Acceleration recognition method, recognition device, recognition equipment and computer-readable storage medium based on low-frequency driving data

Abstract

The present invention relates to an acceleration recognition method, recognition device, recognition equipment, and computer-readable storage medium based on low-frequency driving data. The acceleration recognition method based on low-frequency driving data includes S1, data acquisition; S2, data preprocessing; S3, state recognition model training, which uses classification learning training to obtain an acceleration, deceleration, and uniform speed state recognition model; S4, acceleration calculation model training, which uses regression learning training to obtain an acceleration state acceleration calculation model; and S5, acceleration estimation. The acceleration recognition method, recognition device, recognition equipment, and computer-readable storage medium based on low-frequency driving data proposed by the present invention can effectively obtain the acceleration of low-frequency driving data.

Description

Technical Field

[0001] The present invention relates to the field of new energy vehicle motion control technology, and in particular to an acceleration recognition method, recognition device, recognition equipment and computer-readable storage medium based on low-frequency driving data. Background Art

[0002] Vehicle test cycles serve as a crucial foundational standard for new energy vehicle range, energy consumption certification, and durability assessment. They are a crucial design input into the automotive product development process and the primary benchmark for calibrating and optimizing various vehicle performance indicators. They should closely reflect actual user experience. Generally speaking, range and energy consumption certification utilize standard cycles (WLTC, CLTC, and NEDC), while vehicle manufacturers utilize specialized durability cycles for durability assessment.

[0003] However, the actual working conditions of Chinese users have the following characteristics: First, China has a vast territory, and the road conditions and climate conditions vary greatly between different regions, and the working conditions are complex. A unified standard working condition cannot take into account the differences in road conditions between regions; second, the usage scenarios of different user types and different vehicle types vary greatly, and a unified standard working condition cannot construct a working condition for subdivided scenarios; third, with the development of cities, traffic conditions and road conditions have changed greatly, and their impact on working conditions cannot be ignored. The unified standard working conditions are all based on road data collected several years ago and cannot fully reflect the actual working conditions of current users.

[0004] According to the national standard GBT32960, all new energy vehicles must upload key driving data, including battery SOC (state-of-charge), speed, motor, and battery information. Leveraging this new energy vehicle operating big data, or RTM data, it's possible to construct user scenarios that reflect real-world user driving characteristics. These scenarios can be tailored to specific regions and scenarios for different user types and vehicle models. Using real-world driving data to create scenarios is more user-friendly and consistent with current traffic conditions.

[0005] The operating condition method is used to construct user operating conditions through RTM data. The steps of the operating condition method are as follows: 1. Data preprocessing; 2. Dividing driving segments; 3. Extracting driving segment features; 4. Performing principal component analysis based on segment sample characteristics; 5. Clustering analysis of driving samples, dividing driving segment samples into three categories: low, medium and high speed; 5. Selecting alternative segment samples based on duration, medium, high and low speed ratio; 6. Calculating the av distribution of alternative samples, and selecting the segment with the smallest chi-square value with the overall distribution as the optimal segment; 7. Splicing the optimal segments to form the user operating condition SEVC.

[0006] A key step is to obtain driving characteristics for each driving segment, including speed, distance, average speed, idling time, total time, average acceleration / deceleration, maximum acceleration / deceleration, and acceleration / deceleration time. Since the current RTM data frame rate is 28 seconds, the frequency is too low to directly calculate acceleration from velocity differentials. Therefore, obtaining acceleration information from RTM data is a pressing issue to prepare for the subsequent construction of user driving scenarios. Summary of the Invention

[0007] In response to the above-mentioned problems in the prior art, the present invention proposes an acceleration recognition method, recognition device, recognition equipment and computer-readable storage medium based on low-frequency driving data, which can effectively obtain the acceleration of low-frequency driving data.

[0008] Specifically, the present invention proposes an acceleration recognition method based on low-frequency driving data, comprising the steps of:

[0009] S1, data acquisition, obtains CAN data of the test vehicle;

[0010] S2, data preprocessing, calculating the true value of the acceleration of the CAN data, and marking the speed state of the CAN data as acceleration, deceleration or uniform speed based on the magnitude of the true value of the acceleration;

[0011] S3, state recognition model training, selects the state features of CAN data, and obtains the acceleration, deceleration and uniform speed state recognition model through classification learning training based on the CAN data and the corresponding speed state;

[0012] S4, training the acceleration calculation model, selecting the acceleration feature of the CAN data, and using regression learning training for the CAN data with an acceleration state and the corresponding acceleration true value to obtain an acceleration calculation model for the acceleration state; selecting the deceleration feature of the CAN data, and using regression learning training for the CAN data with a deceleration state and the corresponding acceleration true value to obtain an acceleration calculation model for the deceleration state;

[0013] S5, acceleration estimation, obtaining RTM driving data matching the test vehicle, and calculating the acceleration of the RTM driving data based on an acceleration / deceleration constant speed state recognition model, an acceleration state acceleration calculation model, and a deceleration state acceleration calculation model.

[0014] According to one embodiment of the present invention, in step S2, the calculation formula of the acceleration true value a is:

[0015] a=(V_n-V_(n-1)) / dt;

[0016] Wherein, V_n is the vehicle speed of the current frame of the CAN data, V_(n-1) is the vehicle speed of the previous frame, and dt is the time interval between the current frame and the previous frame.

[0017] According to one embodiment of the present invention, in step S2, it is set that:

[0018] If the true value of acceleration a is greater than 0.15m / s 2 , then the speed state of the CAN data corresponding to the mark is acceleration; if the acceleration true value a is less than -0.15m / s 2 , then the speed state of the CAN data corresponding to the mark is deceleration; if the acceleration true value a is not greater than 0.15m / s 2 And not less than -0.15m / s 2 , then the speed state of the CAN data corresponding to the mark is uniform.

[0019] According to one embodiment of the present invention, in step S3, the state characteristics include at least vehicle speed, accelerator pedal opening, brake pedal opening, SOC, motor torque, battery pack total voltage, and battery pack total current;

[0020] In step S4 , the acceleration characteristics and deceleration characteristics include at least vehicle speed, accelerator pedal opening, brake pedal opening, SOC, motor torque, battery pack total voltage, and battery pack total current.

[0021] According to one embodiment of the present invention, the state feature further includes one or more of motor voltage, motor current, motor torque, motor speed, driving mode and braking energy recovery gear;

[0022] The acceleration characteristics and deceleration characteristics also include one or more of motor voltage, motor current, motor torque, motor speed, driving mode and brake energy recovery gear.

[0023] According to one embodiment of the present invention, the signal frame rate of the CAN data is not less than 1 Hz.

[0024] The present invention further provides an acceleration recognition device based on low-frequency driving data, for implementing the aforementioned acceleration recognition method, the acceleration recognition device comprising:

[0025] A data acquisition unit, used to acquire CAN data of a test vehicle and RTM driving data matching the test vehicle;

[0026] a preprocessing unit, configured to calculate a true value of acceleration of the CAN data, and mark a speed state of the CAN data as acceleration, deceleration, or uniform speed based on the magnitude of the true value of acceleration;

[0027] The first model training unit is used to select state features of CAN data and obtain an acceleration, deceleration and uniform speed state recognition model through classification learning training based on the CAN data and the corresponding speed state;

[0028] The second model training unit is configured to select the acceleration feature of the CAN data, and obtain an acceleration calculation model for the acceleration state through regression learning training based on the CAN data with an acceleration state and the corresponding acceleration true value; and select the deceleration feature of the CAN data, and obtain an acceleration calculation model for the deceleration state through regression learning training based on the CAN data with a deceleration state and the corresponding acceleration true value;

[0029] The acceleration estimation unit is used to calculate the acceleration of the RTM driving data based on the acceleration, deceleration and uniform speed state recognition model, the acceleration state acceleration calculation model and the deceleration state acceleration calculation model.

[0030] According to one embodiment of the present invention, the pre-processing unit includes a comparison module and a marking module;

[0031] The comparison module compares whether the acceleration true value a is greater than 0.15m / s 2 If it is greater than, the marking module marks the speed state of the corresponding CAN data as acceleration; the comparison module compares whether the acceleration true value a is less than -0.15m / s 2 If it is less than, the marking module marks the speed state of the corresponding CAN data as deceleration; the comparison module compares whether the acceleration true value a is not greater than 0.15m / s 2 And not less than -0.15m / s 2 If so, the marking module marks the speed state of the corresponding CAN data as uniform speed.

[0032] The present invention also provides an identification device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the acceleration identification method based on low-frequency driving data provided by the present invention are implemented.

[0033] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the acceleration identification method based on low-frequency driving data provided by the present invention are implemented.

[0034] The present invention provides an acceleration recognition method, recognition device, recognition equipment and computer-readable storage medium based on low-frequency driving data, which can effectively obtain the acceleration of low-frequency driving data.

[0035] It is to be understood that both the foregoing general description and the following detailed description of the present invention are exemplary and explanatory and are intended to provide further explanation of the invention as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings are included to provide further explanation of the present invention and are incorporated into and constitute a part of this application. The accompanying drawings illustrate embodiments of the present invention and together with the description serve to explain the principle of the present invention.

[0037] In the attached figure:

[0038] Figure 1 A flowchart of an acceleration identification method based on low-frequency driving data according to an embodiment of the present invention is shown.

[0039] Figure 2 A schematic structural diagram of an acceleration recognition device based on low-frequency driving data according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0040] It should be noted that, unless there is any conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0041] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is in no way intended to limit the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0042] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0043] Unless otherwise specified, the relative arrangement of the parts and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application. Meanwhile, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to actual proportional relationships. Technology, methods and equipment known to those of ordinary skill in the relevant art may not be discussed in detail, but in appropriate cases, the technology, methods and equipment should be considered as a part of the specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments can have different values. It should be noted that similar numbers and letters represent similar items in the following drawings, and therefore, once an item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0044] Furthermore, it should be noted that the use of terms such as "first" and "second" to define components is solely for the purpose of distinguishing the corresponding components. Unless otherwise stated, these terms have no special meaning and therefore should not be construed as limiting the scope of protection of this application. Furthermore, although the terms used in this application are selected from commonly known and commonly used terms, some terms mentioned in this specification may have been selected by the applicant based on their own comparisons, and their detailed meanings are explained in the relevant sections of this description. Furthermore, this application should be understood not only by the actual terms used, but also by the meaning implied by each term.

[0045] Figure 1 A flowchart of an acceleration recognition method based on low-frequency driving data according to an embodiment of the present invention is shown. As shown in the figure, the present invention provides an acceleration recognition method based on low-frequency driving data, comprising the following steps:

[0046] S1, Data Acquisition, acquires the CAN data of the test vehicle. CAN data refers to communication data carried out in frame format on the vehicle's CAN bus. CAN data consists of multiple frames of data acquired at a certain signal frame rate within a certain time period. Each frame typically includes frame time, vehicle speed, accumulated mileage, accelerator pedal opening, displayed SOC, brake pedal opening, total battery pack voltage, total battery pack current, motor voltage, motor current, motor torque, and motor speed. For some special types of new energy vehicles, each frame of data may also include driving mode and brake energy recovery gear.

[0047] S2, data preprocessing, calculates the true acceleration value of the CAN data and, based on the magnitude of the true acceleration value, labels the CAN data's velocity state as acceleration, deceleration, or uniform velocity. As can be readily understood, the true acceleration value and corresponding velocity state of each frame of CAN data are obtained and added as features to each frame. Based on the different velocity states, the CAN data can be divided into acceleration samples, deceleration samples, and uniform velocity samples.

[0048] S3, state recognition model training, selects the state features of CAN data, and obtains the acceleration, deceleration and uniform speed state recognition model through supervised classification learning training based on CAN data and corresponding speed state.

[0049] S4, Acceleration Calculation Model Training: Select the acceleration features of the CAN data and, using regression training on the CAN data with acceleration and the corresponding true acceleration values, obtain an acceleration calculation model for the acceleration state. Select the deceleration features of the CAN data and, using regression training on the CAN data with deceleration and the corresponding true acceleration values, obtain an acceleration calculation model for the deceleration state. In other words, supervised regression training is used to train the acceleration and deceleration samples and their corresponding true acceleration values ​​to obtain the acceleration calculation model for the acceleration state and the deceleration state, respectively.

[0050] S5: Acceleration estimation: Acquire RTM driving data that matches the test vehicle. It's easy to understand that steps S1 through S4 yield a training model related to a specific vehicle model. Vehicle matching means that the RTM driving data acquired for a specific vehicle should have the same configuration and model as the test vehicle. RTM data is low-frequency driving data with a frame rate of 28 seconds. Each frame of RTM driving data typically includes frame time, vehicle speed, accumulated mileage, accelerator pedal position, displayed SOC, brake pedal position, battery pack total voltage, battery pack total current, motor voltage, motor current, motor torque, and motor speed, essentially consistent with CAN data. Furthermore, based on the acceleration, deceleration, and constant speed state recognition model, the acceleration state of each frame of RTM driving data can be determined. Next, select RTM driving data with an acceleration state and calculate the acceleration of each frame using the acceleration state acceleration calculation model. Select RTM driving data with a deceleration state and calculate the acceleration of each frame using the deceleration state acceleration calculation model.

[0051] The present invention provides an acceleration identification method based on low-frequency driving data, aiming to address the problem of RTM driving data sampling frequency being too low to directly calculate acceleration. Based on vehicle kinematics and electric vehicle control theory, variables such as vehicle speed, accelerator pedal opening, brake pedal opening, and motor torque are closely correlated with the actual vehicle acceleration. Using test data from the same vehicle model, a machine learning model can be established that combines variables such as vehicle acceleration and motor data during operation. This supervised machine learning model can then be used to estimate vehicle acceleration from big data on new energy vehicles, thereby deriving driving characteristics related to acceleration and deceleration, providing a foundation for subsequent user operating condition development.

[0052] Preferably, in step S2, the calculation formula of the acceleration true value a is:

[0053] a=(V_n-V_(n-1)) / dt;

[0054] Where V_n is the vehicle speed of the current frame of CAN data, V_(n-1) is the vehicle speed of the previous frame, and dt is the time interval between the current frame and the previous frame.

[0055] Preferably, in step S2, it is set that if the acceleration true value a is greater than 0.15 m / s 2 , then the speed state of the corresponding CAN data is marked as acceleration, which is used as an acceleration sample; if the acceleration true value a is less than -0.15m / s 2 , then the speed state of the corresponding CAN data is marked as deceleration, which is used as a deceleration sample; if the acceleration true value a is not greater than 0.15m / s 2 And not less than -0.15m / s 2 , then the speed state of the corresponding CAN data is marked as uniform speed, which is used as a uniform speed sample.

[0056] Preferably, in step S3, the state characteristics include at least vehicle speed, accelerator pedal opening, brake pedal opening, SOC, motor torque, total battery pack voltage, and total battery pack current. These characteristics are closely related to the construction of the acceleration and uniform deceleration state recognition model. In step S4, the acceleration characteristics and deceleration characteristics include at least vehicle speed, accelerator pedal opening, brake pedal opening, SOC, motor torque, total battery pack voltage, and total battery pack current. These characteristics are closely related to vehicle acceleration and deceleration and are used to construct the corresponding acceleration calculation model for the acceleration state and the deceleration state.

[0057] Preferably, the state characteristics also include one or more of motor voltage, motor current, motor torque, motor speed, driving mode, and regenerative braking gear. These characteristics are also applicable to the construction of the acceleration / constant speed / deceleration state recognition model. For vehicles with driving modes and regenerative braking gear, these two state characteristics help to establish a more effective acceleration / constant speed / deceleration state recognition model, thereby improving the recognition efficiency of the acceleration / constant speed / deceleration state recognition model.

[0058] Acceleration and deceleration characteristics also include one or more of motor voltage, motor current, motor torque, motor speed, driving mode, and regenerative braking gear. These characteristics are applicable to the construction of acceleration calculation models for acceleration and deceleration. Similarly, for vehicles with driving modes and regenerative braking gears, these two state characteristics help to establish more effective recognition models, thereby improving the estimation efficiency of the acceleration calculation models for acceleration and deceleration.

[0059] Preferably, the signal frame rate of the collected CAN data is not less than 1 Hz.

[0060] It should be noted that the acceleration, deceleration and constant speed state recognition model obtained through classification learning training in the acceleration recognition method based on low-frequency driving data has an accuracy rate of 94%, which can well identify the speed state of the vehicle in acceleration, constant speed or deceleration.

[0061] The average error of the RTM driving data sample acceleration estimated by the acceleration calculation model obtained is ±0.07m / s 2 The statistical characteristic values ​​and the true value of acceleration a are shown in the following table. The percentage error of the mean and median of acceleration is controlled within 2%.

[0062]

[0063] The average error of the RTM driving data sample acceleration estimated by the obtained deceleration state acceleration calculation model is ±0.08m / s 2 The statistical characteristic values ​​and the true value of acceleration a are shown in the following table. The percentage error of the mean and median of acceleration is controlled within 3%.

[0064]

[0065] The present invention also provides an acceleration recognition device based on low-frequency driving data, which is used to implement the above-mentioned acceleration recognition method. Figure 2The schematic diagram of the structure of an acceleration recognition device based on low-frequency driving data according to an embodiment of the present invention is shown. As shown in the figure, the acceleration recognition device 200 mainly includes a data acquisition unit 201, a pre-processing unit 202, a first model training unit 203, a second model training unit 204, and an acceleration estimation unit 205.

[0066] The data acquisition unit 201 is used to obtain the CAN data of the test vehicle and the RTM driving data matched with the test vehicle.

[0067] The pre-processing unit 202 is used to calculate the true value of acceleration of the CAN data, and mark the speed state of the CAN data as acceleration, deceleration or uniform speed based on the magnitude of the true value of acceleration.

[0068] The first model training unit 203 is used to select state features of the CAN data, and obtain an acceleration, deceleration and uniform speed state recognition model through classification learning and training for the CAN data and the corresponding speed state.

[0069] The second model training unit 204 is used to select the acceleration features of the CAN data, and obtain the acceleration calculation model of the acceleration state through regression learning training for the CAN data with an accelerating speed state and the corresponding acceleration true value; select the deceleration features of the CAN data, and obtain the deceleration acceleration calculation model of the deceleration state through regression learning training for the CAN data with a decelerating speed state and the corresponding acceleration true value.

[0070] The acceleration estimation unit 205 is configured to calculate the acceleration of the RTM driving data based on the acceleration, deceleration and uniform speed state recognition model, the acceleration state acceleration calculation model and the deceleration state acceleration calculation model.

[0071] Preferably, the pre-processing unit 202 includes a comparison module and a marking module. The comparison module compares whether the acceleration true value a is greater than 0.15m / s 2 If it is greater than, the marking module marks the corresponding CAN data speed state as acceleration; the comparison module compares whether the acceleration true value a is less than -0.15m / s 2 If it is less than, the marking module marks the speed state of the corresponding CAN data as deceleration; the comparison module compares whether the acceleration true value a is not greater than 0.15m / s 2 And not less than -0.15m / s 2 ,If so, the marking module marks the speed state of the corresponding CAN data as uniform speed.

[0072] The present invention also provides an identification device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the aforementioned acceleration identification method based on low-frequency driving data are implemented.

[0073] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the aforementioned acceleration identification method based on low-frequency driving data are implemented.

[0074] Among them, the specific implementation methods and technical effects of the acceleration recognition device, recognition equipment, and computer-readable storage medium based on low-frequency driving data can all be referred to the embodiments of the acceleration recognition method based on low-frequency driving data provided by the above-mentioned present invention, and will not be repeated here.

[0075] Those skilled in the art will further appreciate that the various illustrative logic blocks, modules, circuits, and algorithmic steps described in conjunction with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or a combination of the two. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps are generally described above in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. A skilled person may implement the described functionality in different ways for each specific application, but such implementation decisions should not be interpreted as resulting in a departure from the scope of the present invention.

[0076] The various illustrative logic modules and circuits described in conjunction with the embodiments disclosed herein may be implemented or executed using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

[0077] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read and write information from / to the storage medium. In an alternative, the storage medium may be integrated into the processor. The processor and storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and storage medium may reside in a user terminal as discrete components.

[0078] In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions may be stored on or transmitted via a computer-readable medium as one or more instructions or codes. Computer-readable media include both computer storage media and communication media, including any media that facilitates the transfer of a computer program from one location to another. A storage medium may be any available medium that can be accessed by a computer. By way of example and not limitation, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Any connection is also properly referred to as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwaves are included in the definition of medium. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc, where disks typically reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0079] The present invention provides an acceleration recognition method, recognition device, recognition equipment, and computer-readable storage medium based on low-frequency driving data. These methods can identify radars with abnormal channels on a production line, preventing them from being released. Furthermore, by performing the acceleration recognition method based on low-frequency driving data before single-point calibration, if the acceleration recognition result based on the low-frequency driving data is abnormal, subsequent steps are omitted, avoiding unnecessary resource waste.

[0080] It will be apparent to those skilled in the art that various modifications and variations may be made to the above exemplary embodiments of the present invention without departing from the spirit and scope of the present invention. Therefore, it is intended that the present invention cover modifications and variations of the present invention that fall within the scope of the appended claims and their equivalent technical solutions.

Claims

1. A method for identifying acceleration based on low-frequency driving data, comprising the steps of: S1, data acquisition, obtains CAN data of the test vehicle; S2, data preprocessing, calculating the true value of the acceleration of the CAN data, and marking the speed state of the CAN data as acceleration, deceleration or uniform speed based on the magnitude of the true value of the acceleration; S3, state recognition model training, selects the state features of CAN data, and obtains the acceleration, deceleration and uniform speed state recognition model through classification learning training based on the CAN data and the corresponding speed state; S4, training the acceleration calculation model, selecting the acceleration feature of the CAN data, and using regression learning training for the CAN data with an acceleration state and the corresponding acceleration true value to obtain an acceleration calculation model for the acceleration state; selecting the deceleration feature of the CAN data, and using regression learning training for the CAN data with a deceleration state and the corresponding acceleration true value to obtain an acceleration calculation model for the deceleration state; S5, acceleration estimation, obtaining RTM driving data matching the test vehicle, and calculating the acceleration of the RTM driving data based on an acceleration / deceleration constant speed state recognition model, an acceleration state acceleration calculation model, and a deceleration state acceleration calculation model; In step S2, the calculation formula of the acceleration true value a is: a=(V_n-V_(n-1)) / dt; Wherein, V_n is the vehicle speed of the current frame of the CAN data, V_(n-1) is the vehicle speed of the previous frame, and dt is the time interval between the current frame and the previous frame.

2. The acceleration recognition method based on low-frequency driving data according to claim 1, characterized in that: In step S2, set: If the true value of acceleration a is greater than 0.15m / s 2 , then the speed state of the corresponding CAN data is marked as acceleration; If the true value of acceleration a is less than -0.15m / s 2 , then the speed state of the corresponding CAN data is marked as deceleration; If the true value of acceleration a is not greater than 0.15m / s 2 And not less than -0.15m / s 2 , the speed state of the CAN data corresponding to the mark is uniform.

3. The acceleration recognition method based on low-frequency driving data according to claim 1, characterized in that: In step S3, the state characteristics include at least vehicle speed, accelerator pedal opening, brake pedal opening, SOC, motor torque, battery pack total voltage and battery pack total current; In step S4 , the acceleration characteristics and deceleration characteristics include at least vehicle speed, accelerator pedal opening, brake pedal opening, SOC, motor torque, battery pack total voltage, and battery pack total current.

4. The acceleration recognition method based on low-frequency driving data according to claim 3, characterized in that: The state characteristics further include one or more of motor voltage, motor current, motor torque, motor speed, driving mode and braking energy recovery gear; The acceleration characteristics and deceleration characteristics also include one or more of motor voltage, motor current, motor torque, motor speed, driving mode and braking energy recovery gear.

5. The acceleration recognition method based on low-frequency driving data according to claim 1, characterized in that: The signal frame rate of the CAN data is not less than 1 Hz.

6. An acceleration recognition device based on low-frequency driving data, used to implement the acceleration recognition method according to any one of claims 1 to 5, characterized in that: The acceleration recognition device comprises: A data acquisition unit, used to acquire CAN data of a test vehicle and RTM driving data matching the test vehicle; a preprocessing unit, configured to calculate a true value of acceleration of the CAN data, and mark a speed state of the CAN data as acceleration, deceleration, or uniform speed based on the magnitude of the true value of acceleration; The first model training unit is used to select state features of CAN data and obtain an acceleration, deceleration and uniform speed state recognition model through classification learning training based on the CAN data and the corresponding speed state; The second model training unit is configured to select the acceleration feature of the CAN data, and obtain an acceleration calculation model for the acceleration state through regression learning training based on the CAN data with an acceleration state and the corresponding acceleration true value; and select the deceleration feature of the CAN data, and obtain an acceleration calculation model for the deceleration state through regression learning training based on the CAN data with a deceleration state and the corresponding acceleration true value; The acceleration estimation unit is used to calculate the acceleration of the RTM driving data based on the acceleration, deceleration and uniform speed state recognition model, the acceleration state acceleration calculation model and the deceleration state acceleration calculation model.

7. The acceleration recognition device based on low-frequency driving data according to claim 6, characterized in that: The pre-processing unit includes a comparison module and a marking module; The comparison module compares whether the acceleration true value a is greater than 0.15m / s 2 , if it is greater than, the marking module marks the speed state of the corresponding CAN data as acceleration; The comparison module compares whether the true value of acceleration a is less than -0.15m / s 2 If it is less than, the marking module marks the speed state of the corresponding CAN data as deceleration; The comparison module compares whether the acceleration true value a is not greater than 0.15m / s 2 And not less than -0.15m / s 2 If so, the marking module marks the speed state of the corresponding CAN data as uniform speed.

8. An identification device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the acceleration recognition method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the acceleration recognition method according to any one of claims 1 to 5 are implemented.

Images