Control strategy verification method and device, electronic equipment and computer readable storage medium

Abstract

The application provides a control strategy verification method and device, electronic equipment and a computer readable storage medium. The method comprises: obtaining real state data of a real vehicle, wherein the real state data is state data generated by applying a real control strategy to the real vehicle under a target working condition; constructing a whole vehicle dynamics model under the target working condition based on the real state data; performing simulation testing on the whole vehicle dynamics model based on the real control strategy to obtain a simulation testing result; when the simulation testing result is simulation success, performing simulation on the whole vehicle dynamics model based on a target control strategy to obtain first simulation state data; and determining a verification result of the target control strategy based on the real state data and the first simulation state data. Through the application, the verification result of the target control strategy under the target working condition can be determined.

Description

Technical Field

[0001] This application relates to the field of new energy vehicle technology, and in particular to a control strategy verification method, apparatus, electronic device and computer-readable storage medium. Background Technology

[0002] New energy vehicles refer to vehicles that use non-traditional fuels as a power source or use advanced technologies to improve the efficiency of traditional fuel use. Software simulation testing (SIL / HIL) has limitations in simulating the coordinated operation of the range extender and power battery under target operating conditions (such as extreme low temperature environments (minus 20 degrees Celsius)). For example, in extreme low temperature environments, due to the low-temperature performance degradation of the power battery and the combustion instability of the range extender, the control strategy of new energy vehicles is difficult to accurately control the actual charging and discharging power of the power battery, which can easily lead to overcharging and over-discharging of the battery, resulting in battery power degradation and seriously affecting the vehicle's power and safety performance. Therefore, related technologies often use real vehicles to test the interaction between the range extender and power battery under target operating conditions, which is costly and difficult. Summary of the Invention

[0003] This application provides a control strategy verification method, apparatus, electronic device, and computer-readable storage medium, which can determine the verification result of the target control strategy under the target operating condition.

[0004] The technical solution of this application embodiment is implemented as follows:

[0005] This application provides a control strategy verification method, the method comprising:

[0006] Acquire real-world state data of a real vehicle, wherein the real-world state data is the state data generated when the real control strategy is applied to the real vehicle under the target operating condition;

[0007] Based on the real-state data, a vehicle dynamics model under the target operating condition is constructed;

[0008] Based on the actual control strategy, the vehicle dynamics model was simulated and tested to obtain simulation test results;

[0009] When the simulation test result is successful, the vehicle dynamics model is simulated based on the target control strategy to obtain the first simulation state data, wherein the target control strategy is a vehicle control strategy to be verified and not embedded in the real vehicle.

[0010] Based on the real-state data and the first-simulated-state data, the verification result of the target control strategy is determined. This application embodiment provides a control strategy verification device, the device comprising:

[0011] The data acquisition module is used to acquire real state data of a real vehicle, wherein the real state data is the state data generated when the real control strategy is applied to the real vehicle under the target working condition.

[0012] The model building module is used to build a vehicle dynamics model under the target working condition based on the real state data.

[0013] The simulation test module is used to perform simulation tests on the vehicle dynamics model based on the actual control strategy and obtain simulation test results.

[0014] The simulation test module is further configured to, when the simulation test result is a successful simulation, simulate the vehicle dynamics model based on the target control strategy to obtain first simulation state data, wherein the target control strategy is a vehicle control strategy to be verified and not embedded in the real vehicle.

[0015] A performance verification model is used to determine the verification result of the target control strategy based on the real state data and the first simulation state data.

[0016] This application provides an electronic device, the electronic device comprising:

[0017] Memory is used to store executable instructions for a computer;

[0018] The processor, when executing computer-executable instructions stored in the memory, implements the control strategy verification method provided in the embodiments of this application.

[0019] This application provides a computer-readable storage medium storing a computer program or computer-executable instructions, which, when executed by a processor, implements the control strategy verification method provided in this application.

[0020] This application provides a computer program product, including a computer program or computer executable instructions. When the computer program or computer executable instructions are executed by a processor, they implement the control strategy verification method provided in this application.

[0021] The embodiments of this application have the following beneficial effects:

[0022] By constructing a vehicle dynamics model using real-world data under target operating conditions, and simultaneously employing a realistic control strategy to simulate and test the vehicle dynamics model, the accuracy of the vehicle dynamics model's simulation of the real vehicle is further verified. This ensures the accurate simulation of the real vehicle under target operating conditions. Based on this, the vehicle dynamics model is simulated using the target control strategy to obtain the first simulation state data. By comparing the real-world data with the first simulation state data, the verification results of the target control strategy under the target operating conditions are determined. This allows for the verification of the target control strategy of the real vehicle under target operating conditions without conducting actual vehicle testing, thereby reducing the cost and difficulty of vehicle testing and improving the safety performance of the real vehicle. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the architecture of the control strategy verification system provided in the embodiments of this application;

[0024] Figure 2 This is a schematic diagram of the structure of an electronic device for control strategy verification provided in an embodiment of this application;

[0025] Figure 3 This is a first flowchart illustrating the control strategy verification method provided in this application embodiment;

[0026] Figure 4 This is a schematic diagram of the second process of the control strategy verification method provided in the embodiments of this application;

[0027] Figure 5 This is a schematic diagram of the third process of the control strategy verification method provided in the embodiments of this application;

[0028] Figure 6 This is a schematic diagram of the fourth process of the control strategy verification method provided in the embodiments of this application;

[0029] Figure 7 This is a schematic diagram of the fifth process of the control strategy verification method provided in the embodiments of this application;

[0030] Figure 8 This is a schematic diagram of the control strategy verification method provided in the embodiments of this application.

[0031] It should be noted that the terms "first" and "second" mentioned above are only used to distinguish between different options and do not represent the degree of superiority or inferiority of the options or their priority in the implementation process. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0034] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0035] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0036] Unless otherwise defined, all technical and scientific terms used in the embodiments of this application have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in the embodiments of this application is for the purpose of describing the embodiments of this application only and is not intended to limit this application.

[0037] In the implementation of this application, the collection and processing of relevant data should strictly comply with the requirements of relevant national laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0038] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0039] 1) Vehicle dynamics model: This is a mathematical model used to simulate the motion state and response of a vehicle in a real environment. It can be used to predict and analyze the performance of a vehicle under various driving conditions, including handling stability, ride comfort, braking performance and acceleration performance.

[0040] 2) Battery Model: The battery model is a sub-model in the vehicle dynamics model used to simulate battery performance. It describes the electrochemical characteristics of the battery, including the charging and discharging process, state of charge (SOC), voltage, current, power output, internal resistance, and temperature rise and temperature effects.

[0041] 3) Range Extender Model: The range extender model is a sub-model in the vehicle dynamics model used to simulate the performance of the range extender. It describes the working principle, energy output, and interaction with the battery and the vehicle power system of the range extender.

[0042] 4) Control Strategy: The control strategy, also known as the vehicle control strategy model, refers to a series of algorithms and rules used to manage and optimize vehicle performance, safety, and fuel efficiency. The control strategy is usually designed by the vehicle manufacturer and embedded in the vehicle's electronic control unit (ECU) to achieve intelligent control of the vehicle's power system, transmission system, braking system, etc. The electronic control unit refers to a highly specialized computer system that is responsible for controlling multiple electronic systems of the vehicle.

[0043] 5) Software Simulation Testing (SIL / HIL): Software simulation testing (SIL / HIL) is two important stages in the development and testing process of automotive electronic control units (ECUs). They refer to Software in the Loop (SIL) testing and Hardware in the Loop (HIL) testing, respectively.

[0044] 6) Actual battery power: Actual battery power refers to the actual power output of the battery at a specific moment. The actual battery power value depends on the current operating state of the battery, including factors such as the battery's state of charge (SOC), current, voltage, and temperature.

[0045] 7) Battery available power: Battery available power refers to the maximum continuous power that a battery can safely and reliably output under specific conditions. Battery available power is usually determined by the battery's design parameters and the manufacturer's specifications.

[0046] 8) Range extender: The range extender is a key component in range-extended electric vehicles. It mainly consists of an engine and a generator. Its core function is to charge the electric vehicle's battery by converting the mechanical energy of the engine into electrical energy when the battery is low.

[0047] 9) Range extender torque: Range extender torque refers to the torque generated by the range extender when it generates electricity.

[0048] This application provides a control strategy verification method, apparatus, electronic device, computer-readable storage medium, and computer program product, which can determine the verification result of the target control strategy under the target operating condition.

[0049] The electronic device for control strategy verification provided in this application can be various types of terminals or servers. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smart TV, smartwatch, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited herein.

[0050] See Figure 1 , Figure 1 This is a schematic diagram of the architecture of the control policy verification system provided in this application embodiment. In the control policy verification system 10 provided in this application embodiment, in order to support a control policy verification application, the terminal 400 connects to the server 200 through the network 300. The network 300 can be a wide area network or a local area network, or a combination of the two.

[0051] Terminal 400 can be used to obtain control policy verification requests containing real state data of real vehicles.

[0052] In some embodiments, a control strategy verification plugin may be embedded in the client running in the terminal 400 to implement the control strategy verification method locally on the client. For example, the terminal 400 calls the control strategy verification plugin to implement the control strategy verification method. Based on the target operating condition, the real control strategy is applied to the real state data generated by the real vehicle to construct the whole vehicle dynamics model under the target operating condition. Based on the real control strategy, the whole vehicle dynamics model is simulated and tested to obtain the simulation test results. When the simulation test result is successful, the whole vehicle dynamics model is simulated based on the target control strategy to obtain the first simulation state data. Based on the real state data and the first simulation state data, the verification result of the target control strategy is determined.

[0053] In some embodiments, after the terminal 400 obtains a control strategy verification request containing real state data of a real vehicle, it calls the control strategy verification interface of the server 200 (which can be provided as a cloud service). The server 200 implements a control strategy verification method through a control strategy verification plugin. Based on the target operating condition, it applies the real control strategy to the real state data generated by the real vehicle to construct a vehicle dynamics model under the target operating condition. Based on the real control strategy, it performs simulation testing on the vehicle dynamics model to obtain simulation test results. When the simulation test result is successful, it simulates the vehicle dynamics model based on the target control strategy to obtain first simulation state data. Based on the real state data and the first simulation state data, it determines the verification result of the target control strategy and returns the verification result of the target control strategy to the terminal 400.

[0054] See Figure 2 , Figure 2 This is a schematic diagram of the structure of an electronic device for control strategy verification provided in an embodiment of this application. Figure 2 The electronic device 500 shown can be Figure 1 The terminal 400 or server 200 in the electronic device 500 includes at least one processor 510, a memory 550, at least one network interface 520, and a user interface 530. The various components in the electronic device 500 are coupled together via a bus system 540. It is understood that the bus system 540 is used to implement communication between these components. In addition to a data bus, the bus system 540 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 2 The general labeled all buses as Bus System 540.

[0055] The processor 510 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0056] User interface 530 includes one or more output devices 531 that enable the presentation of media content, including one or more speakers and / or one or more visual displays. User interface 530 also includes one or more input devices 532, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0057] The memory 550 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 550 may optionally include one or more storage devices physically located away from the processor 510.

[0058] The memory 550 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 550 described in this application embodiment is intended to include any suitable type of memory.

[0059] In some embodiments, memory 550 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, as illustrated below.

[0060] Operating system 551 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, driver layer, etc., for implementing various basic business functions and handling hardware-based tasks;

[0061] The network communication module 552 is used to reach other computing devices via one or more (wired or wireless) network interfaces 520, exemplary network interfaces 520 including: Bluetooth, WiFi, and Universal Serial Bus (USB), etc.

[0062] Presentation module 553 is configured to enable the presentation of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 531 (e.g., a display screen, a speaker, etc.) associated with user interface 530;

[0063] The input processing module 554 is used to detect and translate one or more user inputs or interactions from one or more input devices 532.

[0064] In some embodiments, the apparatus provided in this application can be implemented in software. Figure 2A control strategy verification device 555 stored in memory 550 is shown. This device can be software in the form of programs and plug-ins, and includes the following software modules: a data acquisition module 5551, a model building module 5552, a simulation testing module 5553, and a performance verification module 5554. These modules are logically connected and can therefore be arbitrarily combined or further separated according to the functions they implement. The functions of each module will be described below.

[0065] In other embodiments, the apparatus provided in this application can be implemented in hardware. As an example, the apparatus provided in this application can be a processor in the form of a hardware decoding processor, which is programmed to execute the control strategy verification method provided in this application. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0066] As mentioned above, the electronic device implementing the control strategy verification method of this application embodiment can be a terminal, a server, or a combination of both. Therefore, the executing entity of each step will not be described again below. See [link to relevant documentation]. Figure 3 , Figure 3 This is a first flowchart illustrating the control strategy verification method provided in this application embodiment, which will be combined with... Figure 3 The steps shown are explained.

[0067] In step 101, the real state data of the real vehicle is obtained.

[0068] As an example, a real vehicle refers to an actual physical vehicle, real state data is the state data generated when the real control strategy is applied to a real vehicle under the target operating condition, the real control strategy is the vehicle control strategy embedded in the real vehicle, the target operating condition refers to the target operating state of the vehicle or the target operating environment of the vehicle, and the real state data includes multiple state data arranged in chronological order, and the state data refers to the vehicle performance data and vehicle operating status data during vehicle operation.

[0069] The target operating condition is a condition in which the duration of the vehicle's target state exceeds 100 milliseconds (ms). For example, the target state can be that the lowest temperature of the vehicle's power battery cells is below -20 degrees Celsius, and the actual power of the vehicle's power battery is greater than or equal to 98% of the maximum discharge power. For example, if the sampling period of the vehicle state data is 10ms, and the lowest temperature of the vehicle's power battery cells is below -20 degrees Celsius, then if the actual power of the battery is greater than or equal to 98% of the maximum discharge power at at least 10 consecutive sampling points, then the current vehicle meets the target operating condition.

[0070] In some embodiments, step 101 can be implemented by the following steps: acquiring real vehicle working data under target working conditions; filtering the working data based on the duration information of the working data to obtain filtered working data; and determining real state data based on the filtered working data.

[0071] As an example, the system receives real vehicle operating data under target conditions sent by the vehicle-side triggering program. The operating data consists of vehicle status data received in a single, chronological order. Based on the duration information of the operating data, operating data with a duration less than a duration threshold is filtered out to obtain the filtered operating data. The filtered operating data is stored based on the reception time of the filtered operating data. The filtered operating data is then sequentially determined as real status data according to the chronological order of the reception time. The duration threshold can be a threshold preset based on experience or a threshold predicted based on a duration threshold prediction model, and is not limited here.

[0072] For example, the duration threshold can be set to 20 seconds (s). With a sampling period of 10ms for vehicle status data, work data containing less than 2000 vehicle status data entries is filtered out. Based on the reception time of the filtered work data, the filtered work data from different dates are stored in different file paths. In the order of the reception time of the work data, the work data in different file paths are sequentially determined as the real status data.

[0073] In some embodiments, the real vehicle is equipped with a data collection program, a working condition detection program, and a vehicle-side triggering program. The data collection program is used to collect real-time working data of the real vehicle. The "working data of the real vehicle under the target working condition" is acquired and sent by the real vehicle in the following manner: based on the working condition detection program, the operating status and operating environment of the real vehicle are detected. When the operating status and operating environment of the real vehicle are detected to meet the target working condition, the target working condition flag bit of the real vehicle is activated. In response to the trigger signal reception, the vehicle-side triggering program sends the working data of the real vehicle.

[0074] As an example, the trigger signal can be a timed trigger signal, which executes the trigger program based on a preset time interval or a preset time point. For example, the vehicle-side trigger program executes the "send real vehicle working data" program every hour, provided that the target working condition flag of the real vehicle is activated.

[0075] In this embodiment, in order to ensure the learning accuracy of the subsequent vehicle power model construction algorithm, the duration of the working data is used for filtering, thereby eliminating short-lived or abnormal working data. At the same time, the process of acquiring and sending working data based on trigger signals is highly automated, which improves the efficiency and accuracy of data transmission. Different application scenarios and technical requirements can be met by setting different duration thresholds and trigger conditions.

[0076] In step 102, a vehicle dynamics model under the target operating condition is constructed based on real-state data.

[0077] As an example, the vehicle dynamics model is a mathematical model used to simulate the motion state and response of a vehicle in a real environment. The vehicle dynamics model can include multiple sub-models such as the range extender model, battery model, transmission system model, suspension system model, and steering system model. Real-state data includes range extender characteristic data, real range extender torque, battery characteristic data, and real battery available power. Among them, range extender characteristic data refers to the relevant parameters of the range extender during operation, battery characteristic data refers to the relevant parameters of the battery during operation, real range extender torque is the torque generated by the range extender when generating electricity in the real-state data, and real battery available power is the actual output power of the battery in the real-state data.

[0078] In some embodiments, real-state data includes range extender characteristic data, battery characteristic data, and actual battery power; see [link to relevant documentation]. Figure 4 , Figure 4 This is a schematic diagram of the second process of the control strategy verification method provided in the embodiments of this application. Figure 3 Step 102 shown can be implemented through steps 1021 to 1024, which are explained in detail below.

[0079] In step 1021, an initial range extender model is constructed based on the range extender feature data.

[0080] As an example, range extender feature data may include range extender speed, injection pulse width, ignition advance angle, and engine coolant temperature. The range extender model is a mathematical model used to describe the relationship between range extender feature data and range extender torque. The Limiting Gradient Boosting algorithm (XGBoost algorithm) can be used to select range extender speed, injection pulse width, ignition advance angle, and engine coolant temperature as features, and range extender torque as the target value to fit the initial range extender model. When the difference between the simulated range extender torque output by the initial range extender model and the actual range extender torque is less than a third threshold, the initial range extender model is successfully constructed. The third threshold can be a threshold preset based on experience. The difference between the simulated range extender torque and the actual range extender torque includes mean square error, root mean square error, absolute error, and mean absolute error, etc., which are not limited here.

[0081] When the difference between the simulated range extender torque output by the initial range extender model and the actual range extender torque is greater than or equal to the third threshold, the initial range extender model construction fails, and the control strategy verification method of this application embodiment is re-executed based on the new real state data.

[0082] In step 1022, the actual power of the first simulated battery is determined based on the simulated range extender torque and range extender characteristic data output from the initial range extender model.

[0083] As an example, the mechanical power of the range extender is determined based on the product of the range extender speed and the simulated range extender torque included in the range extender characteristic data. The product of the mechanical power of the range extender and the power generation efficiency of the range extender system is determined as the actual power of the first simulated battery. The power generation efficiency of the range extender system is the energy conversion efficiency of the range extender in the process of converting the chemical energy of fuel into electrical energy.

[0084] In step 1023, an initial battery model is constructed based on battery characteristic data and the actual power of the first simulated battery.

[0085] As an example, battery characteristic data can include battery state of charge (SOC) and cell temperature, etc. The battery model is a mathematical model used to describe the relationship between battery characteristic data and battery available power. The XGBoost algorithm can be used to select battery SOC, cell temperature, and the actual power of the first simulated battery as features, and battery available power as the target value to fit the initial battery model. When the difference between the simulated battery available power output by the initial battery model and the actual battery available power is less than a fourth threshold, the initial battery model is successfully constructed. The fourth threshold can be a threshold preset based on experience. The difference between the simulated battery available power and the actual battery available power includes mean square error, root mean square error, absolute error, and mean absolute error, etc., which are not limited here.

[0086] When the difference between the simulated battery power output by the initial battery model and the actual battery power is greater than or equal to the fourth threshold, the initial battery model construction fails, and the control strategy verification method of this application embodiment is re-executed based on the new real state data.

[0087] In step 1024, when the difference between the actual power of the first simulated battery and the actual power of the real battery is less than a first threshold, a vehicle dynamics model is constructed based on the initial range extender model and the initial battery model.

[0088] As an example, the real state data includes T first state data arranged in chronological order. The first state data is the state data at any time of the real state data. The first state data includes the actual power of the battery and the available power of the battery. The actual power of the real battery is the actual power of the battery in the Tth first state data of the real state data. The difference between the actual power of the first simulated battery and the actual power of the real battery includes mean square error, root mean square error, absolute error and mean absolute error, etc., which are not limited here.

[0089] Taking the mean absolute error as an example, the mean absolute error between the actual power of the first simulated battery and the actual power of the real battery is determined. When the mean absolute error is less than the first threshold, the initial range extender model and the initial battery model are determined as the range extender model and the battery model for constructing the vehicle dynamics model, respectively. The determination of the other sub-models of the vehicle dynamics model other than the range extender model and the battery model is not limited in this application embodiment. The first threshold can be a threshold preset based on experience.

[0090] When the mean absolute error is greater than or equal to the first threshold, the construction of the vehicle dynamics model fails. Based on the new real state data, the control strategy verification method of this application embodiment is re-executed.

[0091] In this embodiment, the extreme gradient boosting algorithm is used to accurately construct the range extender model and the battery model based on the range extender feature data and battery feature data. Then, based on the initial range extender model and the initial battery model, a vehicle dynamics model is constructed. By comparing the difference between the actual power of the first simulated battery and the actual power of the real battery, the difference between the vehicle dynamics model and the real vehicle is quantified, thereby improving the reliability of the vehicle dynamics model and providing a solid foundation for the simulation test of the subsequent control strategy.

[0092] In step 103, the vehicle dynamics model is simulated and tested based on the real control strategy to obtain the simulation test results.

[0093] As an example, the real control strategy is the vehicle control strategy embedded in a real vehicle.

[0094] In some embodiments, real-state data includes the actual power of the battery. Figure 3 Step 103 shown can be implemented through the following steps: based on the real control strategy, simulate the vehicle dynamics model to obtain the second simulation state data; based on the second simulation state data, determine the actual power of the second simulation battery; when the difference between the actual power of the second simulation battery and the actual power of the real battery is less than a second threshold, the simulation is determined to be successful as the simulation test result; when the difference between the actual power of the second simulation battery and the actual power of the real battery is greater than or equal to the second threshold, the simulation is determined to be unsuccessful as the simulation test result.

[0095] As an example, the specific implementation method of simulating the vehicle dynamics model based on the real control strategy to obtain the second simulation state data can be found in step 104 below, and will not be elaborated here.

[0096] The second simulation state data includes T fifth state data arranged in chronological order. The actual battery power of the fifth state data at time T of the second simulation state data is determined as the actual power of the second simulated battery. When the difference between the actual power of the second simulated battery and the actual power of the real battery is less than a second threshold, the simulation is determined as a simulation test result. The second threshold can be a threshold preset based on experience. The difference between the actual power of the second simulated battery and the actual power of the real battery includes mean square error, root mean square error, absolute value error, and mean absolute error, etc., which are not limited here.

[0097] When the difference between the actual power of the second simulated battery and the actual power of the real battery is greater than or equal to the second threshold, the simulation failure is determined as the simulation test result, and the control strategy verification method of this application embodiment is re-executed based on the new real state data.

[0098] In this embodiment, the vehicle dynamics model is simulated based on a real control strategy to obtain second simulation state data. By comparing the difference between the actual power of the second simulated battery and the actual power of the real battery, the accuracy of the vehicle dynamics model in simulating the real vehicle is further verified.

[0099] In step 104, when the simulation test result is successful, the vehicle dynamics model is simulated based on the target control strategy to obtain the first simulation state data.

[0100] As an example, the target control strategy is the control strategy of a vehicle that is to be verified and is not embedded in a real vehicle. The first simulation state data is the state data generated by applying the target control strategy to the vehicle dynamics model under the target operating condition. When the simulation test result is that the simulation is successful, it indicates that the vehicle dynamics model can accurately simulate the state and response of a real vehicle under the target operating condition. Based on the target control strategy, the vehicle dynamics model is simulated to obtain the first simulation state data.

[0101] In some embodiments, the real-state data includes T first-state data points arranged in chronological order. The first-state data includes the actual battery power of the real vehicle and the available battery power of the real vehicle, where T is an integer greater than 1. See [link to documentation]. Figure 5 , Figure 5 This is a schematic diagram of the third process of the control strategy verification method provided in the embodiments of this application. Figure 3 Step 104 shown can be implemented through steps 1041 to 1044, which are explained in detail below.

[0102] In step 1041, the first state data at the first moment in the real state data is determined as the second state data at the first moment output by the vehicle dynamics model.

[0103] The second state data includes the battery's actual power and the battery's available power.

[0104] As an example, the vehicle dynamics model is initialized based on the first state data at the first moment in the real state data, and the first state data at the first moment is used as the second state data at the first moment output by the vehicle dynamics model.

[0105] In step 1042, based on the difference between the actual battery power at time t and the available battery power at time t output by the vehicle dynamics model, the control command for the target control strategy at time t+1 is determined.

[0106] As an example, the second state data includes the actual battery power and the available battery power. Based on the difference between the actual battery power at time t and the available battery power at time t output by the vehicle dynamics model, the control command of the target control strategy at time t+1 is determined, where 1≤t<T.

[0107] In some embodiments, step 1042 can be implemented by the following steps: when the actual battery power at time t is less than or equal to the available battery power at time t, the range extender power increase instruction is determined as the control instruction at time t+1; when the actual battery power at time t is greater than the available battery power at time t, the range extender power decrease instruction is determined as the control instruction at time t+1.

[0108] Among them, the range extender power increase command is used to instruct the range extender power to be increased, and the range extender power decrease command is used to instruct the range extender power to be decreased.

[0109] As an example, when the actual battery power at time t is less than or equal to the battery's available power at time t, it indicates that the battery is in a normal discharge state, and the range extender power increase instruction is determined as the control instruction at time t+1, or the control instruction at time t is determined as the control instruction at time t+1; when the actual battery power at time t is greater than the battery's available power at time t, it indicates that the battery is in an overcharge state, and the range extender power decrease instruction is determined as the control instruction at time t+1.

[0110] In some embodiments, the difference between the actual battery power at time t and the available battery power at time t is determined; when the product of the difference and the battery state of charge is less than the power threshold, the range extender power increase instruction is determined as the control instruction at time t+1; when the product of the difference and the battery state of charge is greater than or equal to the power threshold, the range extender power decrease instruction is determined as the control instruction at time t+1.

[0111] As an example, the battery state of charge (SOC) is obtained from the real-state data. SOC represents the current remaining battery capacity. When the product of the difference and the battery SOC is less than the power threshold, it indicates that the battery is in a normal discharge state or in an overcharge state, and the current remaining battery capacity is low. Therefore, the range extender power increase command is determined as the control command at time t+1. When the product of the difference and the battery SOC is greater than or equal to the power threshold, it indicates that the battery is in an overcharge state. The range extender power decrease command is determined as the control command at time t+1. The power threshold can be a threshold preset based on experience or a threshold predicted based on a power threshold prediction model. No limitation is imposed here.

[0112] In step 1043, based on the control command at time t+1, the vehicle dynamics model is controlled to perform simulation, and the second state data at time t+1 output by the vehicle dynamics model is obtained.

[0113] As an example, the increase and decrease of the range extender power can be achieved by adjusting the parameters of the range extender characteristic data. When the control command at time t+1 is to increase the range extender power, the range extender speed can be increased to control the range extender power of the vehicle dynamics model to increase. Alternatively, when the control command at time t+1 is to decrease the range extender power, the fuel injection pulse width can be reduced to control the range extender power of the vehicle dynamics model to decrease. This allows the vehicle dynamics model to be controlled for simulation, obtaining the second state data at time t+1 output by the vehicle dynamics model.

[0114] In step 1044, the second state data at time T is determined as the first simulation state data.

[0115] As an example, the second states at T time points are sequentially determined as the first simulation state data according to the time sequence.

[0116] In this embodiment, the range extender power is synchronously adjusted based on the battery's state at various times, which can effectively protect the battery, avoid overcharging and over-discharging, extend battery life, ensure energy balance between the battery and the range extender, and improve energy utilization efficiency. At the same time, the target control strategy enables accurate simulation of the vehicle dynamics model, which can accurately obtain the simulation results of vehicle state data under the target operating conditions.

[0117] In step 105, the verification results of the target control strategy are determined based on the real state data and the first simulation state data.

[0118] As an example, the first simulation state data is used to simulate the state data generated when the target control strategy is applied to a real vehicle under the target operating condition. Based on the difference in operating indicators between the real state data and the first simulation state data, the superiority or inferiority between the target control strategy and the real control strategy under the target operating condition is verified.

[0119] In some embodiments, see Figure 6 , Figure 6 This is a schematic diagram of the fourth process of the control strategy verification method provided in the embodiments of this application. Figure 3 Step 105 shown can be implemented through steps 1051 to 1052, which are explained in detail below.

[0120] In step 1051, the operating parameters of the real battery are determined based on the real state data, and the operating parameters of the simulated battery are determined based on the first simulation state data.

[0121] As an example, performance indicators may include average battery power, battery overcharge duration, and number of overcharge suppression cycles. The specific implementation methods of "determining the performance indicators of a real battery based on real state data" and "determining the performance indicators of a simulated battery based on first simulation state data" are the same. Therefore, the following text will only describe the specific implementation method of "determining the performance indicators of a simulated battery based on first simulation state data".

[0122] In some embodiments, the performance metrics include average battery power, battery overcharge duration, and number of overcharge suppression cycles. The first simulation state data includes T second state data points, where T is an integer greater than 1. See [link to documentation]. Figure 7 , Figure 7 This is a schematic diagram of the fifth process of the control strategy verification method provided in this application embodiment. "Determining the working indicators of the simulation battery based on the first simulation state data" can be achieved through the following steps 201 to 203, which will be explained in detail below.

[0123] In step 201, the average battery power is determined by summing the actual battery power of T second state data.

[0124] As an example, the sum of the actual battery power of T second-state data points is determined, and the ratio of the sum of the actual battery power of the T second-state data points to T is determined as the average battery power.

[0125] In step 202, the third and fourth state data are determined from the T second state data.

[0126] Among them, the third state data is the second state data where the actual power of the battery is greater than the available power of the battery, and the fourth state data is the second state data where the actual power of the battery is less than or equal to the available power of the battery.

[0127] As an example, based on the relationship between the actual power and the available power of the battery in the second state data, the third state data and the fourth state data in the T second state data are determined. The third state data is the state data of the battery in the overcharge state, and the fourth state data is the state data of the battery in the normal working state.

[0128] In step 203, based on the third state data and the fourth state data, the battery overcharge duration and the number of battery overcharge suppression attempts for the simulated battery are determined.

[0129] As an example, battery overcharge duration refers to the duration of battery overcharge, battery overcharge suppression count refers to the number of times the control strategy intervenes and suppresses overcharge, and overcharge state refers to the state in which the actual power of the battery is greater than the available power of the battery. The duration of the third state data is determined as the battery overcharge duration of the simulated battery, and the time of each third state data is determined as the battery overcharge time. Based on the battery overcharge time, the number of battery overcharge suppression counts is determined.

[0130] For example, if the second state data for the 100ms period before and the 100ms period after the battery overcharge event are both fourth state data, then one instance of battery overcharge suppression is achieved.

[0131] In this embodiment, by calculating key indicators such as average battery power, statistically analyzing battery overcharge duration and overcharge suppression counts, a comprehensive evaluation of the simulated battery performance is achieved, thereby providing accurate data support for determining the verification results of subsequent target control strategies.

[0132] In step 1052, the verification results of the target control strategy are determined based on the difference between the operating indicators of the real battery and the operating indicators of the simulated battery.

[0133] As an example, the verification results of the target control strategy are determined based on the difference between at least one operating metric of a real battery and the corresponding operating metric of a simulated battery.

[0134] For example, the difference between the overcharge duration of a real battery and the overcharge duration of a simulated battery, or the ratio between the overcharge duration of a real battery and the overcharge duration of a simulated battery, can be used as the verification result of the target control strategy.

[0135] In the embodiments of this application, by comparing the differences between the working indicators of real batteries and simulated batteries, not only can the improvements and shortcomings of the target control strategy compared with the real control strategy be accurately evaluated, but also data support is provided for further optimization of the target control strategy.

[0136] The following will describe an exemplary application of the embodiments of this application in a real-world application scenario.

[0137] In extreme low-temperature environments, such as below -20 degrees Celsius, the performance of new energy vehicle power batteries deteriorates, resulting in limited battery power. During vehicle acceleration, the range extender becomes the main power source. However, due to combustion instability and sluggish power response, the actual output power of the range extender is not constant. The instability of the range extender's output power, coupled with the limited battery power, makes it difficult for the vehicle controller to accurately control the actual charging and discharging power of the power battery. This can easily lead to overcharging and over-discharging of the battery, resulting in battery power degradation and seriously affecting vehicle power and safety.

[0138] Traditional software simulation testing struggles to accurately simulate the collaborative working state of the range extender and battery under such conditions, as well as the complex interactions between them. Consequently, it cannot effectively assess the risk of battery overcharging, increasing the cost and difficulty of real-vehicle testing. This makes it difficult to discover and resolve the problem in the early stages of development. In related technologies, automakers can only rely on spending a significant amount of time and manpower each year on real-vehicle cold-weather calibration to optimize such issues.

[0139] Range-extended electric vehicles (REEVs) are a type of new energy vehicle. In extremely low-temperature environments, the main challenges in REEV energy management are:

[0140] 1) Range extender combustion instability: In extreme low temperature environments, the combustion process of the range extender is easily affected by factors such as air temperature and atmospheric pressure, resulting in large fluctuations in the output power of the range extender, which is difficult to predict accurately.

[0141] 2) Battery performance degradation at low temperatures: In extreme low-temperature environments, the activity of the power battery decreases, resulting in a decrease in the battery's usable power, which limits the response speed and capacity of the range extender.

[0142] 3) Battery overcharge risk: The fluctuation of the range extender's output power, coupled with the reduction of the battery's usable power at low temperatures, can easily lead to misjudgment by the battery management system, causing battery overcharging, resulting in permanent degradation of battery power, and affecting battery life and safety.

[0143] To address the aforementioned issues, this application proposes a control strategy verification method. Relying on a cloud-based big data platform, it aggregates massive data resources to generate numerous test cases, promoting the continuous evolution and optimization of vehicle control strategies. Furthermore, it significantly shortens the verification cycle of pre-research control strategy models (equivalent to the simulation control strategies mentioned above), greatly accelerating the transformation of research results into practical applications.

[0144] See Figure 8 , Figure 8This is a schematic diagram of the control strategy verification method provided in the embodiments of this application. The key steps of the control strategy verification method of the embodiments of this application will be described below, taking the verification of the energy and power related control strategy of a range-extended vehicle under extreme low temperature environment as an example.

[0145] Step 801: Specify the data file path for the date to be verified.

[0146] Prior to step 801, the control strategy verification method of this application embodiment further includes setting target operating conditions and vehicle-side signal uploading.

[0147] Set target operating conditions: Set the target operating conditions to a minimum battery cell temperature below -20℃ and a duration of more than 100ms where the actual battery power reaches 98% of the maximum discharge power. For example, with a sampling period of 10ms, the actual battery power reaches 98% of the maximum discharge power for at least 10 consecutive sampling points.

[0148] After detecting that the target operating condition is met, the target operating condition flag is activated. The vehicle-side trigger uploads the vehicle-side data to the cloud according to the trigger signal, and stores the vehicle-side data as a discharge cycle (equivalent to the real state data mentioned above) according to the date. The discharge cycle includes range extender-related data, such as range extender speed, fuel injection pulse width, ignition advance angle, engine coolant temperature and range extender torque. The discharge cycle also includes battery-related data, such as battery state of charge (SOC), cell temperature, actual power and battery available discharge power.

[0149] Step 802: Extract data from the cloud.

[0150] Based on the data file path of the date to be verified, extract the discharge cycle data of the date to be verified from the cloud data.

[0151] Step 803, working condition screening.

[0152] After receiving the uploaded vehicle-side data, in order to ensure the learning accuracy of the subsequent Extreme Gradient Boosting (XGBoost) algorithm, the uploaded discharge cycle data itself must last for more than 20 seconds. That is, with a sampling period of 10ms, a single discharge cycle must contain at least 2000 data points, and discharge cycle data that do not meet the length requirement will be removed.

[0153] Step 804A: Prepare data by selecting range extender speed, injection pulse width, ignition advance angle, and engine coolant temperature as features, and range extender torque as the target value.

[0154] From the data that meets the target operating conditions, a single discharge cycle data is extracted. Using the XGBoost algorithm, based on the data of a single discharge cycle, the actual torque signal of the range extender output by the controlled object model of the range extender power generation system (equivalent to the vehicle dynamics model mentioned above) is predicted. That is, the actual torque model of the range extender (equivalent to the range extender model mentioned above) of the controlled object model of the range extender power generation system is fitted. Based on past engineering experience, the range extender speed, injection pulse width, ignition advance angle, and engine coolant temperature are selected as features to predict the actual torque value of the range extender.

[0155] Step 805A: Generate the characteristic polynomial of the actual torque model of the range extender.

[0156] Based on the range extender speed, injection pulse width, ignition advance angle, and engine coolant temperature, the actual torque value of the range extender is predicted, and a characteristic polynomial of the range extender model is established.

[0157] Step 806A: Train the actual torque model of the range extender.

[0158] When the difference between the predicted actual torque value of the range extender (equivalent to the simulated range extender torque mentioned above) and the actual torque value of the range extender (equivalent to the real range extender torque mentioned above) is less than the threshold, the actual torque model of the range extender is successfully fitted.

[0159] Step 804B: Prepare data by selecting battery state of charge, cell temperature, and actual power as features, and battery available discharge power as the target value.

[0160] From the data that meets the target operating conditions, data from a single discharge cycle is extracted. Based on the data from a single discharge cycle, the XGBoost algorithm is used to predict the battery's available power output from the controlled object model of the range extender power generation system (equivalent to the vehicle dynamics model mentioned above). That is, the battery's available power model (equivalent to the battery model mentioned above) of the controlled object model of the range extender power generation system is fitted. Battery SOC, cell temperature, and predicted actual battery power are selected as features, and the battery's available charging power is used as the target value for prediction. The predicted actual battery power value is calculated based on the range extender speed, predicted range extender torque, and the power generation efficiency of the range extender system.

[0161] Step 805B: Generate the characteristic polynomial of the battery's available power model.

[0162] Based on the predicted values ​​of battery SOC, cell temperature, and actual battery power, the available charging power of the battery is used as the target value for prediction, and a characteristic polynomial of the battery available power model is established.

[0163] Step 806B: Train the battery available power model.

[0164] When the difference between the predicted battery available charging power (equivalent to the simulated battery available power mentioned above) and the battery available charging power (equivalent to the actual battery available power mentioned above) is less than a threshold, the battery available power model fitting is complete.

[0165] The goal of optimizing the range extender's power generation in this embodiment is to ensure that the actual battery power is less than the available charging power, thereby avoiding the risk of overcharging. Therefore, the mean absolute error (MAE) between the predicted actual battery power and the actual battery power is used to determine the accuracy of the XGBoost algorithm's prediction. After steps 806A and 806B, this embodiment further includes the following step: If the mean absolute error between the predicted actual battery power and the actual battery power is less than a threshold θ... w If the range extender torque model and battery available power model are considered to be accurate, step 807 can be executed; otherwise, if the simulation process of the range extender torque model and battery available power model is deemed not to have converged, the discharge cycle data is discarded, and the process returns to step 802, and new single discharge cycle data is extracted again.

[0166] Step 807: Construct a closed-loop simulation environment for the range extender and power battery (simulation verification of the controlled object model of the range extender system).

[0167] Based on the original vehicle control strategy model (equivalent to the actual control strategy mentioned above), continuous simulation verification is performed on the controlled object model of the range extender system fitted above.

[0168] During continuous simulation, the data at the first moment of the discharge cycle is used as the input of the controlled object model of the range extender system at the first moment. The vehicle status signals such as the available battery power and range extender torque output by the controlled object model of the range extender system at time t are used as the input of the original vehicle control strategy model at time t+1. The original vehicle control strategy model outputs commands (such as range extender speed, fuel injection pulse width, ignition advance angle, etc.) to the controlled object model of the range extender system at time t+2, and so on, where 0 < t ≤ T.

[0169] Step 808: Determine whether the average absolute error between the predicted actual battery power and the actual battery power is less than the threshold (accuracy verification of the controlled object model of the range extender system).

[0170] By comparing real-vehicle data from the cloud with model simulation values, the average absolute error between the predicted actual battery power at time T and the actual battery power at time T of the discharge cycle data is used as a criterion for judging the prediction accuracy of the controlled object model of the range extender power generation system. If the average absolute error between the predicted actual battery power at time T and the actual battery power at time T of the discharge cycle data is less than the threshold θ, the prediction accuracy is determined. pIf the prediction accuracy of the controlled object model of the range extender power generation system is deemed acceptable, proceed to step 809; otherwise, if the cumulative error of the controlled object model of the range extender power generation system is deemed unacceptable and the accuracy verification of the controlled object model of the range extender system fails, return to step 802 to extract the data of a new single discharge cycle.

[0171] Step 809: Closed-loop simulation verification of the pre-researched control strategy model.

[0172] Extract the discharge cycle that has been verified for accuracy by the controlled object model of the range extender system.

[0173] Closed-loop simulations are performed on the pre-research control strategy model and the fitted controlled object model of the range extender system. Only the initial conditions of the simulation are provided. During the continuous simulation, the pre-research control strategy model outputs target commands to control the controlled object model of the range extender system, while the controlled object model of the range extender system simulates the response of the actual vehicle. The vehicle state signals such as the available battery power and range extender torque output by the controlled object model of the range extender system at time t will be used as the input of the pre-research control strategy model at time t+1, thus forming a closed loop. After the continuous simulation is completed, the simulation results of the pre-research control strategy are recorded.

[0174] Step 810: Calculate the battery overcharge index.

[0175] The simulation results of the pre-research control strategy are analyzed. Based on the simulation results of the pre-research control strategy under a single discharge cycle, the actual battery power prediction value obtained from the simulation is compared with the actual vehicle data to verify the effectiveness of the pre-research control strategy model.

[0176] Step 811: Determine whether the simulation verification for all dates is complete.

[0177] If all the discharge cycle data for all dates stored in the cloud have been used for simulation verification of the controlled object model of the range extender system and closed-loop simulation verification of the pre-research control strategy model, then the pre-research control strategy verification fails, and the process returns to step 802 to extract new data for a single discharge cycle. Otherwise, step 812 is executed.

[0178] Step 812: Output the results within the verification period.

[0179] By combining the simulation results of the pre-research control strategy model, and statistically analyzing the average overcharge power, average overcharge duration, and overcharge limitation index per 10,000 cycles of real vehicle data and the pre-research control strategy model, the benefits of the pre-research control strategy model in real-world environments compared to the original vehicle control strategy model are evaluated.

[0180] The calculation method for the average overcharge power and average overcharge duration of the battery is as follows: statistically analyze the data that meets the overcharge condition, and determine the average overcharge duration and average power of the charging cycle for the pre-research control strategy model and the original vehicle control strategy model respectively. The overcharge condition is the condition in which the actual power of the battery is greater than the usable power of the battery under extreme low temperature environment.

[0181] Calculation method of overcharge effective suppression index:

[0182] The overcharge point that meets the overcharge condition is determined. If the actual battery power prediction value of the pre-research control strategy model is less than the battery's available power within 100ms before and after the overcharge point, then it is considered that one overcharge behavior has been effectively suppressed. The number of times overcharge is effectively suppressed in 10,000 discharge cycles is counted.

[0183] In summary, the beneficial effects of the control strategy verification method in this application are as follows:

[0184] 1) A multi-vehicle consistency verification method based on cloud data-driven control strategy model is proposed (the control strategy verification method in this application does not limit the type of real vehicle). It collects massive amounts of data through a cloud big data platform, generates a large number of test cases, promotes the continuous optimization of the pre-research control strategy model, and accelerates the verification cycle of the pre-research control strategy model for practical application.

[0185] 2) Based on the XGBoost data fitting method, a relatively accurate range extender torque model and battery power model were established. A closed-loop environment for verifying the energy power control strategy under extreme low temperature conditions was constructed, which makes the software testing in the software simulation test (SIL / HIL) environment more accurate and flexible. At the same time, the model can be retrained based on data under different operating conditions, and can be quickly updated and expanded without having to build the model from scratch.

[0186] The following description continues to illustrate the exemplary structure of the control strategy verification device 555 provided in the embodiments of this application as a software module. In some embodiments, such as Figure 2 As shown, the software modules stored in the control strategy verification device 555 in the memory 550 may include:

[0187] The data acquisition module 5551 is used to acquire real state data of a real vehicle, wherein the real state data is the state data generated when the real control strategy is applied to the real vehicle under the target working condition.

[0188] The model building module 5552 is used to build a vehicle dynamics model under the target working condition based on the real state data.

[0189] The simulation test module 5553 is used to perform simulation tests on the vehicle dynamics model based on the real control strategy and obtain simulation test results.

[0190] The simulation test module 5553 is further configured to, when the simulation test result is a successful simulation, simulate the vehicle dynamics model based on the target control strategy to obtain first simulation state data, wherein the target control strategy is a vehicle control strategy to be verified and not embedded in the real vehicle.

[0191] The performance verification model 5554 is used to determine the verification result of the target control strategy based on the real state data and the first simulation state data.

[0192] In some embodiments, the real-state data includes range extender characteristic data, battery characteristic data, and actual battery power. The model building module 5552 is used to construct an initial range extender model based on the range extender characteristic data; determine a first simulated battery power based on the simulated range extender torque output by the initial range extender model and the range extender characteristic data; construct an initial battery model based on the battery characteristic data and the first simulated battery power; and construct the vehicle dynamics model based on the initial range extender model and the initial battery model when the difference between the first simulated battery power and the actual battery power is less than a first threshold.

[0193] In some embodiments, the real-state data includes the actual power of the real battery. The simulation test module 5553 is used to simulate the vehicle dynamics model based on the real control strategy to obtain second simulation state data; based on the second simulation state data, determine the second simulated battery actual power; when the difference between the second simulated battery actual power and the actual power of the real battery is less than a second threshold, the simulation is determined to be successful as the simulation test result; when the difference between the second simulated battery actual power and the actual power of the real battery is greater than or equal to the second threshold, the simulation is determined to be unsuccessful as the simulation test result.

[0194] In some embodiments, the real-state data includes T first state data arranged in chronological order, the first state data including the actual battery power and the available battery power of the real vehicle, where T is an integer greater than 1. The simulation test module 5553 is further configured to determine the first state data at the first moment in the real-state data as the second state data at the first moment output by the vehicle dynamics model, wherein the second state data includes the actual battery power and the available battery power; based on the difference between the actual battery power at the t-th moment and the available battery power at the t-th moment output by the vehicle dynamics model, determine the control command at the (t+1)-th moment of the target control strategy, where 1 ≤ t < T; based on the control command at the (t+1)-th moment, control the vehicle dynamics model to perform simulation to obtain the second state data at the (t+1)-th moment output by the vehicle dynamics model; and determine the second state data at the T moments as the first simulation state data.

[0195] In some embodiments, the second state data at time t includes the actual battery power at time t and the available battery power at time t. The simulation test module 5553 is further configured to determine the range extender power increase instruction as the control instruction at time t+1 when the actual battery power at time t is less than or equal to the available battery power at time t, wherein the range extender power increase instruction is used to instruct the increase of the range extender power; and to determine the range extender power decrease instruction as the control instruction at time t+1 when the actual battery power at time t is greater than the available battery power at time t, wherein the range extender power decrease instruction is used to instruct the decrease of the range extender power.

[0196] In some embodiments, the performance verification model 5554 is further configured to determine the operating indicators of the real battery based on the real state data, and to determine the operating indicators of the simulated battery based on the first simulation state data; and to determine the verification result of the target control strategy based on the difference between the operating indicators of the real battery and the operating indicators of the simulated battery.

[0197] In some embodiments, the performance metrics include average battery power, battery overcharge duration, and number of battery overcharge suppression attempts. The first simulation state data includes T second state data points, where T is an integer greater than 1. The performance verification model 5554 is further used to determine the average battery power based on the sum of the actual battery power of the T second state data points; determine third and fourth state data points among the T second state data points, wherein the third state data points are second state data points where the actual battery power is greater than the available battery power, and the fourth state data points are second state data points where the actual battery power is less than or equal to the available battery power; and determine the battery overcharge duration and the number of battery overcharge suppression attempts for the simulated battery based on the third and fourth state data points.

[0198] In some embodiments, the data acquisition module 5551 is further configured to acquire the working data of the real vehicle under the target working condition; filter the working data based on the duration information of the working data to obtain filtered working data; and determine the real state data based on the filtered working data.

[0199] This application provides a computer program product, which includes a computer program or computer-executable instructions stored in a computer-readable storage medium. The processor of an electronic device reads the computer-executable instructions from the computer-readable storage medium and executes the computer-executable instructions, causing the electronic device to perform the control strategy verification method described in this application.

[0200] This application provides a computer-readable storage medium storing computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are executed by a processor, the processor will execute the control strategy verification method provided in this application. For example, ... Figures 3 to 7 The control strategy verification method is shown.

[0201] In some embodiments, the computer-readable storage medium may be a memory such as RAM, ROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0202] In some embodiments, computer-executable instructions may take the form of programs, software, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

[0203] As an example, computer-executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., files that store one or more modules, subroutines, or code sections).

[0204] As an example, computer-executable instructions can be deployed to execute on a single electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed across multiple locations and interconnected via a communication network.

[0205] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A method for verifying control strategies, characterized in that, The method includes: Acquire real-world state data of a real vehicle, wherein the real-world state data is the state data generated when the real control strategy is applied to the real vehicle under the target operating condition; Based on the real-state data, a vehicle dynamics model under the target operating condition is constructed; Based on the actual control strategy, the vehicle dynamics model was simulated and tested to obtain simulation test results; When the simulation test result is successful, the vehicle dynamics model is simulated based on the target control strategy to obtain the first simulation state data, wherein the target control strategy is a vehicle control strategy to be verified and not embedded in the real vehicle. Based on the real state data and the first simulation state data, the verification result of the target control strategy is determined.

2. The method according to claim 1, characterized in that, The real-state data includes range extender characteristic data, battery characteristic data, and actual battery power. The process of constructing a vehicle dynamics model under the target operating condition based on the real-state data includes: Based on the range extender feature data, an initial range extender model is constructed; Based on the simulated range extender torque output from the initial range extender model and the range extender characteristic data, the actual power of the first simulated battery is determined. Based on the battery characteristic data and the actual power of the first simulated battery, an initial battery model is constructed; When the difference between the actual power of the first simulated battery and the actual power of the real battery is less than a first threshold, the vehicle dynamics model is constructed based on the initial range extender model and the initial battery model.

3. The method according to claim 1 or 2, characterized in that, The actual state data includes the actual power of the actual battery. The simulation test is performed on the vehicle dynamics model based on the actual control strategy to obtain the simulation test results, including: Based on the actual control strategy, the vehicle dynamics model is simulated to obtain the second simulation state data; Based on the second simulation state data, determine the actual power of the second simulated battery; When the difference between the actual power of the second simulated battery and the actual power of the real battery is less than the second threshold, the simulation is determined to be successful as the simulation test result. When the difference between the actual power of the second simulated battery and the actual power of the real battery is greater than or equal to the second threshold, the simulation failure is determined as the simulation test result.

4. The method according to claim 1, characterized in that, The real state data includes T first state data arranged in chronological order, the first state data including the actual battery power of the real vehicle and the available battery power of the real vehicle, where T is an integer greater than 1. The simulation of the vehicle dynamics model based on the target control strategy yields first simulation state data, including: The first state data at the first moment in the real state data is determined as the second state data at the first moment output by the vehicle dynamics model, wherein the second state data includes the actual power of the battery and the available power of the battery; Based on the difference between the actual battery power at time t and the available battery power at time t output by the vehicle dynamics model, the control command of the target control strategy at time t+1 is determined, where 1≤t<T; Based on the control command at time t+1, the vehicle dynamics model is controlled to perform simulation, and the second state data at time t+1 output by the vehicle dynamics model is obtained. The second state data at time T is determined as the first simulation state data.

5. The method according to claim 4, characterized in that, The difference between the actual battery power at time t and the available battery power at time t, based on the output of the vehicle dynamics model, determines the control command for the target control strategy at time t+1, including: When the actual battery power at time t is less than or equal to the available battery power at time t, the range extender power increase instruction is determined as the control instruction at time t+1, wherein the range extender power increase instruction is used to instruct the range extender power to be increased; When the actual power of the battery at time t is greater than the available power of the battery at time t, the range extender power reduction instruction is determined as the control instruction at time t+1, wherein the range extender power reduction instruction is used to indicate the reduction of the power of the range extender.

6. The method according to claim 1, characterized in that, The step of determining the verification result of the target control strategy based on the real state data and the first simulation state data includes: Based on the real state data, the operating indicators of the real battery are determined, and based on the first simulation state data, the operating indicators of the simulated battery are determined. The verification result of the target control strategy is determined based on the difference between the operating indicators of the real battery and the operating indicators of the simulated battery.

7. The method according to claim 6, characterized in that, The performance indicators include average battery power, battery overcharge duration, and number of battery overcharge suppression attempts. The first simulation state data includes T second state data points, where T is an integer greater than 1. The step of determining the operating parameters of the simulated battery based on the first simulation state data includes: The average power of the battery is determined by summing the actual power of the T second state data. Determine the third and fourth state data among the T second state data, wherein the third state data is the second state data in which the actual power of the battery is greater than the available power of the battery, and the fourth state data is the second state data in which the actual power of the battery is less than or equal to the available power of the battery; Based on the third state data and the fourth state data, the battery overcharge duration and the number of battery overcharge suppression attempts of the simulated battery are determined.

8. The method according to claim 1, characterized in that, The acquisition of real vehicle status data includes: Obtain the operating data of the real vehicle under the target operating conditions; Based on the duration information of the work data, the work data is filtered to obtain filtered work data; Based on the filtered working data, the actual state data is determined.

9. A control strategy verification device, characterized in that, The device includes: The data acquisition module is used to acquire real state data of a real vehicle, wherein the real state data is the state data generated when the real control strategy is applied to the real vehicle under the target working condition. The model building module is used to build a vehicle dynamics model under the target working condition based on the real state data. The simulation test module is used to perform simulation tests on the vehicle dynamics model based on the actual control strategy and obtain simulation test results. The simulation test module is further configured to, when the simulation test result is a successful simulation, simulate the vehicle dynamics model based on the target control strategy to obtain first simulation state data, wherein the target control strategy is a vehicle control strategy to be verified and not embedded in the real vehicle. A performance verification model is used to determine the verification result of the target control strategy based on the real state data and the first simulation state data.

10. An electronic device, characterized in that, The electronic device includes: Memory is used to store executable instructions for a computer; A processor, when executing computer-executable instructions or computer programs stored in the memory, implements the control strategy verification method according to any one of claims 1 to 8.

11. A computer-readable storage medium storing computer-executable instructions or a computer program, characterized in that, When the computer-executable instructions or computer program are executed by a processor, the control strategy verification method according to any one of claims 1 to 8 is implemented.

12. A computer program product comprising computer-executable instructions or a computer program, characterized in that, When the computer-executable instructions or computer program are executed by a processor, the control strategy verification method according to any one of claims 1 to 6 is implemented.

Images