High and low temperature box refrigeration power closed-loop control method and device

Abstract

The invention relates to the technical field of refrigeration control, in particular to a high-low temperature box refrigeration power closed-loop control method and device. The method comprises the following steps: acquiring multi-dimensional monitoring parameters of a high-low temperature box according to a multi-source sensor; performing three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, performing three-dimensional modeling on the environment in the box, and constructing a three-dimensional parameter feature model; performing dynamic refrigeration airflow dynamic analysis on the three-dimensional parameter characteristic model according to a high-low temperature box, performing three-dimensional track distribution rendering, and constructing a refrigeration flow digital twinborn model; performing multi-stage temperature change trend calculation on the refrigeration flow digital twin model, and performing dynamic refrigeration demand calculation to obtain refrigeration demand characteristics of each stage; and performing refrigeration flow property simulation based on the refrigeration demand characteristics of each stage, performing compressor refrigeration power demand planning, and constructing a dynamic power demand planning strategy. According to the invention, efficient and accurate refrigeration closed-loop control is realized.

Description

Technical Field

[0001] The present invention relates to the field of refrigeration control technology, and in particular to a method and device for closed-loop control of refrigeration power of a high and low temperature box. Background Art

[0002] With the increasing demand for industrial automation and refined control, high and low temperature chambers, as key temperature control equipment, are widely used in precision testing, scientific research, and product storage and transportation in the medical, food and other industries. Its core function is to maintain the stability of the temperature inside the chamber through the refrigeration system to ensure the quality of sensitive items during transportation and storage. However, with the continuous changes in the use environment and load conditions, the working state of the refrigeration system of the high and low temperature chamber is affected by many factors during long-term operation, such as changes in external ambient temperature, load fluctuations in the chamber, uneven airflow distribution, system aging, compressor performance degradation, etc. These factors often cause fluctuations in refrigeration power, thereby affecting the precise control and stability of the temperature inside the chamber.

[0003] Traditional temperature control methods for high and low temperature boxes usually adopt control strategies based on fixed value settings, such as PID control or fuzzy control. Although these methods can maintain the stability of the temperature inside the box within a certain range, their control accuracy and response speed have certain limitations when facing complex working conditions such as load changes and external environmental fluctuations. Especially in terms of power regulation of refrigeration systems, traditional methods are often unable to dynamically adjust the refrigeration power according to real-time demand and load status, resulting in low energy efficiency and inability to quickly respond to changes in the external environment, causing over-refrigeration or insufficient refrigeration, affecting the energy consumption and operational stability of the equipment.

[0004] In order to meet the modern industry's demand for high-precision, high-efficiency and intelligent temperature control systems, traditional control methods can no longer meet the requirements of refined management under complex working conditions. Therefore, a more intelligent closed-loop control method for refrigeration power is urgently needed. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method and device for closed-loop control of refrigeration power of a high and low temperature box to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a closed-loop control method for refrigeration power of a high and low temperature box, comprising the following steps: Step S1: obtaining multi-dimensional monitoring parameters of the high and low temperature box according to the multi-source sensor; performing three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, and performing three-dimensional modeling of the environment inside the box to construct a three-dimensional parameter feature model; Step S2: Perform dynamic refrigeration airflow dynamics analysis on the three-dimensional parameter characteristic model according to the high and low temperature box, and perform three-dimensional trajectory distribution rendering to construct a refrigeration flow digital twin model; Step S3: Calculate the temperature change trend of multiple stages of the refrigeration flow digital twin model, and calculate the dynamic refrigeration demand to obtain the refrigeration demand characteristics of each stage; Step S4: simulating the refrigeration flow performance based on the refrigeration demand characteristics of each stage, planning the compressor refrigeration power demand, and building a dynamic power demand planning strategy; Step S5: Perform multi-stage equipment power output calculation and multi-objective power optimization based on the dynamic power demand planning strategy, thereby generating a multi-objective power optimization demand strategy; Step S6: Perform real-time refrigeration control on the high and low temperature box according to the multi-objective power optimization demand strategy, perform iterative decision optimization, and build an intelligent refrigeration power control model to execute the refrigeration power closed-loop control operation of the high and low temperature box.

[0007] The present invention obtains multi-dimensional monitoring parameters of high and low temperature boxes through multi-source sensors, which is helpful to fully understand the environmental conditions in the box, provide a data basis for subsequent analysis, conduct three-dimensional temperature distribution analysis and modeling, and construct a three-dimensional parameter characteristic model, which can more accurately describe the temperature distribution characteristics in the box, and provide a basis for subsequent refrigeration power control. Dynamic refrigeration airflow dynamics analysis and the construction of a refrigeration flow digital twin model help to understand the working principle and performance of the refrigeration system. Through three-dimensional trajectory distribution rendering, the flow path of the refrigeration airflow is more intuitively displayed, providing visualization support for refrigeration power control. Multi-stage temperature change trend calculation and dynamic refrigeration demand calculation accurately predict the refrigeration demand at each stage, provide basic data for subsequent power planning, and obtain the refrigeration demand characteristics at each stage, which is helpful for more precise Finely regulate the refrigeration system to improve energy efficiency and save energy. Simulate refrigeration flow performance and plan compressor refrigeration power demand to optimize the performance and power utilization of the refrigeration system. The construction of dynamic power demand planning strategies can more effectively adjust the refrigeration power and improve the system operation efficiency. The generation of multi-objective power optimization demand strategies helps to comprehensively consider multiple factors for power optimization and improve the overall performance of the system. Multi-stage equipment power output calculation and power optimization enable the refrigeration system to maintain efficient and stable operation under different working conditions. The construction of real-time refrigeration control and intelligent refrigeration power control models can achieve precise control and adjustment of high and low temperature boxes, iterative decision optimization and refrigeration power closed-loop control operations, and continuously adjust system parameters to ensure that the refrigeration power meets demand while improving system stability.

[0008] Preferably, step S1 comprises the following steps: Step S11: Acquire multi-dimensional monitoring parameters of the high and low temperature chamber according to the multi-source sensor; Step S12: performing a three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box to generate a three-dimensional temperature distribution feature in the box; Step S13: calculating the temperature fluctuation amplitude of the three-dimensional temperature distribution characteristics in the box to generate a three-dimensional temperature distribution fluctuation diagram; Step S14: Analyze the humidity characteristics of each region based on the multi-dimensional monitoring parameters of the high and low temperature chamber, thereby generating multiple regional humidity fields; Step S15: Perform three-dimensional modeling of the in-box environment on the three-dimensional temperature distribution fluctuation diagram and the humidity fields in multiple regions to construct a three-dimensional parameter characteristic model.

[0009] The present invention obtains multi-dimensional monitoring parameters of high and low temperature boxes through multi-source sensors, which is helpful for a comprehensive understanding of the environment inside the box, provides a data basis for subsequent analysis, and performs three-dimensional temperature distribution analysis, which can generate three-dimensional temperature distribution characteristics inside the box and provide detailed temperature distribution information, which is helpful for understanding the temperature distribution law inside the box, calculates the temperature fluctuation amplitude, and generates a three-dimensional temperature distribution fluctuation diagram, which is helpful for identifying the law and amplitude of temperature changes and providing a reference basis for refrigeration power control. It performs humidity characteristic analysis in each area and generates multiple regional humidity fields, which is helpful for understanding the humidity distribution in different areas and is very important for comprehensive environmental control. Based on the three-dimensional temperature distribution fluctuation diagram and multiple regional humidity fields, three-dimensional modeling of the environment inside the box is performed, and a three-dimensional parameter characteristic model is constructed to more accurately describe the environmental characteristics inside the box and provide a basis for subsequent closed-loop control of refrigeration power.

[0010] Preferably, the specific steps of step S2 are: Step S21: monitor the refrigeration operation status of the high and low temperature box and collect refrigeration operation status parameters; Step S22: Identify the refrigeration air outlets of the high and low temperature boxes and mark all the refrigeration air outlets; Step S23: performing spatial layout identification on all cooling air outlets to generate spatial layout features of the cooling air outlets; Step S24: performing dynamic refrigeration airflow dynamics analysis on the refrigeration operation state parameters to extract the refrigeration airflow flow trajectory; Step S25: mining the spatial flow distribution of the cooling air flow trajectory according to the spatial layout characteristics of the cooling air outlet, thereby generating a cooling air flow distribution field; Step S26: Render the three-dimensional trajectory distribution of the three-dimensional parameter feature model based on the cold air flow distribution field to construct a refrigeration flow digital twin model.

[0011] The present invention monitors the refrigeration operation status of the high and low temperature box and collects refrigeration operation status parameters, monitors the refrigeration condition of the box in real time, and provides data support for subsequent analysis and control. The refrigeration operation status monitoring is helpful to understand the temperature change inside the box, and provides key information for the closed-loop control of the refrigeration power. By identifying and marking the refrigeration air outlets of the high and low temperature box, the air outlet positions of the refrigeration system are accurately determined, and basic data for subsequent flow analysis and control are provided. Marking the refrigeration air outlets is helpful to distinguish different air flow outlets, and accurate parameter input is provided for the refrigeration air flow dynamics analysis. The refrigeration air outlets are identified through spatial layout, and the spatial layout characteristics of the refrigeration air outlets are generated, which reveals the air outlet distribution inside the refrigeration system and provides a basis for flow distribution mining. Spatial layout identification is helpful to understand the internal structure of the refrigeration system, and provides spatial layout information for subsequent flow analysis and control. Mechanical analysis extracts the flow trajectory of the refrigeration airflow, deeply understands the flow of the refrigeration airflow inside the box, and provides a basis for refrigeration power control. The refrigeration airflow dynamics analysis helps to understand the movement law of the airflow and provide a reference for optimizing the operation of the refrigeration system. According to the spatial layout characteristics of the refrigeration outlet, the spatial flow distribution of the refrigeration airflow flow trajectory is mined to generate a cold air flow distribution field. The impact of the spatial layout on the airflow is comprehensively considered to improve the efficiency and stability of the refrigeration system. Spatial flow distribution mining helps to optimize the distribution of airflow and provide support for improving the accuracy and performance of the closed-loop control of refrigeration power. A refrigeration flow digital twin model is constructed based on the cold air flow distribution field to realize digital modeling of the refrigeration system and provide a simulation and optimization platform for the closed-loop control of refrigeration power. The refrigeration flow digital twin model helps to simulate the system response under different control strategies and provide a reliable basis for the optimization and adjustment of the refrigeration system.

[0012] Preferably, step S3 specifically comprises the following steps: Step S31: performing a time-series temperature change analysis on the refrigeration flow digital twin model to generate a time-series temperature change curve; Step S32: performing multi-segment temperature change trend calculation on the time series temperature change curve, thereby obtaining the temperature change trends in multiple time periods; Step S33: Predicting the temperature change trend for multiple time periods and constructing a temperature change trend prediction graph; Step S34: Based on the refrigeration operation log and the temperature change trend prediction diagram, the refrigeration flow digital twin model is dynamically calculated to obtain the refrigeration demand characteristics of each stage.

[0013] The present invention generates a time-series temperature change curve by performing a time-series temperature change analysis on the refrigeration flow digital twin model, intuitively showing the temperature change trend over time, and providing a basis for subsequent analysis and prediction. The time-series temperature change curve helps to monitor the dynamic process of temperature change, and provides a real-time reference for the closed-loop control of refrigeration power. The time-series temperature change curve performs multi-segment temperature change trend calculation to obtain the temperature change trend of multiple time periods, refines the characteristics of temperature change, and provides more accurate data support for subsequent prediction and calculation. The multi-segment temperature change trend calculation helps to identify the temperature change law in different time periods, and provides a basis for the timing adjustment of refrigeration power control. The temperature change trend is predicted, and a temperature change trend prediction map is constructed to predict the trend of future temperature changes, providing early warning and guidance for the closed-loop control of refrigeration power. The temperature change trend prediction map helps to predict the temperature change trend in advance, and provides a basis and decision-making support for the dynamic adjustment of refrigeration demand. Based on the refrigeration operation log and the temperature change trend prediction map, the refrigeration flow digital twin model is dynamically calculated to obtain the refrigeration demand characteristics of each stage, and the refrigeration power is adjusted and optimized according to the real-time demand. The refrigeration demand calculation helps to dynamically adjust the refrigeration power according to the actual situation, improve the efficiency and response speed of the refrigeration system, and ensure the temperature stability in the box and the safety of the goods.

[0014] Preferably, the specific steps of step S34 are: Get refrigeration operation log; identifying predetermined refrigeration temperatures for multiple stages based on refrigeration operation logs; extracting a timestamp for each stage based on a predetermined refrigeration temperature for the multiple stages; The temperature change situation prediction diagram is time-matched based on the timestamp of each stage, so as to obtain the temperature change prediction situation matching the timestamp; Based on the predetermined refrigeration temperatures of multiple stages, the current temperature deviation is calculated for the temperature change prediction situation matching the timestamp, thereby generating the temperature deviation parameters for each stage; The dynamic refrigeration demand calculation of the refrigeration flow digital twin model is performed according to the temperature deviation parameters of each stage to obtain the refrigeration demand characteristics of each stage.

[0015] The present invention records the operating status and parameter changes of the refrigeration system through the refrigeration operation log, understands the system operation status by analyzing the log data, provides basic data support for subsequent control and optimization, identifies the predetermined refrigeration temperatures of multiple stages, helps to determine the target temperature settings of different stages, and provides target reference values ​​for the control of the refrigeration system. The timestamp of each stage is extracted to accurately mark the start and end time of each stage, and provides a time reference for subsequent time matching and analysis. The temperature change prediction trend obtained by time matching is combined with the actual time for prediction, and a timing reference is provided for the adjustment of refrigeration demand. The temperature deviation parameter is calculated to evaluate the difference between the actual temperature and the predicted temperature, and a basis is provided for the adjustment and optimization of the refrigeration power. The dynamic refrigeration demand calculation adjusts the refrigeration power according to the actual situation based on the temperature deviation parameter, thereby improving the response speed and stability of the system and ensuring the accuracy of temperature control.

[0016] Preferably, the specific steps of step S4 are: Step S41: Calculating the power of the refrigeration compressor based on the refrigeration operation state parameters, thereby obtaining the real-time power characteristics of the refrigeration compressor; Step S42: simulating the refrigeration flow performance of the refrigeration compressor according to the cold air flow distribution field to generate refrigeration flow simulation data in the box; Step S43: planning the compressor refrigeration power demand for each stage of the refrigeration demand characteristics according to the refrigeration flow simulation data in the box, and constructing a dynamic power demand planning strategy.

[0017] The present invention obtains the power characteristics of the compressor in real time, so that the system can optimize power regulation in time according to the current workload, avoid excessive or low power, and ensure that the system strikes a balance between energy efficiency and refrigeration effect. The calculation of real-time power characteristics can not only help monitor the current state, but also provide accurate data support for subsequent cold air flow distribution field simulation and dynamic power demand planning, ensuring the continuity and consistency of the overall refrigeration process. The real-time power characteristics provide feedback information to ensure that the low-temperature box system can achieve automatic adjustment in different working environments to maintain temperature stability. Based on the simulation of the cold air flow distribution field, the distribution and flow pattern of the cold air flow in the refrigeration system can be more accurately reproduced, thereby helping to understand the impact of different cold air flow patterns on temperature uniformity. This provides a theoretical basis for the balance of temperature inside the low-temperature box and can ensure the stability of temperature in different areas. By simulating the distribution of cold air flow, problem areas in the refrigeration process, such as uneven cold air flow, local overcooling or lack of cooling, are identified, thereby optimizing the design and operation parameters of the system and improving the overall refrigeration efficiency. By simulating the flow performance in the box, Potential temperature control problems can be identified in advance, such as dead corners and hot spots of cold air flow, to avoid these problems from having an adverse effect on the temperature of the goods. Combined with the refrigeration flow simulation data, the precise compressor refrigeration power demand can be calculated for each stage, which helps to flexibly adjust the refrigeration power according to the actual refrigeration needs of different stages to avoid over-refrigeration or insufficient power. At different stages, the power is reasonably allocated according to the actual needs of cold air flow. For example, in the early stage of the refrigeration process, higher power is required to quickly reduce the temperature; after the temperature stabilizes, the refrigeration power is appropriately reduced. Through this planning strategy, the system can achieve optimal energy efficiency and temperature control. The dynamic power demand planning strategy means that the system can respond to temperature changes or changes in cold air flow in real time to ensure that the refrigeration demand of each stage is met in a timely manner, thereby avoiding the lag reaction and energy efficiency waste of the temperature control system. Through precise power demand planning, the system forms a complete closed-loop control. Based on the flow simulation in the box and the actual data feedback, the system continuously adjusts the refrigeration power to ensure that the temperature in the low-temperature box is always kept within the set range to avoid the impact of temperature fluctuations on the goods.

[0018] Preferably, the specific steps of step S5 are: Step S51: performing multi-stage equipment power output calculation based on a dynamic power demand planning strategy, thereby generating equipment output power for each stage; Step S52: performing a time-series cooling efficiency evaluation according to the equipment output power at each stage, thereby obtaining a cooling efficiency evaluation value at each stage; Step S53: Calculate the preset temperature fluctuation constraint threshold value for the refrigeration operation log to obtain the maximum temperature fluctuation amplitude value; Step S54: performing cooling energy consumption analysis on the equipment output power at each stage to generate cooling energy consumption characteristics at each stage; Step S55: Perform multi-objective power optimization on the refrigeration energy consumption characteristics of each stage, the refrigeration efficiency evaluation value of each stage and the maximum temperature fluctuation amplitude value, so as to generate a multi-objective power optimization demand strategy.

[0019] The present invention calculates the power output of multi-stage equipment based on a dynamic power demand planning strategy to generate the output power of the equipment in each stage, which helps to determine the operating status and power demand of the equipment in each stage. The calculation of the output power of the equipment provides a basis for the optimal scheduling of the equipment, ensuring that the equipment operates in the optimal power working range in each stage, improving the efficiency of the system, and performing a time-series refrigeration efficiency evaluation according to the output power of the equipment in each stage to obtain the refrigeration efficiency evaluation value of each stage, which helps to evaluate the performance of the refrigeration system under different working conditions. The acquisition of the refrigeration efficiency evaluation value helps the system optimize the refrigeration efficiency, improve the energy utilization rate, and reduce the refrigeration cost. The predetermined temperature fluctuation constraint threshold is calculated for the refrigeration operation log to obtain the maximum temperature fluctuation amplitude value. It helps to determine the stability requirements of temperature control. The calculation of the maximum temperature fluctuation amplitude value is used as a reference standard for the control system to ensure that the temperature fluctuation in the box is within an acceptable range. The refrigeration energy consumption analysis is performed on the equipment output power at each stage to generate the refrigeration energy consumption characteristics of each stage, which helps to evaluate the energy consumption of the system under different working conditions. The analysis of refrigeration energy consumption characteristics helps the system optimize energy utilization, reduce energy consumption costs, and improve the economy of the system. Multi-objective power optimization is performed on the refrigeration energy consumption characteristics of each stage, the refrigeration efficiency evaluation value of each stage, and the maximum temperature fluctuation amplitude value to generate a multi-objective power optimization demand strategy. The multi-objective power optimization demand strategy balances the refrigeration efficiency, energy consumption, and temperature stability requirements to achieve multi-faceted optimization of the refrigeration system.

[0020] Preferably, the specific steps of step S6 are: Step S61: performing real-time refrigeration control on the high and low temperature boxes according to the multi-objective power optimization demand strategy, and collecting real-time refrigeration control response data; Step S62: performing multi-scenario transfer learning on the real-time refrigeration control response data, thereby obtaining refrigeration control response learning data; Step S63: Iterative decision optimization is performed on the refrigeration control response learning data to construct an intelligent refrigeration power control model to execute the refrigeration power closed-loop control operation of the high and low temperature box.

[0021] The present invention performs real-time refrigeration control on high and low temperature boxes through a multi-objective power optimization demand strategy, responds to multi-objective optimization demands in real time, ensures efficient operation of the system under dynamic working conditions, collects real-time refrigeration control response data to help understand the actual operation of the refrigeration system, and provides data support for subsequent optimization and improvement, performs multi-scenario transfer learning on the real-time refrigeration control response data to obtain refrigeration control response learning data, which helps the system learn and adapt to control responses under different working conditions, and multi-scenario transfer learning improves the system's ability to generalize control responses under various working conditions, enhances the system's stability and adaptability, performs iterative decision optimization on the refrigeration control response learning data, constructs an intelligent refrigeration power control model, executes refrigeration power closed-loop control operations of high and low temperature boxes, and constructs an intelligent refrigeration power control model to adjust the refrigeration power control strategy in real time according to the learned knowledge and data, thereby improving the intelligence level and performance of the control system.

[0022] In this specification, a high and low temperature box refrigeration power closed-loop control device is provided, which is used to execute the high and low temperature box refrigeration power closed-loop control method as described above, including: The three-dimensional distribution module is used to obtain the multi-dimensional monitoring parameters of the high and low temperature box based on multi-source sensors; perform three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, perform three-dimensional modeling of the box environment, and construct a three-dimensional parameter feature model; The digital twin module is used to perform dynamic refrigeration airflow dynamics analysis on the three-dimensional parameter characteristic model according to the high and low temperature box, and perform three-dimensional trajectory distribution rendering to build a digital twin model of refrigeration flow; The cooling demand module is used to calculate the multi-stage temperature change trend of the cooling flow digital twin model, and perform dynamic cooling demand calculation to obtain the cooling demand characteristics of each stage; The power demand planning module is used to simulate the refrigeration flow performance based on the refrigeration demand characteristics of each stage, and to plan the compressor refrigeration power demand and build a dynamic power demand planning strategy; The power optimization module is used to perform multi-stage equipment power output calculation and multi-objective power optimization based on the dynamic power demand planning strategy, thereby generating a multi-objective power optimization demand strategy; The closed-loop control module is used to perform real-time refrigeration control of the high and low temperature chambers according to the multi-objective power optimization demand strategy, perform iterative decision optimization, and build an intelligent refrigeration power control model to perform the refrigeration power closed-loop control operation of the high and low temperature chambers.

[0023] The present invention obtains multi-dimensional monitoring parameters of high and low temperature boxes through multi-source sensors, realizes all-round monitoring of the environment inside the box, and performs three-dimensional temperature distribution analysis and three-dimensional modeling of the environment inside the box, which is helpful to understand the temperature distribution law inside the box and improve the efficiency and stability of the refrigeration system. The construction of a three-dimensional parameter characteristic model can provide basic data for the subsequent digital twin module, and realize a deeper understanding and analysis of the environment inside the box. Through dynamic refrigeration airflow dynamics analysis and three-dimensional trajectory distribution rendering, digital modeling and simulation of refrigeration flow are realized. Construction of a refrigeration flow digital twin model can help simulate the refrigeration effect under various working conditions and provide a visual reference for refrigeration system performance optimization. The digital twin model helps the system to quickly respond to needs under different working conditions and improve the intelligence of the refrigeration system. Multi-segment temperature change trend calculation and dynamic refrigeration demand calculation are performed on the refrigeration flow digital twin model to realize accurate prediction and analysis of refrigeration demand and obtain the refrigeration demand at each stage. The characteristics can help the system better regulate the refrigeration system, improve energy utilization and reduce energy consumption costs. Based on the refrigeration demand characteristics, refrigeration flow performance simulation and compressor refrigeration power demand planning are carried out to realize the planning and optimization of dynamic power demand. The construction of dynamic power demand planning strategy can help the system more effectively regulate power output and improve the efficiency and stability of the system. Based on the dynamic power demand planning strategy, multi-stage equipment power output calculation and multi-objective power optimization are carried out to generate a multi-objective power optimization demand strategy. The multi-objective power optimization demand strategy balances refrigeration efficiency, energy consumption and temperature stability, realizes the optimization of the refrigeration system in many aspects, real-time refrigeration control and iterative decision optimization, builds an intelligent refrigeration power control model, realizes the refrigeration power closed-loop control operation of high and low temperature boxes, and the intelligent refrigeration power control model can adjust the refrigeration power according to real-time demand to ensure the efficient and stable operation of the system under different working conditions and improve the intelligence level of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic flow chart of the steps of a closed-loop control method for refrigeration power of a high and low temperature box according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0025] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0026] The present application example provides a closed-loop control method and device for the refrigeration power of a high-low temperature box. The execution subject of the closed-loop control method and device for the refrigeration power of a high-low temperature box includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload equipment, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0027] See also Figures 1 to 4 The present invention provides a closed-loop control method for the refrigeration power of a high-low temperature box, and the closed-loop control method for the refrigeration power of a high-low temperature box comprises the following steps: Step S1: obtaining multi-dimensional monitoring parameters of the high and low temperature box according to the multi-source sensor; performing three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, and performing three-dimensional modeling of the environment inside the box to construct a three-dimensional parameter feature model; Step S2: Perform dynamic refrigeration airflow dynamics analysis on the three-dimensional parameter characteristic model according to the high and low temperature box, and perform three-dimensional trajectory distribution rendering to construct a refrigeration flow digital twin model; Step S3: Calculate the temperature change trend of multiple stages of the refrigeration flow digital twin model, and calculate the dynamic refrigeration demand to obtain the refrigeration demand characteristics of each stage; Step S4: simulating the refrigeration flow performance based on the refrigeration demand characteristics of each stage, planning the compressor refrigeration power demand, and building a dynamic power demand planning strategy; Step S5: Perform multi-stage equipment power output calculation and multi-objective power optimization based on the dynamic power demand planning strategy, thereby generating a multi-objective power optimization demand strategy; Step S6: Perform real-time refrigeration control on the high and low temperature box according to the multi-objective power optimization demand strategy, perform iterative decision optimization, and build an intelligent refrigeration power control model to execute the refrigeration power closed-loop control operation of the high and low temperature box.

[0028] The present invention obtains multi-dimensional monitoring parameters of high and low temperature boxes through multi-source sensors, which is helpful to fully understand the environmental conditions in the box, provide a data basis for subsequent analysis, conduct three-dimensional temperature distribution analysis and modeling, and construct a three-dimensional parameter characteristic model, which can more accurately describe the temperature distribution characteristics in the box, and provide a basis for subsequent refrigeration power control. Dynamic refrigeration airflow dynamics analysis and the construction of a refrigeration flow digital twin model help to understand the working principle and performance of the refrigeration system. Through three-dimensional trajectory distribution rendering, the flow path of the refrigeration airflow is more intuitively displayed, providing visualization support for refrigeration power control. Multi-stage temperature change trend calculation and dynamic refrigeration demand calculation accurately predict the refrigeration demand at each stage, provide basic data for subsequent power planning, and obtain the refrigeration demand characteristics at each stage, which is helpful for more precise Finely regulate the refrigeration system to improve energy efficiency and save energy. Simulate refrigeration flow performance and plan compressor refrigeration power demand to optimize the performance and power utilization of the refrigeration system. The construction of dynamic power demand planning strategies can more effectively adjust the refrigeration power and improve the system operation efficiency. The generation of multi-objective power optimization demand strategies helps to comprehensively consider multiple factors for power optimization and improve the overall performance of the system. Multi-stage equipment power output calculation and power optimization enable the refrigeration system to maintain efficient and stable operation under different working conditions. The construction of real-time refrigeration control and intelligent refrigeration power control models can achieve precise control and adjustment of high and low temperature boxes, iterative decision optimization and refrigeration power closed-loop control operations, and continuously adjust system parameters to ensure that the refrigeration power meets demand while improving system stability.

[0029] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a closed-loop control method for the refrigeration power of a high and low temperature box of the present invention. In this example, the steps of the closed-loop control method for the refrigeration power of the high and low temperature box include: Step S1: obtaining multi-dimensional monitoring parameters of the high and low temperature box according to the multi-source sensor; performing three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, and performing three-dimensional modeling of the environment inside the box to construct a three-dimensional parameter feature model; In this embodiment, the types of multi-source sensors are determined, including temperature sensors, humidity sensors, air flow sensors, and pressure sensors. These sensors should have high precision and fast response capabilities. The installation locations of the sensors are determined, such as various corners, air outlets, and air inlets of the high and low temperature box to ensure comprehensive coverage of the data. The sensors are installed according to the design plan to ensure that the readings of each sensor can accurately reflect the environmental status of its location. The sensors are calibrated to ensure that their measurement accuracy meets the requirements. A data acquisition system is constructed to connect each sensor to a central control unit. A suitable communication protocol (such as Modbus, RS-485, etc.) is used to achieve real-time data transmission. The data acquisition frequency is set (such as once per second) to ensure that various parameters can be continuously recorded during the operation of the high and low temperature box. The collected data is stored in a database to form a multi-dimensional monitoring parameter data set. The data should include information such as timestamp, temperature, humidity, air flow rate, and pressure. The stored multi-dimensional monitoring parameters are cleaned to remove outliers and missing values ​​to ensure the accuracy and consistency of the data. The temperature data is smoothed, such as using a moving average method or a filter to reduce the impact of noise. A three-dimensional visualization tool (such as MATLAB, Paraview, or Python) is used to visualize the data. The processed temperature data is converted into a three-dimensional model using Matplotlib and Mayavi (such as Matplotlib and Mayavi), the temperature reading of each sensor is mapped to the corresponding position in the three-dimensional space, and a three-dimensional temperature distribution map is generated. The generated three-dimensional temperature distribution map is analyzed to identify the uniformity of temperature distribution and hot spots, and the temperature control effect inside the high and low temperature box is evaluated. The analysis results, including the highest and lowest temperature points and temperature gradient, are recorded. Based on the results of the three-dimensional temperature distribution analysis, a three-dimensional environmental model of the high and low temperature box is constructed. A physical model of the high and low temperature box is created using CAD software (such as AutoCAD and SolidWorks) or three-dimensional modeling software (such as Blender), and the position and related parameters of each sensor are marked in the model to ensure that the environmental model is consistent with the actual situation. Multi-dimensional monitoring parameters, including temperature, humidity, airflow rate and pressure, are integrated into the three-dimensional model to form a complete three-dimensional parameter feature model to ensure that the model can reflect the actual environmental status inside the high and low temperature box, which is convenient for subsequent analysis and optimization. The constructed three-dimensional parameter feature model is verified to ensure that it is consistent with the actual monitoring data. The model output is verified by comparing the actual sensor readings. Necessary adjustments and optimizations are made according to the verification results to ensure the accuracy and reliability of the model.

[0030] Step S2: Perform dynamic refrigeration airflow dynamics analysis on the three-dimensional parameter characteristic model according to the high and low temperature box, and perform three-dimensional trajectory distribution rendering to construct a refrigeration flow digital twin model; In this embodiment, necessary environmental parameter data, including temperature, pressure, airflow rate, etc., are extracted from the three-dimensional parameter characteristic model to facilitate airflow dynamics analysis, and the boundary conditions and initial conditions required for the analysis are determined, such as the inlet and outlet states, temperature field, and pressure field of the fluid. Suitable computational fluid dynamics (CFD) software, such as ANSYS Fluent, OpenFOAM, or COMSOL Multiphysics, is selected to perform airflow dynamics simulation, ensuring that the software can handle complex geometric shapes and multi-physics field problems to adapt to the flow model of the high and low temperature box. The three-dimensional parameter characteristic model of the high and low temperature box is imported into the CFD software, and the physical field of the fluid is set. Physical properties (such as density, viscosity, etc.), define flow boundary conditions, including inlet flow velocity, outlet pressure and wall conditions (such as insulation or thermal conductivity), run airflow dynamics simulation, observe the dynamic behavior of airflow inside the high and low temperature box, record key parameters (such as velocity field, pressure field and temperature field), perform steady-state or transient analysis, depending on the system's operating cycle and target requirements, analyze simulation results, identify important information such as flow characteristics, vortices and dead zones of airflow, record the distribution of airflow in the high and low temperature box, form a flow characteristic data set, use post-processing tools in CFD software to extract flow trajectory data of airflow, and determine the particle origin through methods such as particle tracking or streamline drawing. The starting position is usually selected from the flow inlet and outlet and a specific position in the box. Visualization tools (such as ParaView, Tecplot or MATLAB) are used to render the extracted airflow trajectory data in three dimensions to generate a flow trajectory diagram, showing the movement path and flow pattern of the airflow inside the high and low temperature box. Dynamic effects are added to the trajectory rendering to simulate the real-time flow of the airflow. This is achieved through animation effects or flow rate changes to make the visualization more real-time and intuitive. Based on the results of airflow dynamics analysis and three-dimensional trajectory distribution, a digital twin model of refrigeration flow is defined. The model should be able to dynamically reflect the real-time status of the system. The airflow dynamics simulation results are combined with the three-dimensional trajectory diagram to construct a complete A digital twin model is constructed to ensure that the model contains information such as environmental parameters, airflow status, and flow trajectory. Appropriate modeling tools (such as Unity, Blender, or MATLAB) are used to integrate all data to form a visual digital twin model. The constructed digital twin model is verified by comparing the simulation results with the actual operation data to confirm the accuracy of the model. According to the verification results, necessary optimization and adjustment are performed to ensure that the model can adapt to different working conditions and environmental changes. The application scenarios of the digital twin model, such as fault diagnosis, performance optimization, and intelligent control, are determined. Corresponding usage strategies are formulated to make full use of the digital twin model for real-time monitoring and management of high and low temperature chambers.

[0031] Step S3: Calculate the temperature change trend of multiple stages of the refrigeration flow digital twin model, and calculate the dynamic refrigeration demand to obtain the refrigeration demand characteristics of each stage; In this embodiment, temperature data is extracted from the refrigeration flow digital twin model to ensure that the data covers the required time range and the temperature changes in each stage. The time period to be analyzed is determined, which is usually divided according to the equipment operation cycle, external environment changes or internal demand changes. A suitable temperature change trend calculation method is selected, such as linear regression, polynomial regression or moving average method, to analyze the temperature change trend in each stage, determine the key indicators of the temperature change trend, such as temperature change rate, maximum temperature and minimum temperature, etc., analyze the temperature data of each time period, and calculate the temperature change trend, which includes: calculating the temperature average value, change rate and standard deviation of each stage, drawing a temperature change curve according to the selected method, observing the temperature fluctuations in different stages, and recording the temperature change trend characteristics of each stage, including heating, cooling and its amplitude, and determining the formula and method for calculating the dynamic refrigeration demand. It is usually based on the energy balance principle or the actual refrigeration demand model. A suitable calculation tool is selected, and Excel, Python or MATLAB are used for calculation. The temperature change trend calculation results are combined with the operating status parameters of the high and low temperature box (such as compressor power, refrigerant flow, etc.) to form a complete data set to ensure that the data set contains the temperature of each stage. Temperature changes, cooling demand and equipment operating status. Based on the integrated data, calculate the cooling demand characteristics of each stage, calculate the cooling demand, Q=m⋅C⋅ΔT, where m is the refrigerant flow, C is the specific heat capacity of the refrigerant, and ΔT is the temperature change. Evaluate the changes in dynamic cooling demand, including the demand characteristics of different stages (such as demand, cooling load, etc.). Save the cooling demand characteristics of each stage and generate an analysis report to record key data and trends. According to the calculation results, analyze the changing trend of cooling demand to provide a basis for the optimization of subsequent control strategies. Analyze the cooling demand characteristics of each stage. Identify high-demand and low-demand stages in order to develop corresponding control strategies, evaluate the refrigeration efficiency and energy consumption in different stages, analyze their impact on the overall system performance, and adjust the control strategy of the refrigeration system according to the results of the dynamic refrigeration demand calculation to ensure sufficient refrigeration capacity during peak demand periods while reducing energy consumption during low demand periods. It is necessary to implement intelligent control algorithms or optimization algorithms to achieve dynamic adjustment and optimization, compare the calculation results of dynamic refrigeration demand with actual operating data, evaluate the effectiveness of the control strategy, make necessary adjustments and iterations based on feedback results, and continuously optimize the performance and energy efficiency of the refrigeration system.

[0032] Step S4: simulating the refrigeration flow performance based on the refrigeration demand characteristics of each stage, planning the compressor refrigeration power demand, and building a dynamic power demand planning strategy; In this embodiment, the refrigeration demand characteristics of each stage are extracted from the previous steps, including the required refrigeration capacity, temperature change and operating state parameters, and the boundary conditions and initial conditions required for the simulation, such as ambient temperature, pressure, flow rate and refrigerant type, are determined. Appropriate computational fluid dynamics (CFD) software, such as ANSYS Fluent, OpenFOAM or COMSOL Multiphysics, is selected to perform refrigeration flow performance simulation, and it is ensured that the selected software supports complex flow simulation and multi-physics field analysis. The three-dimensional model of the high and low temperature box is imported into the CFD software, and the flow boundary conditions (such as inlet flow rate, outlet pressure, etc.) and physical properties (such as thermodynamic properties of the refrigerant) are set. According to the refrigeration demand characteristics of each stage, the corresponding flow and temperature field conditions are configured, and the CFD simulation is run to observe the dynamic behavior of the refrigeration flow, and the key parameters (such as velocity field, pressure field and temperature field) are recorded. Steady-state or transient simulation can be performed, depending on the operation of the refrigeration system and the research objectives. The simulation results are analyzed to evaluate the performance of the refrigeration flow, including flow uniformity, temperature distribution and pressure loss, etc. The flow performance characteristics of each stage are recorded to provide a basis for subsequent power demand planning and determine the compression mechanism. The calculation method of cooling power demand is usually based on thermodynamic principles and energy balance models. According to the simulation results of refrigeration flow performance in each stage, the required cooling capacity is calculated. According to the efficiency and thermodynamic characteristics of the compressor, the refrigeration power demand of the compressor in each stage is calculated. The power demand in each stage is combined with temperature changes, external loads and system response time to formulate a dynamic power demand planning strategy. The corresponding control strategy is formulated to ensure that sufficient cooling capacity is provided during high-load periods and energy consumption is reduced during low-load periods. The power demand planning results of each stage are saved and an analysis report is generated to record key data and trends. The changes in power demand in different stages and their impact on system performance are analyzed to provide a basis for subsequent optimization. The constructed dynamic power demand planning strategy is evaluated to identify efficient and inefficient operation stages, record the deviation between actual operation data and planned demand, analyze the reasons for the deviation, and adjust the dynamic power demand planning strategy according to the evaluation results to ensure that it adapts to the actual operating conditions of the system. It is necessary to implement intelligent algorithms or machine learning methods to continuously optimize and adjust the control strategy during operation, continuously collect operation data, and conduct periodic evaluation and optimization to ensure that the dynamic power demand planning strategy is always in the best state.

[0033] Step S5: Perform multi-stage equipment power output calculation and multi-objective power optimization based on the dynamic power demand planning strategy, thereby generating a multi-objective power optimization demand strategy; In this embodiment, the dynamic power demand planning strategy is extracted from the previous step, including the cooling demand, equipment operating conditions and power demand calculation results of each stage, and the entire operation cycle is divided into multiple stages according to the operating status of the equipment, cooling demand and environmental changes (such as startup stage, stable stage, shutdown stage, etc.), and a suitable power output calculation method is determined. According to the dynamic power demand planning strategy, the power output of the equipment in each stage is calculated, which includes: calculating the actual power output required for each stage and recording the results, considering the operating characteristics and efficiency of the equipment to ensure that the calculation results are reasonable, selecting a suitable multi-objective optimization algorithm, such as genetic algorithm (GA), particle swarm optimization (PSO) or nonlinear programming (NLP), etc., and determining the optimization constraints, such as power output. The upper and lower limits, equipment operating time and temperature range, etc. are taken into consideration. The power output data of each stage is combined with the optimization target to run the multi-objective optimization algorithm. The specific steps include: inputting the power output, energy consumption, efficiency and other relevant parameters of each stage into the algorithm, running the optimization algorithm, generating a set of optimal solutions (ParetoFront), representing the compromise between different objectives, analyzing the output results of the optimization algorithm, evaluating the performance of each solution under multiple objectives, selecting the optimization strategy that best meets the needs, and comparing it with the actual operating conditions. Based on the optimization results, a multi-objective power optimization demand strategy is constructed to ensure that the cooling demand can be met at different operating stages, while achieving energy efficiency optimization, and recording the specific content of the multi-objective optimization strategy, including the power output plan and expected results for each stage.

[0034] Step S6: Perform real-time refrigeration control on the high and low temperature box according to the multi-objective power optimization demand strategy, perform iterative decision optimization, and build an intelligent refrigeration power control model to execute the refrigeration power closed-loop control operation of the high and low temperature box.

[0035] In this embodiment, the generated multi-objective power optimization demand strategy is extracted from the previous step, including the power output plan and expected effect of each stage, the current operating state, environmental conditions and equipment status are determined in order to implement control, and the real-time control system is configured to ensure that it can receive the instructions and parameters in the multi-objective optimization demand strategy, confirm the normal operation of sensors and actuators, ensure real-time monitoring of data and accurate execution of control instructions, use sensors to monitor the key parameters of the high and low temperature boxes in real time, such as temperature, humidity, refrigerant flow and equipment power, etc., and transmit the collected data to the central control unit for real-time analysis and decision-making. According to the multi-objective optimization demand strategy, Use control algorithms (such as PID control, fuzzy control or adaptive control) to adjust the working state of the refrigeration system, dynamically adjust the power output of the compressor and the refrigerant flow rate to meet the real-time refrigeration demand, establish a feedback mechanism in the control system, evaluate the control effect based on real-time monitoring data, and if there is a deviation between the actual temperature and the set temperature, adjust the control strategy in real time to correct the deviation. Collect operating data after implementing real-time control, including energy consumption, refrigeration efficiency, temperature control accuracy, etc. Analyze the data to identify the advantages and disadvantages of the control strategy, focus on high-efficiency and inefficient operation stages, and select highly adaptable decision-making optimization models, such as reinforcement learning, regression analysis or genetic algorithms. It can adjust itself according to the feedback data, determine the input parameters (such as ambient temperature, current power output, cooling demand) and output parameters (such as optimized control instructions) of the model, run the optimization model based on the collected data and feedback, generate new control strategies, update the parameters of the control model in each round of optimization to improve the prediction accuracy and control effect, verify the optimized control strategy, use new real-time data for comparison to ensure the effectiveness of the model, further adjust the model parameters according to the verification results to ensure that it can adapt to different operating conditions, and design an intelligent cooling power control model based on the results of iterative optimization, including input, output and decision rules, to ensure Ensure that the model can process input data in real time and generate control instructions according to multi-objective optimization requirements. Integrate the intelligent control model into the real-time control system to ensure that it can work seamlessly with other system components (such as sensors and actuators). Ensure that the system can quickly respond to and execute control instructions in a dynamic environment. Test the intelligent refrigeration power control model under actual operating conditions and evaluate its performance in different scenarios. Analyze the advantages and disadvantages of the model based on the test results, record the operating data for subsequent optimization, regularly evaluate the performance of the intelligent control model, update and optimize the model based on new operating data, and continuously introduce new data and feedback to improve the adaptability and control accuracy of the model.

[0036] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: Acquire multi-dimensional monitoring parameters of the high and low temperature chamber according to the multi-source sensor; Step S12: performing a three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box to generate a three-dimensional temperature distribution feature in the box; Step S13: calculating the temperature fluctuation amplitude of the three-dimensional temperature distribution characteristics in the box to generate a three-dimensional temperature distribution fluctuation diagram; Step S14: Analyze the humidity characteristics of each region based on the multi-dimensional monitoring parameters of the high and low temperature chamber, thereby generating multiple regional humidity fields; Step S15: Perform three-dimensional modeling of the in-box environment on the three-dimensional temperature distribution fluctuation diagram and the humidity fields in multiple regions to construct a three-dimensional parameter characteristic model.

[0037] In this embodiment, a variety of sensors are installed in the high and low temperature box, including temperature sensors, humidity sensors, air pressure sensors and flow sensors. Ensure the accuracy and sensitivity of the sensors to obtain accurate monitoring data. Establish a data acquisition system to collect data from each sensor regularly. Set up a data transmission protocol to ensure that real-time data is transmitted to the central processing unit. The acquired multi-dimensional monitoring parameters are stored in a database, including timestamps, temperature, humidity, air pressure and other information for subsequent analysis. The collected temperature data is preprocessed, including denoising, normalization and missing value completion. Select a suitable analysis method, such as a three-dimensional interpolation method (such as Kriging interpolation or inverse distance weighting) to generate a three-dimensional temperature distribution model. According to the processed temperature data, the three-dimensional temperature distribution characteristics in the high and low temperature box are constructed. Generate a three-dimensional visualization model to show the temperature distribution at different locations. Select a suitable fluctuation amplitude calculation method, such as standard deviation, root mean square (RMS) or absolute amplitude. Analyze the three-dimensional temperature distribution characteristics and calculate the temperature fluctuation amplitude of each area. Generate a three-dimensional temperature distribution fluctuation map to show the temperature changes in each area. Collect humidity data from each area to ensure the accuracy and completeness of the data. Select appropriate humidity analysis methods, such as statistical analysis, regional comparative analysis, etc. Analyze the humidity data of the high and low temperature chambers to generate humidity fields for multiple regions. Record the humidity characteristics of each region, including average humidity, fluctuation range, etc. Select appropriate 3D modeling software or libraries, such as Blender, MATLAB, or VTK (Visualization Toolkit). Construct a 3D parameter characteristic model of the high and low temperature chamber based on the 3D temperature distribution fluctuation diagram and humidity field data of multiple regions. Combine temperature and humidity data to generate an integrated 3D environmental model. Visualize the constructed 3D parameter characteristic model to ensure the accuracy and practicality of the model. Verify the model to ensure that it conforms to the actual monitoring data. Save the 3D parameter characteristic model for subsequent analysis and application.

[0038] In this embodiment, refer to Figure 3, is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: monitor the refrigeration operation status of the high and low temperature box and collect refrigeration operation status parameters; Step S22: Identify the refrigeration air outlets of the high and low temperature boxes and mark all the refrigeration air outlets; Step S23: performing spatial layout identification on all cooling air outlets to generate spatial layout features of the cooling air outlets; Step S24: performing dynamic refrigeration airflow dynamics analysis on the refrigeration operation state parameters to extract the refrigeration airflow flow trajectory; Step S25: mining the spatial flow distribution of the cooling air flow trajectory according to the spatial layout characteristics of the cooling air outlet, thereby generating a cooling air flow distribution field; Step S26: Render the three-dimensional trajectory distribution of the three-dimensional parameter feature model based on the cold air flow distribution field to construct a refrigeration flow digital twin model.

[0039] In this embodiment, a variety of sensors are installed, including temperature sensors, pressure sensors and flow meters. The sensors should be distributed in key positions of the high and low temperature box, such as the inlet, outlet and internal layers of the refrigeration system. Ensure the calibration of the sensors, confirm that their measurement range and accuracy can meet the monitoring requirements, design and build a data acquisition system, connect the sensors to the central control unit, formulate the data acquisition frequency (such as once per minute), adopt a suitable communication protocol (such as Modbus, CAN bus, etc.) to achieve real-time data transmission, ensure that the data is updated in time, store the collected data in the database, and form a time series data set. Each data should contain information such as timestamp, temperature, pressure, flow and current for subsequent analysis. Regular Check the data integrity and accuracy to ensure that the monitoring system is operating normally. Use a high-definition camera to shoot the inside of the high and low temperature box, try to avoid reflections and shadows, and ensure that the air outlet is clearly visible. If conditions permit, use infrared imaging equipment to supplement the shooting to facilitate the observation of areas with uneven temperature distribution. Use computer vision technology and target detection algorithms (such as YOLO or SSD) to analyze the images, identify and mark all refrigeration outlets, determine the type, size and location of each outlet, record its coordinates and related features, and store the identified refrigeration outlet information in the database to form an outlet feature data set for subsequent analysis and reference. Perform spatial layout analysis based on the marked outlet information, and use clustering algorithms to analyze the outlets. The distribution of the outlets is evaluated, the relative position and arrangement of each outlet is evaluated, the distance between the outlets and the area of ​​mutual influence is identified by using the spatial relationship analysis method, the spatial layout characteristics of the outlets are recorded, including the number of outlets, distribution density and location distribution map, and a visualization chart is generated to intuitively display the spatial layout of the refrigeration outlets, which is helpful for subsequent airflow analysis. The airflow model of the high and low temperature box is established using computational fluid dynamics (CFD) software (such as ANSYS Fluent or OpenFOAM), and the boundary conditions of the model are set, including the temperature, flow rate and environmental conditions of the outlet to ensure the accuracy of the simulation. The airflow dynamics simulation is run to observe the flow trajectory, velocity distribution and pressure changes of the refrigeration airflow, and record the airflow. The flow characteristics of the airflow are extracted, the motion trajectory and flow velocity data of the airflow are extracted, the uniformity and stability of the airflow are analyzed, and the airflow trajectory and related parameters are stored in the database for subsequent data analysis and visualization. Based on the airflow trajectory data and the spatial layout characteristics of the air outlet, the distribution of the airflow in the high and low temperature box is analyzed. The flow field visualization technology is used to generate the distribution field of the cold air flow and display the flow pattern of the airflow in space. The data mining technology is used to identify the pattern and characteristics of the airflow distribution, and the intensity and coverage of the airflow in different areas are evaluated. The intensity map and flow direction map of the cold air flow distribution field are recorded to form the airflow distribution characteristic data, and the data of the cold air flow distribution field are saved for subsequent modeling, analysis and visualization.Select appropriate 3D modeling tools (such as Unity, Blender or MATLAB) to build the digital twin model, import the 3D model and airflow distribution data of the high and low temperature box, render the 3D parameter feature model based on the cold air flow distribution field, display the flow trajectory of the refrigeration airflow, combine the airflow speed and pressure data to generate dynamic effects, simulate the actual flow of the airflow in the box, visualize the constructed digital twin model, ensure the accuracy and practicality of the model, verify the effectiveness of the model by comparing it with the actual monitoring data, and record the model's operating parameters and performance indicators for subsequent evaluation and optimization.

[0040] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: performing a time-series temperature change analysis on the refrigeration flow digital twin model to generate a time-series temperature change curve; Step S32: performing multi-segment temperature change trend calculation on the time series temperature change curve, thereby obtaining the temperature change trends in multiple time periods; Step S33: Predicting the temperature change trend for multiple time periods and constructing a temperature change trend prediction graph; Step S34: Based on the refrigeration operation log and the temperature change trend prediction diagram, the refrigeration flow digital twin model is dynamically calculated to obtain the refrigeration demand characteristics of each stage.

[0041] In this embodiment, temperature data is extracted from the digital twin model to ensure that the data covers the required time range and related parameters, ensure that the time series of the data is complete, and arrange them in chronological order to avoid the impact of missing data. A suitable time series analysis method is selected, such as moving average method, exponential smoothing method or time series analysis method (such as ARIMA), the extracted temperature data is analyzed, a time series temperature change curve is generated, the temperature value at each time point is calculated, and a curve of temperature change over time is drawn to intuitively display the temperature fluctuation. According to business needs or the characteristics of temperature change, the entire time series temperature change curve is divided into multiple time periods, and a suitable trend calculation method is selected, such as linear regression, polynomial regression or moving average calculation, the temperature data of each time period is analyzed, the temperature change trend is calculated, and the temperature change trend characteristics of each time period are recorded, including slope, change amplitude and trend direction (increasing or decreasing), and the temperature change trend data of each time period is saved for subsequent analysis and reference. A suitable temperature change trend prediction model is selected, such as linear regression model, time series prediction model (such as AR IMA) or machine learning model (such as LSTM), use the previously calculated temperature change trend data as the training set, train the selected prediction model, optimize the model parameters through cross-validation and other methods to improve the accuracy of the prediction, use the trained model to predict future temperature changes, generate a temperature change trend prediction diagram, record the predicted value and its confidence interval for subsequent analysis and decision-making, display the temperature change trend prediction results in graphical form to intuitively reflect the future temperature change trend, collect the refrigeration operation logs of high and low temperature boxes, including refrigeration start time, operation time, refrigeration power and other information, ensure the integrity of the log data for accurate analysis, select a suitable calculation method, such as a calculation method based on energy balance or a calculation method based on demand prediction, combine the temperature change trend prediction diagram with the refrigeration operation log, analyze the impact of temperature changes at different stages on refrigeration demand, calculate the refrigeration demand characteristics of each stage, including the required refrigeration capacity, refrigeration power, etc., save the refrigeration demand characteristic data of each stage, and generate an analysis report to provide to relevant managers for decision-making reference.

[0042] In this embodiment, the specific steps of step S34 are: Get refrigeration operation log; identifying predetermined refrigeration temperatures for multiple stages based on refrigeration operation logs; extracting a timestamp for each stage based on a predetermined refrigeration temperature for the multiple stages; The temperature change situation prediction diagram is time-matched based on the timestamp of each stage, so as to obtain the temperature change prediction situation matching the timestamp; Based on the predetermined refrigeration temperatures of multiple stages, the current temperature deviation is calculated for the temperature change prediction situation matching the timestamp, thereby generating the temperature deviation parameters for each stage; The dynamic refrigeration demand calculation of the refrigeration flow digital twin model is performed according to the temperature deviation parameters of each stage to obtain the refrigeration demand characteristics of each stage.

[0043] In this embodiment, the source of the operation log of the refrigeration system is determined, which is usually a data acquisition system or a control system. To ensure the integrity and accuracy of the log records, the operation log is regularly extracted from the refrigeration system. The log content includes: Timestamp: the time of recording each log entry, Refrigeration temperature setting value (SetTemperature): the target temperature preset by the system, Actual Refrigeration Temperature (ActualTemperature): the current temperature reading of the system, Refrigeration operating status (OperatingStatus): such as on / off status, operating time, etc. According to the timestamp and set temperature in the operation log, multiple refrigeration stages are identified, and each stage The definition of is based on the difference in set temperature changes and time periods. The predetermined refrigeration temperature of each stage is extracted to form a staged temperature data set, including: Stage 1: set temperature (for example: -18°C), Stage 2: set temperature (for example: 0°C), Stage 3: set temperature (for example: -5°C), and the start and end timestamps of each stage are extracted from the operation log to form a time period list, Stage 1: start time (2024-01-0108:00), end time (2024-01-0112:00), Stage 2: start time (2024-01-0112:00), end time (2024-01-0116:00), and the extracted timestamp data is stored in The temperature change prediction data of the matching timestamps are stored in the database to facilitate subsequent analysis and matching, and ensure that the generated temperature change situation prediction diagram contains detailed timestamps and corresponding temperature prediction values. According to the timestamps of each stage, the temperature change situation prediction diagram is time-matched to find the corresponding predicted temperature value of each stage. The predicted temperature corresponding to the matching timestamp of stage 1 (such as the predicted temperature is -17°C) and the predicted temperature corresponding to the matching timestamp of stage 2 (such as the predicted temperature is -4°C) are stored in the database to form a staged temperature prediction data set. The temperature deviation is calculated according to the predetermined refrigeration temperature of each stage and the temperature change prediction situation of the matching timestamp: temperature deviation = predetermined temperature - matching predicted temperature, stage 1 :-18°C-(-17°C)=-1°C (1°C higher) Stage 2: 0°C-(-4°C)=4°C (4°C lower) Select an appropriate dynamic refrigeration demand calculation model, usually based on an energy balance or demand forecasting model, and consider the following parameters: Temperature deviation: a key factor affecting refrigeration demand, External ambient temperature: affects the workload of the refrigeration system, Number of items in the box: another factor affecting heat load, based on the temperature deviation parameters of each stage, combined with external environmental conditions and the number of items in the box, perform a dynamic calculation of refrigeration demand, for example: record the refrigeration demand characteristics calculated for each stage in a database, generate an analysis report, and provide it to relevant managers for decision-making reference.

[0044] In this embodiment, step S4 includes the following steps: Step S41: Calculating the power of the refrigeration compressor based on the refrigeration operation state parameters, thereby obtaining the real-time power characteristics of the refrigeration compressor; Step S42: simulating the refrigeration flow performance of the refrigeration compressor according to the cold air flow distribution field to generate refrigeration flow simulation data in the box; Step S43: planning the compressor refrigeration power demand for each stage of the refrigeration demand characteristics according to the refrigeration flow simulation data in the box, and constructing a dynamic power demand planning strategy.

[0045] In this embodiment, key operating state parameters of the refrigeration system are collected, including refrigerant flow rate, evaporation temperature, condensation temperature, and operating frequency of the compressor, and a suitable calculation formula is selected. The compressor power is usually calculated based on the principle of thermodynamics using the following formula: P=Q / η, where P is the compressor power, Q is the cooling capacity, and η is the efficiency of the compressor. Flow parameter data, including air flow velocity, temperature distribution, and pressure distribution, are extracted from the cold air flow distribution field. Suitable flow performance simulation tools or software, such as CFD (computational fluid dynamics) software (such as ANSYS Fluent, COMSOL Multiphysics) are selected for simulation. A three-dimensional model of the interior of the high and low temperature box is established in the CFD software, and the cold air flow distribution field data is imported as boundary conditions. Initial conditions and boundary conditions are set, including cold air flow velocity, temperature, and compressor power, and flow performance simulation is run to observe the air flow. The dynamic behavior of the flow in the box, record the simulation data, including pressure, temperature and flow rate, etc., generate the simulation data of the refrigeration flow in the box, provide a basis for subsequent analysis, select the appropriate demand planning method, such as linear programming, dynamic programming or model-based optimization algorithm, combine the compressor refrigeration power demand with the refrigeration flow simulation data in the box, analyze the refrigeration demand characteristics at different stages, and plan the compressor refrigeration power demand at each stage according to the simulation data and real-time power characteristics. This includes: evaluating the workload of the compressor, determining the power demand in each time period, adjusting the operating parameters of the compressor (such as frequency and power) according to the predicted temperature changes and refrigeration flow characteristics, and building a dynamic power demand planning strategy based on the demand planning results to ensure that the refrigeration demand can be met at different stages and optimize energy consumption. The refrigeration power demand planning results of each stage are saved, and a report is generated and provided to relevant managers for decision-making and optimization.

[0046] In this embodiment, step S5 includes the following steps: Step S51: performing multi-stage equipment power output calculation based on a dynamic power demand planning strategy, thereby generating equipment output power for each stage; Step S52: performing a time-series cooling efficiency evaluation according to the equipment output power at each stage, thereby obtaining a cooling efficiency evaluation value at each stage; Step S53: Calculate the preset temperature fluctuation constraint threshold value for the refrigeration operation log to obtain the maximum temperature fluctuation amplitude value; Step S54: performing cooling energy consumption analysis on the equipment output power at each stage to generate cooling energy consumption characteristics at each stage; Step S55: Perform multi-objective power optimization on the refrigeration energy consumption characteristics of each stage, the refrigeration efficiency evaluation value of each stage and the maximum temperature fluctuation amplitude value, so as to generate a multi-objective power optimization demand strategy.

[0047] In this embodiment, the formulated dynamic power demand planning strategy is obtained to ensure that the strategy includes the power demand and operating parameters of each stage, and the entire refrigeration operation cycle is divided into multiple stages (such as the startup stage, the stable stage, and the shutdown stage). The division is performed according to the operating status and demand of the equipment, and a suitable power output calculation method is selected. Usually, an efficiency-based calculation model is used to calculate the equipment output power of each stage according to the dynamic power demand planning strategy and stage division. This includes: considering the efficiency and load of the compressor, calculating the actual power output of each stage, recording the power output value of each stage, forming a power output data set, and determining the indicators for refrigeration efficiency evaluation, such as energy efficiency EER (Eer) or refrigeration COP (Coefficient of Performance), collect the equipment output power and actual cooling capacity data of each stage for efficiency evaluation, calculate the refrigeration efficiency evaluation value of each stage according to the equipment output power and actual cooling capacity data, record the refrigeration efficiency evaluation value of each stage, form an efficiency evaluation data set, collect the operation log of the refrigeration system, obtain temperature change data, including set temperature, actual temperature and temperature change rate, select a suitable fluctuation amplitude calculation method, such as maximum temperature fluctuation amplitude calculation, calculate the maximum temperature fluctuation amplitude value according to the collected temperature data, and calculate the set temperature by The difference between the actual temperature and the maximum temperature is realized by recording the maximum and minimum temperatures in each stage, calculating the difference, generating the maximum temperature fluctuation amplitude value, and using it as a constraint for subsequent analysis. Determine the indicators for refrigeration energy consumption analysis, such as energy consumption (kWh) or energy efficiency ratio (EER) in each stage, integrate the equipment output power data and operating time in each stage, calculate the total energy consumption in each stage, and calculate the refrigeration energy consumption characteristics of each stage based on the equipment output power and operating time. This includes: Energy consumption calculation formula: EnergyConsumption=PowerOutput×Time, record the energy consumption characteristics of each stage, and form energy consumption data The cooling energy consumption characteristics of each stage are saved in the database for subsequent analysis and optimization. The objectives of multi-objective optimization are determined, including minimizing energy consumption, maximizing cooling efficiency and controlling temperature fluctuations. A suitable optimization algorithm is selected. Usually, a multi-objective genetic algorithm, particle swarm optimization or linear programming method is used. The cooling energy consumption characteristics of each stage, the cooling efficiency evaluation value and the maximum temperature fluctuation amplitude value are used as inputs to perform multi-objective optimization calculations, generate multi-objective power optimization demand strategies, provide optimal power output strategies to meet cooling needs, store multi-objective power optimization demand strategies in the database, and generate reports for relevant managers to use for decision-making and implementation.

[0048] In this embodiment, step S6 includes the following steps: Step S61: performing real-time refrigeration control on the high and low temperature boxes according to the multi-objective power optimization demand strategy, and collecting real-time refrigeration control response data; Step S62: performing multi-scenario transfer learning on the real-time refrigeration control response data, thereby obtaining refrigeration control response learning data; Step S63: Iterative decision optimization is performed on the refrigeration control response learning data to construct an intelligent refrigeration power control model to execute the refrigeration power closed-loop control operation of the high and low temperature box.

[0049] In this embodiment, according to the multi-objective power optimization demand strategy, a real-time refrigeration control system is designed. The system should be able to adjust the output power of the refrigeration equipment in real time according to the current temperature, humidity and other environmental factors. The control algorithm (such as PID control, fuzzy control or adaptive control) is determined to ensure that the control system can respond to changes quickly. Sensors are installed inside and around the high and low temperature box to monitor parameters such as temperature, humidity, airflow and equipment power in real time. The collected data is transmitted to the central processing unit in real time through the data acquisition system to ensure the timeliness and accuracy of the data. The real-time refrigeration control response data is recorded, including control instructions, actual power output, environmental parameters and timestamp information. Features are extracted from the real-time refrigeration control response data to form a refrigeration control response learning data set. These features include ambient temperature, humidity, equipment power and time, etc., to ensure the diversity of the data set to cover different environmental conditions and usage scenarios. A suitable transfer learning method is selected, such as transfer learning based on deep learning (for example, convolutional neural network, recursive neural network, etc.), which can effectively learn new scenarios from existing knowledge. The source domain and the target domain are determined. The source domain is the known historical control data, and the target domain is the current real-time control data. The real-time refrigeration control response data is trained using a transfer learning algorithm. The knowledge is transferred to the target field. During the training process, the model parameters are optimized to improve the prediction accuracy in the new scenario, and the refrigeration control response learning data is generated to form the basis for subsequent decision optimization. The refrigeration control response learning data is analyzed to extract key features and patterns, including temperature change trends, power output changes, etc., to identify the key factors affecting refrigeration efficiency and response time to guide decision optimization, and to select suitable decision optimization algorithms, such as reinforcement learning, genetic algorithms or other intelligent optimization algorithms, to achieve iterative decision optimization and determine the optimization goals, such as minimizing energy consumption, maximizing refrigeration efficiency or maintaining temperature stability. Through multiple rounds of iterations, The control strategy is continuously adjusted to optimize the cooling power output. After each iteration, the performance of the model is evaluated and adjusted. In each round of iteration, the model parameters are updated using feedback data to improve the responsiveness and efficiency of the system. Based on the results of iterative optimization, an intelligent cooling power control model is constructed, which can realize closed-loop control of high and low temperature boxes. The model is integrated into the real-time control system to ensure that the system can automatically adjust the cooling power according to environmental changes. The constructed intelligent control model is tested to verify its effect in different scenarios. Necessary adjustments and optimizations are made according to the test results to ensure that the model can adapt to different operating conditions and maintain efficient cooling performance.

[0050] In this embodiment, a closed-loop control device for refrigeration power of a high-low temperature box is provided, which is used to execute the closed-loop control method for refrigeration power of a high-low temperature box as described above, including: The three-dimensional distribution module is used to obtain the multi-dimensional monitoring parameters of the high and low temperature box based on multi-source sensors; perform three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, perform three-dimensional modeling of the environment inside the box, and construct a three-dimensional parameter feature model; The digital twin module is used to perform dynamic refrigeration airflow dynamics analysis on the three-dimensional parameter characteristic model according to the high and low temperature box, and perform three-dimensional trajectory distribution rendering to build a digital twin model of refrigeration flow; The cooling demand module is used to calculate the multi-stage temperature change trend of the cooling flow digital twin model, and perform dynamic cooling demand calculation to obtain the cooling demand characteristics of each stage; The power demand planning module is used to simulate the refrigeration flow performance based on the refrigeration demand characteristics of each stage, and to plan the compressor refrigeration power demand and build a dynamic power demand planning strategy; The power optimization module is used to perform multi-stage equipment power output calculation and multi-objective power optimization based on the dynamic power demand planning strategy, thereby generating a multi-objective power optimization demand strategy; The closed-loop control module is used to perform real-time refrigeration control of the high and low temperature box according to the multi-objective power optimization demand strategy, perform iterative decision optimization, and build an intelligent refrigeration power control model to perform the refrigeration power closed-loop control operation of the high and low temperature box.

[0051] The present invention obtains multi-dimensional monitoring parameters of high and low temperature boxes through multi-source sensors, realizes all-round monitoring of the environment inside the box, and performs three-dimensional temperature distribution analysis and three-dimensional modeling of the environment inside the box, which is helpful to understand the temperature distribution law inside the box and improve the efficiency and stability of the refrigeration system. The construction of a three-dimensional parameter characteristic model can provide basic data for the subsequent digital twin module, and realize a deeper understanding and analysis of the environment inside the box. Through dynamic refrigeration airflow dynamics analysis and three-dimensional trajectory distribution rendering, digital modeling and simulation of refrigeration flow are realized. Construction of a refrigeration flow digital twin model can help simulate the refrigeration effect under various working conditions and provide a visual reference for refrigeration system performance optimization. The digital twin model helps the system to quickly respond to needs under different working conditions and improve the intelligence of the refrigeration system. Multi-segment temperature change trend calculation and dynamic refrigeration demand calculation are performed on the refrigeration flow digital twin model to realize accurate prediction and analysis of refrigeration demand and obtain the refrigeration demand at each stage. The characteristics can help the system better regulate the refrigeration system, improve energy utilization and reduce energy consumption costs. Based on the refrigeration demand characteristics, refrigeration flow performance simulation and compressor refrigeration power demand planning are carried out to realize the planning and optimization of dynamic power demand. The construction of dynamic power demand planning strategy can help the system more effectively regulate power output and improve the efficiency and stability of the system. Based on the dynamic power demand planning strategy, multi-stage equipment power output calculation and multi-objective power optimization are carried out to generate a multi-objective power optimization demand strategy. The multi-objective power optimization demand strategy balances refrigeration efficiency, energy consumption and temperature stability, realizes the optimization of the refrigeration system in many aspects, real-time refrigeration control and iterative decision optimization, builds an intelligent refrigeration power control model, realizes the refrigeration power closed-loop control operation of high and low temperature boxes, and the intelligent refrigeration power control model can adjust the refrigeration power according to real-time demand to ensure the efficient and stable operation of the system under different working conditions and improve the intelligence level of the system.

[0052] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0053] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A closed-loop control method for refrigeration power of a high and low temperature box, characterized in that: The following steps are involved: Step S1: obtaining multi-dimensional monitoring parameters of the high and low temperature box according to the multi-source sensor; performing three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, and performing three-dimensional modeling of the environment inside the box to construct a three-dimensional parameter feature model; Step S2: Perform dynamic refrigeration airflow dynamics analysis on the three-dimensional parameter characteristic model according to the high and low temperature box, and perform three-dimensional trajectory distribution rendering to construct a refrigeration flow digital twin model; Step S3: Calculate the temperature change trend of multiple stages of the refrigeration flow digital twin model, and calculate the dynamic refrigeration demand to obtain the refrigeration demand characteristics of each stage; Step S4: simulating the refrigeration flow performance based on the refrigeration demand characteristics of each stage, planning the compressor refrigeration power demand, and building a dynamic power demand planning strategy; Step S5: Perform multi-stage equipment power output calculation and multi-objective power optimization based on the dynamic power demand planning strategy, thereby generating a multi-objective power optimization demand strategy; Step S6: Perform real-time refrigeration control on the high and low temperature box according to the multi-objective power optimization demand strategy, perform iterative decision optimization, and build an intelligent refrigeration power control model to execute the refrigeration power closed-loop control operation of the high and low temperature box.

2. The closed-loop control method for refrigeration power of a high and low temperature box according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: Acquire multi-dimensional monitoring parameters of the high and low temperature chamber according to the multi-source sensor; Step S12: performing a three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box to generate a three-dimensional temperature distribution feature in the box; Step S13: calculating the temperature fluctuation amplitude of the three-dimensional temperature distribution characteristics in the box to generate a three-dimensional temperature distribution fluctuation diagram; Step S14: Analyze the humidity characteristics of each region based on the multi-dimensional monitoring parameters of the high and low temperature chamber, thereby generating multiple regional humidity fields; Step S15: Perform three-dimensional modeling of the in-box environment on the three-dimensional temperature distribution fluctuation diagram and the humidity fields in multiple regions to construct a three-dimensional parameter characteristic model.

3. The closed-loop control method for refrigeration power of a high and low temperature box according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: monitor the refrigeration operation status of the high and low temperature box and collect refrigeration operation status parameters; Step S22: Identify the refrigeration air outlets of the high and low temperature boxes and mark all the refrigeration air outlets; Step S23: performing spatial layout identification on all cooling air outlets to generate spatial layout features of the cooling air outlets; Step S24: performing dynamic refrigeration airflow dynamics analysis on the refrigeration operation state parameters to extract the refrigeration airflow flow trajectory; Step S25: mining the spatial flow distribution of the cooling air flow trajectory according to the spatial layout characteristics of the cooling air outlet, thereby generating a cooling air flow distribution field; Step S26: Render the three-dimensional trajectory distribution of the three-dimensional parameter feature model based on the cold air flow distribution field to construct a refrigeration flow digital twin model.

4. The closed-loop control method for refrigeration power of a high and low temperature box according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: performing a time-series temperature change analysis on the refrigeration flow digital twin model to generate a time-series temperature change curve; Step S32: performing multi-segment temperature change trend calculation on the time series temperature change curve, thereby obtaining the temperature change trends in multiple time periods; Step S33: Predicting the temperature change trend for multiple time periods and constructing a temperature change trend prediction graph; Step S34: Based on the refrigeration operation log and the temperature change trend prediction diagram, the refrigeration flow digital twin model is dynamically calculated to obtain the refrigeration demand characteristics of each stage.

5. The closed-loop control method for refrigeration power of a high and low temperature box according to claim 4, characterized in that: The specific steps of step S34 are: Get refrigeration operation log; identifying predetermined refrigeration temperatures for multiple stages based on refrigeration operation logs; extracting a timestamp for each stage based on a predetermined refrigeration temperature for the multiple stages; The temperature change situation prediction diagram is time-matched based on the timestamp of each stage, so as to obtain the temperature change prediction situation matching the timestamp; Based on the predetermined refrigeration temperatures of multiple stages, the current temperature deviation is calculated for the temperature change prediction situation matching the timestamp, thereby generating the temperature deviation parameters for each stage; The dynamic refrigeration demand calculation of the refrigeration flow digital twin model is performed according to the temperature deviation parameters of each stage to obtain the refrigeration demand characteristics of each stage.

6. The closed-loop control method for refrigeration power of a high and low temperature box according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: Calculating the power of the refrigeration compressor based on the refrigeration operation state parameters, thereby obtaining the real-time power characteristics of the refrigeration compressor; Step S42: simulating the refrigeration flow performance of the refrigeration compressor according to the cold air flow distribution field to generate refrigeration flow simulation data in the box; Step S43: planning the compressor refrigeration power demand for each stage of the refrigeration demand characteristics according to the refrigeration flow simulation data in the box, and constructing a dynamic power demand planning strategy.

7. The closed-loop control method for refrigeration power of a high and low temperature box according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: performing multi-stage equipment power output calculation based on a dynamic power demand planning strategy, thereby generating equipment output power for each stage; Step S52: performing a time-series cooling efficiency evaluation according to the equipment output power at each stage, thereby obtaining a cooling efficiency evaluation value at each stage; Step S53: Calculate the preset temperature fluctuation constraint threshold value for the refrigeration operation log to obtain the maximum temperature fluctuation amplitude value; Step S54: performing cooling energy consumption analysis on the equipment output power at each stage to generate cooling energy consumption characteristics at each stage; Step S55: Perform multi-objective power optimization on the refrigeration energy consumption characteristics of each stage, the refrigeration efficiency evaluation value of each stage and the maximum temperature fluctuation amplitude value, so as to generate a multi-objective power optimization demand strategy.

8. The closed-loop control method for refrigeration power of a high and low temperature box according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing real-time refrigeration control on the high and low temperature boxes according to the multi-objective power optimization demand strategy, and collecting real-time refrigeration control response data; Step S62: performing multi-scenario transfer learning on the real-time refrigeration control response data, thereby obtaining refrigeration control response learning data; Step S63: Iterative decision optimization is performed on the refrigeration control response learning data to construct an intelligent refrigeration power control model to execute the refrigeration power closed-loop control operation of the high and low temperature box.

9. A closed-loop control device for refrigeration power of a high and low temperature box, characterized in that: The method for executing the closed-loop control method of refrigeration power of a high-low temperature box according to claim 1 comprises: The three-dimensional distribution module is used to obtain the multi-dimensional monitoring parameters of the high and low temperature box based on multi-source sensors; perform three-dimensional temperature distribution analysis on the multi-dimensional monitoring parameters of the high and low temperature box, perform three-dimensional modeling of the environment inside the box, and construct a three-dimensional parameter feature model; The digital twin module is used to perform dynamic refrigeration airflow dynamics analysis on the three-dimensional parameter characteristic model according to the high and low temperature box, and perform three-dimensional trajectory distribution rendering to build a digital twin model of refrigeration flow; The cooling demand module is used to calculate the multi-stage temperature change trend of the cooling flow digital twin model, and perform dynamic cooling demand calculation to obtain the cooling demand characteristics of each stage; The power demand planning module is used to simulate the refrigeration flow performance based on the refrigeration demand characteristics of each stage, and to plan the compressor refrigeration power demand and build a dynamic power demand planning strategy; The power optimization module is used to perform multi-stage equipment power output calculation and multi-objective power optimization based on the dynamic power demand planning strategy, thereby generating a multi-objective power optimization demand strategy; The closed-loop control module is used to perform real-time refrigeration control of the high and low temperature box according to the multi-objective power optimization demand strategy, perform iterative decision optimization, and build an intelligent refrigeration power control model to perform the refrigeration power closed-loop control operation of the high and low temperature box.

Images