Temperature stability control system and method for dynamic internal heat source high-low temperature box
Through the communication between the cloud computing center and the control device, combined with CAE simulation, deep learning and reinforcement learning algorithms, intelligent temperature control of high and low temperature boxes is realized, solving the problems of high labor costs and low accuracy in traditional methods, and improving temperature control efficiency and reliability.
Patent Information
- Application Number
- CN202411021213.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-11-15
AI Technical Summary
The traditional high and low temperature box temperature control method relies on human experience, resulting in high labor costs, low control accuracy and slow response speed, which cannot meet the high-precision and high-speed temperature control needs, and the steps are complex and the time cost is high, so intelligent control cannot be achieved.
采用云计算中心与控制装置的通信连接,利用CAE仿真、深度学习和强化学习算法,构建温度场模型和控制参数优化模型,实时监测和分析温度数据,生成优化策略,自动化控制高低温箱的动态内热源。
It reduces labor cost investment, improves the accuracy and response speed of temperature control, achieves temperature stability and rapid response, reduces time and hardware costs, and improves the reliability of test results.
Smart Images

Figure CN118963441A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of temperature control, and in particular relates to a temperature stability control system and method for a high and low temperature box with a dynamic internal heat source. Background Art
[0002] Thermal vacuum test is one of the most important environmental and reliability tests in the development stage of spacecraft. Its purpose is to verify the coordination and matching between various subsystems and single machines under vacuum high and low temperature environmental stress. At present, the thermal vacuum test cycle is long and the cost is high. When the spacecraft enters the batch production stage, in order to relieve the pressure on the thermal vacuum test equipment, it is necessary to use low-cost and short-cycle normal pressure thermal cycle tests to replace part of the thermal vacuum test. As an important equipment for normal pressure thermal cycle tests, the temperature stability of high and low temperature chambers has an important influence on the test results. However, traditional temperature control methods of high and low temperature chambers mostly rely on human experience and judgment, with high manpower cost investment, low control accuracy, slow response speed, and cannot meet the needs of high-precision and high-speed temperature control; in addition, the steps are complicated, requiring a lot of time and cost, and cannot be intelligently controlled. Summary of the invention
[0003] In order to solve the problems of high manpower cost, low control accuracy, slow response speed, complicated steps and high time cost in the prior art, the present invention aims to provide a temperature stability control system and method for a dynamic internal heat source high and low temperature box.
[0004] The technical solution adopted by the present invention is:
[0005] A temperature stability control system for a high and low temperature box with a dynamic internal heat source comprises a cloud computing center and a plurality of control devices. The cloud computing center is respectively connected to the plurality of control devices for communication. Each control device is arranged at a high and low temperature box in a one-to-one correspondence, and the control device is electrically connected to the corresponding high and low temperature box and its dynamic internal heat source.
[0006] Furthermore, the cloud computing center is provided with a CAE simulation unit, a model building unit, a temperature field reduction unit, a temperature field analysis unit and a control parameter optimization unit which are connected in sequence, and the temperature field reduction unit and the control parameter optimization unit are respectively communicated with a number of control devices.
[0007] Furthermore, the control device includes an industrial computer, a main control module and a sensor module. The industrial computer is communicatively connected to the cloud computing center, and the industrial computer is electrically connected to the sensor module, the main control module and the controllers of the corresponding high and low temperature boxes, respectively. The main control module is electrically connected to the actuator of the dynamic internal heat source of the corresponding high and low temperature box, and the sensor module is arranged at the corresponding high and low temperature box.
[0008] Furthermore, the industrial computer includes a wireless communication module, a processor, a control instruction generating module and a data processing module. The processor is electrically connected to the wireless communication module, the control instruction generating module and the data processing module respectively. The wireless communication module is communicatively connected to the cloud computing center, the control instruction generating module is electrically connected to the main control module, and the data processing module is electrically connected to the sensor module and the controller of the corresponding high and low temperature box respectively.
[0009] Furthermore, the dynamic internal heat source of the high and low temperature box is provided with a liquid nitrogen storage tank, a liquid nitrogen valve, a hot and cold air blower and an electric heater, and the control end of the liquid nitrogen valve, the control end of the hot and cold air blower and the control end of the electric heater are all electrically connected to the main control module;
[0010] The sensor module includes a temperature sensor, a humidity sensor, a pressure sensor, a wind speed sensor and a flow sensor. The temperature sensor and humidity sensor are arranged inside the high and low temperature box, the pressure sensor is arranged inside the liquid nitrogen storage tank, and the flow sensor is arranged at the air outlet of the hot and cold air blower.
[0011] A temperature stability control method for a dynamic internal heat source high and low temperature box, based on a temperature stability control system, the method comprises the following steps:
[0012] Based on the cloud computing center, according to some historical monitoring data, the temperature field CAE simulation is carried out, the temperature field CAE simulation model is constructed, and some historical temperature field data are generated;
[0013] Based on several historical monitoring data and corresponding historical temperature field data, a temperature field reduction model, a temperature field analysis model and a control parameter optimization model are constructed;
[0014] Based on the control device, the real-time sensor data of the high and low temperature box and the real-time operation data of the corresponding dynamic internal heat source are collected to form real-time monitoring data, and uploaded to the cloud computing center;
[0015] Based on the cloud computing center, according to the real-time monitoring data, the temperature field is reduced using the temperature field reduction model to obtain real-time temperature field data;
[0016] According to the real-time temperature field data, the temperature field analysis model is used to perform temperature field analysis and obtain the real-time temperature field analysis results;
[0017] According to the real-time monitoring data and the real-time temperature field analysis results, the control parameter optimization model is used to optimize the control parameters, obtain the real-time control parameter optimization strategy, and return it to the corresponding control device;
[0018] Based on the control device, according to the real-time control parameter optimization strategy, a real-time temperature control instruction is generated, and according to the real-time temperature control instruction, the dynamic internal heat source of the high and low temperature box is controlled.
[0019] Furthermore, based on the cloud computing center, according to a number of historical monitoring data, a temperature field CAE simulation is performed, a temperature field CAE simulation model is constructed, and a number of historical temperature field data are generated, including the following steps:
[0020] Based on the cloud computing center, some historical monitoring data are reduced in dimension to obtain a set of key monitoring indicators and some historical monitoring data after dimension reduction;
[0021] The monitored data after dimension reduction is used as the model input of the temperature field CAE simulation, and the temperature field data is used as the model output of the temperature field CAE simulation;
[0022] According to the model input and model output, use CAE simulation tools to perform temperature field CAE simulation and build a temperature field CAE simulation model;
[0023] According to a number of historical monitoring data after dimensionality reduction, a temperature field CAE simulation model is used to perform temperature field CAE simulation to generate a number of historical temperature field data.
[0024] Furthermore, based on a number of historical monitoring data and corresponding historical temperature field data, a temperature field reduced-order model, a temperature field analysis model and a control parameter optimization model are constructed, including the following steps:
[0025] Based on several historical monitoring data and corresponding historical temperature field data, a temperature field reduced-order model is constructed using the singular value decomposition algorithm;
[0026] Based on several historical temperature field data, a deep learning algorithm is used to build a temperature field analysis model and generate several historical temperature field analysis results;
[0027] Based on several historical dimensionality-reduced monitoring data and the corresponding historical temperature field analysis results, a control parameter optimization model is constructed using the reinforcement learning algorithm.
[0028] Furthermore, the temperature field analysis model is constructed based on the ResNet algorithm.
[0029] Furthermore, the control parameter optimization model is constructed based on the DQN algorithm.
[0030] The beneficial effects of the present invention are:
[0031] The present invention discloses a temperature stability control system and method for a high and low temperature box with a dynamic internal heat source. CAE simulation technology is used to establish a CAE simulation model of the temperature field of the high and low temperature box, and the temperature distribution characteristics in the box are analyzed to provide training data support for subsequent temperature control, thereby avoiding the problem of low prediction accuracy caused by insufficient training data; a temperature field reduced-order model is used to generate corresponding real-time temperature field data according to real-time monitoring data, thereby reducing time cost investment, reducing hardware requirements and the calculation complexity of the control algorithm, and improving control efficiency; deep learning technology is used to learn the temperature field data, extract characteristic information of the temperature field, and realize real-time analysis and prediction of temperature changes, thereby reducing manpower cost investment and improving the accuracy and response speed of temperature control; a reinforcement learning algorithm is used to automatically and intelligently optimize the control parameters of the dynamic internal heat source, thereby optimizing the temperature stability of the high and low temperature box, realizing rapid response and precise control of temperature fluctuations, reducing time costs, and improving the reliability of test results; the system has high versatility and can be applied to the temperature stability control of different types of high and low temperature boxes, thereby having strong practical value.
[0032] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 It is a structural block diagram of a temperature stability control system for a high and low temperature box with a dynamic internal heat source.
[0034] Figure 2 It is a flow chart of a temperature stability control method for a high and low temperature box with a dynamic internal heat source. DETAILED DESCRIPTION
[0035] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.
[0036] Embodiment 1:
[0037] like Figure 1 As shown, this embodiment provides a temperature stability control system for a high and low temperature box with a dynamic internal heat source, including a cloud computing center and a plurality of control devices, wherein the cloud computing center is respectively connected to the plurality of control devices for communication, each control device is arranged at a high and low temperature box in a one-to-one correspondence, and the control device is electrically connected to the corresponding high and low temperature box and its dynamic internal heat source;
[0038] The cloud computing center is used to perform three-dimensional computer-aided engineering (CAE) simulation of the temperature field according to a number of historical monitoring data, construct a CAE simulation model of the temperature field, and generate a number of historical temperature field data; construct a temperature field reduction model, a temperature field analysis model, and a control parameter optimization model according to a number of historical monitoring data and corresponding historical temperature field data; use the temperature field reduction model to reduce the temperature field according to real-time monitoring data to obtain real-time temperature field data; use the temperature field analysis model to perform temperature field analysis according to the real-time temperature field data to obtain real-time temperature field analysis results; use the control parameter optimization model to optimize the control parameters according to the real-time monitoring data and the real-time temperature field analysis results, obtain a real-time control parameter optimization strategy, and return it to the corresponding control device;
[0039] The control device is used to collect real-time sensor data of the high and low temperature box and the real-time operation data of the corresponding dynamic internal heat source, constitute real-time monitoring data, and upload it to the cloud computing center; generate real-time temperature control instructions according to the real-time control parameter optimization strategy, and control the dynamic internal heat source of the high and low temperature box according to the real-time temperature control instructions.
[0040] Preferably, the cloud computing center is provided with a CAE simulation unit, a model building unit, a temperature field order reduction unit, a temperature field analysis unit and a control parameter optimization unit connected in sequence, and the temperature field order reduction unit and the control parameter optimization unit are respectively connected to a plurality of control devices for communication;
[0041] A CAE simulation unit is used to perform a temperature field CAE simulation based on a number of historical monitoring data, construct a temperature field CAE simulation model, and generate a number of historical temperature field data;
[0042] A model building unit, used to build a temperature field reduced-order model, a temperature field analysis model and a control parameter optimization model according to a number of historical monitoring data and corresponding historical temperature field data;
[0043] The temperature field order reduction unit is used to reduce the temperature field order according to the real-time monitoring data using the temperature field order reduction model to obtain the real-time temperature field data;
[0044] The temperature field analysis unit is used to perform temperature field analysis based on real-time temperature field data using a temperature field analysis model to obtain real-time temperature field analysis results;
[0045] The control parameter optimization unit is used to optimize the control parameters according to the real-time monitoring data and the real-time temperature field analysis results, use the control parameter optimization model, obtain the real-time control parameter optimization strategy, and return it to the corresponding control device.
[0046] Preferably, the control device includes an industrial computer, a main control module and a sensor module. The industrial computer is connected to the cloud computing center for communication, and the industrial computer is electrically connected to the sensor module, the main control module and the controller of the corresponding high and low temperature box respectively. The main control module is electrically connected to the actuator of the dynamic internal heat source of the corresponding high and low temperature box, and the sensor module is arranged at the corresponding high and low temperature box.
[0047] The sensor module is used to collect real-time sensor data of the high and low temperature box and send it to the industrial computer;
[0048] The industrial computer is used to receive the real-time sensor data sent by the sensor module, collect the real-time operation data of the dynamic internal heat source, and form the real-time monitoring data; generate the real-time temperature control instruction according to the real-time control parameter optimization strategy, and send the real-time temperature control instruction to the main control module;
[0049] The main control module is used to control the dynamic internal heat source of the high and low temperature box according to real-time temperature control instructions.
[0050] Preferably, the industrial computer includes a wireless communication module, a processor, a control instruction generation module and a data processing module, the processor is electrically connected to the wireless communication module, the control instruction generation module and the data processing module respectively, the wireless communication module is communicatively connected to the cloud computing center, the control instruction generation module is electrically connected to the main control module, and the data processing module is electrically connected to the sensor module and the controller of the corresponding high and low temperature box respectively;
[0051] Wireless communication module, used to establish a communication channel with the cloud computing center to achieve the transmission of real-time monitoring data and real-time control parameter optimization strategy;
[0052] Processor, used to control the normal operation of other modules;
[0053] A control instruction generation module is used to generate real-time temperature control instructions according to the real-time control parameter optimization strategy and send them to the main control module;
[0054] The data processing module is used to perform data format conversion and other processing on the real-time sensor data sent by the sensor module and the real-time operation data of the dynamic internal heat source, and convert them into a data format that can be recognized by the cloud computing center.
[0055] Preferably, the dynamic internal heat source of the high and low temperature box is provided with a liquid nitrogen storage tank, a liquid nitrogen valve, a hot and cold air blower and an electric heater, and the control end of the liquid nitrogen valve, the control end of the hot and cold air blower and the control end of the electric heater are all electrically connected to the main control module;
[0056] Liquid nitrogen storage tanks, liquid nitrogen valves, hot and cold air blowers and electric heaters constitute the dynamic internal heat source of the high and low temperature box, which is used to achieve temperature control of the high and low temperature box;
[0057] The sensor module includes a temperature sensor, a humidity sensor, a pressure sensor, a wind speed sensor and a flow sensor. The temperature sensor and humidity sensor are arranged inside the high and low temperature box, the pressure sensor is arranged inside the liquid nitrogen storage tank, and the flow sensor is arranged at the air outlet of the hot and cold air blower.
[0058] Embodiment 2:
[0059] like Figure 2 As shown, this embodiment provides a temperature stability control method for a dynamic internal heat source high and low temperature box. Based on the temperature stability control system, the method includes the following steps:
[0060] S1: Based on the cloud computing center, according to a number of historical monitoring data, a temperature field CAE simulation is performed, a temperature field CAE simulation model is constructed, and a number of historical temperature field data are generated, including the following steps:
[0061] S1-1: Based on the cloud computing center, the principal component analysis (PCA) algorithm is used to reduce the dimension of some historical monitoring data to obtain a set of key monitoring indicators and some historical monitoring data after dimension reduction;
[0062] The data volume after dimensionality reduction is reduced, and the ability to represent key information is improved, which increases the speed of subsequent model construction and the accuracy of prediction;
[0063] The monitoring data includes the temperature data, humidity data, pressure data, wind speed data, flow data, etc. of the high and low temperature box collected by the sensor module, and the current data, voltage data, power data, frequency data, and cycle data of the dynamic internal heat source in actual operation;
[0064] S1-2: Using the reduced-dimensional monitoring data as the model input of the temperature field CAE simulation, and using the temperature field data as the model output of the temperature field CAE simulation;
[0065] S1-3: Based on the model input and model output, use CAE simulation tools to perform temperature field CAE simulation and build a temperature field CAE simulation model;
[0066] CAE simulation tools can calculate the spatial distribution of temperature field physical quantities under given boundary conditions. By dividing the space into countless small blocks, it is easy to derive the basic physical formulas that each small block obeys. By combining the physical formulas obeyed by all small blocks, a set of equations with dimensions of up to hundreds of millions can be formed. By solving the set of equations, the values of all physical quantities in the physical formulas on each small block can be obtained, thereby realizing the calculation of the spatial distribution of temperature field physical quantities.
[0067] S1-4: Based on a number of historical monitoring data after dimensionality reduction, use the temperature field CAE simulation model to perform temperature field CAE simulation to generate a number of historical temperature field data;
[0068] S2: Based on a number of historical monitoring data and corresponding historical temperature field data, a temperature field reduction model, a temperature field analysis model and a control parameter optimization model are constructed, including the following steps:
[0069] S2-1: Based on several historical monitoring data and corresponding historical temperature field data, a temperature field reduced-order model is constructed using a singular value decomposition algorithm, including the following steps:
[0070] S2-1-1: Convert some historical temperature field data into historical training field matrix;
[0071] Among them, the number of historical monitoring data is m, the number of historical temperature field data is n, and the corresponding historical training field matrix has a dimension of m×n, a rectangular matrix, and the number of rows is much larger than the number of columns;
[0072] S2-1-2: Perform singular value decomposition on the historical training ground matrix to obtain the corresponding historical eigenvector;
[0073] The formula is:
[0074] A=U×Σ×V T
[0075] Where A is the historical training field matrix; U is the historical left eigenvector; Σ is the eigenvalue matrix; V T is the transpose of the historical right eigenvector; U is an m×n unitary matrix, V is an n×n unitary matrix, Σ is an m×n matrix, the values on the diagonal are the eigenvalues, and according to the properties of the singular value decomposition, the eigenvalues gradually decrease from the upper left corner to the lower right corner;
[0076] The column vectors in the U matrix, each column represents an independent temperature field whose contribution to the final temperature field decreases from left to right;
[0077] A T =VΣ T U T
[0078] In the formula, A T is the transpose of the historical training field matrix; Σ T is the transpose of the eigenvalue matrix; U T is the transpose of the historical left eigenvector; V is the historical right eigenvector;
[0079] A T A=VΣ T U T UΣV T =VΣ2 V T
[0080] Since U is a unitary matrix, Σ 2 is a diagonal matrix. The values on the diagonal are the squares of the eigenvalues after singular value decomposition. T A is a square matrix;
[0081] U=A(ΣV T ) -1
[0082] Σ is an m×n matrix, (ΣV T ) -1 It is an n×m matrix. Construct Σ into a diagonal square matrix, and the obtained U is an m×n matrix;
[0083] S2-1-3: According to the contribution of historical characteristic vectors to the temperature field, select the main historical characteristic vectors (U s |s=1,2,...,r), where U s are the sth historical main eigenvectors, s is the main eigenvector indicator, r is the historical main eigenorder, and the corresponding historical main eigenvector coefficients (ΣV T =a s |s=1,2,...,r), where a s is the historical main eigenvector coefficient, Σ is the eigenvalue matrix, V T is the transpose of the historical right eigenvector;
[0084] S2-1-4: Generate some historical temperature field data according to the historical main characteristic vectors and the corresponding historical main characteristic vector coefficients;
[0085] The formula is:
[0086] F=a1U1+a2U2+...+a s U s +...+a r U r
[0087] Where, F is the historical temperature field data matrix composed of several historical temperature field data; a s 、a r are the main eigenvector coefficients of the sth and rth histories respectively; U s , U r are the sth and rth historical main eigenvectors respectively; s is the main eigenvector indicator; r is the historical main eigenorder;
[0088] S2-2: Based on some historical temperature field data, a residual network (ResNet) algorithm is used to build a temperature field analysis model and generate some historical temperature field analysis results;
[0089] ResNet is a deep neural network structure. By adding residual blocks, each residual block only needs to learn the difference (residual) between input and output instead of directly learning the output. In this way, even if the network is very deep, the input information can be directly transmitted to the output, which effectively alleviates the problem of gradient disappearance and improves the accuracy and efficiency of temperature field analysis.
[0090] S2-3: Based on several historical dimensionality-reduced monitoring data and the corresponding historical temperature field analysis results, a control parameter optimization model is constructed using the Deep Q Network (DQN) algorithm, including the following steps:
[0091] S2-3-1: Based on the control parameter optimization problem, define the simulation environment of the reinforcement learning algorithm, and build the intelligent agent and experience replay pool of the reinforcement learning algorithm;
[0092] S2-3-2: Define the state space of the DQN algorithm based on the historical monitoring data after dimensionality reduction, adjust the action based on the control parameters corresponding to the historical temperature field analysis results, and define the action space of the DQN algorithm;
[0093] S2-3-3: Based on the simulation environment, analyze the impact of each control parameter adjustment action in the action space on the historical dimensionality reduction monitoring data, and define the reward function of the DQN algorithm;
[0094] S2-3-4: Construct the input layer, several hidden layers, and output layer of the deep Q network, connect the state space to the input layer, and connect the action space and reward function to the output layer;
[0095] S2-3-5: Integrate the deep Q network, intelligent agent and experience replay pool to build the initial control parameter optimization model;
[0096] S2-3-6: Based on a number of historical dimensionality-reduced monitoring data and the corresponding historical temperature field analysis results, the initial control parameter optimization model is optimized and trained to obtain an optimized control parameter optimization model, and a number of historical control parameter optimization experiences are generated;
[0097] S2-3-7: storing several historical control parameter optimization experiences in the experience playback pool, and evaluating the optimized control parameter optimization model. If several evaluations meet the requirements, the final control parameter optimization model is output;
[0098] S3: Based on the control device, the real-time sensor data of the high and low temperature box and the real-time operation data of the corresponding dynamic internal heat source are collected to form real-time monitoring data, and uploaded to the cloud computing center;
[0099] S4: Based on the cloud computing center, according to the real-time monitoring data, the temperature field is reduced using the temperature field reduction model to obtain real-time temperature field data, including the following steps:
[0100] S4-1: According to the key monitoring indicator set, the real-time monitoring data is subjected to data dimension reduction to obtain real-time reduced-dimensional monitoring data;
[0101] S4-2: input the real-time dimensionality-reduced monitoring data into the temperature field order reduction model, perform temperature field order reduction, and obtain real-time temperature field data;
[0102] S5: Perform temperature field analysis based on real-time temperature field data using a temperature field analysis model to obtain real-time temperature field analysis results;
[0103] S6: Based on the real-time dimensionality reduction monitoring data and the real-time temperature field analysis results, the control parameter optimization model is used to optimize the control parameters, obtain the real-time control parameter optimization strategy, and return it to the corresponding control device, including the following steps:
[0104] S6-1: According to the real-time temperature field analysis results, a number of corresponding target historical control parameter optimization experiences are matched in the experience playback pool of the control parameter optimization model;
[0105] S6-2: Based on several target historical control parameter optimization experiences, the action space of the control parameter optimization model is updated to obtain the updated action space A' = [a'1, ..., a' j" ,...,a' I ], where a' j" is the updated j-th "action value", j" is the action indicator;
[0106] S6-3: Update the state space of the control parameter optimization model according to the real-time dimensionality reduction monitoring data, and obtain the updated state space S'=[s'1,...,s' i" ,...,s' I ], where s' i" is the updated i-th state value, i is the state indicator;
[0107] S6-4: Input the updated state space to the input layer of the deep Q network of the control parameter optimization model, and input the updated action space to the output layer of the deep Q network;
[0108] S6-5: Use the control parameters to optimize the model's agent, controlling the deep Q network to generate updated Q values for each possible action in the action space;
[0109] S6-6: Use the reward function to obtain the reward value of each possible action in the updated action space, and update the Q value of the possible action according to the reward value to obtain the updated Q value of the possible action;
[0110] The formula is:
[0111] Q(s' p' ,a' p' )=(1-α")·Q(s p' ,a p' )+α"·(R(s p' ,a p' ,s' p' )+γ·Q max (s p' ,a p' ))
[0112] In the formula, Q(s' p' ,a' p' ) is the updated state value s' p' and the updated action value a' p' The corresponding updated Q value; Q(s p' ,a p' ) is the state value s p' and action value a p' The corresponding predicted Q value; α" is the learning rate; Q max (s p' ,a p' ) is the highest predicted Q value;
[0113] S6-7: Repeat the above steps until the number of iterations reaches the iteration number threshold, and use the possible action corresponding to the highest updated Q value as the execution action for control parameter optimization;
[0114] S6-8: Generate a corresponding real-time control parameter optimization strategy according to the execution action, and return the real-time control parameter optimization strategy to the corresponding control device;
[0115] The real-time control parameter optimization strategy includes adjusting the control cycle parameters and control frequency parameters of the liquid nitrogen valve, adjusting the opening cycle parameters and working time parameters of the hot and cold air blowers, and adjusting the opening cycle parameters and working time parameters of the electric heater, etc.
[0116] S7: Based on the control device, according to the real-time control parameter optimization strategy, a real-time temperature control instruction is generated, and according to the real-time temperature control instruction, the dynamic internal heat source of the high and low temperature box is controlled.
[0117] The present invention discloses a temperature stability control system and method for a high and low temperature box with a dynamic internal heat source. CAE simulation technology is used to establish a CAE simulation model of the temperature field of the high and low temperature box, and the temperature distribution characteristics in the box are analyzed to provide training data support for subsequent temperature control, thereby avoiding the problem of low prediction accuracy caused by insufficient training data; a temperature field reduced-order model is used to generate corresponding real-time temperature field data according to real-time monitoring data, thereby reducing time cost investment, reducing hardware requirements and the calculation complexity of the control algorithm, and improving control efficiency; deep learning technology is used to learn the temperature field data, extract characteristic information of the temperature field, and realize real-time analysis and prediction of temperature changes, thereby reducing manpower cost investment and improving the accuracy and response speed of temperature control; a reinforcement learning algorithm is used to automatically and intelligently optimize the control parameters of the dynamic internal heat source, thereby optimizing the temperature stability of the high and low temperature box, realizing rapid response and precise control of temperature fluctuations, reducing time costs, and improving the reliability of test results; the system has high versatility and can be applied to the temperature stability control of different types of high and low temperature boxes, thereby having strong practical value.
[0118] The present invention is not limited to the above optional implementations, and anyone can derive other various forms of products under the enlightenment of the present invention. The above specific implementations should not be understood as limiting the scope of protection of the present invention. The scope of protection of the present invention should be based on the definition in the claims, and the description can be used to interpret the claims.
Claims
1. A temperature stability control system for a dynamic internal heat source high and low temperature box, characterized in that: It includes a cloud computing center and several control devices. The cloud computing center is communicated with the several control devices respectively. Each of the control devices is arranged in a high and low temperature box in a one-to-one correspondence, and the control device is electrically connected to the corresponding high and low temperature box and its dynamic internal heat source.
2. A temperature stability control system for a dynamic internal heat source high and low temperature box according to claim 1, characterized in that: The cloud computing center is provided with a CAE simulation unit, a model building unit, a temperature field reduction unit, a temperature field analysis unit and a control parameter optimization unit which are connected in sequence. The temperature field reduction unit and the control parameter optimization unit are respectively connected to a plurality of control devices for communication.
3. A temperature stability control system for a dynamic internal heat source high and low temperature box according to claim 2, characterized in that: The control device includes an industrial computer, a main control module and a sensor module. The industrial computer is communicatively connected to the cloud computing center, and is electrically connected to the sensor module, the main control module and the controller of the corresponding high and low temperature box respectively. The main control module is electrically connected to the actuator of the dynamic internal heat source of the corresponding high and low temperature box, and the sensor module is arranged at the corresponding high and low temperature box.
4. A temperature stability control system for a dynamic internal heat source high and low temperature box according to claim 3, characterized in that: The industrial computer includes a wireless communication module, a processor, a control instruction generation module and a data processing module. The processor is electrically connected to the wireless communication module, the control instruction generation module and the data processing module respectively. The wireless communication module is communicatively connected to the cloud computing center, the control instruction generation module is electrically connected to the main control module, and the data processing module is electrically connected to the sensor module and the corresponding high and low temperature box controller.
5. A temperature stability control system for a dynamic internal heat source high and low temperature box according to claim 3, characterized in that: The dynamic internal heat source of the high and low temperature box is provided with a liquid nitrogen storage tank, a liquid nitrogen valve, a hot and cold air blower and an electric heater. The control end of the liquid nitrogen valve, the control end of the hot and cold air blower and the control end of the electric heater are all electrically connected to the main control module; The sensor module includes a temperature sensor, a humidity sensor, a pressure sensor, a wind speed sensor and a flow sensor. The temperature sensor and humidity sensor are arranged inside the high and low temperature box, the pressure sensor is arranged inside the liquid nitrogen storage tank, and the flow sensor is arranged at the air outlet of the hot and cold air blower.
6. A temperature stability control method for a dynamic internal heat source high and low temperature box, based on the temperature stability control system according to any one of claims 1 to 5, characterized in that: The method comprises the following steps: Based on the cloud computing center, according to some historical monitoring data, the temperature field CAE simulation is carried out, the temperature field CAE simulation model is constructed, and some historical temperature field data are generated; Based on several historical monitoring data and corresponding historical temperature field data, a temperature field reduction model, a temperature field analysis model and a control parameter optimization model are constructed; Based on the control device, the real-time sensor data of the high and low temperature box and the real-time operation data of the corresponding dynamic internal heat source are collected to form real-time monitoring data, and uploaded to the cloud computing center; Based on the cloud computing center, according to the real-time monitoring data, the temperature field is reduced using the temperature field reduction model to obtain real-time temperature field data; According to the real-time temperature field data, the temperature field analysis model is used to perform temperature field analysis and obtain the real-time temperature field analysis results; According to the real-time monitoring data and the real-time temperature field analysis results, the control parameter optimization model is used to optimize the control parameters, obtain the real-time control parameter optimization strategy, and return it to the corresponding control device; Based on the control device, according to the real-time control parameter optimization strategy, a real-time temperature control instruction is generated, and according to the real-time temperature control instruction, the dynamic internal heat source of the high and low temperature box is controlled.
7. A temperature stability control method for a dynamic internal heat source high and low temperature box according to claim 6, characterized in that: Based on the cloud computing center, according to some historical monitoring data, the temperature field CAE simulation is performed, the temperature field CAE simulation model is constructed, and some historical temperature field data are generated, including the following steps: Based on the cloud computing center, some historical monitoring data are reduced in dimension to obtain a set of key monitoring indicators and some historical monitoring data after dimension reduction; The monitored data after dimension reduction is used as the model input of the temperature field CAE simulation, and the temperature field data is used as the model output of the temperature field CAE simulation; According to the model input and model output, use CAE simulation tools to perform temperature field CAE simulation and build a temperature field CAE simulation model; According to a number of historical monitoring data after dimensionality reduction, a temperature field CAE simulation model is used to perform temperature field CAE simulation to generate a number of historical temperature field data.
8. A temperature stability control method for a dynamic internal heat source high and low temperature box according to claim 6, characterized in that: According to several historical monitoring data and corresponding historical temperature field data, a temperature field reduction model, a temperature field analysis model and a control parameter optimization model are constructed, including the following steps: Based on several historical monitoring data and corresponding historical temperature field data, a temperature field reduced-order model is constructed using the singular value decomposition algorithm; Based on several historical temperature field data, a deep learning algorithm is used to build a temperature field analysis model and generate several historical temperature field analysis results; Based on several historical dimensionality-reduced monitoring data and the corresponding historical temperature field analysis results, a control parameter optimization model is constructed using the reinforcement learning algorithm.
9. A temperature stability control method for a dynamic internal heat source high and low temperature box according to claim 8, characterized in that: The temperature field analysis model is built based on the ResNet algorithm.
10. A temperature stability control method for a dynamic internal heat source high and low temperature box according to claim 8, characterized in that: The control parameter optimization model is constructed based on the DQN algorithm.
Citation Information
Patent Citations
Satellite component thermal layout temperature field prediction method based on semi-supervised learning
CN112733275A
Transformer digital twin cloud platform and implementation method thereof
CN113485156A
Terminal area meteorological scene recognition system
CN113989747A
Indoor space temperature and humidity regulation and control method and device
CN114020079A
Cloud data center heuristic task scheduling and energy consumption optimization method based on deep Q network
CN115686788A
