Calibration model generation method, cooling flow calibration method, device, equipment and medium

Abstract

The invention discloses a calibration model generation method, a cooling flow calibration method, a device, equipment and a medium. The method comprises the following steps: acquiring training data corresponding to the electromechanical coupling system, wherein the training data comprises a test working condition and a calorific value corresponding to the test working condition; determining a key part temperature corresponding to the test working condition based on the test working condition and the calorific value corresponding to the test working condition; generating a target reduced-order model based on the test working condition and the key part temperature corresponding to the test working condition; and determining a target calibration model based on the target reduced-order model and the control strategy model. According to the method, physical calibration does not need to be carried out on the electromechanical coupling system, the calibration efficiency of the cooling flow corresponding to the electromechanical coupling system can be effectively improved through the target calibration model, the purpose of improving the calibration efficiency of the cooling flow is achieved, convenience and high efficiency are achieved, manpower and material resources are saved, and high application value is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual calibration, and in particular to a calibration model generation method, a cooling flow calibration method, a device, equipment and a medium. Background Art

[0002] In the prior art, when calibrating the cooling flow of an electromechanical coupling system, the electromechanical coupling system is usually placed on a test bench for physical calibration, which not only takes a long calibration time but also has a large calibration workload, resulting in low calibration efficiency. Therefore, how to improve the calibration efficiency of the cooling flow of the electromechanical coupling system is a technical problem that needs to be solved urgently at present. Summary of the Invention

[0003] Embodiments of the present invention provide a calibration model generation method, a cooling flow calibration method, a device, equipment and a medium to solve the problem of how to improve the calibration efficiency of the cooling flow of the electromechanical coupling system.

[0004] A calibration model generation method includes: Obtaining training data corresponding to an electromechanical coupling system, where the training data includes test conditions and calorific values corresponding to the test conditions; Based on the test conditions and the calorific values corresponding to the test conditions, determining the temperatures of key components corresponding to the test conditions; Based on the test conditions and the temperatures of key components corresponding to the test conditions, generating a target reduced-order model; Based on the target reduced-order model and a control strategy model, determining a target calibration model.

[0005] Preferably, the electromechanical coupling system includes at least one key component; The determining the temperatures of key components corresponding to the test conditions based on the test conditions and the calorific values corresponding to the test conditions includes: Using the temperature model corresponding to each key component to process each test condition and the calorific value corresponding to the test condition, and determining the temperature of each key component corresponding to each key component under each test condition.

[0006] Preferably, the generating a target reduced-order model based on the test conditions and the temperatures of key components corresponding to the test conditions includes: Inputting the test conditions into an original neural network to determine the temperatures of test components corresponding to the test conditions; Based on the temperatures of key components and test components corresponding to the test conditions, determining the target loss of the original neural network; If the target loss meets a preset convergence condition, determining the original neural network as the target reduced-order model.

[0007] Preferably, determining the target calibration model based on the target reduced-order model and the control strategy model includes: Performing format conversion on the target reduced-order model to generate a target format file; Integrating the target format file with the control strategy model to obtain the target calibration model.

[0008] A cooling flow calibration method includes: Obtaining the current working condition of the electromechanical coupling system; Processing the current working condition using the target calibration model to determine the cooling flow requirement of the electromechanical coupling system; Wherein, the target calibration model is a model obtained by training with the above calibration model generation method.

[0009] Preferably, processing the current working condition using the target calibration model to determine the cooling flow requirement of the electromechanical coupling system includes: Processing the current working condition using the target reduced-order model to determine the current component temperature corresponding to the key components in the electromechanical coupling system; Performing calibration processing on the current component temperature corresponding to the key components using the control strategy model to determine the cooling flow requirement of the electromechanical coupling system.

[0010] A calibration model generation device includes: A training data acquisition module, configured to acquire training data corresponding to the electromechanical coupling system, where the training data includes test working conditions and the heat generation corresponding to the test working conditions; A key component temperature determination module, configured to determine the key component temperature corresponding to the test working condition based on the test working condition and the heat generation corresponding to the test working condition; A target reduced-order model generation module, configured to generate a target reduced-order model based on the test working condition and the key component temperature corresponding to the test working condition; A target calibration model determination module, configured to determine a target calibration model based on the target reduced-order model and the control strategy model.

[0011] A cooling flow calibration device includes: A current working condition acquisition module, configured to acquire the current working condition of the electromechanical coupling system; A cooling flow requirement determination module, configured to process the current working condition using the target calibration model to determine the cooling flow requirement of the electromechanical coupling system.

[0012] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned calibration model generation method is implemented, or when the processor executes the computer program, the above-mentioned cooling flow calibration method is implemented.

[0013] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the above-mentioned calibration model generation method is implemented, or when the computer program is executed by the processor, the above-mentioned cooling flow calibration method is implemented.

[0014] The above-mentioned calibration model generation method, cooling flow calibration method, device, equipment and medium are based on training data such as test conditions and calorific values corresponding to the test conditions, determine the temperatures of key components corresponding to the test conditions, generate a target reduced-order model based on the test conditions and the temperatures of key components corresponding to the test conditions, and determine a target calibration model capable of virtual calibration of the cooling flow of the mechatronic coupling system based on the target reduced-order model and the control strategy model. This method generates a target calibration model based on a large amount of data corresponding to different test conditions of the mechatronic coupling system and the temperatures of key components corresponding to each test condition, facilitating the relatively efficient and accurate calibration of the cooling flow of the mechatronic coupling system according to the target calibration model. This method does not require physical calibration of the mechatronic coupling system, can effectively improve the calibration efficiency of the cooling flow corresponding to the mechatronic coupling system through the target calibration model, achieve the purpose of improving the calibration efficiency of the cooling flow, is relatively convenient and efficient, saves human and material resources, and has high application value. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 is a flowchart of a calibration model generation method in an embodiment of the present invention; Figure 2 is another flowchart of a calibration model generation method in an embodiment of the present invention; Figure 3 is another flowchart of a calibration model generation method in an embodiment of the present invention; Figure 4 is a flowchart of a cooling flow calibration method in an embodiment of the present invention; Figure 5 It is another flowchart of the cooling flow rate calibration method in an embodiment of the present invention; Figure 6 It is a schematic diagram of a calibration model generation device in an embodiment of the present invention; Figure 7 It is a schematic diagram of a cooling flow rate calibration device in an embodiment of the present invention; Figure 8 It is a schematic diagram of a computer device in an embodiment of the present invention. Detailed implementation manners

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] The embodiment of the present invention provides a calibration model generation method, and this calibration model generation method is used to achieve the purpose of improving the efficiency of cooling flow rate calibration of the electromechanical coupling system.

[0019] In one embodiment, as Figure 1 shown, a calibration model generation method is provided. Taking the computer device in Figure 8 as an example for description, the method includes the following steps: S101: Obtain training data corresponding to the electromechanical coupling system, where the training data includes test working conditions and the calorific value corresponding to the test working conditions; S102: Based on the test working conditions and the calorific value corresponding to the test working conditions, determine the temperature of key components corresponding to the test working conditions; S103: Based on the test working conditions and the temperature of key components corresponding to the test working conditions, generate a target reduced-order model; S104: Based on the target reduced-order model and the control strategy model, determine the target calibration model.

[0020] Among them, the training data refers to the data used to train the model. The test working condition refers to different working conditions of the electromechanical coupling system.

[0021] As an example, in step S101, the computer device obtains the heat generation of the electromechanical coupling system under different test conditions, and determines each test condition and the heat generation corresponding to each test condition as the training data corresponding to the electromechanical coupling system. In this example, the electromechanical coupling system is usually assembled on a vehicle. Since the vehicle travels under different conditions such as different speeds and road conditions, the electromechanical coupling system also operates under different conditions. Various conditions during the operation of the electromechanical coupling system are determined as test conditions to obtain training data with a certain quantity scale and corresponding to the conditions during the operation of the electromechanical coupling system, so that the model trained based on the training data is more accurate.

[0022] Among them, the temperature of key components refers to the temperature of the components in the electromechanical coupling system that affect the calibration of the cooling flow rate.

[0023] As an example, in step S102, the computer device analyzes and processes each test condition and the heat generation corresponding to each test condition to determine the temperature of the key components corresponding to the test condition. In this example, under different test conditions of the electromechanical coupling system, different heat generations correspond to different temperatures of the key components in the electromechanical coupling system. Therefore, by analyzing and processing different test conditions and the heat generations corresponding to different test conditions, the temperature of the key components corresponding to different test conditions can be determined more accurately.

[0024] Among them, the target reduced-order model refers to the model trained to determine the temperature of key components according to the test conditions.

[0025] As an example, in step S103, the computer device uses the test conditions and the temperatures of the key components corresponding to the test conditions for model training until the model converges, and obtains a target reduced-order model for directly determining the temperature of the key components according to the test conditions. In this example, the data volume of the temperatures of the key components corresponding to the test conditions is at least N times that of the test conditions (where N≥100), so as to train a more accurate target reduced-order model based on a larger data volume.

[0026] Among them, the control strategy model refers to a control model used for calibrating the cooling flow rate of the mechatronic coupling system. The target calibration model refers to a model used for calibrating the cooling flow rate of the mechatronic coupling system. Cooling flow rate calibration refers to the process of determining the flow rate of the coolant required by the mechatronic coupling system under different working conditions. Under different working conditions of the mechatronic coupling system, the key component temperatures corresponding to each key component are different. Therefore, coolants with different flow rates are required to cool down the mechatronic coupling system. It is necessary to pre-calibrate the coolant requirements of the mechatronic coupling system under different working conditions, that is, to calibrate the cooling flow rate of the mechatronic coupling system, so that when determining the working conditions of the mechatronic coupling system, the mechatronic coupling system can be directly cooled down according to the pre-calibrated cooling flow rate.

[0027] As an example, in step S104, the computer device determines the trained target reduced-order model and the control strategy model preset in the system as the target calibration model. The target calibration model is trained based on a large amount of data corresponding to different test conditions of the mechatronic coupling system and the key component temperatures corresponding to each test condition, and can calibrate the cooling flow rate of the mechatronic coupling system more efficiently and accurately. This method does not require physical calibration of the mechatronic coupling system, and can achieve the purpose of improving the calibration efficiency of the cooling flow rate through the obtained target calibration model. It is relatively convenient and fast, saving human and material resources.

[0028] In this embodiment, based on the test conditions and training data such as the calorific value corresponding to the test conditions, the key component temperatures corresponding to the test conditions are determined. Based on the test conditions and the key component temperatures corresponding to the test conditions, based on the test conditions and the key component temperatures corresponding to the test conditions, a target reduced-order model is generated. Based on the target reduced-order model and the control strategy model, a target calibration model capable of virtually calibrating the cooling flow rate of the mechatronic coupling system is determined. This method trains and generates a target calibration model based on a large amount of data corresponding to different test conditions of the mechatronic coupling system and the key component temperatures corresponding to each test condition, which is convenient for calibrating the cooling flow rate of the mechatronic coupling system more efficiently and accurately according to the target calibration model. This method does not require physical calibration of the mechatronic coupling system, and can effectively improve the calibration efficiency of the cooling flow rate corresponding to the mechatronic coupling system through the target calibration model, achieving the purpose of improving the calibration efficiency of the cooling flow rate. It is relatively convenient and efficient, saving human and material resources, and has high application value.

[0029] In one embodiment, the electromechanical coupling system includes at least one critical component. Step S102, that is, based on the test conditions and the heat generation corresponding to the test conditions, determining the temperature of the critical component corresponding to the test conditions, includes: using the temperature model corresponding to each critical component to process each test condition and the heat generation corresponding to the test condition, and determining the temperature of the critical component corresponding to each critical component under each test condition.

[0030] Wherein, the temperature model refers to a model used to determine the temperature of the critical component.

[0031] As an example, the computer device obtains the temperature model corresponding to the critical component established in advance, takes each test condition and the heat generation corresponding to the electromechanical coupling system under each test condition as model inputs, inputs them into the temperature model, and outputs the temperature of the critical component corresponding to each critical component in the electromechanical coupling system under each test condition. In this example, the critical components include, but are not limited to, the motor, motor controller, gears, bearings, electronic pump, electromechanical coupling housing, shaft, and clutch in the electromechanical coupling system. For example, if each critical component corresponds to a temperature model for determining the temperature of the critical component, the computer device takes each test condition and the heat generation corresponding to the electromechanical coupling system under each test condition as model inputs, inputs them into the temperature model corresponding to each critical component, and outputs the temperature of the critical component corresponding to each critical component under each test condition. Another example is that if all critical components correspond to a temperature model for comprehensively determining the temperature of each critical component according to the heat generation corresponding to the electromechanical coupling system, the computer device takes each test condition and the heat generation corresponding to the electromechanical coupling system under each test condition as model inputs, inputs them into a temperature model, and outputs the temperature of the critical component corresponding to each critical component under each test condition.

[0032] In this example, the heat generation corresponding to the electromechanical coupling system includes, but is not limited to, motor loss heat, switching loss heat, and transmission loss heat. Generally speaking, the electromechanical coupling system includes a motor controller, a motor, and a transmission component (including gears and bearings). The motor controller controls whether the motor works through a switch. The switch includes, but is not limited to, an IGBT (Insulated Gate Bipolar Transistor) switch. The motor controller, the motor, and the transmission component (including gears and bearings) will generate corresponding loss heat during operation, and through heat transfer, the critical components in the electromechanical coupling system will generate corresponding temperatures of the critical components.

[0033] Wherein, the motor loss heat refers to the loss generated due to the operation of the motor. The switching loss heat refers to the loss of the switch in the motor controller. The transmission loss heat refers to the loss generated during the operation of transmission components such as gears and bearings.

[0034] The computer device obtains the temperature model corresponding to the pre-established key components, and takes each test condition and the motor loss heat, switch loss heat, and transmission loss heat under each test condition as model inputs, and inputs them into the temperature model to output the key component temperature corresponding to each key component under each test condition.

[0035] In this embodiment, by processing various test conditions and the heat generation corresponding to the electromechanical coupling system under each test condition based on the temperature model corresponding to the key components, the key component temperature corresponding to each key component under each test condition can be obtained more accurately. Moreover, this method can obtain the key component temperature corresponding to each key component under different test conditions during the development stage of the electromechanical coupling system without collecting the key component temperature corresponding to the key components, which not only saves human and material resources but also is relatively efficient.

[0036] In one embodiment, as Figure 2 shown, step S103, that is, generating a target reduced-order model based on the test condition and the key component temperature corresponding to the test condition, includes: S201: Input the test condition into the original neural network to determine the test component temperature corresponding to the test condition; S202: Determine the target loss of the original neural network based on the key component temperature and the test component temperature corresponding to the test condition; S203: If the target loss meets the preset convergence condition, determine the original neural network as the target reduced-order model.

[0037] Among them, the original neural network refers to the original model used for training. The test component temperature refers to the temperature of the key component output according to the test condition during the training process of the original neural network.

[0038] As an example, in step S201, the computer device inputs different test conditions into the original neural network to obtain the test component temperature corresponding to each key component in the electromechanical coupling system under each test condition. In this example, the original neural network includes but is not limited to a black-box model driven by a neural network or a gray-box model driven by a differential neural network, which is used to train a large amount of data corresponding to different test conditions and the key component temperature under each test condition to obtain a relatively accurate target reduced-order model.

[0039] Among them, the target loss refers to the loss generated during the training process of the original neural network.

[0040] As an example, in step S202, the computer device calculates the losses of the temperatures of the key components corresponding to the test conditions and the temperatures of the tested components, and obtains the target loss generated during the original neural network training process. In this example, the computer device uses the temperature of the key component corresponding to each key component under each test condition as the training label, and calculates the loss of the temperature of the tested component corresponding to each key component generated by training the original neural network under each test condition, to obtain the target loss. In this example, the target loss includes, but is not limited to, the mean square error loss function, the cross entropy loss function, and the label loss function. The computer device can weight the mean square error loss function, the cross entropy loss function, and the label loss function respectively through preset weights to obtain the target loss.

[0041] In this example, if the temperature of the j-th key component corresponding to the i-th test condition is , and the temperature of the tested component corresponding to the i-th test condition is , where i ≥ 1, j ≥ 1, then the mean square error loss function is: = , the cross entropy loss function is: = , and the label loss function is: = . The computer device uses the preset weights to weight the mean square error loss function, the cross entropy loss function, and the label loss function respectively to obtain the target loss function : = + +(1 - - ) .

[0042] That is = + +(1 - - ) where K is the number of test conditions, and M is the number of key components in the electromechanical coupling system. 0 ≤ ≤ 1, 0 ≤ ≤ 1.

[0043] Among them, the preset convergence condition refers to the preset condition for judging whether the target loss converges.

[0044] ​​As an example, in step S203, when the computer device determines that the target loss meets the preset convergence condition, it determines that the original neural network training is completed, and determines the trained original neural network as the target reduced-order model, which is used to more efficiently and accurately determine the key component temperature corresponding to each key component in the electromechanical coupling system under the test conditions. When the computer device determines that the target loss does not meet the preset convergence condition, it continues to execute steps S201 to S202 until the target loss meets the preset convergence condition, and then determines that the original neural network training is completed to obtain the target reduced-order model. In this example, the preset convergence condition can be that the target loss is less than the preset loss value, or the difference between the target losses corresponding to two adjacent trainings is within the preset range; where the preset loss value is not greater than 2%, and the preset range is [0, 2%].

[0045] In this embodiment, based on the key component temperature corresponding to the test conditions and the test component temperature, the target loss of the original neural network is determined. When the target loss meets the preset convergence condition, a relatively accurate target reduced-order model is obtained, which can more efficiently and accurately determine the key component temperature corresponding to each key component in the electromechanical coupling system under different test conditions.

[0046] In one embodiment, as Figure 3 shown, step S104, that is, determining the target calibration model based on the target reduced-order model and the control strategy model, includes: S301: Perform format conversion on the target reduced-order model to generate a target format file; S302: Integrate the target format file with the control strategy model to obtain the target calibration model.

[0047] Among them, the target format file refers to the format of the target reduced-order model that can be integrated with the control strategy model.

[0048] As an example, in step S301, the computer device performs format conversion on the target reduced-order model according to the preset conversion format to obtain the target format file corresponding to the format-converted target reduced-order model. It can be understood that when calibrating the cooling flow demand of the electromechanical coupling system, it is necessary to more accurately determine the key component temperature corresponding to each key component under different working conditions through the target reduced-order model, so that the control strategy model can calibrate the cooling flow demand according to the key component temperature. If the format of the target reduced-order model is not converted, the control strategy model cannot recognize the key component temperature determined by the target reduced-order model. Therefore, it is necessary to perform format conversion on the target reduced-order model. In this example, the preset conversion format includes but is not limited to the FMU standard format, which is used to convert the target reduced-order model into a target format file in the FMU standard format.

[0049] As an example, in step S302, the computer device associates and integrates the target format file corresponding to the target reduced-order model with the control strategy model to obtain a target calibration model. The target calibration model is used to link the interface of the system database according to the key component temperatures of the test conditions output by the target reduced-order model, query the cooling flow requirements corresponding to the key component temperatures of different test conditions, and implement the cooling flow calibration for the mechatronic coupling system. The computer device integrates the target reduced-order model and the control strategy model into a target calibration model, which can simplify the cooling flow calibration process and improve the efficiency of cooling flow calibration. In this example, the computer device sets the target format file corresponding to the target reduced-order model as the first execution sub-model of the target calibration model, and sets the control strategy model as the second execution sub-model of the target calibration model to implement the integration of the target reduced-order model and the control strategy model and obtain the target calibration model.

[0050] In this embodiment, the format of the target reduced-order model is converted to generate a target format file, so as to integrate the target format file with the control strategy model to obtain a target calibration model. This method integrates the target reduced-order model and the control strategy model into a target calibration model, which can simplify the cooling flow calibration process and make the cooling flow calibration more accurate and efficient.

[0051] In one embodiment, as Figure 4 shown, a cooling flow calibration method is provided. Taking the computer device in Figure 8 as an example, the method includes the following steps: S401: Obtain the current condition of the mechatronic coupling system; S402: Process the current condition using the target calibration model to determine the cooling flow requirement of the mechatronic coupling system.

[0052] Among them, the target calibration model is the model obtained by training with the above calibration model generation method.

[0053] Among them, the current condition refers to the condition of the mechatronic coupling system corresponding to the cooling flow calibration to be performed.

[0054] As an example, in step S401, the computer device obtains the current working condition of the electromechanical coupling system that needs to be calibrated for the cooling flow rate, so as to determine the cooling flow rate requirement of the electromechanical coupling system according to the current working condition. Understandably, since the electromechanical coupling system is usually assembled in a vehicle, in order to adapt to different speeds and different road conditions of the vehicle, the electromechanical coupling system also corresponds to different working conditions. The computer device models the different speeds and different road conditions of the vehicle, and simulates the different working conditions corresponding to the electromechanical coupling system to determine a variety of working conditions corresponding to the electromechanical coupling system. In this example, the computer device sequentially determines each working condition as the current working condition, so as to calibrate the cooling flow rate requirement for each working condition corresponding to the electromechanical coupling system.

[0055] Among them, the cooling flow rate requirement refers to the flow rate of the coolant required by the electromechanical coupling system under different working conditions.

[0056] As an example, in step S402, the computer device inputs the obtained current working condition into the target calibration model generated by the above calibration model generation method, and outputs the cooling flow rate requirement of the electromechanical coupling system under the current working condition. The computer device inputs each current working condition into the target calibration model separately in the above manner, and obtains the cooling flow rate requirement corresponding to each current working condition, so as to realize the calibration of the cooling flow rate of the electromechanical coupling system. This method does not require physical calibration of the electromechanical coupling system, and moreover, this method can realize the calibration of the cooling flow rate of the electromechanical coupling system without complex operations, which is relatively efficient and convenient.

[0057] In this embodiment, the target calibration model is used to process the current working condition to determine the cooling flow rate requirement of the electromechanical coupling system under each current working condition, so as to realize the calibration of the cooling flow rate of the electromechanical coupling system. This method does not require physical calibration of the electromechanical coupling system, and moreover, this method can realize the calibration of the cooling flow rate of the electromechanical coupling system without complex operations, which is relatively efficient and convenient, and can achieve the purpose of improving the efficiency of the cooling flow rate calibration of the electromechanical coupling system.

[0058] In one embodiment, as Figure 5 shown, step S402, that is, using the target calibration model to process the current working condition to determine the cooling flow rate requirement of the electromechanical coupling system, includes: S501: Use the target reduced-order model to process the current working condition to determine the current component temperature corresponding to the key components in the electromechanical coupling system; S502: Use the control strategy model to perform calibration processing on the current component temperature corresponding to the key components to determine the cooling flow rate requirement of the electromechanical coupling system.

[0059] Among them, the current component temperature refers to the temperature of each key component under the current working condition output by the target reduced-order model.

[0060] As an example, in step S501, the computer device processes the current working condition using the target reduced-order model to accurately determine the current component temperature corresponding to each key component in the electromechanical coupling system. In this example, the target reduced-order model can directly output the relatively accurate current component temperature corresponding to each key component under the current working condition according to the current working condition, without the need for complex analysis and processing, which is relatively efficient and convenient.

[0061] As an example, in step S502, the computer device calibrates the current component temperature corresponding to the key component using the control strategy model to determine the cooling flow rate requirement of the electromechanical coupling system. In this example, when the computer device obtains the current component temperature corresponding to each key component under the current working condition, it queries the mapping relationship table of the combination of different key component temperatures and the cooling flow rate requirement of the electromechanical coupling system stored in the system database in advance through the interface corresponding to the control strategy model linking the system database, and obtains the cooling flow rate requirement corresponding to the combination of the current component temperature of each key component under the current working condition, and determines this cooling flow rate requirement as the cooling flow rate requirement of the electromechanical coupling system, thus completing the calibration of the cooling flow rate requirement of the electromechanical coupling system under the current working condition.

[0062] In this embodiment, the target reduced-order model obtained by training with a large amount of data is used to process the current working condition, which can accurately determine the current component temperature corresponding to the key components in the electromechanical coupling system. The control strategy model is used to calibrate the current component temperature corresponding to the key components to determine the cooling flow rate requirement of the electromechanical coupling system. This method can achieve the calibration of the cooling flow rate of the electromechanical coupling system without complex operations, is relatively efficient and convenient, can achieve the purpose of improving the efficiency of the cooling flow rate calibration of the electromechanical coupling system, and has high application value.

[0063] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0064] In one embodiment, a calibration model generation device is provided, and this calibration model generation device corresponds one-to-one with the calibration model generation method in the above embodiment. As Figure 6 shown, this calibration model generation device includes a training data acquisition module 601, a key component temperature determination module 602, a target reduced-order model generation module 603, and a target calibration model determination module 604. The detailed description of each functional module is as follows: A training data acquisition module 601, configured to acquire training data corresponding to an electromechanical coupling system, where the training data includes test conditions and calorific values corresponding to the test conditions; A key component temperature determination module 602, configured to determine the temperature of key components corresponding to a test condition based on the test condition and the calorific value corresponding to the test condition; A target reduced-order model generation module 603, configured to generate a target reduced-order model based on the test condition and the temperature of key components corresponding to the test condition; A target calibration model determination module 604, configured to determine a target calibration model based on the target reduced-order model and a control strategy model.

[0065] In one embodiment, the key component temperature determination module 602 includes: A key component temperature determination sub-module, configured to process each test condition and the calorific value corresponding to the test condition by using a temperature model corresponding to each key component, and determine the temperature of the key component corresponding to each key component under each test condition.

[0066] In one embodiment, the target reduced-order model generation module 603 includes: A test component temperature determination sub-module, configured to input a test condition into an original neural network to determine the temperature of a test component corresponding to the test condition; A target loss determination sub-module, configured to determine a target loss of the original neural network based on the temperature of key components corresponding to the test condition and the temperature of the test component; A target reduced-order model determination sub-module, configured to determine the original neural network as the target reduced-order model if the target loss meets a preset convergence condition.

[0067] In one embodiment, the target calibration model generation module 604 includes: A target format file generation sub-module, configured to perform format conversion on the target reduced-order model to generate a target format file; A target calibration model determination sub-module, configured to integrate the target format file with the control strategy model to obtain a target calibration model.

[0068] In another embodiment, a cooling flow rate calibration device is provided, and the cooling flow rate calibration device corresponds one-to-one to the cooling flow rate calibration method in the above embodiment. As Figure 7 shown, the cooling flow rate calibration device includes a current condition acquisition module 701 and a cooling flow rate demand determination module 702. The detailed description of each functional module is as follows: A current condition acquisition module 701, configured to acquire the current condition of the electromechanical coupling system; The cooling flow rate demand determination module 702 is configured to process the current working condition by using a target calibration model to determine the cooling flow rate demand of the electromechanical coupling system.

[0069] In one embodiment, the cooling flow rate demand determination module 702 includes: The current component temperature determination sub-module is configured to process the current working condition by using a target reduced-order model to determine the current component temperature corresponding to the key components in the electromechanical coupling system; The cooling flow rate demand calibration sub-module is configured to perform calibration processing on the current component temperature corresponding to the key components by using a control strategy model to determine the cooling flow rate demand of the electromechanical coupling system.

[0070] For the specific limitations of the calibration model generation device and the cooling flow rate calibration device, reference may be made to the limitations of the calibration model generation method and the cooling flow rate calibration method in the foregoing text, which will not be elaborated here. Each module in the above-mentioned calibration model generation device and cooling flow rate calibration device can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in the processor of the computer device in hardware form or be independent of it, or can be stored in the memory of the computer device in software form, so as to facilitate the processor to call and execute the operations corresponding to the above-mentioned modules.

[0071] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 8 shown. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the data used or generated during the execution of the calibration model generation method, or is used to store the data used or generated during the execution of the cooling flow rate calibration method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a calibration model generation method, or when the computer program is executed by the processor, it implements a cooling flow rate calibration method.

[0072] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the calibration model generation method in the above-mentioned embodiment, such as Figure 1 S101-S104 shown, or Figures 2 to 3As shown, to avoid repetition, it will not be elaborated here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the calibration model generation device. For example Figure 6 the functions of the training data acquisition module 601, the key component temperature determination module 602, the target reduced-order model generation module 603, and the target calibration model determination module 604 shown. To avoid repetition, it will not be elaborated here. Alternatively, when the processor executes the computer program, it implements the cooling flow calibration method in the above embodiment. For example Figure 4 S401 - S402 shown, or Figure 5 as shown, to avoid repetition, it will not be elaborated here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the cooling flow calibration device. For example Figure 7 the functions of the current working condition acquisition module 701 and the cooling flow demand determination module 702 shown. To avoid repetition, it will not be elaborated here.

[0073] In one embodiment, a computer-readable storage medium is provided. A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, it implements the calibration model generation method in the above embodiment. For example Figure 1 S101 - S104 shown, or Figures 2 to 3 as shown, to avoid repetition, it will not be elaborated here. Alternatively, when the computer program is executed by the processor, it implements the functions of each module / unit in this embodiment of the calibration model generation device. For example Figure 6 the functions of the training data acquisition module 601, the key component temperature determination module 602, the target reduced-order model generation module 603, and the target calibration model determination module 604 shown. To avoid repetition, it will not be elaborated here. Alternatively, when the computer program is executed by the processor, it implements the cooling flow calibration method in the above embodiment. For example Figure 4 S401 - S402 shown, or Figure 5 as shown, to avoid repetition, it will not be elaborated here. Alternatively, when the computer program is executed by the processor, it implements the functions of each module / unit in this embodiment of the cooling flow calibration device. For example Figure 7 the functions of the current working condition acquisition module 701 and the cooling flow demand determination module 702 shown. To avoid repetition, it will not be elaborated here. The computer-readable storage medium can be non-volatile or volatile.

[0074] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. This computer program can be stored in a non-volatile computer-readable storage medium. When this computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in this application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0075] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0076] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A calibration model generation method, characterized in that: include: Acquire training data corresponding to the electromechanical coupling system, wherein the training data includes a test condition and a heat value corresponding to the test condition; Based on the test condition and the heat generation corresponding to the test condition, determining the temperature of key components corresponding to the test condition; Generate a target reduced-order model based on the test conditions and the temperatures of key components corresponding to the test conditions; Based on the target reduced-order model and the control strategy model, a target calibration model is determined.

2. The calibration model generation method according to claim 1, characterized in that: The electromechanical coupling system includes at least one key component; The determining, based on the test condition and the heat generation corresponding to the test condition, the temperature of key components corresponding to the test condition includes: A temperature model corresponding to each of the key components is used to process each of the test conditions and the heat generation corresponding to the test conditions, and the key component temperature corresponding to each of the key components under each of the test conditions is determined.

3. The calibration model generation method according to claim 1, characterized in that: The generating a target reduced-order model based on the test condition and the temperature of key components corresponding to the test condition includes: Inputting the test condition into the original neural network to determine the temperature of the test component corresponding to the test condition; Determining a target loss of the original neural network based on the temperature of the key components and the temperature of the test components corresponding to the test condition; If the target loss satisfies a preset convergence condition, the original neural network is determined as a target reduced-order model.

4. The calibration model generation method according to claim 1, characterized in that: The step of determining a target calibration model based on the target reduced-order model and the control strategy model includes: Performing format conversion on the target reduced-order model to generate a target format file; The target format file is integrated with the control strategy model to obtain a target calibration model.

5. A cooling flow calibration method, characterized in that: include: Obtain the current working condition of the electromechanical coupling system; Processing the current operating condition using a target calibration model to determine a cooling flow requirement of the electromechanical coupling system; Wherein, the target calibration model is a model trained by the calibration model generation method described in any one of claims 1-4.

6. The cooling flow calibration method according to claim 5, characterized in that: The adopting the target calibration model to process the current working condition to determine the cooling flow demand of the electromechanical coupling system includes: The current operating condition is processed by using a target reduced-order model to determine the current component temperature corresponding to the key component in the electromechanical coupling system; The control strategy model is used to calibrate the current component temperature corresponding to the key component to determine the cooling flow demand of the electromechanical coupling system.

7. A calibration model generation device, characterized in that: include: A training data acquisition module, used to acquire training data corresponding to the electromechanical coupling system, wherein the training data includes a test condition and a calorific value corresponding to the test condition; A key component temperature determination module, which determines the key component temperature corresponding to the test condition based on the test condition and the calorific value corresponding to the test condition; A target reduced-order model generation module generates a target reduced-order model based on the test conditions and the temperatures of key components corresponding to the test conditions; The target calibration model determination module determines the target calibration model based on the target reduced-order model and the control strategy model.

8. A cooling flow calibration device, characterized in that: include: A current working condition acquisition module is used to acquire the current working condition of the electromechanical coupling system; The cooling flow demand determination module is used to process the current working condition using a target calibration model to determine the cooling flow demand of the electromechanical coupling system.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the calibration model generation method as described in any one of claims 1 to 4 is implemented, or when the processor executes the computer program, the cooling flow calibration method as described in any one of claims 5 to 6 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the calibration model generation method as described in any one of claims 1 to 4 is implemented, or, when the computer program is executed by a processor, the cooling flow calibration method as described in any one of claims 5 to 6 is implemented.

Images