A combined control method and system based on water storage and phase change storage in cooling system

Abstract

The present application relates to a joint control method based on water storage and phase change storage in a cooling system, which includes the following steps: when receiving user demand information, obtaining electricity price fluctuation information and meteorological information and inputting them into a prediction model for trend prediction; obtaining equipment status information and matching the corresponding federal storage strategy model; inputting the trend feature set output by the prediction model and the equipment status information into the federal storage strategy model to generate a storage strategy; solving the storage strategy generated by the federal storage strategy model to output an optimal storage strategy set; the present application realizes efficient joint control of water storage and phase change storage equipment through the application of a joint control method and a federal storage strategy model; by deploying a federal storage strategy model, the cooling system can quickly match the federal storage strategy model when facing new storage needs, thereby generating a highly personalized and precise storage strategy to ensure the effectiveness and efficiency of the strategy.

Description

Technical Field

[0001] The present application relates to the technical field of cold storage control, and in particular to a combined control method and system based on water cold storage and phase change cold storage in a cooling system. Background Art

[0002] In energy management and cooling systems, cold storage technology, as an effective means of energy regulation, is of great significance for improving energy efficiency, reducing operating costs, and promoting green and sustainable development. Traditional cold storage technologies mainly include water storage and phase change storage, each of which has unique advantages and application scenarios. However, with the complexity of the energy market and the increasing requirements of users for the economy and environmental protection of cooling systems, a single cold storage method can no longer meet the increasingly diverse needs.

[0003] In the prior art, water storage and phase change storage systems usually operate independently and lack effective joint control strategies. This decentralized control method has the following significant defects:

[0004] First, traditional cold storage systems usually make decisions based only on the current equipment status or simple preset rules, while ignoring the impact of electricity price fluctuations on operating costs and the long-term prediction of weather conditions on cold storage demand. In actual operation, traditional cold storage systems do not fully utilize low electricity price periods for cold storage, and it is difficult to adjust cold storage strategies in advance according to weather changes, thereby increasing operating costs and reducing energy efficiency.

[0005] Secondly, the existing technology lacks an effective collaborative control mechanism to integrate the operation of water storage and phase change storage equipment. In fact, different storage equipment has its own unique characteristics and advantages. However, under the current technical framework, when the cooling system faces new storage needs, it is difficult to quickly match the most suitable storage strategy. It often adopts a one-size-fits-all approach and cannot flexibly adjust and optimize the storage strategy according to the real-time status information of the equipment, resulting in the cooling strategy being not personalized and accurate enough. As a result, in the actual operation process, it is difficult for water storage and phase change storage equipment to fully exert their synergy, and thus it is impossible to form an efficient and economical storage solution. Summary of the invention

[0006] In order to solve the above-mentioned defects, the present application provides a combined control method and system based on water storage and phase change storage in a cooling system.

[0007] The above-mentioned invention objective of the present application is achieved through the following technical solutions:

[0008] A combined control method based on water cold storage and phase change cold storage in a cooling system, comprising the steps of:

[0009] When receiving user demand information from the interactive terminal, obtaining electricity price fluctuation information and meteorological information within a preset interval, wherein the user demand information includes cold storage demand information and green economy demand information;

[0010] Input the cold storage demand information, electricity price fluctuation information and meteorological information into the pre-trained prediction model to make predictions based on the meteorological environment trend, electricity price change trend and demand change trend;

[0011] The device status information of the water cooling device and the phase change cooling device is obtained respectively, and the corresponding pre-trained federated cooling storage strategy model is matched based on the preset matching conditions;

[0012] Inputting the trend feature set output by the prediction model and the equipment status information into the matching federal cold storage strategy model to generate a cold storage strategy, wherein the trend feature set includes meteorological environment trend features, electricity price change trend features, and demand change trend features;

[0013] Based on the green economic demand information and the preset optimization algorithm, the cold storage strategy generated by the federal cold storage strategy model is solved to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system.

[0014] By adopting the above technical solution, user demand information is received from the interactive terminal in real time, including specific cold storage demand and user demand for green economy, and electricity price fluctuation information and meteorological information within the preset interval are dynamically obtained. The cold storage demand information, electricity price fluctuation information and meteorological information are comprehensively input into the pre-trained prediction model. The model predicts the meteorological environment trend, electricity price change trend and cold storage demand change trend based on the input data; the equipment status information is obtained, and according to the preset matching conditions, the corresponding pre-trained federal cold storage strategy model is matched. The federal cold storage strategy model is a model trained by federated learning, which has stronger generalization ability. By aggregating data from multiple cooling systems with similar characteristics, highly personalized and precise strategy formulation is achieved; the trend feature set output by the prediction model is input into the selected federal cold storage strategy model together with the equipment status information to generate a specific cold storage strategy. The cold storage strategy not only takes into account the current cold storage demand and electricity price conditions, but also incorporates future meteorological and Electricity price prediction, so as to adjust the water storage parameters and phase change storage parameters in a targeted manner; based on the user's green economic demand information and the preset group optimization algorithm, the generated storage strategy is further optimized and solved, and the optimal storage strategy set based on the joint deployment of water storage parameters and phase change storage parameters is output to the cooling system, so as to ensure that while meeting the cooling demand, the economic and environmental benefits are maximized; this application realizes the efficient joint control of water storage and phase change storage equipment through the application of the joint control method and the federal storage strategy model. On the basis of the joint deployment of water storage parameters and phase change storage parameters, it not only significantly improves the cooling efficiency, but also promotes green economy and energy efficiency optimization; by deploying the federal storage strategy model, the cooling system can quickly match the federal storage strategy model that best suits the current situation when facing new cooling demand, thereby generating a highly personalized and precise cooling strategy, ensuring the effectiveness and efficiency of the strategy, and meeting the user's dual needs for comfort and economy, and improving the user experience.

[0015] In a preferred example, the present application can be further configured as follows: the prediction model includes an identification layer, an adjustment layer, an association layer, and an analysis layer, and the step of inputting the cold storage demand information, the electricity price fluctuation information, and the meteorological information into the pre-trained prediction model to perform prediction based on the meteorological environment trend, the electricity price change trend, and the demand change trend includes the steps of:

[0016] The recognition layer performs multimodal data type recognition on cold storage demand information, electricity price fluctuation information, and meteorological information;

[0017] The adjustment layer dynamically adjusts the weight of each modal data type based on the preset attention mechanism and cold storage demand information;

[0018] The correlation layer performs cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, and generates correlation features containing spatiotemporal correlation information;

[0019] The analysis layer predicts the meteorological environment trend, electricity price change trend and demand change trend based on the associated features and pre-stored historical data, and outputs a set of trend features.

[0020] By adopting the above technical solution, within the framework of the prediction model, the cold storage demand information, electricity price fluctuation information and meteorological information are input and the prediction model is used to process them; the recognition layer performs modal recognition on the input multi-source data to distinguish their respective data types, the adjustment layer uses the preset attention mechanism to dynamically adjust the weights of various types of data in the overall analysis according to the cold storage demand information, the association layer performs deep cross-modal information fusion and spatiotemporal correlation analysis on each modal data, and extracts the correlation features containing spatiotemporal correlation information; the analysis layer, based on the correlation features and combined with the pre-stored historical data, comprehensively predicts the meteorological environment trend, electricity price change trend and demand change trend, and outputs the characteristics of meteorological environment trend, electricity price change trend and demand change. A set of trend features of trend features; This application realizes the comprehensive integration and accurate prediction of cold storage demand, electricity price fluctuations and meteorological information by constructing a comprehensive prediction model including an identification layer, an adjustment layer, a correlation layer and an analysis layer. The model can not only automatically identify and distinguish data types of different modes, but also dynamically adjust the weight of each data type in the analysis according to actual needs, ensuring the effective extraction of key information, and through cross-modal information fusion and spatiotemporal correlation analysis, it explores the intrinsic connections and spatiotemporal distribution laws between data of each mode, thereby generating more representative correlation features, which not only improves the accuracy and timeliness of the prediction, but also significantly enhances the pertinence and practicality of strategy formulation, providing strong support for energy conservation and consumption reduction, optimizing resource allocation and improving the overall performance of the system.

[0021] In a preferred example, the present application can be further configured as follows: the steps of respectively obtaining the device status information of the water cooling device and the phase change cooling device, and matching the corresponding pre-trained federated cooling strategy model based on the preset matching conditions, include the steps of:

[0022] Construct a cooling feature vector based on the equipment status information and user demand information of the water cooling equipment and the phase change cooling equipment;

[0023] The federal cold storage strategy model corresponding to the cooling feature vector is matched through the preset similarity matching rules.

[0024] By adopting the above technical solution, after obtaining the real-time status information of the water-cooled storage equipment and the phase-change cold storage equipment, combined with the user's demand information, a feature vector based on the equipment status that comprehensively reflects the cooling demand status is constructed. The feature vector not only contains key information such as the equipment's operating status and performance parameters, but also incorporates the user's actual needs and preferences; using the preset similarity matching rules, the federal cold storage strategy model library is searched and matched to the federal cold storage strategy model that best matches the current cooling feature vector; this application integrates the status information of the water-cooled storage equipment and the phase-change cold storage equipment and the user's demand information to construct an accurate cooling feature vector, and with the help of the similarity matching mechanism, it achieves fast and accurate matching of the federal cold storage strategy model.

[0025] In a preferred example, the present application can be further configured as follows: the federal cold storage strategy model includes a water cold storage matching layer, a phase change cold storage matching layer and an integration layer, and the trend feature set output by the prediction model and the device status information are input into the matching federal cold storage strategy model to generate a cold storage strategy, wherein the trend feature set includes meteorological environment trend features, electricity price change trend features and demand change trend features, including the steps of:

[0026] The water storage matching layer matches several water storage strategy items from the pre-stored water storage strategy library based on the meteorological environment trend characteristics, demand change trend characteristics and equipment status information;

[0027] The phase change cold storage matching layer matches several phase change cold storage strategy items from the pre-stored phase change cold storage strategy library based on the meteorological environment trend characteristics, electricity price change trend characteristics and equipment status information;

[0028] The integration layer integrates several water cooling strategy items and several phase change cooling strategy items based on the demand change trend characteristics, thereby generating several cooling strategies.

[0029] By adopting the above technical solution, in the federal cold storage strategy model, the water cold storage matching layer receives the meteorological environment trend characteristics, demand change trend characteristics and equipment status information output by the prediction model, and matches a number of suitable water cold storage strategy items from the pre-stored rich water cold storage strategy library. At the same time, the phase change cold storage matching layer matches a number of suitable phase change cold storage strategy items from the phase change cold storage strategy library based on the meteorological environment trend characteristics, electricity price change trend characteristics and equipment status information. The integration layer receives the water cold storage strategy items and the phase change cold storage strategy items, integrates them based on the demand change trend characteristics, and finally generates a number of cold storage strategies. The present application constructs a federal cold storage strategy model including a water cold storage matching layer, a phase change cold storage matching layer and an integration layer to achieve comprehensive consideration and precise matching of multiple trend characteristics and equipment status, improve the pertinence and effectiveness of strategy generation, and improve the processing efficiency and response speed of the system through layered processing.

[0030] In a preferred example, the present application can be further configured as follows: the step of solving the cold storage strategy generated by the federal cold storage strategy model based on the green economic demand information and the preset optimization algorithm to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system includes the steps of:

[0031] A. Randomly select several cold storage strategies as the initial cold storage strategy solutions;

[0032] B. Determine a simulated annealing parameter set and define an economic objective function based on the green economic demand information, wherein the simulated annealing parameter set includes an initial temperature, a cooling function, a termination temperature, a maximum number of iterations, and a Metropolis parameter;

[0033] C. Obtain the neighborhood of each initialization cold storage strategy solution respectively, generate the candidate cold storage strategy corresponding to each initialization cold storage strategy solution through the preset random perturbation rule, and set each initialization cold storage strategy solution and its corresponding candidate cold storage strategy as the same group of verification sets;

[0034] D. Calculate the difference between the same validation set based on the economic objective function , and through the Metropolis parameters To verify;

[0035] E. If If the verification is successful, the current candidate cold storage strategy is updated to a new initialization cold storage strategy solution, and the current temperature is updated based on the cooling function. The current temperature when verification is performed is the initial temperature;

[0036] F. Taking steps C, D, and E as one iteration, repeat the iteration until a termination condition is reached, wherein the termination condition includes that the number of iterations reaches the maximum number of iterations and the initial temperature is reduced to below the termination temperature by a cooling function;

[0037] G. Unpack the current initialized cold storage strategy in each group of the same verification set into the optimal cold storage strategy set and output it to the cooling system.

[0038] By adopting the above technical scheme, several cold storage strategies are randomly selected as the initialization cold storage strategy solutions, and the simulated annealing parameter set is determined, which includes the initial temperature, the cooling function, the termination temperature, the maximum number of iterations and the Metropolis parameter, and the economic objective function is defined based on the green economic demand information; for each initialization cold storage strategy solution, its neighborhood is obtained, and the corresponding candidate cold storage strategy is generated through the preset random perturbation rule, and each initialization cold storage strategy solution and its candidate cold storage strategy are grouped into the same verification set; the difference of the same verification set is calculated based on the economic objective function, and the Metropolis parameter is used for verification. If the verification passes, the current candidate cold storage strategy is updated to the new initialization cold storage strategy solution, and the cooling function is used to generate the corresponding candidate cold storage strategy. The function updates the current temperature, and this process constitutes an iteration, and the iteration is repeated until the termination condition is reached, that is, the number of iterations reaches the maximum number of iterations or the initial temperature drops below the termination temperature; finally, the current initialized cold storage strategy in each group of the same group of verification sets is decomposed into an optimal cold storage strategy set, and output to the cooling system; this application solves the federal cold storage strategy model by adopting a simulated annealing algorithm combined with green economic demand information, realizes the optimal selection of cold storage strategies, and effectively improves the economy and feasibility of cold storage strategies. At the same time, by setting the termination condition to control the iterative process, the efficiency and stability of the algorithm are ensured, which provides a more scientific and reasonable cold storage strategy set for the cooling system, with significant energy-saving and consumption-reducing and economic benefit-enhancing effects.

[0039] In a preferred example, the present application can be further configured as follows: the difference between the same group of verification sets is calculated based on the economic objective function , and through the Metropolis parameters Perform verification, including:

[0040] like <0, then Verification passed;

[0041] like > 0, the acceptance probability P is calculated based on the Metropolis criterion, and its calculation formula is: , where K is the Boltzmann constant and T is the current temperature;

[0042] Generate a random number r in the interval [0,1) and compare r with P. If r < P, then Verification passed, if r ≥ P, then Verification failed.

[0043] By adopting the above technical solution, the difference in economic objective function between the current initialized cold storage strategy solution and its candidate cold storage strategy is calculated. If the difference is less than 0, it means that the candidate cold storage strategy is better than the current initialized cold storage strategy solution in terms of economy, and the verification is passed directly; if the difference is greater than 0, the acceptance probability P is calculated according to the Metropolis criterion, which reflects the probability that the system accepts a worse solution at the current temperature. A random number r is generated in the interval [0,1) and compared with the acceptance probability P; if r is less than P, it means that even if the candidate cold storage strategy is slightly inferior to the current initialized cold storage strategy in terms of economy. solution, but according to the Metropolis criterion, it still has a certain probability of being accepted, and the verification passes at this time; if r is greater than or equal to P, the candidate cold storage strategy is not accepted and the verification fails; the application realizes the optimal selection of the cold storage strategy by combining the difference of the economic objective function and the Metropolis parameter for verification. It not only directly compares the economy, but also introduces the Metropolis criterion, so that even if a solution with slightly worse economy is encountered during the iteration process, it has a certain probability of being accepted, thereby increasing the search space and diversity of the algorithm, which is helpful to find the global optimal solution or the approximate optimal solution.

[0044] The second object of the invention is achieved by the following technical solutions:

[0045] A combined control system based on water storage and phase change storage in a cooling system, comprising:

[0046] An information acquisition module, configured to acquire electricity price fluctuation information and meteorological information within a preset interval when receiving user demand information from an interactive terminal, wherein the user demand information includes cold storage demand information and green economy demand information;

[0047] The first input module is used to input the cold storage demand information, electricity price fluctuation information and meteorological information into the pre-trained prediction model to make predictions based on the meteorological environment trend, electricity price change trend and demand change trend;

[0048] A matching module, used to obtain the device status information of the water cooling device and the phase change cooling device respectively, and match the corresponding pre-trained federated cooling storage strategy model based on the preset matching conditions;

[0049] A second input module is used to input the trend feature set output by the prediction model and the equipment status information into the matching federal cold storage strategy model to generate a cold storage strategy, wherein the trend feature set includes meteorological environment trend features, electricity price change trend features, and demand change trend features;

[0050] The strategy solving module is used to solve the cold storage strategy generated by the federal cold storage strategy model based on the green economic demand information and the preset optimization algorithm, so as to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system.

[0051] By adopting the above technical solution, the information acquisition module is used to obtain the electricity price fluctuation information and meteorological information within a preset interval time when receiving the user demand information from the interactive terminal, and the user demand information includes the cold storage demand information and the green economy demand information; the first input module is used to input the cold storage demand information, the electricity price fluctuation information and the meteorological information into the pre-trained prediction model to make predictions based on the meteorological environment trend, the electricity price change trend and the demand change trend; the matching module is used to respectively obtain the equipment status information of the water cold storage equipment and the phase change cold storage equipment, and match the corresponding pre-trained federal cold storage strategy model based on the preset matching conditions; the second input module is used to input the trend feature set and the equipment status information output by the prediction model into the matching federal cold storage strategy model to generate the cold storage strategy, and the trend feature set includes the meteorological environment trend feature, the electricity price change trend feature and the demand change trend feature; the strategy solving module is used to solve the cold storage strategy generated by the federal cold storage strategy model based on the green economy demand information and the preset optimization algorithm, so as to output the optimal cold storage strategy set based on the joint deployment of the water cold storage parameters and the phase change cold storage parameters to the cooling system.

[0052] In a preferred example, the present application can be further configured as follows:

[0053] The recognition layer module is used to identify multimodal data types of cold storage demand information, electricity price fluctuation information and meteorological information respectively;

[0054] The adjustment layer module is used to dynamically adjust the weight of each modal data type based on the preset attention mechanism and cold storage demand information;

[0055] The association layer module is used to perform cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, and generate association features containing spatiotemporal correlation information;

[0056] The analysis layer module is used to predict the meteorological environment trend, electricity price change trend and demand change trend based on the associated features and pre-stored historical data, and output a trend feature set.

[0057] By adopting the above technical scheme, the identification layer module is used to identify the multimodal data types of cold storage demand information, electricity price fluctuation information and meteorological information respectively; the adjustment layer module is used to dynamically adjust the weights of each modal data type based on a preset attention mechanism and cold storage demand information; the association layer module is used to perform cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, and generate correlation features containing spatiotemporal correlation information; the analysis layer module is used to predict the meteorological environment trend, electricity price change trend and demand change trend based on the correlation features and pre-stored historical data, and output a trend feature set.

[0058] In summary, the present application includes at least one of the following beneficial technical effects:

[0059] 1. This application realizes efficient joint control of water cold storage and phase change cold storage equipment through the application of joint control method and federal cold storage strategy model, which not only significantly improves cold storage efficiency, but also promotes green economy and energy efficiency optimization; by deploying federal cold storage strategy model, when facing new cold storage demand, the cooling system can quickly match it with the federal cold storage strategy model that best suits the current situation, thereby generating a highly personalized and precise cold storage strategy, ensuring the effectiveness and efficiency of the strategy, meeting the dual needs of users for comfort and economy, and improving the user experience;

[0060] 2. This application realizes the comprehensive integration and accurate prediction of cold storage demand, electricity price fluctuations and meteorological information by constructing a comprehensive prediction model including an identification layer, an adjustment layer, a correlation layer and an analysis layer. The model can not only automatically identify and distinguish data types of different modes, but also dynamically adjust the weight of each data type in the analysis according to actual needs, ensuring the effective extraction of key information, and through cross-modal information fusion and spatiotemporal correlation analysis, the intrinsic connection and spatiotemporal distribution law between the modal data are mined, thereby generating more representative correlation features, which not only improves the accuracy and timeliness of the prediction, but also significantly enhances the pertinence and practicality of strategy formulation, providing strong support for energy saving and consumption reduction, optimizing resource allocation and improving the overall performance of the system;

[0061] 3. This application constructs an accurate cooling feature vector by integrating the status information of water cooling equipment and phase change cooling equipment and user demand information, and uses a similarity matching mechanism to achieve fast and accurate matching of the federal cooling strategy model;

[0062] 4. This application constructs a federal cold storage strategy model that includes a water cold storage matching layer, a phase change cold storage matching layer, and an integration layer to achieve comprehensive consideration and precise matching of multiple trend characteristics and equipment status, improve the pertinence and effectiveness of strategy generation, and improve the processing efficiency and response speed of the system through layered processing;

[0063] 5. This application uses a simulated annealing algorithm combined with green economic demand information to solve the federal cold storage strategy model, achieves optimal selection of cold storage strategies, and effectively improves the economy and feasibility of cold storage strategies. At the same time, by setting termination conditions to control the iteration process, the efficiency and stability of the algorithm are ensured, providing a more scientific and reasonable cold storage strategy set for the cooling system, which has significant effects of energy saving, consumption reduction and economic benefit improvement;

[0064] 6. This application realizes the optimal selection of the cold storage strategy by combining the difference of the economic objective function and the Metropolis parameter for verification. It not only directly compares the economic performance, but also introduces the Metropolis criterion, so that even if a solution with slightly worse economic performance is encountered during the iteration process, it has a certain probability of being accepted, thereby increasing the search space and diversity of the algorithm, which is helpful to find the global optimal solution or the approximate optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a flow chart of an embodiment of a combined control method based on water cold storage and phase change cold storage in a cooling system of the present application;

[0066] Figure 2 This is a flowchart for implementing step S20 in an embodiment of a combined control method based on water cold storage and phase change cold storage in a cooling system of the present application;

[0067] Figure 3 It is a flowchart for implementing step S40 in an embodiment of a combined control method based on water cold storage and phase change cold storage in a cooling system of the present application. DETAILED DESCRIPTION

[0068] The following is combined with Figure 1-3 This application is described in further detail.

[0069] In one embodiment, if Figure 1 As shown, the present application discloses a combined control method based on water cold storage and phase change cold storage in a cooling system, which specifically includes the following steps:

[0070] S10: when receiving user demand information from the interactive terminal, obtaining electricity price fluctuation information and meteorological information within a preset interval, the user demand information including cold storage demand information and green economy demand information;

[0071] In this embodiment, water cold storage utilizes the sensible heat of water to achieve cold storage; during the night when electricity consumption is low, the cold is stored in the water tank in the form of low-temperature chilled water through refrigeration by a chiller; during the daytime when electricity consumption is peak, the stored low-temperature chilled water is used to provide a cold source for the air-conditioning system; it can utilize the fire water tank of a large building to store cold, so it has a wide range of application prospects. However, water cold storage technology also has some disadvantages, such as large space occupation, large cold loss, high requirements for cold preservation and waterproof measures of the cold storage water tank, and because the water tank is partially open, the circulating water is easily contaminated; phase change cold storage utilizes the principle that phase change materials release or absorb a large amount of latent heat when the temperature changes to achieve cold storage; when the ambient temperature rises to the phase change temperature, the phase change material changes from solid to liquid to absorb heat, and when the ambient temperature drops below the phase change temperature, the phase change material changes from liquid to solid to release heat; the phase change cold storage system has a higher heat storage density and a more stable indoor temperature control effect, The high heat storage density of phase change materials enables the system to store heat more efficiently. The heat storage and release process of phase change materials is controllable, which can achieve rapid response and stable control of indoor temperature. Phase change materials generally use non-toxic and non-polluting substances at normal temperature and will not pollute the environment. The interactive terminal is an interface or device for information exchange between users and the cooling system. Users can input cooling demand information, green economic demand information, etc. through the interactive terminal, and the cooling system displays relevant information or control results to users through the interactive terminal. Electricity price fluctuation information is information reflecting the fluctuation of electricity prices. Meteorological information is information reflecting the current environmental weather within the preset range. Cooling requirement information is the user's specific demand for cooling, including detailed parameters such as the start time and end time of cooling, the size of cooling capacity, and the cooling temperature. Green economic demand information is the user's requirement information on environmental protection and energy saving, involving the user's restrictions on energy consumption, preference for renewable energy, and budget for operating costs.

[0072] Specifically, it receives user demand information from interactive terminals in real time, including specific cold storage demand and user demand for green economy, and dynamically obtains electricity price fluctuation information and meteorological information within a preset interval;

[0073] S20: Inputting the cold storage demand information, electricity price fluctuation information and meteorological information into the pre-trained prediction model to perform prediction based on the meteorological environment trend, electricity price change trend and demand change trend;

[0074] In this embodiment, the prediction model is a mathematical model constructed using machine learning or statistical learning methods. Specifically, the prediction model is trained to predict future meteorological conditions, electricity price changes, and changing trends in cold storage demand; the meteorological environment trend is the change characteristics of meteorological conditions in a future period of time predicted by the prediction model through analyzing input meteorological information (such as temperature, humidity, precipitation, etc.); the electricity price change trend is the high and low change characteristics of future electricity prices predicted by the prediction model through analyzing input electricity price fluctuation information; the demand change trend is the change trend characteristics of future cold storage demand predicted by the prediction model through analyzing historical cold storage demand information and current user demand;

[0075] Specifically, the cold storage demand information, electricity price fluctuation information and meteorological information are comprehensively input into the pre-trained prediction model. The model predicts the meteorological environment trend, electricity price change trend and cold storage demand change trend based on the input data;

[0076] S30: respectively obtaining device status information of the water cooling device and the phase change cooling device, and matching the corresponding pre-trained federated cooling storage strategy model based on a preset matching condition;

[0077] In this embodiment, the water cooling device is a device that uses the sensible heat characteristics of water to store and release cold. During periods when the cooling demand is low or the electricity price is cheap (such as at night), the device stores cold in water through a refrigerator. When the cooling demand increases or the electricity price is high, the stored cold can be released to meet the needs of air conditioning or industrial cooling, thereby helping to save energy and reduce operating costs; the phase change cooling device is a device that uses the property of a substance to absorb or release a large amount of heat during phase change (such as the transition between solid and liquid, liquid and gas) to store and release cold. Common phase change materials include ice, salt mixtures, waxy substances, etc. The phase change cooling device stores cold when the cooling demand is low and releases it when the demand is high, so as to achieve efficient energy utilization; the device status information is detailed information about the current operating status, performance parameters, maintenance status, and possible faults or problems of the water cooling device and the phase change cooling device. information; the preset matching conditions are a series of standards or rules for selecting the federal cold storage strategy model that best suits the current equipment status and user needs, including equipment type, specifications, performance parameters, current operating status, historical operating data, user demand characteristics, electricity price fluctuations, meteorological conditions, and environmental protection and economic benefit goals; the federal cold storage strategy model is a distributed machine learning model based on machine learning or artificial intelligence algorithms and constructed by federal training, which is used to generate optimal or approximately optimal cold storage strategies based on input equipment status information, prediction results, and user needs, etc. It allows independent training on multiple devices or cooling systems without the need to centralize the data of these devices and cooling systems to a central location, wherein the matched federal cold storage strategy model corresponds to the equipment status information and matching conditions of the water cold storage device and the phase change cold storage device, and is the federal cold storage strategy model that best suits the current cooling system;

[0078] Specifically, the device status information is obtained, and the corresponding pre-trained federated cold storage strategy model is matched according to the preset matching conditions. The federated cold storage strategy model is a model trained through federated learning, which has stronger generalization ability. By aggregating data from multiple cooling systems with similar characteristics, highly personalized and precise strategy formulation can be achieved.

[0079] S40: inputting the trend feature set output by the prediction model and the device status information into the matching federal cold storage strategy model to generate a cold storage strategy, wherein the trend feature set includes meteorological environment trend features, electricity price change trend features, and demand change trend features;

[0080] In this embodiment, the meteorological environment trend feature is a feature that reflects the possible change trend of future meteorological conditions (such as temperature, humidity, precipitation, etc.); the electricity price change trend feature is a feature that describes the possible change trend of future electricity prices, including the peak and valley periods of electricity prices; the demand change trend feature is a feature that indicates the possible change of future users or market demand for cooling, including the increase or decrease of demand, the time of occurrence of demand peak, etc.;

[0081] Specifically, the trend feature set output by the prediction model is input into the selected federal cold storage strategy model together with the equipment status information to generate a specific cold storage strategy, which not only considers the current cold storage demand and electricity price situation, but also incorporates future weather and electricity price forecasts;

[0082] S50: solving the cold storage strategy generated by the federal cold storage strategy model based on the green economic demand information and the preset optimization algorithm, so as to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system;

[0083] In this embodiment, the preset optimization algorithm is a method used in mathematics or computer science to find the optimal solution or the approximate optimal solution. In the formulation of the cold storage strategy, the optimization algorithm will iteratively calculate and optimize the cold storage strategy generated by the federal cold storage strategy model according to the input green economic demand information, equipment status information, and the trend feature set output by the prediction model, so as to find the optimal strategy that meets specific goals (such as minimum cost, minimum energy consumption, maximum benefit, etc.); the optimal cold storage strategy set is a set of cold storage strategies that meet the constraints such as green economic demand information, equipment status information and the trend feature set output by the prediction model after solving by the optimization algorithm, and achieve the optimal or approximate optimal cold storage strategy on certain goals (such as cost, energy consumption, benefit, etc.), which includes parameters of water cold storage and parameters of phase change cold storage, such as the start and end time of cold storage, cold storage capacity, operation mode of cold storage equipment and parameter settings and other specific information; the cooling system is a system for providing cooling services, including components such as refrigerators, cold storage equipment, transmission pipelines, and control systems;

[0084] Specifically, based on the user's green economic demand information and the preset group optimization algorithm, the generated cold storage strategy is further optimized and solved, and the optimal cold storage strategy set is output to the cooling system, thereby ensuring that the economic and environmental benefits are maximized while meeting the cold storage needs.

[0085] In one embodiment, the prediction model includes a recognition layer, an adjustment layer, a correlation layer, and an analysis layer. Figure 2 As shown, step S20 includes the steps of:

[0086] S21: The recognition layer performs multimodal data type recognition on the cold storage demand information, electricity price fluctuation information and meteorological information respectively;

[0087] S22: The adjustment layer dynamically adjusts the weight of each modal data type based on the preset attention mechanism and cold storage demand information;

[0088] S23: The correlation layer performs cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, and generates correlation features containing spatiotemporal correlation information;

[0089] S24: The analysis layer predicts the meteorological environment trend, the electricity price change trend, and the demand change trend based on the associated features and the pre-stored historical data, and outputs a trend feature set;

[0090] In this embodiment, the recognition layer is the first layer of the prediction model, which is used to receive input cold storage demand information, electricity price fluctuation information and meteorological information, and perform multimodal data type recognition on this information, so as to distinguish and process different types of data, such as text, numerical values, time series, etc.; multimodal data type recognition is the process of the recognition layer classifying and identifying the input cold storage demand, electricity price fluctuation and meteorological information; the adjustment layer is the second layer of the prediction model, which is used to dynamically adjust the weights of each modal data type based on the preset attention mechanism and cold storage demand information; the attention mechanism is a mechanism that allows the model to pay attention to important information and ignore irrelevant information, which helps to improve the prediction accuracy of the model; the weights of each modal data type are dynamically adjusted. The adjustment layer dynamically adjusts the weights of relevant data in the model according to the importance and urgency of cold storage demand, as well as the correlation between electricity price fluctuations and meteorological information; the association layer is the third layer of the prediction model, which is used to perform cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, where cross-modal information fusion combines different types of data to extract common features and related information between them, and spatiotemporal correlation analysis focuses on the relationship and change rules of data in time and space; the analysis layer is the last layer of the prediction model, which is used to predict meteorological environment trends, electricity price change trends, and demand change trends based on the correlation features generated by the association layer and pre-stored historical data;

[0091] Specifically, within the framework of the prediction model, cold storage demand information, electricity price fluctuation information and meteorological information are input and processed by the prediction model; the recognition layer performs modal recognition on the input multi-source data to distinguish their respective data types, the adjustment layer uses a preset attention mechanism to dynamically adjust the weights of various types of data in the overall analysis according to the cold storage demand information, the association layer performs deep cross-modal information fusion and spatiotemporal correlation analysis on each modal data, and extracts correlation features containing spatiotemporal correlation information; the analysis layer, based on the correlation features and combined with pre-stored historical data, comprehensively predicts the meteorological environment trend, electricity price change trend and demand change trend, and outputs a trend feature set including meteorological environment trend features, electricity price change trend features and demand change trend features.

[0092] In one embodiment, step S30 includes the steps of:

[0093] S31: constructing a cooling feature vector based on the equipment status information of the water cooling equipment and the phase change cooling equipment and the user demand information;

[0094] S32: matching the federal cold storage strategy model corresponding to the cooling feature vector through a preset similarity matching rule;

[0095] In this embodiment, the cooling feature vector is a numerical representation of the equipment status information of the water cooling equipment and the phase change cooling equipment and the user demand information, which includes multiple dimensions, each dimension represents a specific feature, such as the temperature, pressure, cooling capacity, operation status (on / off), peak / valley time period of user demand, demand size, etc. of the equipment; the similarity matching rule is a rule or algorithm for comparing and matching the similarity between different objects. Specifically, the similarity matching rule is used to compare the similarity between the cooling feature vector and the federal cooling strategy model;

[0096] Specifically, after obtaining the real-time status information of the water-cooled storage equipment and the phase-change cool storage equipment, combined with the user's demand information, a feature vector based on the equipment status is constructed to comprehensively reflect the cooling demand status. This feature vector not only includes key information such as the equipment's operating status and performance parameters, but also incorporates the user's actual needs and preferences. Using the preset similarity matching rules, the federal cool storage strategy model library is searched and matched to the federal cool storage strategy model that best matches the current cooling feature vector.

[0097] In one embodiment, the federated cold storage strategy model includes a water cold storage matching layer, a phase change cold storage matching layer, and an integration layer, such as Figure 3 As shown, step S40 includes the steps of:

[0098] S41: The water storage cooling matching layer matches a number of water storage cooling strategy items from a pre-stored water storage cooling strategy library based on the meteorological environment trend characteristics, demand change trend characteristics and equipment status information;

[0099] S42: The phase change cold storage matching layer matches a number of phase change cold storage strategy items from a pre-stored phase change cold storage strategy library based on the meteorological environment trend characteristics, the electricity price change trend characteristics and the equipment status information;

[0100] S43: the integration layer integrates a number of water cooling strategy items and a number of phase change cooling strategy items based on the demand change trend characteristics, thereby generating a number of cooling strategies;

[0101] In this embodiment, the water storage matching layer is the first layer of the federal cold storage strategy model, which is used to process the strategy matching related to water storage. It selects a number of water storage strategy items suitable for the current situation from the pre-stored water storage strategy library according to the trend characteristics of the meteorological environment (such as future temperature, humidity, etc. prediction), demand change trend characteristics (such as changes in user cooling demand) and equipment status information (such as the current status and performance of the water storage equipment). The water storage strategy library is a database or collection that stores a variety of water storage strategies. The water storage strategy is based on historical data, experience or simulation experiments, and is used to guide the operation of water storage equipment under different conditions. The phase change cold storage matching layer is the second layer of the federal cold storage strategy model, which is used for phase change cold storage strategy matching. It is based on the trend characteristics of the meteorological environment and the trend characteristics of the electricity price change (because the phase change cold storage equipment may be more affected by the fluctuation of the electricity price). The integration layer is the last layer of the federal cold storage strategy model, which is used to integrate the strategy items selected by the water cold storage matching layer and the phase change cold storage matching layer. It is based on the demand change trend characteristics (i.e., the user's expectation of future cooling demand changes), comprehensively considers the advantages and limitations of water cold storage and phase change cold storage, and generates several comprehensive cold storage strategies. The cold storage strategy is the strategy information output by the integration layer, that is, the specific cold storage plan or scheme formulated according to the current and future meteorological environment, electricity price, user demand and equipment status, including detailed information such as the start and end time of cold storage, cold storage capacity, and equipment operation mode.

[0102] Specifically, in the federal cold storage strategy model, the water cold storage matching layer receives the meteorological environment trend characteristics, demand change trend characteristics and equipment status information output by the prediction model, and matches a number of suitable water cold storage strategy items from the pre-stored rich water cold storage strategy library. At the same time, the phase change cold storage matching layer matches a number of suitable phase change cold storage strategy items from the phase change cold storage strategy library based on the meteorological environment trend characteristics, electricity price change trend characteristics and equipment status information. The integration layer receives the water cold storage strategy items and the phase change cold storage strategy items, integrates them based on the demand change trend characteristics, and finally generates a number of cold storage strategies.

[0103] In one embodiment, step S50 includes the steps of:

[0104] A. Randomly select several cold storage strategies as the initial cold storage strategy solutions;

[0105] B. Determine a simulated annealing parameter set and define an economic objective function based on the green economic demand information, wherein the simulated annealing parameter set includes an initial temperature, a cooling function, a termination temperature, a maximum number of iterations, and a Metropolis parameter;

[0106] C. Obtain the neighborhood of each initialization cold storage strategy solution respectively, generate the candidate cold storage strategy corresponding to each initialization cold storage strategy solution through the preset random perturbation rule, and set each initialization cold storage strategy solution and its corresponding candidate cold storage strategy as the same group of verification sets;

[0107] D. Calculate the difference between the same validation set based on the economic objective function , and through the Metropolis parameters To verify;

[0108] E. If If the verification is successful, the current candidate cold storage strategy is updated to a new initialization cold storage strategy solution, and the current temperature is updated based on the cooling function. The current temperature when verification is performed is the initial temperature;

[0109] F. Taking steps C, D, and E as one iteration, repeat the iteration until a termination condition is reached, wherein the termination condition includes that the number of iterations reaches the maximum number of iterations and the initial temperature is reduced to below the termination temperature by a cooling function;

[0110] G. Unpack the current initialization cold storage strategy in each group of the same verification set into an optimal cold storage strategy set and output it to the cooling system;

[0111] In this embodiment, the simulated annealing parameter set is a key parameter for controlling the operation of the simulated annealing algorithm, including the initial temperature, the cooling function, the termination temperature, the maximum number of iterations and the Metropolis parameter. These parameters jointly determine the algorithm's search efficiency, convergence speed and the quality of the final solution; the economic objective function is a function defined based on the green economic demand information, which is used to evaluate the economic efficiency of the cold storage strategy; the neighborhood is a set of solutions adjacent to or similar to each initialized cold storage strategy solution; the random perturbation rule is a rule or method for generating candidate cold storage strategies. By applying random perturbations to the initialized cold storage strategy solutions, new and possibly better solutions can be generated; the same group verification set is a set formed by combining each initialized cold storage strategy solution with its corresponding candidate cold storage strategy; the Metropolis parameter is an important parameter in the simulated annealing algorithm, which is used to control the probability of accepting a new solution. Even if the economic performance of the new solution is worse than that of the current solution, there is a certain probability that it will be accepted, which helps the algorithm to jump out of the local optimal solution and explore a broader solution space; The economic performance difference between the current candidate cold storage strategy and the current initialized cold storage strategy solution (or the current optimal solution) calculated by the economic objective function; the cooling function is a function or rule used to gradually reduce the current temperature. As the temperature decreases, the probability of the algorithm accepting a new solution will gradually decrease, thereby gradually converging to the optimal solution; the optimal cold storage strategy set is the final solution set output after the algorithm runs, which includes all the cold storage strategies with the best economic performance obtained through simulated annealing optimization;

[0112] Specifically, several cold storage strategies are randomly selected as the initialization cold storage strategy solutions, and a simulated annealing parameter set is determined, which includes the initial temperature, the cooling function, the termination temperature, the maximum number of iterations and the Metropolis parameter, and the economic objective function is defined based on the green economic demand information; for each initialization cold storage strategy solution, its neighborhood is obtained, and the corresponding candidate cold storage strategy is generated through a preset random perturbation rule, and each initialization cold storage strategy solution and its candidate cold storage strategy are combined into a verification set of the same group; the difference of the same group of verification sets is calculated based on the economic objective function, and the Metropolis parameter is used for verification. If the verification passes, the current candidate cold storage strategy is updated to a new initialization cold storage strategy solution, and the current temperature is updated according to the cooling function. This process constitutes an iteration, and the iteration is repeated until the termination condition is reached, that is, the number of iterations reaches the maximum number of iterations or the initial temperature is reduced to below the termination temperature; finally, the current initialization cold storage strategy solution in each group of verification sets is integrated into the optimal cold storage strategy set, and output to the cooling system.

[0113] In one embodiment, step D comprises:

[0114] like <0, then Verification passed;

[0115] like > 0, the acceptance probability P is calculated based on the Metropolis criterion, and its calculation formula is: , where K is the Boltzmann constant and T is the current temperature;

[0116] Generate a random number r in the interval [0,1) and compare r with P. If r < P, then Verification passed, if r ≥ P, then Verification failed.

[0117] In this embodiment, the Metropolis criterion is a criterion used in the simulated annealing algorithm to decide whether to accept a poor solution (i.e., a candidate cold storage strategy with poor economic performance). It is based on the annealing process in physics and allows the algorithm to accept poor solutions with a certain probability. The acceptance probability P is calculated according to the Metropolis criterion and is used to decide whether to accept a poor solution. The Boltzmann constant is a constant in physics that is used to describe the relationship between the temperature and energy of a particle system. In the simulated annealing algorithm, it is usually used as an adjustment parameter to control the acceptance probability. The probability of accepting a poor solution, the larger the value of K, the higher the probability of accepting a poor solution, conversely, the smaller the value of K, the lower the probability of accepting a poor solution; the current temperature is an important parameter in the simulated annealing algorithm, which gradually decreases with the iteration of the algorithm. The temperature determines the probability of the algorithm accepting a poor solution. At the beginning of the algorithm, the temperature is high and the probability of accepting a poor solution is also high. As the temperature gradually decreases, the probability of accepting a poor solution also gradually decreases, and the algorithm gradually converges to the optimal solution; the random number r is a random number generated in the interval [0,1) and is used to compare with the acceptance probability P;

[0118] Specifically, the difference in economic objective function between the current initialized cold storage strategy solution and its candidate cold storage strategy is calculated. If the difference is less than 0, it means that the candidate cold storage strategy is better than the current initialized cold storage strategy solution in terms of economy, and the verification is passed directly; if the difference is greater than 0, the acceptance probability P is calculated according to the Metropolis criterion, which reflects the probability that the system accepts a worse solution at the current temperature. A random number r is generated in the interval [0,1) and compared with the acceptance probability P; if r is less than P, it means that even if the candidate cold storage strategy is slightly inferior to the current initialized cold storage strategy solution in terms of economy, it still has a certain probability of being accepted according to the Metropolis criterion, and the verification is passed at this time; if r is greater than or equal to P, the candidate cold storage strategy is not accepted and the verification fails.

[0119] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0120] In one embodiment, a joint control system based on water storage and phase change storage in a cooling system is provided, and the joint control system based on water storage and phase change storage in a cooling system corresponds one-to-one to a joint control method based on water storage and phase change storage in a cooling system in the above embodiment. The joint control system based on water storage and phase change storage in a cooling system includes:

[0121] An information acquisition module, configured to acquire electricity price fluctuation information and meteorological information within a preset interval when receiving user demand information from an interactive terminal, wherein the user demand information includes cold storage demand information and green economy demand information;

[0122] The first input module is used to input the cold storage demand information, electricity price fluctuation information and meteorological information into the pre-trained prediction model to make predictions based on the meteorological environment trend, electricity price change trend and demand change trend;

[0123] A matching module, used to obtain the device status information of the water cooling device and the phase change cooling device respectively, and match the corresponding pre-trained federated cooling storage strategy model based on the preset matching conditions;

[0124] A second input module is used to input the trend feature set output by the prediction model and the equipment status information into the matching federal cold storage strategy model to generate a cold storage strategy, wherein the trend feature set includes meteorological environment trend features, electricity price change trend features, and demand change trend features;

[0125] A strategy solving module is used to solve the cold storage strategy generated by the federal cold storage strategy model based on the green economic demand information and the preset optimization algorithm, so as to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system;

[0126] Optionally, also include:

[0127] The recognition layer module is used to identify multimodal data types of cold storage demand information, electricity price fluctuation information and meteorological information respectively;

[0128] The adjustment layer module is used to dynamically adjust the weight of each modal data type based on the preset attention mechanism and cold storage demand information;

[0129] The association layer module is used to perform cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, and generate association features containing spatiotemporal correlation information;

[0130] The analysis layer module is used to predict the meteorological environment trend, electricity price change trend and demand change trend based on the associated features and pre-stored historical data, and output a trend feature set;

[0131] Optionally, matching modules include:

[0132] A feature vector construction submodule is used to construct a cooling feature vector based on the equipment status information and user demand information of the water cooling equipment and the phase change cooling equipment;

[0133] A model matching submodule, used to match the federal cold storage strategy model corresponding to the cooling feature vector through a preset similarity matching rule;

[0134] Optionally, also include:

[0135] The water storage cooling matching layer module is used to match a number of water storage cooling strategy items from a pre-stored water storage cooling strategy library based on the meteorological environment trend characteristics, demand change trend characteristics and equipment status information;

[0136] The phase change cold storage matching layer module is used to match a number of phase change cold storage strategy items from a pre-stored phase change cold storage strategy library based on the trend characteristics of the meteorological environment, the trend characteristics of the electricity price change, and the equipment status information;

[0137] An integration layer module is used to integrate a number of water cooling strategy items and a number of phase change cooling strategy items based on demand change trend characteristics, thereby generating a number of cooling strategies;

[0138] Optionally, the strategy solving module includes:

[0139] The initialization submodule is used to randomly select several cold storage strategies as the initialization cold storage strategy solutions;

[0140] A parameter determination submodule, used to determine a simulated annealing parameter set and define an economic objective function based on green economic demand information, wherein the annealing parameter set includes an initial temperature, a cooling function, a termination temperature, a maximum number of iterations, and a Metropolis parameter;

[0141] The same group verification set determination submodule is used to obtain the neighborhood of each initialized cold storage strategy solution, generate a candidate cold storage strategy corresponding to each initialized cold storage strategy solution through a preset random perturbation rule, and set each initialized cold storage strategy solution and its corresponding candidate cold storage strategy as a same group verification set;

[0142] The verification submodule is used to calculate the difference between the same group of verification sets based on the economic objective function. , and through the Metropolis parameters To verify;

[0143] The strategy update submodule is used if If the verification is successful, the current candidate cold storage strategy is updated to a new initialization cold storage strategy solution, and the current temperature is updated based on the cooling function. The current temperature when verification is performed is the initial temperature;

[0144] An iterator module is used to repeat iterations until a termination condition is reached, wherein the termination condition includes that the number of iterations reaches a maximum number of iterations and the initial temperature is reduced to below the termination temperature through a cooling function;

[0145] The output submodule is used to unpack the current initialized cold storage strategy in each group of the same verification set into an optimal cold storage strategy set and output it to the cooling system.

[0146] For the specific definition of a joint control system based on water storage and phase change storage in a cooling system, please refer to the definition of a joint control method based on water storage and phase change storage in a cooling system mentioned above, which will not be repeated here. Each module in the above-mentioned joint control system based on water storage and phase change storage in a cooling system can be fully or partially implemented by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0147] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A combined control method based on water storage and phase change storage in a cooling system, characterized in that: Includes steps: When receiving user demand information from the interactive terminal, obtaining electricity price fluctuation information and meteorological information within a preset interval, wherein the user demand information includes cold storage demand information and green economy demand information; Input the cold storage demand information, electricity price fluctuation information and meteorological information into the pre-trained prediction model to make predictions based on the meteorological environment trend, electricity price change trend and demand change trend; The device status information of the water cooling device and the phase change cooling device is obtained respectively, and the corresponding pre-trained federated cooling storage strategy model is matched based on the preset matching conditions; Inputting the trend feature set output by the prediction model and the equipment status information into the matching federal cold storage strategy model to generate a cold storage strategy, wherein the trend feature set includes meteorological environment trend features, electricity price change trend features, and demand change trend features; Based on the green economy demand information and the preset optimization algorithm, the cold storage strategy generated by the federal cold storage strategy model is solved to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system; The step of solving the cold storage strategy generated by the federal cold storage strategy model based on the green economic demand information and the preset optimization algorithm to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system includes the following steps: A. Randomly select several cold storage strategies as the initial cold storage strategy solutions; B. Determine a simulated annealing parameter set and define an economic objective function based on the green economic demand information, wherein the simulated annealing parameter set includes an initial temperature, a cooling function, a termination temperature, a maximum number of iterations, and a Metropolis parameter; C. Obtain the neighborhood of each initialization cold storage strategy solution respectively, generate the candidate cold storage strategy corresponding to each initialization cold storage strategy solution through the preset random perturbation rule, and set each initialization cold storage strategy solution and its corresponding candidate cold storage strategy as the same group of verification sets; D. Calculate the difference between the same validation set based on the economic objective function , and through the Metropolis parameters To verify; E. If If the verification is successful, the current candidate cold storage strategy is updated to a new initialization cold storage strategy solution, and the current temperature is updated based on the cooling function. The current temperature when verification is performed is the initial temperature; F. Taking steps C, D, and E as one iteration, repeat the iteration until a termination condition is reached, wherein the termination condition includes that the number of iterations reaches the maximum number of iterations and the initial temperature is reduced to below the termination temperature by a cooling function; G. Unpack the current initialized cold storage strategy in each group of the same verification set into the optimal cold storage strategy set and output it to the cooling system.

2. A combined control method based on water cold storage and phase change cold storage in a cooling system according to claim 1, characterized in that: The prediction model includes an identification layer, an adjustment layer, an association layer and an analysis layer. The step of inputting the cold storage demand information, the electricity price fluctuation information and the meteorological information into the pre-trained prediction model to perform prediction based on the meteorological environment trend, the electricity price change trend and the demand change trend includes the steps of: The recognition layer performs multimodal data type recognition on cold storage demand information, electricity price fluctuation information, and meteorological information; The adjustment layer dynamically adjusts the weight of each modal data type based on the preset attention mechanism and cold storage demand information; The correlation layer performs cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, and generates correlation features containing spatiotemporal correlation information; The analysis layer predicts the meteorological environment trend, electricity price change trend and demand change trend based on the associated features and pre-stored historical data, and outputs a set of trend features.

3. A combined control method based on water cold storage and phase change cold storage in a cooling system according to claim 1, characterized in that: The steps of respectively obtaining the device status information of the water cooling device and the phase change cooling device, and matching the corresponding pre-trained federated cooling strategy model based on the preset matching conditions, include the steps of: Construct a cooling feature vector based on the equipment status information and user demand information of the water cooling equipment and the phase change cooling equipment; The federal cold storage strategy model corresponding to the cooling feature vector is matched through the preset similarity matching rules.

4. The combined control method based on water cold storage and phase change cold storage in a cooling system according to claim 1, characterized in that: The federal cold storage strategy model includes a water cold storage matching layer, a phase change cold storage matching layer and an integration layer. The trend feature set output by the prediction model and the device status information are input into the matching federal cold storage strategy model to generate a cold storage strategy. The trend feature set includes meteorological environment trend features, electricity price change trend features and demand change trend features. The steps include: The water storage matching layer matches several water storage strategy items from the pre-stored water storage strategy library based on the meteorological environment trend characteristics, demand change trend characteristics and equipment status information; The phase change cold storage matching layer matches several phase change cold storage strategy items from the pre-stored phase change cold storage strategy library based on the meteorological environment trend characteristics, electricity price change trend characteristics and equipment status information; The integration layer integrates several water cooling strategy items and several phase change cooling strategy items based on the demand change trend characteristics, thereby generating several cooling strategies.

5. The combined control method based on water cold storage and phase change cold storage in a cooling system according to claim 1, characterized in that: The difference between the same group of validation sets is calculated based on the economic objective function , and through the Metropolis parameters Perform verification, including: like <0, then Verification passed; like > 0, the acceptance probability P is calculated based on the Metropolis criterion, and its calculation formula is: , where K is the Boltzmann constant and T is the current temperature; Generate a random number r in the interval [0,1) and compare r with P. If r < P, then Verification passed, if r ≥ P, then Verification failed.

6. A combined control system based on water storage and phase change storage in a cooling system, used in the steps of a combined control method based on water storage and phase change storage in a cooling system as claimed in any one of claims 1 to 5, characterized in that: include: An information acquisition module, configured to acquire electricity price fluctuation information and meteorological information within a preset interval when receiving user demand information from an interactive terminal, wherein the user demand information includes cold storage demand information and green economy demand information; The first input module is used to input the cold storage demand information, electricity price fluctuation information and meteorological information into the pre-trained prediction model to make predictions based on the meteorological environment trend, electricity price change trend and demand change trend; A matching module, used to obtain the device status information of the water cooling device and the phase change cooling device respectively, and match the corresponding pre-trained federated cooling storage strategy model based on the preset matching conditions; A second input module is used to input the trend feature set output by the prediction model and the equipment status information into the matching federal cold storage strategy model to generate a cold storage strategy, wherein the trend feature set includes meteorological environment trend features, electricity price change trend features, and demand change trend features; The strategy solving module is used to solve the cold storage strategy generated by the federal cold storage strategy model based on the green economic demand information and the preset optimization algorithm, so as to output the optimal cold storage strategy set based on the joint deployment of water cold storage parameters and phase change cold storage parameters to the cooling system.

7. A combined control system based on water storage and phase change storage in a cooling system according to claim 6, characterized in that: Also includes: The recognition layer module is used to identify multimodal data types of cold storage demand information, electricity price fluctuation information and meteorological information respectively; The adjustment layer module is used to dynamically adjust the weight of each modal data type based on the preset attention mechanism and cold storage demand information; The association layer module is used to perform cross-modal information fusion and spatiotemporal correlation analysis on each modal data based on the dynamically adjusted weights, and generate association features containing spatiotemporal correlation information; The analysis layer module is used to predict the meteorological environment trend, electricity price change trend and demand change trend based on the associated features and pre-stored historical data, and output a trend feature set.

Images