of the cooling station.

[0035] Exemplarily, the power prediction data and comfort prediction data of the cooling station can be predicted by the cooling station prediction model and the comfort prediction model, for example, based on 96 cooling station overall control parameters and environmental parameters. Therefore, the number of power prediction data and comfort prediction data is also 96, and corresponds to the time nodes of the 96 cooling station overall control parameters.

[0036] Step 140 : Obtain basic control parameter data of the cooling station according to the power prediction data, the comfort prediction data and the overall control parameters of the cooling station.

[0037] For example, in step 130, when the power prediction data and comfort prediction data for 96 cooling stations are predicted using the cooling station prediction model and the comfort prediction model, the 96 cooling station overall control parameters are all used. If it is determined that the 96 predicted power prediction data are all within the target control power range obtained in step 110, and the 96 comfort prediction data are all within a reasonable range of human comfort, all cooling station overall control parameters can be determined as the cooling station control parameter basic data. Otherwise, if any data point in the 96 power prediction data and the 96 comfort prediction data does not meet the above range, the method of step 120 is used to re-infer the data point that does not meet the above range, up to n times, where n can be set to 3 or other values. For the purpose of illustration, n=3 is used as an example. The data point that does not meet the range is selected at least once and, together with the other data points that meet the range, serves as the 96 cooling station control parameter basic data for the next 24 hours.

[0038] Step 150: Control the operation of the cooling station based on the control parameter basic data.

[0039] For example, the basic data of 96 control parameters of the cooling station for the next 24 hours are sent to the feedback control module. The module can control the operation of the cooling station based on the basic data of the control parameters. It can also monitor the changes in the power output and comfort of the cooling station in real time through sensors and data acquisition systems during the operation of the cooling station.

[0040] The cooling station power control method provided in the embodiment of the present application can achieve high-precision adjustment of the cooling station power by obtaining the power control curve in advance, ensuring that the power output meets the target control power range. At the same time, based on environmental parameters and cooling station operation data, it dynamically predicts the overall control parameters of the cooling station for the next 24 hours, significantly improving the system response speed and prediction accuracy. Through the comfort prediction model, this method effectively maintains the indoor environment within a reasonable range of human comfort while meeting the target control power range. When the power or comfort prediction data does not meet the requirements, the overall control parameters of the cooling station are optimized through multiple reasoning to achieve a dynamic balance between power and comfort.

[0041] In another embodiment of the present application, basic data of control parameters of the cooling station are obtained based on power prediction data, comfort prediction data and overall control parameters of the cooling station, including: when the power prediction data is within the range of target control power and the comfort prediction data is within a reasonable comfort range, the overall control parameters of the cooling station are determined as the basic data of control parameters of the cooling station; when the power prediction data is not within the range of target control power, or the comfort prediction data is not within a reasonable comfort range, the basic data of control parameters of the cooling station are re-predicted based on the power prediction data and the comfort prediction data.

[0042] For example, when power prediction data and comfort prediction data for 96 cooling stations are predicted, all 96 cooling station overall control parameters are used. If it is determined that all 96 predicted power prediction data are within the target control power range obtained in step 110, and all 96 comfort prediction data are within a reasonable range for human comfort, all cooling station overall control parameters can be determined as the cooling station control parameter basic data. Otherwise, if any of the 96 power prediction data and the 96 comfort prediction data do not meet the above ranges, the method of step 120 is used to re-infer the nodes that do not meet the above ranges, up to three times of inference, and the node with the least number of nodes that do not meet the ranges is selected and used together with the other nodes that meet the ranges as the 96 cooling station control parameter basic data for the next 24 hours.

[0043] In this embodiment, power and comfort levels at 96 time points are double-verified to ensure that the predicted data is within the target control power range and a reasonable range of human comfort, thereby generating reliable basic control parameter data. If the predicted data does not meet the requirements, up to three rounds of inference optimization are performed, and the optimal result is selected and combined with other nodes that meet the conditions to ultimately form the basic control parameter data for the next 24 hours. This method achieves the dual goals of power control and comfort assurance, while also improving the accuracy and reliability of control parameters through an adaptive optimization mechanism, providing technical support for the efficient and stable operation of the cooling station.

[0044] In another embodiment of the present application, the control parameter basic data includes multiple control parameter basic values corresponding to multiple time periods; controlling the operation of the cold station based on the control parameter basic data includes: obtaining the overall control parameters corresponding to the current time period based on the control parameter basic values corresponding to the current time period; when controlling the operation of the cold station based on the overall control parameters corresponding to the current time period, obtaining the overall control parameters corresponding to the next time period according to the actual operating power and current environmental parameters; when the next time period arrives, controlling the operation of the cold station based on the overall control parameters corresponding to the next time period.

[0045] In another example, when the operation of the cooling station is controlled based on the overall control parameters corresponding to the current time period, the overall control parameters corresponding to the next time period are obtained according to the actual operating power and the current environmental parameters, including: when the actual power is within the range of the target control power and the current environmental parameters are within the preset environmental threshold range, the basic value of the control parameter corresponding to the next time period is determined as the overall control parameter corresponding to the next time period; when the actual power is not within the range of the target control power, or the current environmental parameters are not within the preset environmental threshold range, the overall control parameters corresponding to the next time period are re-predicted based on the target data from the base time period to the current time period, wherein the target data includes at least one of the control parameter basic data, the operation data of the cooling station, the environmental data, and the comfort data.

[0046] Exemplarily, the control parameter base data may include 96 control parameter base values at 15-minute intervals. For example, when the cooling station operates based on the third control parameter base value corresponding to the third 15-minute time period, the actual power and current environmental parameters, such as humidity and temperature data of the current environment during cooling station operation, can be monitored in real time. If the actual power of the cooling station monitored in real time deviates from the previously acquired target control power by ±5%, or if the outdoor temperature in the current environmental parameters deviates from a preset environmental threshold by ±3°C, or the humidity deviates from a preset environmental threshold by ±3RH (relative humidity), then a fourth control parameter base value corresponding to the fourth 15-minute time period needs to be re-predicted. For example, the control parameter base value can be re-predicted three times. Among these three times, the power prediction data and comfort prediction data obtained based on the predicted control parameter base value are selected, and the two values with the smallest deviation from the target control power and the preset environmental threshold are selected. This re-predicted control parameter base value is used as the fourth control parameter base value corresponding to the fourth 15-minute time period.

[0047] In an embodiment of the present application, by real-time monitoring of the actual power and environmental parameters of the cooling station, when the deviation exceeds a preset threshold (such as power deviation ±5%, temperature deviation ±3°C, humidity deviation ±3RH), the basic data of the control parameters are dynamically optimized, and the basic values of the control parameters that minimize the deviation between the actual power of the cooling station and the target control power are re-predicted and selected. Through this embodiment, adaptive adjustment can be achieved during the operation of the cooling station, ensuring real-time matching of power control accuracy with changes in environmental parameters, significantly improving the stability and energy efficiency of the cooling station operation, and providing reliable technical support for precise control in complex environments.

[0048] In another embodiment of the present application, the power prediction data of the cold station is predicted by a power prediction model, which includes multiple long and short-term memory network layers and fully connected layers. The power prediction model is trained in the following manner: obtaining power sample data and power labels corresponding to the power sample data, wherein the power sample data includes outdoor temperature, outdoor humidity, and overall control parameters; using multiple long and short-term memory network layers and fully connected layers to process the power sample data to obtain a power prediction result; determining the error between the power prediction result and the power label by means of a root mean square error, and adjusting the model parameters of the power prediction model based on the error to obtain a trained power prediction model.

[0049] For example, the power prediction data for a cooling station is obtained using a power prediction model. This model utilizes a deep learning network architecture, including an input layer, a long short-term memory (LSTM) network layer, a fully connected layer, and an output layer. The input layer can be used to receive power sample data, such as outdoor temperature, outdoor humidity, and overall cooling station control parameters. The LSTM network layer can be configured with, for example, three layers, with 128, 64, and 32 neurons per layer, respectively, to extract time series features. The fully connected layer can be configured with, for example, two layers, with 16 and 1 neurons per layer, respectively, to generate power prediction data. The output layer is used to output the power prediction results.

[0050] For example, during the training phase of the power prediction model, the overall control parameters of the cooling station, cooling station operation data, outdoor temperature, outdoor humidity and comfort data of the cooling station in the recent period (e.g., the previous 365 days) can be obtained from the operation log of the cooling station as basic training data to train the power prediction model. The basic training data is input through the long short-term memory network layer and the fully connected layer of the power prediction model to generate the power prediction results of the training phase. The root mean square error (RMSE) can also be used to calculate the error between the power prediction result and the power label. For example, the total time consumed for model training does not exceed 2 hours.

[0051] For example, during the optimization stage of the power prediction model, power sample data and its corresponding power tags can be obtained from the operation log of the cooling station at 22:00 at the end of each month, and the data time resolution can be 15 minutes. The power prediction model is optimized and evaluated using the acquired power sample data and its corresponding power tags, and the entire process takes no more than 2 hours. The cooling station operation data of the previous 365 days can also be further read, including the power of the cooling station, outdoor temperature, outdoor humidity, overall control parameters of the cooling station, and comfort data, to optimize the power prediction model again. When using RMSE to evaluate the optimized power prediction model, the model parameters of the power prediction model can be adjusted based on the error. If the RMSE of the new model is lower than that of the current model, the new model can be used to replace the current model.

[0052] In an embodiment of the present application, a power prediction model based on a deep learning network architecture is constructed, and the historical operating data of the cold station (such as the power, outdoor temperature, humidity, overall control parameters, etc. of the previous 365 days) is used to train the model, and the prediction accuracy is evaluated by RMSE; the model is dynamically optimized at 22:00 at the end of each month, and the latest data is used to retrain and evaluate. If the accuracy of the new model is improved, the current model is replaced, and the training and optimization processes are controlled to be completed within 2 hours. This technical means can effectively improve the accuracy of the cold station power prediction and dynamically optimize it, and can adapt to the dynamic changes of different external variables. The embodiment provided in this application can enhance the adaptability of the model to different external variables by unified coding and integrating multiple external variables. At the same time, through efficient data processing and model update mechanisms, as well as the introduction of multiple external variables and improved model structure, the embodiment provided in this application can more accurately predict the power of the cold station. Compared with traditional methods, the prediction error is reduced by more than 20% in typical scenarios.

[0053] In another embodiment of the present application, the comfort prediction data of the cooling station is obtained through a comfort prediction model, which includes multiple long and short-term memory network layers and fully connected layers. The comfort prediction model is trained in the following manner: obtaining comfort sample data and comfort labels corresponding to the comfort sample data, wherein the comfort sample data includes at least one of outdoor temperature, outdoor humidity, overall control parameters, and weighted number of manual operations, wherein the weighted number of manual operations includes the weighted average number of heating operations, cooling operations, and wind speed adjustment operations; using multiple long and short-term memory network layers and fully connected layers to process the comfort sample data to obtain a comfort prediction result; determining the error between the comfort prediction result and the comfort label by means of a root mean square error method, and adjusting the model parameters of the comfort prediction model based on the error to obtain a trained comfort prediction model.

[0054] For example, the structure, training, and optimization process of the comfort prediction model are similar to those of the power prediction model described above and will not be further elaborated here. However, the data used in the comfort prediction model training and optimization process includes not only outdoor temperature, humidity, and overall cooling station control parameters, but also historical data on the weighted number of manual terminal operations (historical comfort data). The weighted number of manual terminal operations includes the weighted average of the number of heating operations, the number of cooling operations, and the number of wind speed adjustment operations.

[0055] In this application's embodiment, a comfort prediction model based on LSTM and fully connected layers is constructed. This model is trained and optimized using historical data, and prediction accuracy is evaluated using RMSE. The model is dynamically updated at the end of each month to improve performance. This significantly improves cooling plant power control accuracy while minimizing the impact on building comfort, achieving the dual goals of flexible power control and ensuring building comfort.

[0056] In another embodiment of the present application, the basic data of the control parameters of the cooling station is obtained through a reinforcement learning model, and the reinforcement learning model is trained in the following manner: obtaining historical data of the cooling station, wherein the historical data of the cooling station includes at least one of historical basic data of control parameters, historical operation data, historical outdoor temperature, historical outdoor humidity and historical comfort data; and training the reinforcement learning model based on the historical data of the cooling station.

[0057] The embodiment of the present application adopts a reinforcement learning model to automatically optimize the overall control parameters through outdoor temperature, outdoor humidity, target control power of the cooling station, etc., while meeting the power requirements of the cooling station, maximizing the efficiency of the cooling station and minimizing the weighted number of terminal manual operations.

[0058] In another embodiment of the present application, a reinforcement learning model is trained based on historical data of the cold station, including: based on the historical data of the cold station, a reinforcement learning model is trained according to a reward function, wherein the reward function is associated with at least one of the cold station efficiency reward function, the comfort sample data penalty function, the constraint penalty function, and the final reward function; wherein the cold station efficiency reward function is associated with the real-time cooling capacity, real-time power, and reward coefficient of the cold station; wherein the comfort sample data penalty function is associated with the weighted number of manual operations and the weighted number of operations penalty coefficient; wherein the constraint penalty function is associated with the demand power, real-time power, and penalty coefficient of the cold station.

[0059] Exemplarily, the design of a reinforcement learning model also includes a state space, an action space, and a reward function. The state space describes the current operating state of the cooling station system, including: chilled water supply temperature, chilled water return temperature, cooling water supply temperature, cooling water return temperature, cooling water flow rate, chilled water flow rate, outdoor temperature, outdoor humidity, weighted number of terminal manual operations, cooling station power demand, cooling station real-time power, and cooling station real-time cooling capacity. The action space describes adjustable control parameters, i.e., overall control parameters.

[0060] The reward function is used to guide the optimization goal of the reinforcement learning model and is designed as follows:

[0061] Cold station efficiency reward function:

[0062] R cop =C real / P real *W c

[0063] Among them, C real is the real-time cooling capacity of the cooling station, P real is the real-time power of the cooling station, W c is the bonus coefficient for cooling station efficiency, and the goal is to maximize it.

[0064] Comfort sample data penalty function:

[0065] R op =-O*W o

[0066] Among them, C is the weighted number of manual operations at the end, W o is the penalty coefficient for the weighted number of operations, the goal is to minimize it.

[0067] Constrained penalty function:

[0068] R v =-P set / P real *W v

[0069] Among them, P set / P real is the degree to which the cooling station power demand is not met, P set is the required power, P real is the real-time power of the cooling station, W v is the penalty coefficient.

[0070] Final reward function:

[0071] R=R cop +R op +R V

[0072] The embodiments of the present application can also provide a feedback control model. The model receives the overall control parameters output by the reinforcement learning model, monitors the power output and comfort changes of the cooling station in real time, and adjusts the start and stop and frequency of equipment such as the cooling station, cooling pump, freezing pump, cooling tower, and the like, as well as the chilled water inlet and return water, cooling water inlet and return water parameter settings, etc. Based on the monitored data, the operating status of the cooling station system equipment is adjusted through the feedback control algorithm, thereby achieving feedback fine-tuning of the deviation of the predicted power from the target control power, ensuring that the actual power of the cooling station meets the target control power range. In the process of adjusting the cooling station system equipment, the time response characteristics of different equipment can also be classified, and the actual power of the cooling station, comfort data, equipment energy consumption and other aspects are comprehensively considered to optimize and select multiple output targets.

[0073] Exemplarily, the reinforcement learning model can also be used to adjust the baseline control parameter values for the next time period. For example, after obtaining the baseline control parameter values for the next day, the feedback control model controls the operation of the cooling station equipment (such as chillers, cooling pumps, refrigeration pumps, and cooling towers) based on these baseline control parameter values. During operation, the model checks the deviation between the cooling station's actual power and the power control target, as well as the deviation between the outdoor temperature and humidity data and the environmental parameters, every 15 minutes. If the power deviation is ±5%, the outdoor temperature deviation is ±3 degrees, or the humidity deviation is ±3RH, the overall control parameters for the next 15 minutes are re-inferred using the overall control parameters, operating data, outdoor temperature and humidity, and the weighted number of manual operations performed by the terminal from midnight to the current time. Similarly, the model performs a maximum of three inferences, and the overall control parameter with the smallest deviation between the predicted power data and the target control power and the smallest weighted number of manual operations performed by the terminal is selected as the control parameter for the next 15 minutes and sent to the feedback control model. If the change in the external data does not exceed the control threshold, the overall control parameter for that time point is directly sent to the feedback control model.

[0074] In the embodiments of this application, the overall control parameters are automatically optimized based on outdoor temperature, humidity, and target control power, maximizing cooling station efficiency and minimizing the weighted number of end-user manual operations while meeting the cooling station's power requirements. By designing a state space, action space, and reward function, a multi-objective optimization mechanism is constructed, taking into account the cooling station's real-time operating status, cooling efficiency, manual operation frequency, and power demand constraints. This approach significantly improves cooling station operating efficiency, reduces the frequency of manual intervention, and ensures accurate power control.

[0075] Figure 2 A flowchart illustrating the implementation of a cooling station power control method according to an embodiment of the present application.

[0076] This application provides an example and combines Figure 2 To better illustrate the content of the technical solution provided by this application:

[0077] A centralized cooling station in an office park consists of n chillers, chilled water pumps, cooling pumps, and cooling towers. The original control system set a fixed chilled water outlet temperature and adjusted the chilled and cooling water pump frequencies based on pressure. The number of cooling towers was generally controlled based on the number of chillers. This system only provided basic cooling control, adjusting power by increasing or decreasing the number of chillers in operation. Power adjustment was done in 1 / n steps, lacking flexibility and unable to achieve optimal system efficiency while maintaining comfort.

[0078] After the cold station adds the cold station power flexible control method based on this invention, it can achieve 1kW precision regulation through predictive optimization adjustment based on weather and comfort feedback, while ensuring the operating efficiency of the cold station and the comfort of the terminal space.

[0079] During the model training phase, the original system's historical operation records or personnel-recorded control records (such as environmental parameters, real-time power of the cooling station, chilled water supply temperature, chilled water return temperature, chilled water flow, cooling water supply temperature, cooling water return temperature, chiller set temperature, chiller on / off status, freezing pump frequency, cooling pump frequency, cooling tower on / off status, cooling tower frequency, etc.) are first entered into the system. After the entry is completed, the power prediction model, comfort prediction model, and reinforcement learning model are automatically trained. After the training is completed, flexible power control can begin.

[0080] like Figure 2 As shown, during the stage of using the model to flexibly control the power of the power station, for example, the demand reading module can obtain the target control power of the cooling station in advance. Then, the predictive control module can output the overall control parameters of the cooling station based on factors such as environmental parameters, target control power, and cooling station operating data, and predict changes in power and comfort. The predictive control module can include a power prediction model, a comfort prediction model, and a reinforcement learning model. When the predictive control module determines, based on the power prediction model and the comfort prediction model, that the predicted power prediction data are all within the range of the target control power and that the comfort prediction data are all within a reasonable range of human comfort, the power prediction data, comfort prediction data, and overall control parameters of the cooling station can be combined to obtain the basic control parameter data of the cooling station, which is determined as the overall control parameters, and the cooling station system is controlled to operate based on the overall control parameters.

[0081] During the operation of the cooling station system, the predictive control module can also check the actual power and current environmental parameters of the cooling station system at intervals of a preset time period (e.g., 15 minutes). If the actual power is not within the target control power range, or the environmental parameters are not within the preset environmental threshold range, the corresponding overall control parameters for the next time period are re-predicted. At this time, the power prediction model and the comfort prediction model are used to predict the re-predicted overall control parameters to obtain power prediction data and comfort prediction data, which have power variables and comfort variables compared to the previous data.

[0082] During the operation of the cold station system, the feedback control module can also be used to monitor the power output and comfort changes of the cold station system in real time. By adjusting the start and stop and frequency of equipment such as the chiller, cooling pump, freezing pump, cooling tower, and the setting of chilled water inlet and return parameters, feedback and fine-tuning can be performed on the deviation between the actual power of the cold station system and the target control power. During the adjustment process, the time response characteristics of different equipment can be classified to obtain the adjustable parameters of the cold station, and the multiple output targets of the cold station power can be optimized and selected. Through the collaborative work of the above modules, this application can achieve flexible adjustment of the cold station power, which not only meets the power demand of the power system, but also ensures the comfort level in the building, and contributes to the stable operation of the power system and energy conservation and emission reduction.

[0083] The power prediction model, comfort prediction model and reinforcement learning model can be continuously and automatically trained and optimized during operation according to the method of the above embodiment of this application; the overall control parameters output by the reinforcement learning model (including the set temperature of chilled water, the number of chillers in operation, the frequency of the chilled water pump, the frequency of the cooling pump, the number of cooling towers, the frequency of the cooling tower, etc.) are transmitted to the reinforcement learning model, and the reinforcement learning model dynamically fine-tunes the control parameters according to the actual data changes and outputs them to the feedback control model, and the feedback control model is sent to the controlled equipment (chiller, chilled water pump, cooling pump, cooling tower).

[0084] Figure 3 A schematic diagram of a power control device for a cooling station provided in an embodiment of the present application.

[0085] The embodiment of the present application further provides a power control device 300 for a cooling station, characterized in that the device 300 includes:

[0086] The acquisition module 310 is configured to acquire the target control power of the cooling station.

[0087] The prediction module 320 is used to predict the future overall control parameters of the cooling station based on the environmental parameters, the target control power and the cooling station operation data, and is also used to obtain the power prediction data and comfort prediction data of the cooling station based on the overall control parameters of the cooling station.

[0088] The inference module 330 is used to obtain basic control parameter data of the cooling station based on the cooling station power prediction data, the comfort prediction data and the cooling station overall control parameters.

[0089] The control module 340 is used to control the operation of the cooling station based on the cooling station control parameter basic data.

[0090] It can be understood that for a detailed description of the power control device 300 of the cooling station, reference can be made to the description of the power control method 100 applied to the cooling station above.

[0091] In some embodiments, the reasoning module 330 is also used to: when the power prediction data is within the range of the target control power, and the comfort prediction data is within the reasonable comfort range, determine the overall control parameters of the cooling station as the basic data of the control parameters of the cooling station; when the power prediction data is not within the range of the target control power, or the comfort prediction data is not within the reasonable comfort range, re-predict the basic data of the control parameters of the cooling station based on the power prediction data and the comfort prediction data.

[0092] In some embodiments, the control parameter basic data includes multiple control parameter basic values corresponding to multiple time periods; the control module 340 is also used to: obtain the overall control parameters corresponding to the current time period based on the control parameter basic values corresponding to the current time period; when the operation of the cold station is controlled based on the overall control parameters corresponding to the current time period, the overall control parameters corresponding to the next time period are obtained according to the actual power of the operation and the current environmental parameters; when the next time period arrives, the operation of the cold station is controlled based on the overall control parameters corresponding to the next time period.

[0093] In some embodiments, when the operation of the cooling station is controlled based on the overall control parameters corresponding to the current time period, the overall control parameters corresponding to the next time period are obtained according to the actual operating power and the current environmental parameters, including: when the actual power is within the range of the target control power and the current environmental parameters are within the preset environmental threshold range, the basic value of the control parameter corresponding to the next time period is determined as the overall control parameter corresponding to the next time period; when the actual power is not within the range of the target control power, or the current environmental parameters are not within the preset environmental threshold range, the overall control parameters corresponding to the next time period are re-predicted based on the target data from the base time period to the current time period, wherein the target data includes at least one of the basic control parameter data, the operation data of the cooling station, the environmental data, and the comfort data.

[0094] In some embodiments, the power prediction data of the cold station is obtained by prediction through a power prediction model, and the power prediction model includes multiple long short-term memory network layers and fully connected layers. The device 300 also includes: a first acquisition module, used to obtain power sample data and power labels corresponding to the power sample data, wherein the power sample data includes outdoor temperature, outdoor humidity, and overall control parameters; a first prediction module, used to use multiple long short-term memory network layers and fully connected layers to process the power sample data to obtain a power prediction result; a first adjustment module, used to determine the error between the power prediction result and the power label by means of a root mean square error, and adjust the model parameters of the power prediction model based on the error to obtain a trained power prediction model.

[0095] In some embodiments, the comfort prediction data of the cooling station is obtained through a comfort prediction model, and the power prediction model includes multiple long short-term memory network layers and fully connected layers. The device 300 also includes: a second acquisition module, used to obtain comfort sample data and comfort labels corresponding to the comfort sample data, wherein the comfort sample data includes at least one of outdoor temperature, outdoor humidity, overall control parameters, and weighted number of manual operations, wherein the weighted number of manual operations includes the weighted average number of heating operations, cooling operations, and wind speed adjustment operations; a second prediction module, using multiple long short-term memory network layers and fully connected layers to process the comfort sample data to obtain a comfort prediction result; a second adjustment module, used to determine the error between the comfort prediction result and the comfort label by means of a root mean square error, and adjust the model parameters of the comfort prediction model based on the error to obtain a trained comfort prediction model.

[0096] In some embodiments, the basic data of the control parameters of the cold station are obtained through a reinforcement learning model, and the device 300 also includes: a third acquisition module, used to obtain historical data of the cold station, wherein the historical data of the cold station includes at least one of historical basic data of control parameters, historical operation data, historical outdoor temperature, historical outdoor humidity and historical comfort data; a third training module, used to train the reinforcement learning model based on the historical data of the cold station.

[0097] In some embodiments, the third training module is also used to: train a reinforcement learning model based on the historical data of the cold station according to the reward function, wherein the reward function is associated with at least one of the cold station efficiency reward function, the comfort sample data penalty function, the constraint penalty function, and the final reward function; wherein the cold station efficiency reward function is associated with the real-time cooling capacity, real-time power, and reward coefficient of the cold station; wherein the comfort sample data penalty function is associated with the weighted number of manual operations and the weighted number of operations penalty coefficient; wherein the constraint penalty function is associated with the demand power, real-time power, and penalty coefficient of the cold station.

[0098] An embodiment of the present application provides an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions that can be executed by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method of any of the above embodiments.

[0099] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method in any of the above embodiments are implemented.

[0100] An embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed by a processor of a computer device, the computer device is enabled to perform the steps of the method of any of the above embodiments.

[0101] Figure 3 A block diagram of an electronic device provided in an embodiment of the present application.

[0102] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the method in any of the above embodiments when executing the computer program.

[0103] like Figure 4 As shown, for ease of understanding, the embodiment of the present application shows a specific electronic device 400.

[0104] The electronic device 400 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are intended to be examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0105] like Figure 4 As shown, the device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. Various programs and data required for the operation of the electronic device 400 can also be stored in the RAM 403. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0106] Multiple components in the electronic device 400 are connected to the I / O interface 405, including an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a magnetic disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0107] The computing unit 401 can be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods described above. For example, in some embodiments, any one or more of the above methods can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of any one or more of the above methods can be performed. Alternatively, in other embodiments, the computing unit 401 can be configured to perform any one or more of the above methods by any other appropriate means (e.g., by means of firmware).

[0108] It should be noted that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device, or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device, or apparatus and execute the instructions), or in conjunction with such instruction execution systems, devices, or apparatuses. For purposes of this application, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by an instruction execution system, device, or apparatus, or in conjunction with such instruction execution systems, devices, or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion having one or more wires (electronic device), a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing it in a suitable manner if necessary, and then storing it in a computer memory.

[0109] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0110] In the description of this application, reference to the terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of this application. In this application, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples.

[0111] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0112] In addition, the terms "first" and "second" used in the embodiments of the present application are for descriptive purposes only and should not be understood as indicating or implying relative importance, or implicitly indicating the number of technical features indicated in the embodiments. Therefore, the features defined in the embodiments of the present application by terms such as "first" and "second" can explicitly or implicitly indicate that at least one of the features is included in the embodiment. In the description of the present application, the word "multiple" means at least two or two or more, such as two, three, four, etc., unless otherwise clearly and specifically defined in the embodiments.

[0113] In this application, unless otherwise specified or limited in the embodiments, the terms "installed", "connected", "connected", and "fixed" appearing in the embodiments should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integral connection. It can also be a mechanical connection, an electrical connection, etc.; of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two elements, or the interaction between two elements. For those skilled in the art, the specific meanings of the above terms in this application can be understood based on the specific implementation.

[0114] In this application, unless otherwise expressly specified or limited, when a first feature is "above" or "below" a second feature, it may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Furthermore, when a first feature is "above," "above," or "above" a second feature, it may mean that the first feature is directly above or diagonally above the second feature, or simply means that the first feature is at a higher level than the second feature. When a first feature is "below," "below," or "below" a second feature, it may mean that the first feature is directly below or diagonally below the second feature, or simply means that the first feature is at a lower level than the second feature.

Claims

1. A cooling station power control method, characterized in that: The method comprises: Obtaining a target control power of the cooling station; predicting future overall control parameters of the cooling station based on environmental parameters, the target control power, and cooling station operation data; Based on the overall control parameters of the cooling station, obtaining power prediction data and comfort prediction data of the cooling station; Obtaining basic control parameter data of the cooling station according to the power prediction data, the comfort prediction data and the overall control parameters of the cooling station; The operation of the cooling station is controlled based on the control parameter basic data.

2. The method according to claim 1, characterized in that The step of obtaining basic control parameter data of the cooling station according to the power prediction data, the comfort prediction data and the overall control parameters of the cooling station includes: When the power prediction data is within the range of the target control power, and the comfort prediction data is within a reasonable comfort range, determining the overall control parameters of the cooling station as basic control parameter data of the cooling station; When the power prediction data is not within the range of the target control power, or the comfort prediction data is not within the reasonable comfort range, the control parameter basic data of the cooling station is re-predicted based on the power prediction data and the comfort prediction data.

3. The method according to claim 1, characterized in that The control parameter basic data includes a plurality of control parameter basic values corresponding to a plurality of time periods; and controlling the operation of the cooling station based on the control parameter basic data includes: Based on the basic value of the control parameter corresponding to the current time period, the overall control parameter corresponding to the current time period is obtained; In the case where the operation of the cooling station is controlled based on the overall control parameters corresponding to the current time period, the overall control parameters corresponding to the next time period are obtained according to the actual operating power and the current environmental parameters; When the next time period arrives, the operation of the cooling station is controlled based on the overall control parameters corresponding to the next time period.

4. The method according to claim 3, characterized in that When the operation of the cooling station is controlled based on the overall control parameters corresponding to the current time period, the overall control parameters corresponding to the next time period are obtained according to the actual operating power and the current environmental parameters, including: When the actual power is within the range of the target control power and the current environmental parameter is within the preset environmental threshold range, the control parameter base value corresponding to the next time period is determined as the overall control parameter corresponding to the next time period; When the actual power is not within the range of the target control power, or the current environmental parameter is not within the preset environmental threshold range, the overall control parameter corresponding to the next time period is re-predicted based on the target data from the reference time period to the current time period, wherein the target data includes at least one of the control parameter basic data, the operating data of the cooling station, the environmental data, and the comfort data.

5. The method according to any one of claims 1 to 4, characterized in that The power prediction data of the cooling station is obtained by predicting a power prediction model. The power prediction model includes multiple long short-term memory network layers and a fully connected layer. The power prediction model is trained in the following manner: Acquire power sample data and a power tag corresponding to the power sample data, wherein the power sample data includes outdoor temperature, outdoor humidity, and overall control parameters; Processing the power sample data using multiple long short-term memory network layers and fully connected layers to obtain a power prediction result; The error between the power prediction result and the power label is determined by a root mean square error method, and the model parameters of the power prediction model are adjusted based on the error to obtain a trained power prediction model.

6. The method according to any one of claims 1 to 4, characterized in that The comfort prediction data of the cooling station is obtained by a comfort prediction model. The power prediction model includes multiple long short-term memory network layers and a fully connected layer. The comfort prediction model is trained in the following way: Acquire comfort sample data and a comfort label corresponding to the comfort sample data, wherein the comfort sample data includes at least one of outdoor temperature, outdoor humidity, an overall control parameter, and a weighted number of manual operations, wherein the weighted number of manual operations includes a weighted average number of heating operations, a cooling operations, and a wind speed adjustment operations; Using multiple long short-term memory network layers and fully connected layers to process the comfort sample data to obtain a comfort prediction result; The error between the comfort prediction result and the comfort label is determined by a root mean square error method, and the model parameters of the comfort prediction model are adjusted based on the error to obtain a trained comfort prediction model.

7. The method according to any one of claims 1 to 4, characterized in that The basic data of the control parameters of the cooling station are obtained by a reinforcement learning model, and the reinforcement learning model is trained in the following manner: Acquiring historical data of the cooling station, wherein the historical data of the cooling station includes at least one of historical control parameter basic data, historical operation data, historical outdoor temperature, historical outdoor humidity, and historical comfort data; The reinforcement learning model is trained based on the cold station historical data.

8. The method according to claim 7, characterized in that Training the reinforcement learning model based on the cold station historical data includes: Based on the cooling station historical data, the reinforcement learning model is trained according to a reward function, wherein the reward function is associated with at least one of a cooling station efficiency reward function, a comfort sample data penalty function, a constraint penalty function, and a final reward function; The cooling station efficiency reward function is associated with the real-time cooling capacity, real-time power, and reward coefficient of the cooling station; The comfort sample data penalty function is associated with the weighted number of manual operations and the weighted number of operations penalty coefficient; The constraint penalty function is associated with the required power, real-time power and penalty coefficient of the cooling station.

9. A power control device for a cooling station, characterized in that: The device comprises: An acquisition module is used to obtain the target control power of the cooling station; a prediction module, configured to predict future overall control parameters of the cooling station based on environmental parameters, the target control power, and cooling station operation data, and further configured to obtain power prediction data and comfort prediction data of the cooling station based on the overall control parameters of the cooling station; an inference module, configured to obtain basic control parameter data of the cooling station based on the cooling station power prediction data, the comfort prediction data, and the overall control parameters of the cooling station; A control module is used to control the operation of the cooling station based on the cooling station control parameter basic data.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

Images