Intelligent cold supply control method for refrigeration house based on phase change cold storage

Abstract

The invention relates to the technical field of intelligent control, in particular to an intelligent cold supply control method for a refrigeration house based on phase change cold storage. The multiple sensors collect equipment operation data in real time and store the equipment operation data in the terminal equipment control unit. A data transmission channel is established through the equipment interface unit, the terminal equipment control unit and the central control unit, and data is operated synchronously. After the central control unit receives the data, the execution program module performs data analysis: firstly, the load, the phase change state and the energy consumption are predicted through the multi-dimensional information prediction module; then, the program generates an optimal control instruction according to a prediction result by using a multi-target rolling time domain dynamic optimization strategy, integrates a self-adaptive controller based on fuzzy logic, and adjusts instruction parameters in real time; finally, the adjusted control instruction is transmitted to the terminal equipment control unit through the equipment interface unit, and program control over the equipment running state or running parameters is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and particularly to an intelligent cooling control method for cold storage based on phase change energy storage Background Art

[0002] At present, the industrial control system has limitations in the non-linear modeling of cold storage cooling control. The existing technology that executes the preset PID algorithm cannot handle the non-linear thermal physical property changes. Secondly, the existing program control architecture has insufficient adaptability to complex production scenarios. Especially when dealing with different mode switches, it lacks a self-learning optimization algorithm based on real-time data and is difficult to achieve intelligent matching of parameter dynamic correction and equipment operation mode. In addition, the existing PLC program control architecture lacks a feed-forward compensation mechanism and cannot adjust the control strategy in advance according to the external disturbance prediction information, resulting in system response lag and reduced temperature control accuracy. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent cooling control method for cold storage based on phase change energy storage. Multiple sensors are used to collect the equipment operation data in real time and store it in the control unit of the terminal device. Through the equipment interface unit, the control unit of the terminal device establishes a data transmission channel with the central control unit to synchronize the operation data. After receiving the data, the central control unit executes the program module for data analysis: First, through the multi-dimensional information prediction module, the load, phase change state and energy consumption are predicted. Then, the program uses the multi-objective rolling horizon dynamic optimization strategy to generate the optimal control instruction according to the prediction result, and integrates an adaptive controller based on fuzzy logic to adjust the instruction parameters in real time. Finally, the adjusted control instruction is transmitted to the control unit of the terminal device through the equipment interface unit to realize the program control of the equipment operation state or operation parameters.

[0004] To achieve the above purpose, the present invention provides the following technical solutions:

[0005] An intelligent cooling control method for cold storage based on phase change energy storage, comprising:

[0006] Collect the equipment operation data in real time through multiple sensors and store the data in the corresponding memory in the control unit of the terminal device;

[0007] Establish a data transmission channel between the control unit of the terminal device and the central control unit by using the equipment interface unit, communicate the relevant equipment operation data stored in the memory of the control unit of the terminal device with the central control unit in real time, and transmit the equipment operation data and output control instruction signals through the equipment interface unit;

[0008] The central control unit receives the device operation data for processing and control optimization, outputs control instruction signals, and transmits them to the terminal device control unit via the device interface unit. The process of the central control unit processing the device operation data includes: using the multi-dimensional information prediction module to process the device operation data to obtain load prediction, phase change state prediction information, and energy consumption prediction information; based on the prediction information, adopting a multi-objective rolling horizon dynamic optimization strategy to generate optimal control instructions; introducing an adaptive controller based on fuzzy logic to adjust the optimal control instructions in real time.

[0009] The terminal device control unit controls the operation state of the terminal device or adjusts the operation parameters of the device according to the control instruction signal of the central control unit.

[0010] Furthermore, the device operation data includes: device operation environment data, device operation state data, energy consumption data, and device operation data.

[0011] Furthermore, the process of the central control unit using the multi-dimensional information prediction module to process the device operation data to obtain the load prediction, the phase change state prediction information, and the energy consumption prediction information includes:

[0012] The central control unit obtains the device operation data through the data transmission channel established by the device interface unit.

[0013] Preprocess the device operation data to obtain prediction demand data information.

[0014] Use the historical device operation data to process the multi-dimensional information prediction module to obtain a pre-trained module.

[0015] Parallelly input the first prediction demand information, the second prediction demand information, and the third prediction demand information in the prediction demand data information into the pre-trained module to respectively obtain the load prediction, the phase change state prediction information, and the energy consumption prediction information.

[0016] Among them, the memory of the central control unit is used to store the numerical prediction information for subsequent execution of the optimization control strategy based on the prediction information.

[0017] Furthermore, the process of the central control unit adopting the multi-objective rolling horizon dynamic optimization strategy based on the prediction information to generate the optimal control instructions includes:

[0018] Extract the prediction information from the memory of the central control unit.

[0019] Model the cooling control problem based on phase change cold storage as a multi-objective constrained optimization problem, and determine the objective function, constraint conditions, and optimization scheme for solving.

[0020] Determine the prediction horizon, control horizon, and control period for rolling horizon optimization;

[0021] At the start of each control period, based on the prediction information and the device operation data, the central control unit solves the objective function in combination with the optimization scheme, thereby generating an optimal control instruction sequence; only the optimal control instruction corresponding to the current control period in the sequence is executed. Subsequently, the device operation data is updated, entering the next control period, and based on the updated device operation data and updated prediction information, the optimization solution process is repeatedly executed to achieve the rolling update and execution of control instructions.

[0022] Furthermore, the process of real-time adjustment of the optimal control instruction according to the fuzzy logic-based adaptive controller includes:

[0023] Fuzzify the sensor data collected in real time and convert it into a set of fuzzy sets;

[0024] Perform fuzzy inference based on the set of fuzzy sets and the fuzzy rule base to obtain a fuzzy output;

[0025] Defuzzify the fuzzy output to obtain a control instruction adjustment amount;

[0026] Use the control instruction adjustment amount to adjust the optimal control instruction obtained by the multi-objective rolling horizon dynamic optimization strategy to obtain a final control instruction, and the central control unit sends it to the terminal device control unit.

[0027] Furthermore, the operating state includes the device start / stop state and the device working mode state; the operating parameters include: device operating temperature, device operating frequency, device operating speed, device operating rotation speed, device operating flow rate, and device operating power.

[0028] Compared with the prior art, the beneficial effects of the present invention are:

[0029] 1. The present invention proposes a multi-dimensional information prediction module, which obtains device operation data from the terminal device control unit for predicting information in different dimensions; it not only relies on the training process of historical data, but also predicts future cooling load states, PCM states, and energy consumption states by integrating multi-dimensional device data; this module provides accurate state prediction results for subsequent dynamic control optimization and is the basis and prerequisite for realizing forward-looking and refined program control decisions.

[0030] 2. The present invention proposes a multi-objective rolling horizon dynamic optimization strategy for users to generate optimal control instructions. This optimization strategy models the cooling control problem based on phase change thermal energy storage as a multi-objective constrained optimization problem. Combining the multi-objective function model and the prediction results of the multi-dimensional information prediction module, it can not only consider the conventional factors of the equipment, but also incorporate the factors affecting the non-linear characteristics of phase change thermal energy storage into the control optimization process, thus ensuring the feasibility and safety of the program control strategy.

[0031] 3. The present invention proposes an adaptive control adjustment method for adaptively adjusting control instructions. This method uses a fuzzy logic controller to perform online adjustment through real-time monitored operation data and in combination with a rule base, effectively compensating for prediction errors and suppressing external disturbances to ensure the stable operation of the control strategy. Through the real-time fine-tuning of fuzzy logic, this method can respond faster and more precisely to small changes in actual parameters, further reducing the parameter fluctuation range and achieving high-precision program control. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic flow chart of a cold storage intelligent cooling control method based on phase change thermal energy storage according to the present invention;

[0033] Figure 2 It is a schematic structural diagram of the multi-dimensional information prediction module according to the present invention;

[0034] Figure 3 It is a schematic flow chart of generating optimal control instructions according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0036] The current intelligent control system of automated stereoscopic warehouses has limitations in accurately modeling and controlling non-linear characteristics. The existing technology executes the preset PID algorithm and cannot handle non-linear thermal property changes. Secondly, the existing program control architecture has insufficient adaptability to complex production scenarios. Especially when dealing with different mode switches, it lacks a self-learning optimization algorithm based on real-time data and is difficult to achieve intelligent matching of parameter dynamic correction and equipment operation modes. In addition, the existing PLC program control architecture lacks a feed-forward compensation mechanism and cannot adjust the control strategy in advance according to external disturbance prediction information, resulting in system response lag and reduced temperature control accuracy.

[0037] Please refer to Figures 1 to 3, the present invention provides an intelligent cooling control method for cold storage based on phase change cold storage, and the technical solution is as follows:

[0038] Example 1:

[0039] In order to improve the effectiveness and reliability of the cooling control strategy, an enterprise uses an intelligent cooling control method for cold storage based on phase change cold storage proposed by the present invention. This method combines the terminal device control unit, device interface unit, central control unit, and terminal device control unit in the control system, and its process schematic is as Figure 1 shown, specifically including:

[0040] Real-time collect device operation data through multiple sensors and store the data in the corresponding memory in the terminal device control unit;

[0041] Furthermore, the device operation data includes: device operation environment data, device operation status data, energy consumption data, and device operation data;

[0042] Furthermore, the device operation environment data includes in-store environment data and out-of-store environment data; the environment data includes environmental temperature and environmental humidity;

[0043] Furthermore, the device operation status data includes phase change material (PCM) parameter data and refrigeration equipment parameter data; the PCM parameter data includes: thermophysical properties (phase change temperature, phase change latent heat, specific heat capacity, density, etc.), cold storage capacity, and state of charge (SoC) of the PCM; the refrigeration equipment parameter data includes: condensation temperature, refrigeration capacity, refrigerant flow rate, fan speed, heat exchange rate, charge / discharge rate, etc.;

[0044] Furthermore, the energy consumption data includes device power data and electricity price data;

[0045] Furthermore, the device operation data includes: inbound and outbound cargo data, frequency of warehouse door opening, equipment maintenance logs, etc.

[0046] Using multi-dimensional device operation data can provide a reliable data source for subsequent multi-task prediction, multi-objective optimization, and adaptive adjustment operations, thereby ensuring the effectiveness of the intelligent cooling program control.

[0047] Establish a data transmission channel between the terminal device control unit and the central control unit through the device interface unit, communicate the relevant device operation data stored in the memory of the terminal device control unit with the central control unit in real time, and transmit the device operation data and output control instruction signals through the device interface unit;

[0048] The central control unit receives the device operation data for processing and control optimization, outputs control instruction signals, and transmits them to the terminal device control unit via the device interface unit. The process of the central control unit processing the device operation data includes: using the multi-dimensional information prediction module to process the device operation data to obtain load prediction, phase change state prediction information, and energy consumption prediction information; based on the prediction information, adopting a multi-objective rolling horizon dynamic optimization strategy to generate optimal control instructions; introducing an adaptive controller based on fuzzy logic to adjust the optimal control instructions in real time.

[0049] Further, the process of the central control unit using the multi-dimensional information prediction module to process the device operation data to obtain load prediction, phase change state prediction information, and energy consumption prediction information includes:

[0050] The central control unit obtains the device operation data through the data transmission channel established by the device interface unit.

[0051] Preprocess the device operation data to obtain prediction demand data information.

[0052] Use the historical device operation data to process the multi-dimensional information prediction module to obtain a pre-trained module.

[0053] Parallelly input the first prediction demand information, the second prediction demand information, and the third prediction demand information in the prediction demand data information into the pre-trained module to respectively obtain load prediction, phase change state prediction information, and energy consumption prediction information.

[0054] Among them, the memory of the central control unit is used to store the numerical prediction information for subsequent execution of the optimization control strategy based on the prediction information.

[0055] Further, the preprocessing of the device operation data includes: data alignment, data cleaning, data normalization, and data classification.

[0056] Further, data classification is carried out according to three different prediction input requirements in the multi-dimensional information prediction module to respectively obtain the first prediction demand information, the second prediction demand information, and the third prediction demand information. For example: the second prediction demand information is mainly used to predict the state of the phase change material, so the second prediction demand information includes PCM parameter data and environmental data.

[0057] Further, it is possible that a certain preprocessed data is used for multiple prediction tasks, that is, the first prediction demand information, the second prediction demand information, and the third prediction demand information have the same data. For example: the environmental data will be used as the input for load prediction, phase change state prediction, and energy consumption prediction, that is, the first prediction demand information, the second prediction demand information, and the third prediction demand information all include environmental data.

[0058] Further, the historical device operation data is obtained by calling the device management platform;

[0059] Further, the structure of the multi-dimensional information prediction module is as Figure 2 shown, including: an input layer, a multi-dimensional feature extraction layer, a parallel multi-task prediction layer, and an output layer;

[0060] The input layer and the multi-dimensional feature extraction layer are used to receive multi-dimensional prediction requirement information and convert it into feature representations, namely the first prediction requirement feature representation, the second prediction requirement feature representation, and the third prediction requirement feature representation;

[0061] The multi-dimensional feature extraction layer is used to input the first prediction requirement feature representation, the second prediction requirement feature representation, and the third prediction requirement feature representation into the load prediction unit, the phase change state prediction unit, and the energy consumption prediction unit respectively, to obtain the load prediction feature, the phase change state prediction feature, and the energy consumption prediction feature;

[0062] Further, the prediction output of each unit is used as the input of the subsequent unit. For example, the prediction result of the load prediction unit is used as the input of the phase change state prediction unit and the energy consumption prediction unit. That is, it is necessary to combine the load prediction feature obtained by the load prediction unit with the second prediction requirement feature representation and input them together into the phase change state prediction unit to obtain the phase change state prediction result;

[0063] This embodiment adopts such a prediction scheme because the load is the key influencing factor affecting the PCM state and the energy consumption state. At the same time, the cooling load and the PCM state can jointly affect the prediction of energy consumption. Adopting the prediction scheme proposed by the present invention can improve the prediction accuracy.

[0064] The output layer uses a fully connected layer and an activation function to convert the load prediction feature, the phase change state prediction feature, and the energy consumption prediction feature into load prediction, phase change state prediction information, and energy consumption prediction information respectively;

[0065] Further, the load prediction unit, the phase change state prediction unit, and the energy consumption prediction unit respectively adopt the LSTM, CNN, and Transformer structures for prediction.

[0066] By using the historical device operation data for training and integrating multi-dimensional device data, it is used to predict the future cooling load state, PCM state, and energy consumption state; it can provide accurate state prediction results for subsequent dynamic program control optimization, and is the basis and prerequisite for realizing forward-looking and refined program control decisions.

[0067] Further, the process of the central control unit generating the optimal control instruction based on the prediction information using the multi-objective rolling horizon dynamic optimization strategy is as Figure 3 shown, specifically as follows:

[0068] Extract the prediction information from the memory of the central control unit;

[0069] Model the cooling control problem based on phase change cold storage as a multi-objective constrained optimization problem, and determine the objective function, constraints, and optimization scheme for solving;

[0070] Determine the prediction horizon, control horizon, and control period of the rolling horizon optimization;

[0071] In this embodiment, the prediction horizon, control horizon, and control period are set to 24 hours, 6 hours, and 1 hour respectively.

[0072] At the start of each control period, based on the prediction information and equipment operation data, the central control unit combines the optimization scheme to solve the objective function, thereby generating an optimal control instruction sequence; only the optimal control instruction corresponding to the current control period in the sequence is executed, and then the equipment operation data is updated, entering the next control period, and based on the updated equipment operation data and updated prediction information, the optimization solution process is repeated to achieve the rolling update and execution of the control instructions;

[0073] Furthermore, the objective function can be expressed as:

[0074]

[0075] where J represents the objective function; H c represents the control horizon; ω1 represents the cost weight factor; P ele (t + k) represents the electricity price at time t + k; t and k represent the current time step and future time step indices respectively; P ch (t + k) represents the power of the refrigeration equipment at time t + k; Δt represents the control period; ω2 represents the temperature weight factor; T in (t + k + 1) represents the temperature inside the warehouse at time t + k + 1; T set represents the set value of the temperature inside the warehouse;

[0076] Furthermore, the temperature inside the warehouse at time t + k + 1 can be predicted through the following state transition equation, and the state transition equation is expressed as:

[0077]

[0078] where T in (t + k) represents the temperature inside the warehouse at time t + k; mc p represents the heat capacity coefficient; Q ch (t + k) represents the refrigeration capacity at time t + k; Q pc (t + k) represents the heat exchange rate at time t + k; Q load(t + k + 1) represents the load prediction information at time t + k + 1; T out (t + k) represents the outdoor temperature at time t + k;

[0079] Furthermore, the constraint conditions include: the indoor temperature control range, the PCM state of charge range, the refrigeration equipment power range, and the heat exchange rate range; the limitation of these parameter ranges can be flexibly adjusted according to actual needs.

[0080] By modeling the cooling control problem based on phase change energy storage as a multi-objective constrained optimization problem, and combining the multi-objective function model and the prediction results of the multi-dimensional information prediction module, it is possible to take into account both the conventional factors of the equipment and the factors affecting the non-linear characteristics of phase change energy storage in the process of program control optimization, thus ensuring the feasibility and safety of the program control strategy.

[0081] Furthermore, the process of real-time adjustment of the optimal control instruction according to the fuzzy logic-based adaptive controller includes:

[0082] Fuzzify the sensor data collected in real time and convert it into a group of fuzzy sets;

[0083] Furthermore, the sensor data includes environmental sensor data, PCM state sensor data, and equipment state sensor data, etc.;

[0084] Furthermore, the fuzzification process includes: determining the input variables of the fuzzy controller and performing preprocessing; defining a group of fuzzy sets for each input variable, where each fuzzy set represents a different degree of description of the variable; substituting the preprocessed input value into the membership function of each fuzzy set to calculate the membership degree of the value to each fuzzy set, and forming a complete group of fuzzy sets with the fuzzy sets containing the corresponding membership degrees.

[0085] Perform fuzzy inference based on the group of fuzzy sets and the fuzzy rule base to obtain a fuzzy output;

[0086] Furthermore, the fuzzy inference adopts the Mamdani method to calculate the fuzzy output by combining the fuzzy rules and the group of fuzzy sets.

[0087] Defuzzify the fuzzy output to obtain the control instruction adjustment amount;

[0088] Furthermore, the defuzzification method can adopt the centroid method or the mean maximum method.

[0089] Use the control instruction adjustment amount to adjust the optimal control instruction obtained by the multi-objective rolling horizon dynamic optimization strategy to obtain the final control instruction, and send it to the terminal device control unit by the central control unit.

[0090] By using the fuzzy logic controller to perform online adjustment through real-time monitored operation data and in combination with the rule base, the prediction error can be effectively compensated and external disturbances can be suppressed, ensuring the stable operation of the control strategy; the real-time fine-tuning of the fuzzy logic can respond to the minute changes of actual parameters faster and more precisely, further reducing the parameter fluctuation range and achieving high-precision program control.

[0091] The terminal device control unit controls the operation state of the terminal device or adjusts the operation parameters of the device according to the control instruction signal of the central control unit.

[0092] Furthermore, the operation state includes the device start / stop state and the device working mode state; the operation parameters include: the device operation temperature, the device operation frequency, the device operation speed, the device operation rotation speed, the device operation flow rate, and the device operation power.

[0093] This embodiment proposes an intelligent cooling control method for cold storage based on phase change cold storage; the operation data of the device is collected in real time by multiple sensors and stored in the terminal device control unit. Through the device interface unit, the terminal device control unit establishes a data transmission channel with the central control unit to synchronize the operation data. After receiving the data, the central control unit executes the program module for data analysis: first, through the multi-dimensional information prediction module, the load, the phase change state, and the energy consumption are predicted; then, the program uses the multi-objective rolling horizon dynamic optimization strategy to generate the optimal control instruction according to the prediction result, and integrates the fuzzy logic-based adaptive controller to adjust the instruction parameters in real time; finally, the adjusted control instruction is transmitted to the terminal device control unit through the device interface unit to achieve the program control of the device operation state or operation parameters. This method can effectively improve the accuracy and reliability of the control strategy.

[0094] Embodiment 2:

[0095] The present invention proposes an intelligent cooling control method for cold storage based on phase change cold storage. In order to further verify the effectiveness of the method proposed by the present invention in the multi-dimensional information prediction module and the overall control scheme, the present invention selects two enterprises, A and B, for the effectiveness test of the multi-dimensional information prediction module and the effectiveness test of the overall control scheme.

[0096] The present invention collects the historical device operation data of enterprise A in the past year as the test data for the effectiveness test of the multi-dimensional information prediction module, which is divided into Data 1, Data 2, and Data 3, and each group of data is different; the historical data of enterprise A in the past 2-3 years is selected as the training data for the multi-dimensional information prediction module; the sampling of the data set refers to the following rules: taking a week as a unit, randomly extracting 5 days of data per week; 2 groups of data are extracted in the morning, at noon, and in the evening every day.

[0097] Each set of test data is respectively input into the pre-trained Module 1 and Module 2 to obtain their respective module prediction results, which are divided into load, phase change state, and energy consumption; then, by manually comparing the differences between the module prediction values and the actual values, the average proportion of the prediction results within a reasonable range for each group is obtained; among them, Module 1 is the multi-dimensional information prediction module proposed by the present invention; Module 2 is to change the structure of the parallel multi-task prediction layer in the multi-dimensional information prediction module, that is, each unit in the parallel multi-task prediction layer is completely independent, and the input of each unit is only the corresponding prediction demand feature representation.

[0098] The test results of the effectiveness of the multi-dimensional information prediction module are shown in Table 1.

[0099] Table 1 Test results of the effectiveness of the multi-dimensional information prediction module

[0100]

[0101] It can be seen from the results in Table 1 that the multi-dimensional information prediction module proposed by the present invention, that is, Module 1, has better test results of the effectiveness of the multi-dimensional information prediction module than Module 2; thus, it can be shown that the module proposed by the present invention can accurately predict the future state of different types of data, further improving the reliability of the cooling program control strategy.

[0102] In order to further test the effectiveness of the overall control scheme, this embodiment selects the historical equipment operation data of Company B in the past two years as test data; three different control schemes are used to process the same test data respectively, obtain the control instructions under each group of schemes and execute the control operations, and obtain the proportion of the control results within a reasonable range for each group of schemes through manual investigation; among them, Test Scheme 1 is the overall scheme proposed by the present invention, that is, "multi-dimensional information prediction + multi-objective dynamic optimization + adaptive control adjustment"; Test Scheme 2 is to remove the adaptive control adjustment process in the present invention; Test Scheme 3 is to change the multi-objective dynamic optimization process in the present invention to output control instructions by combining experience summary and multi-dimensional information prediction results, that is, convert "module automation" to "manual operation".

[0103] The test results of the effectiveness of the overall control scheme are shown in Table 2.

[0104] Table 2 Test results of the effectiveness of the overall control scheme

[0105]

[0106] It can be seen from the results in Table 2 that the test results of the effectiveness obtained by using Test Scheme 1, that is, the overall scheme proposed by the present invention, are better than those of other schemes, which shows that it is necessary to combine "multi-dimensional information prediction + multi-objective dynamic optimization + adaptive control adjustment" to improve the effectiveness and reliability of the cooling program control strategy.

[0107] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent cooling control method for cold storage based on phase change cold storage, characterized in that It includes: Real-time collect the device operation data through sensors and store it in the corresponding memory in the terminal device control unit; Use the device interface unit to transmit the device operation data in the memory and the control instruction signals generated by the central control unit; The central control unit receives the device operation data for processing and control optimization, generates control instruction signals, and transmits them to the terminal device control unit via the device interface unit; The processing and control optimization process of the central control unit includes: the central control unit uses the pre-trained multi-dimensional information prediction module to process the device operation data to obtain load prediction, phase change state prediction information, and energy consumption prediction information; the central control unit adopts a multi-objective rolling horizon dynamic optimization strategy based on the load prediction, phase change state prediction information, and energy consumption prediction information to generate optimal control instructions; the central control unit combines the sensor data with an adaptive controller based on fuzzy logic to adjust the optimal control instructions in real time to obtain control instruction signals; The terminal device control unit responds to the control instruction signals of the central control unit, executes program control, and controls the operation state of the device or adjusts the operation parameters of the device.

2. The intelligent cooling control method for a cold storage based on phase change cold storage according to claim 1, wherein, The device operation data includes: device operation environment data, device operation state data, energy consumption data, and device operation data.

3. The intelligent cooling control method for cold storage based on phase change cold storage according to claim 1, characterized in that, The process in which the central control unit uses the multi-dimensional information prediction module to process the device operation data to obtain the load prediction, the phase change state prediction information, and the energy consumption prediction information includes: The central control unit obtains the device operation data through the data transmission channel established by the device interface unit; Preprocess the device operation data to obtain prediction demand data information; Use the historical device operation data to process the multi-dimensional information prediction module to obtain a pre-trained module; Parallelly input the first prediction demand information, the second prediction demand information, and the third prediction demand information in the prediction demand data information into the pre-trained module to respectively obtain the load prediction, the phase change state prediction information, and the energy consumption prediction information; Among them, the memory of the central control unit is used to store numerical prediction information for subsequent execution of the optimization control strategy based on the prediction information.

4. A method for intelligent cooling control of a cold storage based on phase change cold storage, as claimed in claim 1, wherein The process in which the central control unit adopts the multi-objective rolling horizon dynamic optimization strategy based on the load prediction, the phase change state prediction information, and the energy consumption prediction information to generate the optimal control instructions includes: Extract prediction information from the memory of the central control unit; among them, the prediction information includes: the load prediction, the phase change state prediction information, and the energy consumption prediction information; Model the cooling control problem based on phase change energy storage as a multi-objective constrained optimization problem, and determine the objective function, constraint conditions, and optimization scheme for solving; Determine the prediction horizon, control horizon, and control period of the rolling horizon optimization; At the start of each control period, based on the prediction information and the device operation data, the central control unit combines the optimization scheme to solve the objective function, thereby generating an optimal control instruction sequence; only the optimal control instruction corresponding to the current control period in the sequence is executed, and then the device operation data is updated, entering the next control period, and based on the updated device operation data and updated prediction information, the optimization solution process is repeatedly executed to achieve the rolling update and execution of the control instructions.

5. The intelligent cooling control method for cold storage based on phase change cold storage according to claim 1, characterized in that The process of the central control unit combining sensor data with an adaptive controller based on fuzzy logic to adjust the optimal control instruction in real time includes: Performing fuzzification processing on the sensor data collected in real time and converting it into a fuzzy set group; Performing fuzzy inference based on the fuzzy set group and the fuzzy rule base to obtain a fuzzy output; Performing defuzzification processing on the fuzzy output to obtain a control instruction adjustment amount; Using the control instruction adjustment amount to adjust the optimal control instruction obtained by the multi-objective rolling horizon dynamic optimization strategy to obtain a final control instruction, and the central control unit sends it to the terminal device control unit.

6. The intelligent cooling control method for cold storage based on phase change cold storage according to claim 5, wherein, The process of the fuzzification processing includes: determining the input sensor variables of the fuzzy controller and performing preprocessing; defining a set of fuzzy sets for each input sensor variable, and each fuzzy set represents a different degree of description of the variable; substituting the preprocessed input variable values into the membership functions of each fuzzy set to calculate the membership degrees of the variable values to each fuzzy set, and forming a complete fuzzy set group with the fuzzy sets containing the corresponding membership degrees.

7. A method for intelligent cooling control of a cold storage based on phase change cold storage, as claimed in claim 1, wherein The operating state includes the device start-stop state and the device working mode state; the operating parameters include: device operating temperature, device operating frequency, device operating speed, device operating rotation speed, device operating flow rate, and device operating power.

Images