Compilation integration method and system for vehicle thermal management model

Abstract

The invention discloses a complete vehicle thermal management model compiling and integrating method and system, and belongs to the technical field of vehicle analogue simulation, and the method comprises the steps: obtaining a high-precision complete vehicle thermal management offline simulation model, and decomposing sub-models; establishing a signal interaction channel and other input and output signal channels; reconstructing a sub-model to realize reconstruction of a whole vehicle model; off-line testing is carried out, and compiling is carried out to adapt to a development platform; constructing a required matching project file, and performing a model integration test; and a thermal management work development platform containing a high-precision whole vehicle thermal management model and a real-time hardware control unit at the same time is formed. The simulation requirement can be reduced by decomposing the model and setting the signal channel, the whole test can be realized by reconstructing the whole vehicle model, the correctness of the simulation structure can be verified by the off-line test, the requirements of the development platform can be met by compiling, the practicability is improved, and the proper parameters are determined by the integration test. Therefore, a thermal management work development platform is formed.

Description

A method and system for compiling and integrating a vehicle thermal management model Technical Field

[0001] This invention relates to the field of vehicle simulation technology, and in particular to a method and system for compiling and integrating a whole vehicle thermal management model. Background Technology

[0002] In engine research and development, simulation calculation, as a means of reproducing the engine's working process and performance results through computational software, and to a certain extent, providing high-confidence prediction results through technical control, has become an important development tool. However, simulation still has certain limitations. In the simulation calculation process, the closer the model is to the real process, the more accurate the calculation results will be. However, this often increases the difficulty and time cost of actual execution. In actual development work, it is often difficult to find a good balance between these two points. In addition, in the modern engine development process, with the increasing depth of engine performance development, the engine is not only evaluated as an independent power output source, but also as part of a complex powertrain system, and evaluated based on the system's performance in the whole vehicle. Current technologies lack a holistic consideration. Summary of the Invention

[0003] This invention proposes a method and system for compiling and integrating a whole vehicle thermal management model to address the limitations of existing simulation technologies and the lack of overall consideration.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] A method for compiling and integrating a vehicle thermal management model includes: acquiring a high-precision offline simulation model of vehicle thermal management and decomposing it into several sub-models; establishing signal interaction channels between the sub-models, as well as other input and output signal channels of the sub-models; reconstructing the sub-models based on real-time performance and physical memory, and reconstructing the entire vehicle model; performing offline testing on the reconstructed full model, and compiling the full model that passes the test to adapt to the corresponding development platform; constructing the matching project files required by the full model to the digital vehicle assembly virtual development platform, and performing model integration testing; and forming a thermal management working development platform that simultaneously contains a high-precision vehicle thermal management model and a real-time hardware control unit.

[0006] Furthermore, the sub-models include an engine model, a vehicle model, a thermal network model, an engine compartment model, and a cooling and lubrication network model, which are used to constitute the full model. Each sub-model undergoes individual or joint calibration to meet certain accuracy requirements and enable the vehicle thermal management offline simulation model to accurately simulate the calculations corresponding to vehicle thermal management in an offline state. It can generate corresponding actions based on ECU control commands, calculate and feedback the signals required by the ECU in real time, and, based on the corresponding calibration, calculate the performance results under different actions according to the ECU control signals.

[0007] Furthermore, establishing signal interaction channels between the sub-models and other input / output signal channels for the sub-models includes: the sub-models are integrated into the virtual development platform, and interaction between models is achieved through engineering configuration files within the virtual development platform; other input / output signals include interaction signals between the virtual development platform and the ECU control unit, and monitoring signals; correspondingly, establishing the other input / output signal channels changes the computational control logic within the model from offline control to online control.

[0008] Furthermore, the reconstruction of the sub-model based on real-time performance and physical memory, and the reconstruction of the whole vehicle model, includes: dividing the sub-model into a separate platform system, wherein the shared components are shared component modules used for interconnection between the two sub-models that cannot be divided into a specific platform system, and the platform system is the software and hardware environment corresponding to the virtual development platform; after confirming the number of sub-model decompositions according to the actual configuration of the virtual development platform, splitting and reconstructing the shared components between the sub-models, and modifying the corresponding signals; constructing different sub-models into higher-level sub-models, and recompiling and integrating the sub-models and / or the higher-level sub-models into the virtual development platform, so that the recompiled full model and the undecomposed full model obtain essentially the same calculation results under the same calculation conditions.

[0009] Furthermore, the reconstructed full model is subjected to offline testing to verify that the structure and signal transmission are correct and that there are no internal operational errors; the full model that passes the test is then compiled to adapt to the corresponding development platform, including: compiling the full model into the corresponding required format based on the type of the virtual development platform.

[0010] Furthermore, the matching engineering files required for constructing the full model to the digital vehicle assembly virtual development platform and performing model integration testing include: converting the physical / digital signals received and emitted by the real control unit into digital / physical signals through hardware and software to meet the environment required for the model and actuators to work normally in a hardware-in-the-loop state; integrating the full model into the digital vehicle assembly virtual development platform and connecting it to the real ECU control unit and other actuators to form a hardware-in-the-loop system and conducting online model testing.

[0011] Furthermore, the model integration test also includes: if the physical memory of the digital vehicle assembly virtual development platform exceeds the limit or the CPU performance is insufficient due to the specifications of the full model, resulting in poor real-time performance or running failure, then the whole vehicle model is rebuilt.

[0012] Furthermore, the process of performing model integration testing also includes: if the full model continuously exhibits abnormal behavior in a hardware-in-the-loop state, then adjusting the full model and / or the matching project file.

[0013] Furthermore, the thermal management development platform that simultaneously includes a high-precision vehicle thermal management model and a real-time hardware control unit includes: if the full model continuously performs normally in the hardware-in-the-loop state, then the full model and the matching engineering file are solidified.

[0014] A vehicle thermal management model compilation and integration system includes: a first module for acquiring a high-precision offline simulation model of vehicle thermal management and decomposing it into several sub-models; a second module for establishing signal interaction channels between the sub-models and other input / output signal channels of the sub-models; a third module for reconstructing the sub-models based on real-time performance and physical memory, and realizing the reconstruction of the whole vehicle model; a fourth module for performing offline testing on the reconstructed whole vehicle model, and compiling the tested whole vehicle model to adapt to the corresponding virtual development platform; a fifth module for constructing the matching project files required for the full model to the digital vehicle assembly virtual development platform, and performing model integration testing; and a sixth module for forming a thermal management working development platform that simultaneously contains a high-precision vehicle thermal management model and a real-time hardware control unit.

[0015] By adopting the above technical solution, the present invention has the following beneficial effects:

[0016] 1. This invention reduces simulation requirements by decomposing the model and setting signal channels, enables overall testing by reconstructing the whole vehicle model, verifies the correctness of the simulation structure through offline testing, adapts to the needs of the development platform through compilation, improves practicality, and determines suitable parameters through integration testing to form a thermal management work development platform. Attached Figure Description

[0017] Figure 1 is a schematic diagram of a vehicle thermal management model compilation and integration method proposed in this invention;

[0018] Figure 2 is a schematic diagram of the general method proposed in this invention for applying to a virtual development platform for digital vehicle powertrain through compilation and integration. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Figure 1 illustrates a method for compiling and integrating a vehicle thermal management model, including:

[0021] S1. Obtain a high-precision offline simulation model of the whole vehicle thermal management and decompose it into several sub-models;

[0022] S2. Establish signal interaction channels between the sub-models, as well as other input / output signal channels of the sub-models;

[0023] S3. Reconstruct the sub-model based on real-time performance and physical memory, and reconstruct the whole vehicle model.

[0024] S4. Perform offline testing on the reconstructed full model, and compile the full model that passes the test to adapt it to the corresponding development platform;

[0025] S5. Construct the matching project files required for the full model to the digital vehicle assembly virtual development platform, and perform model integration testing;

[0026] S6. A thermal management development platform that simultaneously includes a high-precision vehicle thermal management model and a real-time hardware control unit.

[0027] Different simulation targets have different structures, making it difficult to broadly define their structures. Furthermore, both physical prototypes and designed structures on blueprints can be analyzed for their structure and function. Therefore, by analyzing the structure based on physical prototypes or designed structures, high-precision offline simulation models of vehicle thermal management can be obtained. These high-precision offline simulation models are those built in an offline environment and have undergone complete subsystem and full model calibration. For any subsystem model and the full model, accuracy testing has been performed, verifying that accurate calculation results can be obtained either independently or when running the full model. For example, for the engine model, the calculation results under all universal operating conditions and the measured data should have extremely high accuracy, with explosion pressure differences not exceeding 5 bar, fuel consumption differences not exceeding 1%, and intake air volume differences not exceeding 3%. For the vehicle model, the fuel consumption difference per 100 kilometers under typical road conditions is less than 3%. For the cooling and lubrication network model, the transient temperature rise characteristics and flow rate should be basically consistent with the measured warm-up data. The remaining subsystem models are not further elaborated. Meanwhile, the integrated full model calculation results not only meet the accuracy requirements of the aforementioned subsystem models, but also exhibit extremely high consistency with the measured vehicle data. The vehicle thermal management offline simulation model is a simulation model that includes the vehicle structure, enables thermal management, and can be performed offline (after reconstruction, it is marked as the full model, which includes sub-models, combined models, and the overall model). Sub-models are several components obtained by decomposing the simulation model.

[0028] The implementation and transmission of functions between sub-models involve signals, or signals can be used to describe non-signal functions; for example, signal A can be used to describe function B or action C. By establishing signal interaction channels between these sub-models, different sub-models can be combined instead of isolated from each other, thus ensuring the overall reliability of the system. In addition to communication between sub-models, there are also situations where external input signals affect sub-models; these belong to other input / output signals, and the corresponding communication channels are other input / output signal channels.

[0029] The model reconstruction process based on real-time performance and physical memory refers to the process of determining the number of models that need to be decomposed according to the actual virtual development platform configuration, splitting and reconstructing the shared components between models, and modifying the corresponding signals.

[0030] Offline testing is performed on the reconstructed full model to confirm that the model's structure and signal transmission are correct before compilation, and that there are no internal runtime errors. The corresponding test platform and software can be selected according to actual needs; therefore, compilation is required to adapt to the requirements of the test platform and software.

[0031] During the testing process, a lot of empirical records (engineering files, including values, parameters, formulas, etc.) will be generated, which can be used for analysis. These files can be used to build a test environment that matches the test objectives. If the test results meet the preset objectives, the corresponding engineering files can be used to establish a thermal management development platform that includes a high-precision vehicle thermal management model and a real-time hardware control unit.

[0032] The sub-models include an engine model, a vehicle model, a thermal network model, an engine compartment model, and a cooling and lubrication network model. Each sub-model undergoes individual or joint calibration to achieve a certain level of accuracy, enabling the offline simulation model of vehicle thermal management to accurately simulate the calculations corresponding to vehicle thermal management in an offline state. It can generate corresponding actions based on ECU control commands, calculate and feedback the signals required by the ECU in real time, and, based on the corresponding calibration, calculate the performance results under different actions according to the ECU control signals.

[0033] The establishment of signal interaction channels between the sub-models and other input / output signal channels of the sub-models includes: the sub-models are integrated into the virtual development platform, and the interaction between the models is realized through the project configuration file in the virtual development platform; other input / output signals include interaction signals between the virtual development platform and the ECU control unit, and monitoring signals; correspondingly, the establishment of the other input / output signal channels changes the calculation control logic in the model from offline control to online control.

[0034] The reconstruction of the sub-model based on real-time performance and physical memory, and the reconstruction of the whole vehicle model, includes: dividing the sub-model into a separate platform system, wherein the shared components are shared component modules used for interconnection between the two sub-models that cannot be divided into a specific platform system; after confirming the number of sub-model decompositions according to the actual configuration of the virtual development platform, splitting and reconstructing the shared components between the sub-models, and modifying the corresponding signals; constructing different sub-models into higher-level sub-models, and recompiling and integrating the sub-models and / or the higher-level sub-models into the virtual development platform, so that the recompiled whole vehicle model and the whole vehicle model before decomposition can obtain essentially the same calculation results under the same calculation conditions.

[0035] The offline testing of the reconstructed vehicle model is used to test the correctness of the structure and signal transmission and the absence of internal operational errors; the compilation of the tested vehicle model to adapt to the corresponding virtual development platform includes: compiling the vehicle model into the corresponding required format based on the type of the virtual development platform.

[0036] The process of constructing the full model to the digital vehicle assembly virtual development platform and performing model integration testing includes: converting the physical / digital signals received and emitted by the real control unit into digital / physical signals through hardware and software to create engineering files that meet the environment required for the model and actuators to operate normally in a hardware-in-the-loop state; integrating the full model into the digital vehicle assembly virtual development platform and connecting it to the real ECU control unit and other actuators to form a hardware-in-the-loop system and conducting online model testing.

[0037] The process of performing model integration testing also includes: if the physical memory of the digital vehicle assembly virtual development platform exceeds the limit or the CPU performance is insufficient due to the specifications of the full model, resulting in poor real-time performance or running failure, then the whole vehicle model is rebuilt.

[0038] The process of performing model integration testing also includes: if the full model continuously exhibits abnormal behavior in a hardware-in-the-loop state, then adjusting the full model and / or the matching project file.

[0039] The thermal management development platform that simultaneously includes a high-precision vehicle thermal management model and a real-time hardware control unit includes: if the full model continuously performs normally in the hardware-in-the-loop state, then the full model and the matching engineering file are solidified.

[0040] A vehicle thermal management model compilation and integration system includes: a first module for acquiring a high-precision offline simulation model of vehicle thermal management and decomposing it into several sub-models; a second module for establishing signal interaction channels between the sub-models and other input / output signal channels of the sub-models; a third module for reconstructing the sub-models based on real-time performance and physical memory, and realizing the reconstruction of the whole vehicle model; a fourth module for performing offline testing on the reconstructed whole vehicle model, and compiling the tested whole vehicle model to adapt to the corresponding virtual development platform; a fifth module for constructing the matching project files required for the full model to the digital vehicle assembly virtual development platform, and performing model integration testing; and a sixth module for forming a thermal management working development platform that simultaneously contains a high-precision vehicle thermal management model and a real-time hardware control unit.

[0041] Figure 2 illustrates a general method for compiling and integrating a conventional offline vehicle thermal management model into a digital vehicle powertrain virtual development platform. This method, based on AVL CruiseM software, modifies a high-precision vehicle thermal management model powered by a diesel engine and compiles and integrates it into the digital vehicle powertrain virtual development platform. The method includes:

[0042] 101. Obtain a high-precision offline simulation model of the vehicle's thermal management;

[0043] The high-precision offline simulation model for vehicle thermal management refers to a complete model that includes all six components required for vehicle thermal management simulation calculations. These six components are: engine model, vehicle model, thermal network model, engine compartment model, and cooling and lubrication network model (including the coolant circuit model and oil circuit model, which will be collectively referred to as the cooling and lubrication network model below). Each of these six components should undergo necessary individual or joint calibration to ensure the entire vehicle thermal management model has high accuracy and can accurately simulate vehicle thermal management calculations offline. It is important to note that, since the engine model will ultimately be used on the digital vehicle powertrain virtual development platform, the model must simultaneously possess the following characteristics: 1) high real-time performance; 2) the ability to generate corresponding actions based on ECU control commands, calculate and feedback the signals required by the ECU in real time (meeting the interaction requirements with the ECU); 3) the engine model must undergo special calibration to accurately calculate performance results under different actions based on ECU control signals, including but not limited to the performance impact caused by changes in injection strategies under different combustion modes. The parameterization process for other models in the engine model and vehicle thermal management model is not the main focus of this solution, and there are similar methods in various fields, so it will not be elaborated on here.

[0044] 102. Model Decomposition

[0045] Traditional vehicle thermal management models obtained through 101 typically involve directly connecting six parts (engine model, vehicle model, thermal network model, engine compartment model, and cooling / lubrication network model) during the modeling process to enable interaction between models. For example, the engine model's output shaft can be used as the vehicle model's clutch input shaft; the solid parts of the thermal network model can be directly connected to the heat exchange elements of the cooling / lubrication network model; and the intercooler of the engine compartment model can be directly connected to the intercooler of the engine model. To facilitate subsequent model reconstruction based on real-time performance and physical memory, the vehicle thermal management model (referred to as the full model) needs to be decomposed. The principle is that the vehicle thermal management model subsequently applied to the digital vehicle powertrain virtual development platform will run on the platform through compilation and integration. The performance of the vehicle thermal management model is limited by the virtual development platform's physical memory and CPU units. If the six parts of the vehicle thermal management model are merged and compiled into a single model, insufficient CPU single-core performance may lead to extremely poor real-time performance or even failure. If the vehicle thermal management model is split and compiled multiple times so that each part of the model (i.e., sub-model) is controlled by a different CPU core, the running performance will be greatly improved, but the model's physical memory usage will also increase significantly. Therefore, the final integration method of the vehicle thermal management model will differ for different virtual development platform hardware configurations, but generally it is necessary to first decompose the vehicle thermal management model and then compile and integrate it according to the actual configuration. In this embodiment, the specific operation of decomposing the vehicle thermal management model is as follows: the 6 parts of the vehicle thermal management model are divided into 6 separate subsystems. In this process, the shared component modules used for connecting the two models that cannot be divided into a certain system (the hardware and software foundation supporting the implementation of the virtual development platform system, including hardware structure and running software) will be identified, and the interaction signals between each independent subsystem and other subsystems can be obtained. These shared component modules and interaction signals are the components that need to be reconstructed and the signals that need to be reallocated. The reconstruction and allocation operations will be explained in detail in the following sections 103 (Construction of Inter-Model Interaction Signal Channels) and 105 (Model Reconstruction Based on Real-Time Performance and Physical Memory).

[0046] 103. Construction of inter-model interaction signal channels;

[0047] After completing the model decomposition work described in 102, the signals that need to be reconstructed and reassigned among the sub-models are basically clear. As described in 102, to ensure good real-time performance after the model is finally integrated into the virtual development platform, the vehicle thermal management model will be decomposed into multiple sub-models and allocated to different CPU cores after integration into the virtual development platform. The interaction between models is changed from the original offline model-level interaction to interaction through the project configuration file within the virtual development platform after integration. For example, in this embodiment, by building a Simulink model on the virtual development platform and configuring signal interaction channels between different sub-models, the model interaction channels in different CPU cores are realized. However, the Simulink model cannot directly realize the interaction of arbitrary signals. It is necessary to configure the signal channels and corresponding interfaces to be interacted in each decomposed sub-model in advance. For example, in the vehicle thermal management model built based on AVL CruiseM described in this embodiment, it is necessary to establish a CMC Interface element in the model, predefine the signal interfaces to be interacted in the element, and connect the corresponding signals to the CMC Interface element. Obviously, the method described in this embodiment is only one embodiment based on the method proposed in the application. If the vehicle thermal management model used is a model built by other software, the sub-model should also be configured with signal conversion elements similar to the CMC Interface.

[0048] 104. Construction of other input and output signal channels for the model;

[0049] The other input / output signals of the model mentioned above refer to signals not used for inter-model signal interaction, but rather for satisfying the model's application on the virtual development platform. These signals need to include not only interaction signals between the virtual development platform and the ECU control unit (such as receiving actuator action signals from the ECU, such as fuel injectors and throttle valves, and feedback signals like intake air temperature, intake air pressure, and intake air flow required by the ECU), but also intuitive monitoring signals required for subsequent development work (such as turbocharger efficiency and engine thermal efficiency, which the ECU does not need to receive but developers need to observe). The method for constructing the signal channel is the same as described in 103. In this embodiment, it also involves establishing a CMC Interface element, configuring the signal interface, and connecting the corresponding signals to the CMC Interface element. After completing the content described in 104, the calculation control logic within the sub-model changes from offline control to online control. For example, in the original offline calculation, the actuator's actions were controlled by instructions directly defined in the model, but after completing the content described in 104, the actuator's actions will change to be controlled by receiving control instructions from the ECU through the interface.

[0050] 105. Model reconstruction based on real-time performance and physical memory;

[0051] The model reconstruction process based on real-time performance and physical memory refers to the process of determining the number of sub-models to be decomposed according to the actual virtual development platform configuration, splitting and reconstructing the shared components among the sub-models, and modifying the corresponding signals. In this embodiment, the virtual development platform's physical memory supports a maximum of three models simultaneously. When testing showed that more than two (excluding two) of the six sub-models included in the vehicle thermal management model were merged and compiled into a single higher-level sub-model before being integrated into the virtual development platform, integration failures due to poor real-time performance occurred. Therefore, the overall vehicle thermal management model containing six models was ultimately decomposed and reconstructed into three sub-models in a 2+2+2 format, which were then compiled and integrated into the virtual development platform. Clearly, the decomposition and reconstruction method described in this embodiment is only one approach under the corresponding hardware configuration. Theoretically, if a very high-performance CPU is used, model decomposition may not be necessary at all. If a large amount of physical memory is used, it is entirely possible to decompose the model into six sub-models to maintain good real-time performance. When configuration necessitates model decomposition and reconstruction, components that cannot be shared due to decomposition must be reconstructed in separate models. In this embodiment, the model is split into an engine model + engine compartment model, a cooling and lubrication network model (cooling water circuit model + oil circuit model), and a vehicle model + thermal network model. Since the engine compartment model and engine model remain within the same model, the heat exchange connection between the engine compartment model's intercooler and the engine model's intercooler does not need to be reconstructed. However, because the engine model and vehicle model are not in the same model, the power output connection shaft of the clutch in both models must be reconstructed. In this embodiment, the method involves establishing different control Flange components and configuring control signals in the engine model and vehicle model respectively, enabling both models to achieve normal power transmission and coaxial speed control. A similar approach should be used for the decomposition and reconstruction of other components, which will not be repeated here. Through the operation process described in 105, the reconstructed sub-models should ultimately achieve the necessary model interaction process through the reconstructed components and signals, and the model should ultimately be able to perform the same calculations and obtain essentially the same calculation results as before the decomposition.

[0052] The basic definition should allow for a certain amount of error. According to the test results, this error is very small, less than 3%, and generally around 1%.

[0053] 106. Reconstruct the entire vehicle model;

[0054] The vehicle model integrated into the digital vehicle assembly virtual development platform differs in control method from the offline vehicle model. In offline operation, the driver and road map are defined within the model. In online operation, a data channel interface needs to be established to interact with the virtual development platform's control system, changing the vehicle model from offline control mode to online interface control mode. For example, in this embodiment, in offline operation, the throttle signal output by the vehicle model determines the real-time torque demand, and based on the torque demand, determines the required fuel quantity, timing, rail pressure, and other control signals. When changed to online operation mode, the vehicle model is controlled by the driver model of the virtual development platform's control system, which controls the accelerator pedal, brake pedal, and gear shifting actions. The engine model directly receives control signals output by the ECU and provides real-time feedback of monitoring data signals required by the ECU, such as intake air temperature and intake air flow. At this point, the vehicle thermal management model becomes a model that can run in a hardware-in-the-loop manner on the virtual development platform.

[0055] 107. Offline testing of the decomposed model;

[0056] After completing the steps described in 106, a simple offline test needs to be performed on each part of the model that has been decomposed and reconstructed through the operations described in 102-106. The purpose of the test is to confirm that the model's structure and signal transmission are correct before compilation and that there are no internal runtime errors. Since the initial model is a high-precision offline simulation model of the whole vehicle thermal management obtained through 101, the internal connections and structure of the model are correct and no additional testing is required. The only part that needs to be tested is the part that has been changed through the operations described in 102-106. Therefore, the testing method is as follows: in the signal interface (CMC Interface element in this embodiment) added through the operations described in 102-106, each signal is defined as a reasonable default value, so that each part of the model is calculated under fixed input and the results are checked for reasonableness. For example, in this embodiment, the engine speed, rail pressure, timing, oil temperature, and coolant temperature are defined as fixed values ​​to test whether the engine calculation results are normal, and the heat results obtained by defining the thermal network model are used to test whether the solid block temperature results are normal. Other tests will not be illustrated one by one.

[0057] 108. Model compilation;

[0058] The model compilation here refers to compiling the model (including the model mentioned in 105 above, which may be a single, undecomposed model (not needed when hardware configuration is high), or it may be six decomposed sub-models, or several decomposed and merged higher-level sub-models; the specific format is not limited and depends on actual needs) into the form required by the digital vehicle assembly virtual development platform. The compiled format may differ for different virtual development platform systems. In this embodiment, because the integrated platform is the dSPACE system, the compiled format is the dSPACE system requirement format. If the system is the ETAS system or another system, it needs to be compiled into the format required by that system.

[0059] 109. Construct the matching project files required to integrate the full model into the digital vehicle assembly virtual development platform;

[0060] After completing the 108 tasks described above, it is necessary to further construct matching engineering files required for integrating the entire model into the digital vehicle assembly virtual development platform, so that the model has the environment required for hardware-in-the-loop operation. The matching engineering files are engineering files that, through hardware and software, convert the physical / digital signals received and emitted by the real control unit into digital / physical signals to meet the environment required for the model and actuators to operate normally in a hardware-in-the-loop state. In this embodiment, after the completed model is integrated into the digital vehicle assembly virtual development platform, it will receive instructions from the real ECU control unit and feed back the signals required by the ECU. Simultaneously, the ECU will also issue instructions to control the real injectors, throttle, and other hardware actions. In this embodiment, for the fuel injection quantity signal required by the model, the ECU determines the current required fuel injection quantity based on the engine status fed back by the model and the received driver demand, and generates an electrophysical signal to activate the real injectors. At this time, the generated electrical signal is monitored and converted into a digital signal, which is synchronously transmitted to the model to activate it. The matching engineering files should meet all the signals required by the hardware-in-the-loop system formed by the model and external hardware; the conversion process of other signals will not be elaborated further. Obviously, the signal conversion content and matching project files described here will vary depending on the type of hardware actually connected to the hardware-in-the-loop system.

[0061] 110. Model integration testing;

[0062] After completing 109 tasks, the model (including the model mentioned in 105 above, which may be a single, undecomposed model (not needed when hardware configuration is high), or six decomposed sub-models, or several decomposed and merged higher-level sub-models; the specifics are not limited and depend on actual needs) can be integrated into the digital vehicle assembly virtual development platform and connected to the real ECU control unit and other actuators to form a hardware-in-the-loop system and conduct online model testing. Due to the content described in 102 above, the model integration stage may encounter problems such as physical memory exceeding limits due to excessively large models, poor real-time performance due to insufficient CPU performance, or runtime failures. If such problems occur, model decomposition and reconstruction operations as described in 102-106 need to be performed based on the actual situation.

[0063] 111. The hardware-in-the-loop system based on the virtual development platform tested normally;

[0064] After completing the 110 tasks described above, the entire model can be tested online (integrating the real hardware control unit and actuator states). For example, in the vehicle thermal management model described in this embodiment, the model can be run on the digital vehicle assembly virtual development platform using commands issued by AVL PUMA to execute vehicle test cycles such as CHTC and C-WTVC, testing whether the actions, logic, and accuracy during driving can continuously perform normally and meet accuracy requirements. Clearly, AVL PUMA is only one control system software of the digital vehicle assembly virtual development platform; any other controllable software is applicable to this method.

[0065] 112. The continuously debugged model performs normally in the online environment;

[0066] If the test model in item 111 cannot perform normally continuously in the hardware-in-the-loop state, the model or the matching project file described in item 110 needs to be continuously debugged until the model logic and accuracy are normal.

[0067] 113. Develop a thermal management platform that simultaneously includes a high-precision vehicle thermal management model and a real-time hardware control unit;

[0068] After completing 112 debugging tasks, or confirming in 111 that the model continuously performs normally in the hardware-in-the-loop state, task 1113 can be performed to solidify the model and related matching engineering files, forming a thermal management development platform that simultaneously contains a high-precision vehicle thermal management model and a real-time hardware control unit. This platform enables the development and simulation verification of high-precision vehicle thermal management strategies based on a digital vehicle powertrain virtual development platform, efficiently supporting engine development.

[0069] The above description is a detailed description of the preferred embodiments of the present invention. However, the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications made under the technical spirit of the present invention should fall within the patent scope covered by the present invention.

Claims

1. A method for compiling and integrating a vehicle thermal management model, characterized in that, include: A high-precision offline simulation model of vehicle thermal management is obtained and decomposed into several sub-models; signal interaction channels between the sub-models and other input / output signal channels of the sub-models are established. The sub-model is reconstructed based on real-time performance and physical memory, and the whole vehicle model is reconstructed. The reconstructed whole model is tested offline, and the whole model that passes the test is compiled to adapt to the corresponding development platform. The whole model is built into the matching project files required by the digital vehicle assembly virtual development platform, and the model integration test is performed. A thermal management working development platform containing both a high-precision whole vehicle thermal management model and a real-time hardware control unit is formed.

2. The method for compiling and integrating a vehicle thermal management model according to claim 1, characterized in that, The sub-models include an engine model, a vehicle model, a thermal network model, an engine compartment model, and a cooling and lubrication network model, which are used to constitute the full model. Each sub-model undergoes individual or joint calibration to meet certain accuracy requirements and enable the vehicle thermal management offline simulation model to accurately simulate the calculations corresponding to vehicle thermal management in an offline state. It can generate corresponding actions based on ECU control commands, calculate and feedback the signals required by the ECU in real time, and, based on the corresponding calibration, calculate the performance results under different actions according to the ECU control signals.

3. The method for compiling and integrating a vehicle thermal management model according to claim 2, characterized in that, The establishment of signal interaction channels between the sub-models and other input / output signal channels of the sub-models includes: the sub-models are integrated into the virtual development platform, and the interaction between the models is realized through the project configuration file in the virtual development platform; other input / output signals include interaction signals between the virtual development platform and the ECU control unit, and monitoring signals; correspondingly, the establishment of the other input / output signal channels changes the calculation control logic in the model from offline control to online control.

4. The method for compiling and integrating a vehicle thermal management model according to claim 3, characterized in that, The reconstruction of the sub-model based on real-time performance and physical memory, and the subsequent reconstruction of the whole vehicle model, includes: dividing the sub-model into a separate platform system, wherein the shared components are shared component modules used for interconnection between two sub-models that cannot be divided into a specific platform system, and the platform system is the hardware and software environment corresponding to the virtual development platform; after confirming the number of sub-model decompositions according to the actual configuration of the virtual development platform, splitting and reconstructing the shared components between the sub-models, and modifying the corresponding signals; constructing different sub-models into higher-level sub-models, and recompiling and integrating the sub-models and / or the higher-level sub-models into the virtual development platform, so that the recompiled full model and the undecomposed full model obtain essentially the same calculation results under the same calculation conditions.

5. The method for compiling and integrating a vehicle thermal management model according to claim 4, characterized in that, The offline testing of the reconstructed full model is used to test whether the structure and signal transmission are correct and whether there are any internal operational errors. The process of compiling the tested full model to adapt it to the corresponding development platform includes: compiling the full model into the required format based on the type of the virtual development platform.

6. The method for compiling and integrating a vehicle thermal management model according to claim 5, characterized in that, The process of constructing the full model to the digital vehicle assembly virtual development platform and performing model integration testing includes: converting the physical / digital signals received and emitted by the real control unit into digital / physical signals through hardware and software to create engineering files that meet the environment required for the model and actuators to operate normally in a hardware-in-the-loop state; integrating the full model into the digital vehicle assembly virtual development platform and connecting it to the real ECU control unit and other actuators to form a hardware-in-the-loop system and conducting online model testing.

7. The method for compiling and integrating a vehicle thermal management model according to claim 6, characterized in that, The process of performing model integration testing also includes: if the physical memory of the digital vehicle assembly virtual development platform exceeds the limit or the CPU performance is insufficient due to the specifications of the full model, resulting in poor real-time performance or running failure, then the whole vehicle model is rebuilt.

8. The method for compiling and integrating a vehicle thermal management model according to claim 7, characterized in that, The process of performing model integration testing also includes: if the full model continuously exhibits abnormal behavior in a hardware-in-the-loop state, then adjusting the full model and / or the matching project file.

9. The method for compiling and integrating a vehicle thermal management model according to claim 8, characterized in that, The thermal management development platform that simultaneously includes a high-precision vehicle thermal management model and a real-time hardware control unit includes: if the full model continuously performs normally in the hardware-in-the-loop state, then the full model and the matching engineering file are solidified.

10. A vehicle thermal management model compilation and integration system, characterized in that, include: The first module is used to acquire a high-precision offline simulation model of the vehicle's thermal management and decompose it into several sub-models; The second module is used to establish signal interaction channels between the sub-models, as well as other input and output signal channels of the sub-models; The third module is used to rebuild the sub-model based on real-time performance and physical memory, and to rebuild the whole vehicle model; the fourth module is used to perform offline testing on the reconstructed whole vehicle model, and to compile the whole vehicle model that passes the test to adapt it to the corresponding virtual development platform. The fifth module is used to build the matching engineering files required for the full model to the digital vehicle assembly virtual development platform and to perform model integration testing; the sixth module is used to form a thermal management work development platform that simultaneously contains a high-precision whole vehicle thermal management model and a real-time hardware control unit.

Images