Temperature control method and device for heat flow test based on dynamic setting of PID parameters by machine learning

Abstract

The application discloses a heat flow test temperature control method and device based on machine learning dynamic PID parameter setting, belongs to the machine learning and heat flow test technical field, and the method comprises the steps of collecting multiple state parameters in a heat flow test process; inputting the collected state parameters into a pre-trained neural network model; outputting the dynamically set PID parameters according to the input state parameters by using the pre-trained neural network model; adjusting the output power of a heating device in real time according to the dynamically set PID parameters to realize temperature control; continuously correcting the neural network model through a temperature feedback signal to realize adaptive optimization of the PID parameters. The heat flow test temperature control method and device can effectively improve the accuracy and efficiency of temperature control in the heat flow test.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of machine learning and heat flow test, in particular to a heat flow test temperature control method and device based on machine learning dynamic setting PID parameters. BACKGROUND

[0002] The heat flow test is mainly used for simulating the thermal characteristics of materials, components and products in complex thermal environment, and evaluating the reliability and stability of the materials, components and products under high temperature thermal conditions. The heat flow test has important significance in the research and engineering application of the fields of aviation, aerospace and electronic equipment. In different stages of the heat flow test, such as the heating, steady state and cooling processes, environmental disturbances and material characteristic parameters (such as specific heat capacity and thermal conductivity) will all cause the fixed PID parameters to be difficult to adapt to dynamic adjustment, thereby affecting the temperature control accuracy. Whether the method of setting PID parameters can be differentiated according to the stages of the heat flow test and the material characteristics to realize dynamic adjustment of the PID parameters to adapt to various working conditions to complete high-precision temperature control determines its application value. However, the traditional PID setting method does not have dynamic setting, which leads to poor temperature control effect. SUMMARY

[0003] The present application provides a heat flow test temperature control method and device based on machine learning dynamic setting PID parameters, to solve the technical problems that the traditional PID setting method does not have dynamic setting, the existing method does not fully consider the differentiated processing of the stages of the heat flow test and the material characteristics, and leads to poor temperature control effect.

[0004] To solve the above technical problems, the present application provides the following technical solutions:

[0005] On the one hand, the present application provides a heat flow test temperature control method based on machine learning dynamic setting PID parameters, which comprises:

[0006] Collecting a plurality of state parameters in the heat flow test process;

[0007] Inputting the collected state parameters into a pre-trained neural network model, and using the pre-trained neural network model to output the dynamically set PID parameters according to the input state parameters;

[0008] Adjusting the output power of the heating device in real time according to the dynamically set PID parameters to realize temperature control;

[0009] Continuously correcting the neural network model through the temperature feedback signal to realize self-adaptive optimization of the PID parameters.

[0010] Further, the state parameters include: current temperature, target temperature, temperature change gradient, material specific heat capacity, material thermal conductivity and current test stage in the heat flow test process; wherein, the test stage is divided into three different stages of heating stage, steady state stage and cooling stage, and is represented by one-hot encoding.

[0011] Further, the neural network model is a mixed architecture multi-branch dynamic neural network, which includes: an input layer, a material encoding module, a stage perception module, a heating branch, a steady state branch, a cooling branch, a feature fusion module and an output layer.

[0012] The process of obtaining the dynamically set PID parameters by using the multi-branch dynamic neural network includes:

[0013] The collected state parameters enter the multi-branch dynamic neural network through the input layer;

[0014] The material encoding module processes the material specific heat capacity and the material thermal conductivity through the fully connected layer to generate a material feature vector; the stage perception module concatenates the one-hot encoding of the test stage with the material feature vector and sends it to the gate unit; the gate unit calculates a three-dimensional probability distribution through a preset weight matrix and bias, which is used to represent the activation weight of the heating branch, the steady state branch and the cooling branch; the branch with the maximum probability is taken as the main path, and the outputs of other branches are set to zero; wherein, the heating branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send it to the first hidden layer, which is processed by the fully connected layer and activated by LeakyReLU, and then sent to the second hidden layer, which is processed by mapping and activated by LeakyReLU to obtain the output of the heating branch; the steady state branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send it to the sparse self-attention layer; Query, Key and Value are calculated by using the sparse self-attention layer, and only the first three attention weights are reserved during self-attention calculation, and the remaining weights are set to zero to obtain attention features; then, the attention features are mapped by the fully connected layer and activated by ReLU to obtain the output of the steady state branch; the cooling branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send it to the one-dimensional convolution layer to extract the time sequence features of temperature change, and further reduce the dimension by the max-pooling layer, and finally process by the fully connected layer and activated by Tanh to obtain the output of the cooling branch.

[0015] The feature fusion module fuses the output of the main path with the material feature vector to obtain a fusion feature vector; in the output layer, the fusion feature vector is mapped to a three-dimensional output by the fully connected layer, and the three-dimensional output directly corresponds to the parameters of the PID controller in the temperature control system, including Kp, Ki and Kd.

[0016] Further, the training process of the neural network model comprises:

[0017] Collecting state parameters in the historical heat flow test process, and performing outlier filtering and smoothing processing on the collected state parameters to obtain effective state parameters, and constructing a historical test data set by using the effective state parameters;

[0018] Using the historical test data set, the network weights of the neural network model are optimized through supervised learning, so that the neural network model can dynamically adjust the PID parameters under different heat flow conditions.

[0019] Further, the PID parameter dynamic adjustment strategy is optimized according to the material characteristics, for materials with a thermal conductivity lower than a preset thermal conductivity threshold, the integral parameter is increased to enhance the stability of temperature adjustment; for materials with thermal inertia greater than a preset thermal inertia threshold, the differential parameter is adjusted to reduce the overshoot phenomenon.

[0020] On the other hand, the present application also provides a heat flow test temperature control device based on machine learning dynamic setting of PID parameters, which comprises:

[0021] A data acquisition module for acquiring a plurality of state parameters in the heat flow test process;

[0022] A neural network calculation module for inputting the collected state parameters into a pre-trained neural network model, and outputting the dynamically set PID parameters according to the input state parameters by using the pre-trained neural network model;

[0023] A temperature control execution module for adjusting the output power of the heating device in real time according to the dynamically set PID parameters to realize temperature control;

[0024] A feedback correction module for continuously correcting the neural network model through the temperature feedback signal to realize adaptive optimization of the PID parameters.

[0025] Further, the state parameters include: current temperature, target temperature, temperature change gradient, material specific heat capacity, material thermal conductivity and current test stage in the heat flow test process; wherein the test stage is divided into three different stages of heating stage, steady state stage and cooling stage, and is represented by one-hot encoding.

[0026] Further, the neural network model is a mixed architecture multi-branch dynamic neural network, which comprises: an input layer, a material encoding module, a stage perception module, a heating branch, a steady state branch, a cooling branch, a feature fusion module and an output layer.

[0027] The process of obtaining the dynamically set PID parameters by using the multi-branch dynamic neural network comprises:

[0028] The collected state parameters enter the multi-branch dynamic neural network through the input layer;

[0029] The material coding module processes the material specific heat capacity and the material thermal conductivity through the fully connected layer to generate a material feature vector; the stage perception module concatenates the one-hot encoding of the test stage with the material feature vector and sends the result to the gating unit; the gating unit calculates a three-dimensional probability distribution through a preset weight matrix and bias, which is used to represent the activation weights of the warming branch, the steady-state branch and the cooling branch; the branch with the maximum probability is taken as the main path, and the outputs of other branches are set to zero; wherein, the warming branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send the result to the first hidden layer, which is processed through the fully connected layer and activated by LeakyReLU, and then sent to the second hidden layer, which is processed through mapping and activated by LeakyReLU to obtain the output of the warming branch; the steady-state branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send the result to the sparse self-attention layer; Query, Key and Value are calculated using the sparse self-attention layer, and only the first three attention weights are retained during self-attention calculation, and the remaining weights are set to zero to obtain attention features; then, the attention features are mapped through the fully connected layer and activated by ReLU to obtain the output of the steady-state branch; the cooling branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send the result to the one-dimensional convolution layer to extract the time sequence features of temperature change, and further reduce the dimension through the max-pooling layer, and finally process through the fully connected layer and activate by Tanh to obtain the output of the cooling branch;

[0030] The feature fusion module fuses the output of the main path with the material feature vector to obtain a fusion feature vector; in the output layer, the fusion feature vector is mapped to a three-dimensional output through the fully connected layer, and the three-dimensional output directly corresponds to the parameters of the PID controller in the temperature control system, including Kp, Ki and Kd.

[0031] Further, the training process of the neural network model comprises:

[0032] Collect the state parameters during the historical heat flow test process, and perform outlier filtering and smoothing processing on the collected state parameters to obtain effective state parameters, and use the effective state parameters to construct a historical test data set;

[0033] Using the historical test data set, the network weights of the neural network model are optimized through supervised learning, so that the neural network model can dynamically adjust the PID parameters under different heat flow conditions.

[0034] Further, the PID parameter dynamic adjustment strategy is optimized according to the material characteristics, for the material with a thermal conductivity lower than a preset thermal conductivity threshold, the integral parameter is increased to enhance the stability of temperature adjustment; for the material with a thermal inertia greater than a preset thermal inertia threshold, the differential parameter is adjusted to reduce the overshoot phenomenon.

[0035] In another aspect, the present application also provides an electronic device, comprising a processor and a memory; wherein the memory stores at least one instruction, which is loaded and executed by the processor to implement the above method.

[0036] In another aspect, the present application also provides a computer readable storage medium, which stores at least one instruction, which is loaded and executed by the processor to implement the above method.

[0037] The technical scheme provided by the present application has at least the following beneficial effects:

[0038] The present application adopts a machine learning method to realize dynamic adjustment of PID parameters through a neural network, avoids manual intervention, and improves temperature control accuracy and stability; through differential processing in the heat flow test stage, the temperature control strategy can be automatically adjusted in the test process according to different stages, and the material thermal characteristics (such as specific heat capacity) are also used as one of the parameters, so that the temperature control is more targeted, and the overshoot and steady-state error are reduced; temperature error feedback closed loop optimization is adopted to realize adaptive learning of the temperature control system and enhance the ability to adapt to different test conditions; a hardware and software combined architecture is adopted and modular design is adopted, which has good expansibility and integration, and can adapt to different application requirements. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0040] Figure 1 is an execution flow schematic diagram of the heat flow test temperature control method based on machine learning dynamic setting of PID parameters provided by the embodiment of the present application;

[0041] Figure 2 is a structure schematic diagram of the mixed architecture multi-branch dynamic neural network provided by the embodiment of the present application;

[0042] Figure 3 is a principle schematic diagram based on machine learning dynamic setting of PID parameters provided by the embodiment of the present application;

[0043] Figure 4 is a structural schematic diagram of a heat flow test temperature control device based on machine learning dynamic setting of PID parameters provided by an embodiment of the present application.

[0044] Figure 5 is a system block diagram of an electronic device. DETAILED DESCRIPTION

[0045] To make the objects, technical solutions and advantages of the present application clearer, the following will further describe the embodiments of the present application in combination with the drawings.

[0046] First of all, it should be noted that in the embodiments of the present application, the words such as "exemplarily", "for example" and the like are used to represent as an example, illustration or explanation. Any embodiment or design scheme described as "exemplary" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "exemplarily" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be either one of the two.

[0047] First embodiment

[0048] In view of the problems that the existing PID parameter dynamic setting method is difficult to adapt to the differentiated requirements in the heat flow test stage and the influence of the characteristics of different test materials on temperature control stability, the present embodiment fully considers key factors such as material thermal characteristics, temperature change trend and test stage, and integrates them into machine learning to solve the influence of nonlinear factors in the test and improve the temperature control effect. Based on this, the present embodiment combines the advantages of machine learning in dynamic setting of PID parameters, the differentiated processing of the heat flow test stage and the characteristics of the test materials, and adopts a software and hardware combined architecture to provide a heat flow test temperature control method based on machine learning dynamic setting of PID parameters. The method can be implemented by an electronic device, and its execution process is as shown in Figure 1 , including the following steps:

[0049] S1, collecting a plurality of state parameters in the heat flow test process;

[0050] It should be noted that in the present embodiment, a temperature control instrument and a thermocouple are used to collect a plurality of state parameters in the heat flow test process, including: current temperature, target temperature, temperature change gradient, specific heat capacity of the material, thermal conductivity of the material and current test stage information in the heat flow test process.

[0051] S2, inputting the collected state parameters into a pre-trained neural network model, and using the pre-trained neural network model to output the dynamically set PID parameters according to the input state parameters;

[0052] Wherein, the neural network adopted in the embodiment is a mixed-architecture multi-branch dynamic neural network, that is, the embodiment utilizes the mixed-architecture multi-branch dynamic neural network to construct a PID dynamic tuning model, and trains the neural network through historical test data. The trained neural network can perform online calculation and output PID parameters in the test process, so as to perform PID parameter prediction and ensure that the temperature control system has good adaptability to external interference. In the neural network training stage, multiple sets of heat flow test data are used, and the input parameters include the current test temperature, the test target temperature, the temperature change gradient, the specific heat capacity of the test material, and the current test stage and other factors. Through supervised learning, the network weight is optimized, so that it can dynamically adjust the PID parameters under different heat flow conditions. The dynamically tuned PID parameters are parameters that can meet the preset temperature control target under the current test conditions.

[0053] The mixed-architecture multi-branch dynamic neural network is used for dynamic tuning of PID parameters in a heat flow test temperature control system, and can adaptively select different sub-network branches according to temperature changes, material properties and test stage information in the heat flow test process, so as to achieve the purposes of fast response, efficient calculation and anti-interference control. It should be noted that the parameters and modules in the embodiment can be adjusted according to specific application requirements.

[0054] Specifically, in the embodiment, as shown in Figure 2 The mixed-architecture multi-branch dynamic neural network includes an input layer, a material encoding module, a stage perception module, a temperature rising branch, a steady state branch, a temperature falling branch, a feature fusion module and an output layer. The process of obtaining the dynamically tuned PID parameters by using the network is as follows:

[0055] The system first collects temperature-related parameters including the current temperature, the target temperature and the temperature change gradient from the heat flow test, collects the specific heat capacity and the thermal conductivity which describe the material properties, and adopts one-hot encoding representation according to the test stage (for example, [1, 0, 0] represents the temperature rising stage, [0, 1, 0] represents the steady state, and [0, 0, 1] represents the temperature falling stage), to constitute a total of six-dimensional input parameters. The six-dimensional input is the initial input of the entire network, which enters the subsequent processing module through the input layer.

[0056] Specifically, the embodiment designs and adopts a material characteristic coding module to map the original two-dimensional material parameters of specific heat capacity and thermal conductivity to a 32-dimensional high-dimensional feature space through a fully connected layer, and the fully connected layer adopts a ReLU activation function to enhance the sensitivity of the model to material characteristics, thereby generating a 32-dimensional material feature vector H_material. At the same time, the system is provided with a stage perception module, which splices the one-hot encoding of the test stage with the aforementioned 32-dimensional material features into a 35-dimensional input vector and sends it to a gating unit. The gating unit calculates a 3-dimensional probability distribution through a preset weight matrix and bias to represent the activation weights of the three sub-network branches of the temperature rise, steady state and temperature drop. In this process, by comparing the sizes of the probabilities of each branch, the branch with the maximum probability is taken as the main path, and the outputs of the other branches are set to zero to reduce redundant calculation and ensure the on-demand allocation of computing resources.

[0057] In addition, the embodiment designs three sub-network branches for different test stages. For the temperature rise stage, the application provides a high-speed response branch, which mainly aims to quickly raise the temperature and preferentially optimizes the proportional term (Kp) and the differential term (Kd) to speed up the response. The branch splices the temperature-related parameters (3-dimensional) and the 32-dimensional material features into a 35-dimensional vector, which is further processed through a first hidden layer using a fully connected layer (mapped from 35-dimensional to 64-dimensional) and activated using LeakyReLU, and then further processed through a second hidden layer (64-dimensional mapped to 32-dimensional), also activated using a LeakyReLU activation function, and finally outputs a 32-dimensional feature vector H_rise as the output of the branch. For the steady state stage, the application sets up a sparse attention branch, which aims to focus only on key temperature node information while keeping a low computational overhead. The branch also splices the temperature parameters and material features into a 35-dimensional vector, and then uses a sparse self-attention layer to calculate Query, Key and Value (each dimension is set to 16). During self-attention calculation, only the top-3 attention weights are retained, and the rest are set to zero, thereby significantly reducing the computational load. Next, the 16-dimensional attention features are mapped to 32-dimensional through a fully connected layer, and a ReLU activation function is used to output 32-dimensional attention features H_steady. For the temperature drop stage, to solve the problem of temperature fluctuations and hysteresis, the application provides an anti-interference branch. The branch also splices the temperature parameters and material features into a 35-dimensional vector, and then extracts the time sequence features of temperature changes through a one-dimensional convolution layer (convolution window size is 5, and the number of output channels is set to 16), and further reduces the dimension through a max-pooling layer (pooling size is 2), and finally outputs a 32-dimensional feature vector H_cool as the result of the branch through a fully connected layer (mapping from 16-dimensional to 32-dimensional) and a Tanh activation function.

[0058] Further, after each branch is processed independently, the system fuses the activated branch output with the original 32-dimensional material feature vector to form a 64-dimensional fused feature vector H_fused. Finally, the fused feature is mapped to a 3-dimensional output through a fully connected layer, which directly corresponds to the parameters [Kp, Ki, Kd] of the PID controller in the temperature control system.

[0059] Further, the embodiment provides a loss function Loss suitable for the neural network to optimize the adaptive adjustment of PID parameters in the temperature control process, improve the temperature tracking accuracy, and ensure the stability of the adjustment process. The loss function is composed of a temperature error loss term L temperature , a material adaptive loss term L material , and a PID parameter stability loss term L stab and the experimental phase factor , and the specific definitions are as follows:

[0060]

[0061] Among them, the temperature error loss term L temperature adopts mean square error (MSE) to measure the deviation between the current temperature Tcurrent and the target temperature Ttarget:

[0062]

[0063] Among them, N represents the total number of time steps; T target (i) represents the target temperature of the i-th time step; T current (i) represents the current actual temperature of the i-th time step.

[0064] This loss term ensures that the neural network can accurately track the set temperature, reduce the steady-state error, and improve the temperature control accuracy.

[0065] The material adaptive loss term L material sets an adaptive adjustment mechanism according to the threshold of the material thermal conductivity λ, to ensure that the temperature change rate adapts to different material characteristics, to optimize the heating and cooling rate. The definition is as follows:

[0066]

[0067] Among them, α and β are material characteristic adjustment coefficients (which need to be experimentally calibrated). λ threshold represents the preset thermal conductivity threshold, used to distinguish between low thermal conductivity and high thermal conductivity materials; T represents the actual temperature at the current time step; t represents the current time step; dT / dt represents the actual temperature change rate at the current time step t.

[0068] The introduction of this loss term can effectively adapt to different material characteristics.

[0069] PID parameter stability loss term L stab By constraining the time variation of PID parameters, the influence of parameter drastic fluctuation on temperature control stability is avoided:

[0070] L stab = (K P (t) - K P (t-1)) 2 + (K I (t) - K I (t-1)) 2 + (K D (t) - K D (t-1)) 2

[0071] Where K P (t) represents the PID control parameter proportional term at the current time step t; K P (t-1) represents the PID control parameter proportional term at the previous time step t-1; K I (t) represents the PID control parameter integral term at the current time step t; K I (t-1) represents the PID control parameter integral term at the previous time step t-1; K D (t) represents the PID control parameter derivative term at the current time step t; K D (t-1) represents the PID control parameter derivative term at the previous time step t-1.

[0072] At the same time, in order to adapt to different control stages, the loss function introduces a stage factor to dynamically adjust the stability weight:

[0073]

[0074] This test stage factor can be adjusted adaptively according to preset rules or historical data to ensure optimal temperature control effect in different experimental stages.

[0075] The above loss function can make the neural network accurately converge through multi-objective optimization strategy, so as to optimize PID parameters under different material conditions and control stages, improve temperature control precision, reduce overshoot, and enhance system stability, which is suitable for intelligent adjustment tasks in complex heat flow environment.

[0076] S3, according to the dynamically set PID parameters, real-time adjust the output power of the heating device to realize temperature control;

[0077] S4, through the temperature feedback signal, continuously correct the neural network model to realize adaptive optimization of PID parameters.

[0078] It should be noted that after obtaining the dynamically tuned PID parameters, the embodiment sets the PID controller according to the dynamically tuned PID parameters to control the output percentage of the temperature controller device for heating, while feeding back the temperature error to correct the neural network, thereby completing the closed-loop control of the temperature.

[0079] Specifically, as shown in the figure, Figure 3 The embodiment adopts a temperature controller device and a thermocouple to obtain real-time temperature data of the controlled object in the heat flow test process, establishes a temperature control target, filters and smooths the effective temperature data set by using a data preprocessing module, inputs the effective temperature data and system historical response data into the designed PID parameter tuning model based on the neural network, and adjusts the PID control parameters in real time to obtain the dynamically optimal PID parameters. The system historical response data mainly refers to various data reflecting the dynamic characteristics of the temperature control process of the heat flow test, such as temperature error, error change rate, error integral, overshoot and test stage switching timestamp (such as the critical point of heating to steady state). The system historical response data mainly refers to various data reflecting the dynamic characteristics of the temperature control process of the heat flow test, such as temperature error, error change rate, error integral, overshoot and test stage switching timestamp (such as the critical point of heating to steady state). Real-time PID temperature control is performed by using a combination of software and hardware, and data monitoring and remote adjustment are performed by using the host computer software. According to the simulation training, the weight of the neural network model is adjusted by using the learning algorithm adjustment strategy, so that the system can maintain high-precision temperature control effect under different test conditions. Further, through the test feedback correction mechanism, the weight of the neural network model is adaptively adjusted, so that the temperature control effect is continuously optimized with the progress of the test, and the temperature control precision and system stability are improved.

[0080] Further, the heat flow test temperature control method of the embodiment optimizes the neural network training data set in combination with the test material characteristics, generates a unique identifier (such as a hash code) for each test material, and stores it together with the temperature data and the PID parameters to form a structured data set, so that the PID parameters can be adaptively adjusted, thereby improving the temperature control stability and precision.

[0081] Further, in the embodiment, the temperature control adopts a closed-loop control mode, and different PID parameter adjustment strategies are set according to different test stages (heating, steady state, and cooling). In addition, it should be noted that in the embodiment, the PID parameter dynamic adjustment strategy is optimized according to the test material characteristics, such as appropriately increasing the integral parameter for materials with low thermal conductivity to enhance the stability of temperature adjustment, and appropriately adjusting the differential parameter for materials with large thermal inertia to reduce overshoot.

[0082] In summary, the embodiment provides a heat flow test temperature control method based on dynamic setting of PID parameters by machine learning. The method uses machine learning to realize dynamic adjustment of PID parameters through a neural network, avoids manual intervention, and improves temperature control accuracy and stability. Through differentiated processing during the heat flow test phase, the method can automatically adjust the temperature control strategy during the test process according to different phases, and also uses material thermal properties (such as specific heat capacity) as one of the parameters, making the temperature control more targeted and reducing overshoot and steady-state error. The method uses temperature error feedback closed-loop optimization to realize adaptive learning of the temperature control system and enhance the ability to adapt to different test conditions. The method uses a combination of software and hardware architecture and modular design, has good scalability and integration, and can adapt to different application requirements.

[0083] Second embodiment

[0084] The embodiment provides a heat flow test temperature control device based on dynamic setting of PID parameters by machine learning. The system structure of the heat flow test temperature control device is shown in Figure 4 The device is composed of a data acquisition module, a neural network calculation module, a temperature control execution module, and a feedback correction module. Through the information processing capabilities of different modules, the temperature in the heat flow test is controlled with high precision. The functions of each module are as follows:

[0085] The data acquisition module is used to acquire multiple state parameters during the heat flow test.

[0086] Specifically, in this embodiment, the data acquisition module includes a temperature control instrument device, a thermocouple, and a communication interface, which can realize real-time monitoring and recording of temperature and heat flow test conditions.

[0087] The neural network calculation module is used to input the acquired state parameters into a pre-trained neural network model, and use the pre-trained neural network model to output dynamically set PID parameters according to the input state parameters.

[0088] The temperature control execution module is used to adjust the temperature control instrument device in real time according to the dynamically set PID parameters to drive the temperature control instrument device to control the heating device, thereby realizing precise temperature control.

[0089] Specifically, in this embodiment, the temperature control execution module includes a PID controller, a temperature control instrument device, and a power adjustment unit, which is used to adjust the power output of the heating device to realize temperature adjustment.

[0090] The feedback correction module is used to receive real-time temperature feedback signals. Based on temperature error, adjustment time, overshoot, and other parameters, the neural network model is continuously corrected through the temperature feedback signals to realize adaptive optimization of the PID parameters, so that the test environment reaches the expected temperature and improves the temperature control accuracy and system stability.

[0091] In addition, it should be noted that, for the sake of convenience, Figure 4 Only the main components of the device are shown. Moreover, the heat flow test temperature control device based on machine learning dynamic setting of PID parameters of the present embodiment corresponds to the heat flow test temperature control method based on machine learning dynamic setting of PID parameters of the first embodiment described above; wherein the functions realized by each functional module in the heat flow test temperature control device based on machine learning dynamic setting of PID parameters of the present embodiment correspond one by one to each process step in the heat flow test temperature control method based on machine learning dynamic setting of PID parameters of the first embodiment described above; therefore, it will not be described here.

[0092] In summary, the present embodiment provides a heat flow test temperature control device based on machine learning dynamic setting of PID parameters, which adopts a neural network model to adjust PID parameters in real time according to current temperature, target temperature, PID parameters, temperature change gradient, specific heat capacity of thermal conductive material and current test stage information, to achieve dynamic optimal control; combined with a multi-stage control strategy, differential PID parameter adjustment is performed in different test stages such as heating, steady state and cooling, to improve temperature control precision; through a system architecture combining software and hardware, the temperature control device composed of algorithms, software and temperature control instrument hardware uses the temperature control instrument to execute PID temperature control, and combines the host computer software to perform data monitoring and remote control; the feedback correction mechanism is used to optimize the weight of the neural network model, effectively improving the temperature control stability and adaptive ability.

[0093] Third embodiment

[0094] The present embodiment provides an electronic device, such as Figure 5 As shown in the figure, the electronic device comprises a processor and a memory; wherein the processor and the memory can be connected through a communication bus; the memory stores at least one instruction, which is loaded and executed by the processor to implement the method of the first embodiment described above. In addition, the electronic device can also include a transceiver, and the processor and the transceiver can be connected through a communication bus, and the transceiver is used to communicate with other devices.

[0095] Next, the electronic device will be described in detail in combination with Figure 5 The various constituent components of the electronic device will be described in detail:

[0096] The processor is the control center of the electronic device. The electronic device can include multiple processors. Each of the processors can be a single-CPU or a multi-CPU. The processor can be one processor or a collective term of multiple processing elements. For example, the processor can be one or more central processing units (CPUs), other general purpose processors, application specific integrated circuits (ASICs), or one or more integrated circuits configured to implement one or more embodiments of the present application, such as one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. The general purpose processor can be a microprocessor or any conventional processor, or the like. The processor can perform various functions of the electronic device by running or executing software programs stored in the memory and calling data stored in the memory.

[0097] In a specific implementation, as an embodiment, the processor can include one or more CPUs, such as CPU0 and CPU1 shown in FIG. 1, of course, this is only an exemplary description. Figure 5

[0098] The memory is used to store software programs for implementing the solution of the present application, and is controlled by the processor to perform the implementation. The specific implementation can refer to the above-mentioned method embodiments, which will not be described here.

[0099] ​Optionally, the memory can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, a magnetic disk storage or other magnetic storage devices, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory can be integrated with the processor or exist independently and be coupled to the processor through the interface circuit (not shown in the figure) of the electronic device, and the embodiments of the present application do not make specific limitations hereon. Figure 5

[0100] The transceiver can include a receiver and a transmitter (not shown separately in the figure). The receiver is configured to implement the receiving function, and the transmitter is configured to implement the transmitting function. The transceiver can be integrated with the processor or exist independently and be coupled to the processor through the interface circuit (not shown in the figure) of the electronic device, and the embodiments of the present application do not make specific limitations hereon. Figure 5 Figure 5

[0101] In addition, it should be noted that the structure of the electronic device shown in the figure does not constitute a limitation on the device, and the actual device can include more or fewer components than shown in the figure, or combine certain components, or different component arrangements. In addition, the technical effects achieved by the electronic device when executing the method of the first embodiment can refer to the technical effects described in the first embodiment, and therefore, will not be described here. Figure 5

[0102] Fourth embodiment

[0103] The present embodiment provides a computer readable storage medium, the storage medium stores at least one instruction, the instruction is loaded and executed by the processor to realize the method of the first embodiment. Wherein, the computer readable storage medium can be ROM, random access memory, CD-ROM, magnetic tape, floppy disk and optical data storage device, etc. The instructions stored therein can be loaded and executed by the processor in the terminal to execute the above method.

[0104] ​​​​Moreover, it should be noted that the present application can be provided as a method, an apparatus, or a computer program product. Therefore, the embodiments of the present application can take the form of an entirely or partially hardware embodiment, an entirely or partially software embodiment, or an embodiment combining software and hardware aspects. Furthermore, when implemented in software, the embodiments of the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, a computer diskette, an optical storage medium, a magnetic storage medium, and a semiconductor memory device). The computer program product includes one or more computer instructions that when loaded and executed by a computer, cause the computer to carry out the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, such as from a website, a computer, a server, or a data center to another website, computer, server, or data center through a wired (for example, infrared, wireless, microwave, or the like) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, data center, or the like, including one or more collections of available media. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0105] The embodiments of the present application are described with reference to flowcharts and / or block diagrams of the method, terminal device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate a device that implements the flowcharts and / or block diagrams. Figure 1 The flowcharts and / or block diagrams Figure 1 The device that implements the functions specified in one or more flows and / or blocks.

[0106] These computer program instructions can also be stored in a computer-readable storage medium that can direct the computer or other programmable data processing terminal device to work in a specific manner, so that the instructions stored in the computer-readable storage medium produce a product including instruction devices that implement the flowcharts and / or block diagrams. Figure 1 The flowcharts and / or block diagrams Figure 1the functions specified in the individual block or blocks. Such computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable devices to generate a computer-implemented process, thus the instructions executed on the computer or other programmable devices provide a process for implementing the functions specified in the flowchart block(s) or block(s). Figure 1 the functions specified in the individual block or blocks. Such computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable devices to generate a computer-implemented process, thus the instructions executed on the computer or other programmable devices provide a process for implementing the functions specified in the flowchart block(s) or block(s). Figure 1 Figure 1 the functions specified in the individual block or blocks. Such computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable devices to generate a computer-implemented process, thus the instructions executed on the computer or other programmable devices provide a process for implementing the functions specified in the flowchart block(s) or block(s).

[0107] It should also be noted that, in the present document, the terms such as first and second, etc. are merely used to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or terminal device. Without more limitations, the element defined by the statement "including a…", does not exclude the presence of other identical elements in the process, method, article or terminal device including the element. In addition, the term "and / or" is merely a description of the association relationship between the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the existence of A alone, the existence of A and B together, and the existence of B alone, where A and B can be singular or plural. In addition, the character " / " in the present document generally represents an "or" relationship between the preceding and following associated objects, but it can also represent an "and / or" relationship, which can be understood in the context before and after. "At least one" means one or more, and "multiple" means two or more. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can represent a, b, c, a-b, a-c, b-c or a-b-c, where a, b and c can be single or multiple.

[0108] In addition, it can be understood that in various embodiments of the present application, the size of the sequence number of the above processes does not mean the order of execution, and the execution order of the processes should be determined by their functions and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0109] Those skilled in the art can appreciate that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware or in a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solutions. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0110] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely schematic, for example, the division of functional modules / units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another device, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms. The units described as separate components can be or can not be physically separated, and the components displayed as units can be or can not be physical units, that is, can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment. In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present, or two or more units can be integrated in one unit.

[0111] If the method is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0112] Finally, it should be noted that the above description is only the preferred embodiment of the application, it should be pointed out that although the preferred embodiment of the application has been described, for those skilled in the art, once the basic creative concept of the application is known, several improvements and refinements can be made without departing from the principles of the application, and these improvements and refinements should also be considered as the protection scope of the application. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the application.

Claims

1. A heat flow test temperature control method based on dynamically setting PID parameters by machine learning, characterized in that, The heat flow test temperature control method based on machine learning dynamic PID parameter setting comprises the following steps: Collecting multiple state parameters in the heat flow test process; Inputting the collected state parameters into a pre-trained neural network model, and outputting the dynamically set PID parameters according to the input state parameters by using the pre-trained neural network model; Adjusting the output power of the heating device in real time according to the dynamically set PID parameters to realize temperature control; Continuously correcting the neural network model through temperature feedback signals to realize adaptive optimization of the PID parameters; The state parameters include the current temperature, target temperature, temperature change gradient, material specific heat capacity, material thermal conductivity and current test stage in the heat flow test process; wherein the test stage is divided into three different stages of heating stage, steady state stage and cooling stage, and is represented by one-hot encoding; The neural network model is a mixed architecture multi-branch dynamic neural network, which comprises an input layer, a material encoding module, a stage perception module, a heating branch, a steady state branch, a cooling branch, a feature fusion module and an output layer; The process of obtaining the dynamically set PID parameters by using the multi-branch dynamic neural network comprises the following steps: The collected state parameters enter the multi-branch dynamic neural network through the input layer; The material encoding module processes the material specific heat capacity and the material thermal conductivity through a fully connected layer to generate a material feature vector; the stage perception module concatenates the one-hot encoding of the test stage with the material feature vector and sends it to a gate unit; the gate unit calculates a three-dimensional probability distribution by using a preset weight matrix and bias to represent the activation weights of the heating branch, the steady state branch and the cooling branch; the branch with the maximum probability is taken as the main path, and the outputs of other branches are set to zero; wherein the heating branch is used to concatenate the current temperature, target temperature and temperature change gradient with the material feature vector, and then send them to a first hidden layer; after being processed by a fully connected layer in the first hidden layer and activated by LeakyReLU, the first hidden layer sends the output to a second hidden layer; after being processed by a mapping in the second hidden layer and activated by LeakyReLU, the second hidden layer obtains the output of the heating branch; the steady state branch is used to concatenate the current temperature, target temperature and temperature change gradient with the material feature vector, and then send them to a sparse self-attention layer; Query, Key and Value are calculated by using the sparse self-attention layer; only the first three attention weights are reserved during self-attention calculation, and the remaining weights are set to zero to obtain attention features; then, the attention features are mapped by a fully connected layer and activated by ReLU to obtain the output of the steady state branch; the cooling branch is used to concatenate the current temperature, target temperature and temperature change gradient with the material feature vector, and then send them to a one-dimensional convolution layer to extract the time sequence features of temperature change, and further reduce the dimension by using a max pooling layer; finally, the output of the cooling branch is obtained by processing through a fully connected layer and activating by Tanh; The feature fusion module fuses the output of the main path with the material feature vector to obtain a fused feature vector; at the output layer, the fused feature vector is mapped to a three-dimensional output through a fully connected layer, and the three-dimensional output directly corresponds to parameters of a PID controller in the temperature control system, including Kp, Ki and Kd.

2. The heat run test temperature control method based on machine learning dynamic setting of PID parameters according to claim 1, characterized in that, The training process of the neural network model comprises: Collecting state parameters in a historical heat flow test process, and performing outlier filtering and smoothing processing on the collected state parameters to obtain effective state parameters, and constructing a historical test data set by using the effective state parameters; Using the historical test data set, the network weights of the neural network model are optimized through supervised learning, so that the neural network model can dynamically adjust the PID parameters under different heat flow conditions.

3. The heat run test temperature control method based on machine learning dynamic setting of PID parameters according to claim 2, characterized in that, The PID parameter dynamic adjustment strategy is optimized according to the material characteristics. For materials with a thermal conductivity lower than a preset thermal conductivity threshold, the integral parameter is increased to enhance the stability of temperature adjustment; for materials with thermal inertia greater than a preset thermal inertia threshold, the differential parameter is adjusted to reduce the overshoot phenomenon.

4. A heat flow test temperature control device based on dynamically setting PID parameters by machine learning, characterized in that, The heat flow test temperature control device based on machine learning dynamic setting of PID parameters comprises: A data acquisition module for acquiring a plurality of state parameters in a heat flow test process; A neural network calculation module for inputting the collected state parameters into a pre-trained neural network model, and outputting dynamically set PID parameters according to the input state parameters by using the pre-trained neural network model; A temperature control execution module for adjusting the output power of the heating device in real time according to the dynamically set PID parameters to realize temperature control; A feedback correction module for continuously correcting the neural network model through a temperature feedback signal to realize adaptive optimization of the PID parameters; The state parameters include: current temperature, target temperature, temperature change gradient, specific heat capacity of the material, thermal conductivity of the material and current test stage in the heat flow test process; wherein the test stage is divided into three different stages of heating stage, steady state stage and cooling stage, and is represented by one-hot encoding; The neural network model is a multi-branch dynamic neural network with a hybrid architecture, which comprises: an input layer, a material coding module, a stage perception module, a heating branch, a steady state branch, a cooling branch, a feature fusion module and an output layer; The process of obtaining the dynamically set PID parameters by using the multi-branch dynamic neural network comprises: The collected state parameters enter the multi-branch dynamic neural network through the input layer; The material coding module processes the material specific heat capacity and the material thermal conductivity coefficient through a full connection layer to generate a material feature vector; the stage perception module concatenates the one-hot encoding of the test stage and the material feature vector and sends them to a gate unit; the gate unit calculates a three-dimensional probability distribution through a preset weight matrix and bias to represent the activation weights of the warming branch, the steady-state branch and the cooling branch; the branch with the maximum probability is taken as the main path, and the outputs of other branches are set to zero; wherein the warming branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send them to a first hidden layer, which is processed through a full connection layer and activated by LeakyReLU, and then sent to a second hidden layer, which is processed by mapping and activated by LeakyReLU to obtain the output of the warming branch; the steady-state branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send them to a sparse self-attention layer; Query, Key and Value are calculated using the sparse self-attention layer, and only the top three attention weights are retained during self-attention calculation, and the remaining weights are set to zero to obtain attention features; then, the attention features are mapped through a full connection layer and activated by ReLU to obtain the output of the steady-state branch; the cooling branch is used to concatenate the current temperature, the target temperature and the temperature change gradient with the material feature vector, and then send them to a one-dimensional convolution layer to extract the time sequence features of temperature change, and further reduce the dimension through a max-pooling layer, and finally process them through a full connection layer and activate them by Tanh to obtain the output of the cooling branch; The feature fusion module fuses the output of the main path with the material feature vector to obtain a fused feature vector; in the output layer, the fused feature vector is mapped to a three-dimensional output through a full connection layer, and the three-dimensional output directly corresponds to the parameters of the PID controller in the temperature control system, including Kp, Ki and Kd.

5. The heat flow test temperature control device based on dynamically setting PID parameters by machine learning of claim 4, wherein, The training process of the neural network model includes: Collecting state parameters during the historical heat flow test process, and performing outlier filtering and smoothing processing on the collected state parameters to obtain effective state parameters, and using the effective state parameters to construct a historical test data set; Using the historical test data set, the network weights of the neural network model are optimized through supervised learning, so that the neural network model can dynamically adjust the PID parameters under different heat flow conditions.

6. The heat flow test temperature control device based on dynamically setting PID parameters by machine learning of claim 5, wherein, The PID parameter dynamic adjustment strategy is optimized according to the material characteristics, for materials with a thermal conductivity coefficient lower than a preset thermal conductivity coefficient threshold, the integral parameter is increased to enhance the stability of temperature adjustment; for materials with thermal inertia greater than a preset thermal inertia threshold, the differential parameter is adjusted to reduce the overshoot phenomenon.

Images