Vehicle working condition mode identification method, computer device and storage medium

Abstract

The application provides a vehicle working condition mode identification method, computer equipment and a storage medium, and the method comprises the following steps: acquiring first vehicle data of a to-be-identified vehicle in a first time period; acquiring a historical working condition mode of the to-be-identified vehicle in the first time period by using a pre-trained vehicle working condition mode identification model according to the first vehicle data; determining a target working condition mode of the to-be-identified vehicle in a future time period according to the historical working condition mode of the to-be-identified vehicle in the first time period; issuing the target working condition mode to the to-be-identified vehicle; and making the to-be-identified vehicle dynamically adjust an energy control strategy in the future time period to an energy control strategy corresponding to the target working condition mode. According to the scheme, the working condition mode of the vehicle is identified by using a cloud platform, and the working condition mode is issued to the vehicle end, so that the energy control strategy is dynamically adjusted, and efficient utilization of the battery and optimization of the vehicle energy consumption are realized.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and more specifically, to a vehicle operating condition pattern recognition method, computer equipment, and storage medium. Background Technology

[0002] New energy commercial vehicles refer to commercial vehicles that use new energy technologies. These vehicles use new power systems such as batteries, electric motors, and hydrogen fuel cells to reduce dependence on traditional petroleum fuels. New energy commercial vehicles are mainly used in logistics transportation, urban distribution, public transportation and other fields.

[0003] In related technologies, for new energy commercial vehicles, during vehicle operation, the vehicle control unit (VCU) typically employs a unified energy control strategy to regulate the power of the fuel cell stack.

[0004] However, the aforementioned unified energy control strategy failed to fully utilize the high efficiency of fuel cells, resulting in high overall vehicle energy consumption. Summary of the Invention

[0005] In view of this, embodiments of this application provide a vehicle operating condition pattern recognition method, computer equipment, and storage medium to solve the problem that existing unified energy control strategies fail to fully utilize the high utilization rate of fuel cells, resulting in high energy consumption of the entire vehicle.

[0006] In a first aspect, embodiments of this application provide a vehicle operating condition pattern recognition method, applied to a cloud platform, the method comprising: Obtain the first vehicle data of the vehicle to be identified within the first time period; Based on the first vehicle data, a pre-trained vehicle operating condition pattern recognition model is used to obtain the historical operating condition pattern of the vehicle to be identified in the first time period. Based on the historical operating mode of the vehicle to be identified in the first time period, the target operating mode of the vehicle to be identified in the future time period is determined. The target operating condition mode is sent to the vehicle to be identified, so that the vehicle to be identified dynamically adjusts its energy control strategy for the future time period to the energy control strategy corresponding to the target operating condition mode.

[0007] In an optional implementation, the method further includes: Obtain the target driving route of the vehicle to be identified within the first time period; Acquire second vehicle data of multiple other vehicles on the target driving route during the second time period prior to the first time period; Based on the second vehicle data of multiple other vehicles, the vehicle operating condition pattern recognition model is used to obtain the historical operating condition patterns of multiple other vehicles in the second time period. The step of determining the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle to be identified in the first time period includes: The target operating condition mode is determined based on the historical operating condition mode of the vehicle to be identified in the first time period and the historical operating condition modes of multiple other vehicles in the second time period.

[0008] In an optional implementation, determining the target operating condition mode based on the historical operating condition mode of the vehicle to be identified in the first time period and the historical operating condition modes of the other vehicles in the second time period includes: The operating mode with the most occurrences is selected as the candidate operating mode from among the operating modes of the other vehicles. If the number of other vehicles exceeds a preset threshold, the candidate operating condition mode is not the historical operating condition mode of the vehicle to be identified in the first time period, and the proportion of the number of candidate operating condition modes exceeds a preset proportion threshold, then the candidate operating condition mode is determined as the target operating condition mode, and the proportion of the number of candidate operating condition modes is the ratio of the number of candidate operating condition modes to the number of operating condition modes of all other vehicles.

[0009] In an optional implementation, determining the target operating mode of the vehicle to be identified in a future time period based on the vehicle's historical operating mode during the first time period includes: If the number of other vehicles does not exceed a preset threshold, then the historical operating mode of the vehicle to be identified in the first time period is determined to be the target operating mode.

[0010] In an optional implementation, determining the target operating mode of the vehicle to be identified in a future time period based on the vehicle's historical operating mode during the first time period includes: Obtain the confidence level of the vehicle to be identified in the historical operating condition mode during the first time period; If the confidence level exceeds a preset confidence threshold, then the historical operating condition mode of the vehicle to be identified in the first time period is determined as the target operating condition mode.

[0011] In an optional implementation, the vehicle operating condition pattern recognition model is trained using the following steps: Acquire historical vehicle data for multiple historical vehicles within a historical time period; Based on the historical vehicle data of each historical vehicle, the labeled operating condition mode of each historical vehicle in the historical time period is obtained respectively. A training dataset is generated based on the historical vehicle data of multiple historical vehicles and the labeled operating condition patterns of multiple historical vehicles during the historical time period. Based on the training dataset, the initial random forest model is trained to obtain the vehicle operating condition pattern recognition model.

[0012] In an optional implementation, after sending the target operating mode to the vehicle to be identified, the method further includes: The system receives and stores execution information sent by the vehicle to be identified for the target operating mode. The execution information is used to instruct the vehicle to be identified to dynamically adjust the energy control strategy for the future time period to the energy control strategy corresponding to the target operating mode.

[0013] In an optional implementation, the method further includes: Receive the operational status query request for the vehicle to be identified sent by the query device; In response to the operating condition query request, the target operating condition mode is sent to the querying device based on the execution information.

[0014] Secondly, embodiments of this application also provide a computer device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the method described in any of the first aspects.

[0015] Thirdly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the method described in any of the first aspects.

[0016] This application provides a vehicle operating condition pattern recognition method, computer device, and storage medium. The method includes: acquiring first vehicle data of a vehicle to be identified within a first time period; based on the first vehicle data, using a pre-trained vehicle operating condition pattern recognition model, acquiring the historical operating condition patterns of the vehicle to be identified within the first time period; based on the historical operating condition patterns of the vehicle to be identified within the first time period, determining the target operating condition pattern of the vehicle to be identified within a future time period; and sending the target operating condition pattern to the vehicle to be identified, so that the vehicle to be identified dynamically adjusts its energy control strategy for the future time period to the energy control strategy corresponding to the target operating condition pattern. This solution uses a cloud platform to identify the vehicle's operating condition pattern and sends it to the vehicle to dynamically adjust the energy control strategy, achieving efficient battery utilization and optimized vehicle energy consumption. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 1 ; Figure 2 A schematic diagram illustrating the driving trajectory of the vehicle to be identified within a first time period is provided for the embodiments of this application; Figure 3 A scatter plot of preprocessed trajectory points provided in the embodiments of this application; Figure 4 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 2 ; Figure 5 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 3 ; Figure 6 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 4 ; Figure 7 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 5 ; Figure 8 This is an overall framework diagram for vehicle operation mode recognition provided in the embodiments of this application; Figure 9 This is a schematic diagram of the overall process of vehicle operation mode recognition provided in the embodiments of this application; Figure 10 This is a schematic diagram of the vehicle operating condition pattern recognition device provided in the embodiments of this application; Figure 11 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] To address the issue of high vehicle energy consumption resulting from the failure to fully utilize the high utilization rate of fuel cells when adopting a unified energy control strategy, this application provides a vehicle operation mode recognition and energy optimization control scheme based on a vehicle-cloud integrated architecture. This scheme aggregates vehicle data through a cloud platform, uses a pre-trained vehicle operating condition recognition model to identify the vehicle's operation mode, and sends this information to the vehicle. The vehicle dynamically adjusts the energy control strategy, thereby achieving efficient utilization of the fuel cell and optimization of overall vehicle energy consumption. Here, the energy control strategy refers to the power control strategy of the fuel cell stack.

[0021] Figure 1 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 1 In this embodiment, the execution entity can be a cloud platform.

[0022] like Figure 1 As shown, the method may include: S101. Obtain the first vehicle data of the vehicle to be identified within the first time period.

[0023] The vehicle to be identified can be a new energy commercial vehicle, and the first time period can be a historical time period before the current moment, such as yesterday. The first vehicle data is the vehicle data of the vehicle to be identified within the first time period.

[0024] In some embodiments, the first vehicle data includes: number of stops, number of times the air pump is enabled, the longest distance the vehicle travels, average speed, mileage, trip information, average trip length, air pump frequency, high-frequency parking indication information, and distance segmentation information.

[0025] The number of stops includes the total number of stops and the number of stops per kilometer, where the number of stops per kilometer is expressed as... , Total number of stops This represents the total mileage traveled within the first time period.

[0026] Total number of stops Represented as:

[0027] in, For the first time period The speed at each point in time The parking speed threshold (e.g., 0.5 km / h). For the first time period -1 point in time speed, The value of is a positive integer from 1 to n, where 1 represents the indicator function. When the speed at the (i-1)th time point is greater than the stopping speed threshold, and the... The value is 1 when the speed at any given time point is less than the parking speed threshold, and 0 otherwise. That is, it is counted once each time the speed changes from driving to parking.

[0028] The number of times the air pump is activated includes the total number of times the air pump is activated and the number of times the air pump is activated per kilometer. The total number of times the air pump is activated... This indicates the total number of times the air pump of the vehicle to be identified switched from the off state (0) to the on state (1) within the first time period. Each time the air pump switches from the off state (0) to the on state (1), it is counted as one operation. The number of times the air pump is enabled per kilometer is also indicated. Represented as:

[0029] in, This represents the total mileage traveled within the first time period.

[0030] Maximum distance traveled by vehicle The maximum distance traveled by the vehicle to be identified within the first time period under its own operating mode (historical operating mode).

[0031] It should be noted that for vehicles, whether in the long-reverse-short working mode or the short-reverse-short working mode, there is a possibility that the vehicle may drive out of the short-reverse working area. The positioning data (trajectory points) deviates significantly from the positioning data of the normal working area. If the distance between the starting trajectory point and the ending trajectory point is directly calculated as the farthest distance the vehicle has traveled, it will cause misjudgment. Therefore, the trajectory points need to be preprocessed to remove abnormal trajectory points.

[0032] Assuming that the trajectory points of the vehicles to be identified in the first time period follow a normal distribution, most trajectory points will be concentrated in a certain range. For example, about 68% of the trajectory points fall within the mean ± 1 standard deviation, about 95% fall within the mean ± 2 standard deviations, and about 99.7% fall within the mean ± 3 standard deviations. Trajectory points far from the mean are considered outliers. Based on big data analysis, trajectory points that are more than 2 standard deviations away from the mean are set as outlier trajectory points.

[0033] In some embodiments, the latitude mean, latitude standard deviation, longitude mean, and longitude standard deviation of all trajectory points of the vehicle to be identified within the first time period are calculated. Furthermore, the upper and lower latitude limits that differ from the latitude mean by ±2 times the latitude standard deviation, and the upper and lower longitude limits that differ from the longitude mean by ±2 times the longitude standard deviation, are calculated, resulting in four boundary values, expressed as follows:

[0034]

[0035] in, The average longitude. For the first Longitude value of the trajectory point The mean of the dimension. For the first The latitude value of the trajectory point The value of is a positive integer from 1 to n.

[0036]

[0037]

[0038] in, , , , These are the lower limit of longitude, the upper limit of longitude, the lower limit of latitude, and the upper limit of latitude, respectively. , These are the standard deviations of longitude and latitude, respectively.

[0039] Figure 2 This application provides a schematic diagram illustrating the driving trajectory of the vehicle to be identified within a first time period, as shown in the embodiments. Figure 2 As shown, the red dashed line divides the trajectory into two parts: the right side is the vehicle's working area (the driving area under its own working conditions), and the left side is the area that the vehicle only visited once (non-vehicle working area). The trajectory points in the left area are removed by using the standard deviation boundary to avoid large deviations and prevent misjudgments.

[0040] Figure 3A scatter plot of preprocessed trajectory points provided in the embodiments of this application, such as Figure 3 As shown, the horizontal axis represents longitude, the vertical axis represents latitude, the blue solid line represents the trajectory formed by the preprocessed trajectory points, the red dashed line represents the upper and lower limits of longitude (the red dashed line on the left is the lower limit of longitude, and the red dashed line on the right is the upper limit of longitude), and the green dashed line represents the upper and lower limits of latitude (the green dashed line on the bottom is the lower limit of latitude, and the green dashed line on the top is the upper limit of latitude).

[0041] The optimal driving distance is Figure 3 The distance between the solid red lines (diagonal distance), that is, the distance between the first intersection point between the lower longitude and lower latitude limits and the second intersection point between the upper longitude and upper latitude limits, assuming the first intersection point... Second intersection , Indicates longitude. Representing latitude, the maximum distance traveled by the vehicle is calculated using the semi-versus formula. Convert the intersection points to radians, as follows:

[0042] The longitude and latitude differences are calculated and expressed as follows: ,

[0043] but , Correspondingly, the vehicle's maximum travel distance R is the Earth's radius.

[0044] average speed This refers to the average speed and mileage of the vehicle to be identified within the first time period. Represented as:

[0045] in, This indicates the starting mileage of the interval for the vehicle to be identified within the first time period. This indicates the end mileage of the interval for the vehicle to be identified within the first time period.

[0046] Trip information is used to indicate the vehicle's journey from start to stop and from stop to start. When the vehicle to be identified starts from stop, its status changes from 0 to 1; when it starts from stop, its status changes from 1 to 0. The trip k is represented as: ,time The vehicle status is 0 or 1, and the time is... The vehicle status is 1 or 0.

[0047] In some embodiments, if the time interval between two trips is less than a preset duration (e.g., 5 minutes), the two trips are merged into one trip.

[0048] Average travel length is the ratio of the vehicle's optimal travel distance to the number of stops per kilometer, expressed as: This is used to distinguish between frequent short stops and long-distance travel with few stops. The longer the average trip length, the more frequent the long-distance travel with few stops. The shorter the average trip length, the more frequent the short stops.

[0049] The air pump frequency is the ratio of the number of times the air pump is activated to the maximum distance the vehicle travels, expressed as: / This is used to reflect frequent starts and stops or load changes. The higher the frequency of the air pump, the more frequent the starts and stops or load changes may be.

[0050] The high-frequency parking indication information is used to indicate whether the vehicle to be identified frequently stops within a first time period. If the number of stops per kilometer exceeds a preset parking threshold, it is considered high-frequency parking, as shown below: If the number of stops per kilometer is less than or equal to the preset parking threshold, it is not considered high-frequency parking and is indicated as follows: .

[0051] Distance segmentation information is used to indicate the distance level at which the vehicle has traveled the farthest distance; for example, distance segments. The containers are divided into [0-20, 20-80, 80+], with distance levels including 0-20, 20-80, and 80+, and the distance unit is km.

[0052] In some embodiments, when the vehicle to be identified is driving within a first time period, the vehicle can collect data during its driving process at a preset collection frequency (e.g., 1Hz) and synchronize it to the cloud platform in real time. The cloud platform can preprocess and clean the collected data, including handling missing values, removing outliers, and standardizing the format, to ensure the accuracy and consistency of subsequent analysis. This real-time synchronization to the cloud platform can be achieved through Flink.

[0053] After preprocessing is complete, the cloud platform can use a distributed computing framework to summarize and perform initial calculations on the preprocessed data to obtain the first vehicle data of the vehicle to be identified in the first time period, providing a high-quality data foundation for public pattern recognition and energy consumption optimization.

[0054] S102. Based on the first vehicle data, a pre-trained vehicle operating condition pattern recognition model is used to obtain the historical operating condition pattern of the vehicle to be identified in the first time period.

[0055] A pre-trained vehicle operating condition pattern recognition model is used to process the data of the first vehicle to obtain the historical operating condition pattern of the vehicle to be identified in the first time period. This historical operating condition pattern can be, for example, a long-distance operating condition pattern, a long-to-short-distance operating condition pattern, or a short-to-short-distance operating condition pattern. A long-distance operating condition pattern refers to a vehicle traveling directly at high speed from a logistics center in city A (or region) to a logistics center in city B (or region) at a great distance. Its characteristics include fixed loading and unloading points with little or no stops in between. A long-to-short-distance operating condition pattern refers to a vehicle transferring goods from a fixed central hub (such as a city logistics park, railway freight yard, or port terminal) to multiple dispersed points within a surrounding radiation area. Its characteristics include long-distance trucks delivering goods to the hub, which are then distributed by short-distance trucks. A short-to-short-distance operating condition pattern refers to vehicles conducting extremely high-frequency round-trip transportation on fixed, closed, or semi-closed routes, typically with fixed loading and unloading points and very short distances.

[0056] S103. Based on the historical operating mode of the vehicle to be identified in the first time period, determine the target operating mode of the vehicle to be identified in the future time period.

[0057] The future time period can be a time period after the current moment, such as today.

[0058] Since the operating route of the vehicle to be identified is relatively fixed, the target operating mode of the vehicle to be identified in the future time period can be determined by referring to the historical operating mode of the vehicle to be identified in the first time period. The target operating mode can be the historical operating mode.

[0059] S104. Send the target operating mode to the vehicle to be identified, so that the vehicle to be identified will dynamically adjust its energy control strategy for future time periods to the energy control strategy corresponding to the target operating mode.

[0060] The cloud platform sends the target operating condition mode to the vehicle to be identified. The vehicle receives the target operating condition mode and dynamically adjusts its energy control strategy for future time periods to the energy control strategy corresponding to the target operating condition mode. Here, the dynamic energy control strategy refers to the vehicle's battery management strategy.

[0061] It should be noted that different operating modes correspond to different energy control strategies. By identifying the target operating mode for a future time period, energy consumption can be managed in the future time period using the energy control strategy corresponding to the target operating mode, thereby achieving efficient utilization of fuel cells and optimization of vehicle energy consumption.

[0062] In some embodiments, the cloud platform sends a target operating mode to the vehicle to be identified, displays the target operating mode information on the vehicle's display screen, and responds to a confirmation operation for the target operating mode information. The vehicle then dynamically adjusts its energy control strategy for future time periods to the energy control strategy corresponding to the target operating mode. This allows for more accurate confirmation through the design of in-vehicle terminal user interaction functions and human intervention.

[0063] In this embodiment, by utilizing cloud-based big data computing capabilities and a distributed processing architecture, massive amounts of vehicle data can be processed quickly, enabling efficient identification and dynamic optimization strategy deployment. This allows for optimization of fuel cell power control at the vehicle end, resulting in reduced vehicle energy consumption and improved operating efficiency.

[0064] Figure 4 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 2 ,like Figure 4 As shown, in an optional implementation, the method may further include: S201. Obtain the target driving route of the vehicle to be identified within the first time period.

[0065] The target driving route is the driving route of the vehicle to be identified in the first time period. The starting point and ending point of each trip of the vehicle to be identified in the first time period are obtained, and the target driving route is generated based on each trip and the starting point and ending point of each trip. S202. Obtain second vehicle data of multiple other vehicles on the target driving route within the second time period before the first time period.

[0066] The second time period refers to the time period before the first time period, such as the day before yesterday. Other vehicles refer to vehicles whose routes are the target routes within the second time period. The second vehicle data refers to the vehicle data of other vehicles within the second time period.

[0067] Obtain the start and end points of each journey for all vehicles in the second time period, and use a clustering algorithm to identify routes based on the start and end points of each journey for all vehicles in the second time period, as well as the start and end points of each journey for the vehicle to be identified in the first time period. Vehicles that meet the preset clustering conditions (such as the position difference between the start and end points of each journey being less than the preset position difference) with the start and end points of each journey of the vehicle to be identified are identified as other vehicles.

[0068] In some embodiments, the second vehicle data may include the number of stops, the number of times the air pump is enabled, the maximum distance the vehicle travels, the average speed, the mileage, the trip information, the average trip length, the air pump frequency, the high-frequency parking indication information, and the distance segmentation information.

[0069] S203. Based on the second vehicle data of multiple other vehicles, a vehicle operating condition pattern recognition model is used to obtain the historical operating condition patterns of multiple other vehicles in the second time period.

[0070] A vehicle operating condition pattern recognition model is used to process the second vehicle data of each other vehicle to obtain the historical operating condition pattern of each other vehicle in the second time period. The historical operating condition pattern can be, for example, long-line operating condition pattern, long-to-short operating condition pattern, and short-to-short operating condition pattern.

[0071] Step S103 above, which determines the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle in the first time period, may include: S204. Determine the target operating condition mode based on the historical operating condition mode of the vehicle to be identified in the first time period and the historical operating condition modes of multiple other vehicles in the second time period.

[0072] Since the driving routes of other vehicles and the vehicle to be identified are the same, the target operating condition mode can be determined by combining the historical operating condition mode of the vehicle to be identified in the first time period and the historical operating condition mode of multiple other vehicles in the second time period. The target operating condition mode is either the historical operating condition mode of the vehicle to be identified in the first time period or the historical operating condition mode of other vehicles in the second time period.

[0073] In this embodiment, the operating conditions of other vehicles are compared and analyzed horizontally. The target operating condition of the vehicle to be identified is determined through group analysis, which breaks through the limitations of single vehicles and realizes group analysis. By aggregating historical vehicle data of multiple vehicles in the cloud, the limitations of the traditional single-vehicle identification mode are broken through, and comprehensive analysis across vehicles and time periods is realized, making the operating condition identification results more comprehensive and accurate.

[0074] Figure 5 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 3 ,like Figure 5 As shown, in an optional implementation, step S204 above, determining the target operating condition mode based on the historical operating condition modes of the vehicle to be identified in the first time period and the historical operating condition modes of multiple other vehicles in the second time period, may include: S301. Select the operating mode with the most occurrences from multiple operating modes of other vehicles as the candidate operating mode.

[0075] Among them, the candidate operating condition mode is the operating condition mode with the largest number among the operating condition modes of multiple other vehicles. In other words, the candidate operating condition mode has the largest number among all other vehicle operating condition modes, such as operating condition mode B.

[0076] S302. If the number of other vehicles exceeds a preset threshold, the candidate operating condition mode is not the historical operating condition mode of the vehicle to be identified in the first time period, and the proportion of candidate operating condition modes exceeds a preset proportion threshold, then the candidate operating condition mode is determined as the target operating condition mode.

[0077] The proportion of candidate operating condition modes is the ratio of the number of candidate operating condition modes to the number of operating condition modes of all other vehicles, denoted as P(B).

[0078] If the number of other vehicles exceeds a preset threshold, the candidate operating condition mode is different from the historical operating condition mode of the vehicle to be identified in the first time period, and the proportion of the candidate operating condition mode exceeds the threshold proportion threshold, then the candidate operating condition mode is determined as the target operating condition mode.

[0079] For example, if the historical operating condition mode of the vehicle to be identified in the first time period is operating condition mode A, and the number of other vehicles is M, the preset number threshold can be, for example, 5, and the preset proportion threshold can be, for example, 90% or 99%. This embodiment does not impose any special limitations on these. Then, when M≥5, B!=A, and P(B)>90%, the target operating condition mode is determined to be operating condition mode B.

[0080] In this embodiment, when there are enough other vehicles and the proportion of candidate operating condition modes is large enough, the candidate operating condition mode can be determined as the target operating condition mode, thereby improving the accuracy of the target operating condition mode.

[0081] In an optional implementation, step S104, which determines the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle in the first time period, may include: If the number of other vehicles does not exceed the preset threshold, the historical operating mode of the vehicle to be identified in the first time period is determined as the target operating mode.

[0082] If the number of other vehicles does not exceed the preset threshold (e.g., M < 5), it means that the historical operating conditions of other vehicles in the second time period are not relevant. In this case, the historical operating conditions of the vehicle to be identified in the first time period will be used as the target operating conditions.

[0083] In an optional implementation, step S104, which determines the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle in the first time period, may include: Obtain the confidence level of the historical operating condition patterns of the vehicle to be identified in the first time period; If the confidence level exceeds the preset confidence threshold, the historical operating condition mode of the vehicle to be identified in the first time period is determined as the target operating condition mode.

[0084] Specifically, based on the first vehicle data and the vehicle operating condition pattern recognition model, the historical operating condition patterns of the vehicle to be identified within the first time period and the confidence level corresponding to the historical operating condition patterns are obtained. The confidence level is used to indicate the probability that the operating condition pattern of the vehicle to be identified within the first time period is the historical operating condition pattern.

[0085] If the confidence level exceeds the preset confidence threshold, it indicates that the historical operating condition pattern of the identified vehicle in the first time period has high credibility, and the historical operating condition pattern of the identified vehicle in the first time period is taken as the target operating condition pattern.

[0086] It should be noted that, for situations other than those mentioned above, the historical operating mode of the vehicle to be identified in the first time period can be determined as the target operating mode.

[0087] Figure 6 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 4 ,like Figure 6 As shown, in an optional implementation, the vehicle operating condition pattern recognition model is trained using the following steps: S401. Obtain historical vehicle data for multiple historical vehicles within a historical time period.

[0088] The historical time period can be any period of time prior to the current moment. There can be multiple historical time periods, such as each day within the last three months as a historical time period. The historical vehicle data refers to the vehicle data of historical vehicles within the historical time period.

[0089] In some embodiments, historical vehicle data may include the number of stops, the number of times the air pump is enabled, the maximum distance the vehicle travels, the average speed, the mileage, the trip information, the average trip length, the air pump frequency, the high-frequency parking indication information, and the distance segmentation information.

[0090] S402. Based on the historical vehicle data of each historical vehicle, obtain the labeled operating condition mode of each historical vehicle in the historical time period.

[0091] The labeled operating condition mode for each historical vehicle within a historical time period can be determined by querying a preset labeled operating condition mode table based on the historical vehicle data of each historical vehicle. This preset labeled operating condition mode table includes the correspondence between vehicle data and labeled operating condition modes.

[0092] Historical vehicle data for each vehicle may include: number of stops per kilometer (ParkCount), number of times the air pump is activated per kilometer (AirPumpCount), and the furthest distance the vehicle has traveled. Based on the number of stops per kilometer, the number of times the air pump is activated per kilometer, and the farthest distance the vehicle travels, the preset labeled operating condition mode table is queried to determine the operating condition mode of each historical vehicle in the historical time period as the labeled operating condition mode.

[0093] Table 1 shows the preset annotation working condition mode. As shown in Table 1, if ParkCount <= 2, AirPumpCount < 1, and distance > 20, the annotation working condition mode is the long-line working condition mode. <ParkCount<=5、AirPumpCount> If ParkCount is 1, AirPumpCount is 1, and distance is 20, then the labeling mode is long-to-short. If ParkCount is 5, AirPumpCount is 1, and distance is 10, then the labeling mode is short-to-short.

[0094] Table 1

[0095] It should be noted that if no labeled working condition mode is found in the preset labeled working condition mode table, then the long-term working condition mode will be determined as the labeled working condition mode.

[0096] S403. Generate a training dataset based on the historical vehicle data of multiple historical vehicles and the labeled operating conditions of multiple historical vehicles in historical time periods.

[0097] S404. Based on the training dataset, train the initial random forest model to obtain the vehicle operating condition pattern recognition model.

[0098] The training dataset includes: historical vehicle data of multiple historical vehicles and labeled operating condition patterns of multiple historical vehicles in historical time periods.

[0099] After generating the training dataset, the initial random forest model is trained based on the training dataset to obtain the vehicle condition recognition model.

[0100] In some embodiments, training datasets, validation datasets, and test datasets can be generated based on historical vehicle data from multiple historical vehicles and labeled operating condition patterns of multiple historical vehicles over historical time periods. The training dataset comprises 70% of the samples, the validation dataset comprises 15%, and the test dataset comprises 15%. The model is trained using the training dataset, its hyperparameters are tuned using the validation dataset, and it is tested using the test set. A macro average F1 score of ≥ 0.80 is required for the model to be deployed online. The processes for model training, validation, and testing can be found in existing descriptions and will not be repeated here.

[0101] In an optional implementation, step S504 above, which involves training the initial random forest model based on the training dataset to obtain a vehicle condition pattern recognition model, may include: The optimal hyperparameters for the random forest are searched in the pre-defined search space using the Bayesian optimization algorithm. Based on the training dataset, the initial random forest model corresponding to the optimal hyperparameters of the random forest is trained to obtain the vehicle operating condition pattern recognition model.

[0102] The preset search space can include: the range of the number of decision trees in the random forest (n_estimators), the range of the maximum depth of each decision tree (max_depth), the range of the minimum number of samples required for node splitting (min_samples_split), and the range of the minimum number of samples required for leaf nodes (min_samples_leaf). For example, the range of n_estimators can be 100-200, the range of max_depth can be 8-15, the range of min_samples_split can be 5-20, and the range of min_samples_leaf can be 2-10.

[0103] It should be noted that the optimization objective of the Bayesian optimization algorithm is the F1-macro of 5-fold cross-validation. In other words, the Bayesian optimization algorithm searches for the optimal hyperparameters of the random forest by using the mean of the F1-macro of 5-fold cross-validation as the optimization objective.

[0104] The Bayesian optimization method is used to automatically search for the optimal hyperparameters of the random forest in the preset search space. Then, the initial random forest model is configured with these optimal hyperparameters, and the random forest model with the optimal hyperparameters is trained using the training dataset to obtain the vehicle working condition pattern recognition model.

[0105] It should be noted that after the vehicle operating condition pattern recognition model is launched, it continues to collect historical vehicle data, compare changes in vehicle energy consumption, and regularly optimize the model. The model is fed back and evaluated through a data closed-loop platform, so that the model's capabilities can be continuously evolved. It supports rapid iteration and deployment in scenarios with new vehicle data types and operating condition patterns.

[0106] In this embodiment, high-precision operation pattern recognition based on machine learning uses rule patterns to generate an initial training set, and employs a random forest model with strong interpretability and good generalization ability for training and prediction, which effectively reduces the anomaly in pattern boundary value recognition and significantly improves recognition accuracy.

[0107] Figure 7 A flowchart illustrating the vehicle operating condition pattern recognition method provided in this application embodiment. Figure 5 ,like Figure 7 As shown, in an optional implementation, after step S104, which sends the target operating condition mode to the vehicle to be identified, the method may further include: S501. Receive and store the execution information sent by the vehicle to be identified for the target operating mode.

[0108] After the vehicle to be identified dynamically adjusts its energy control strategy for the future time period to the energy control strategy corresponding to the target operating mode, the vehicle to be identified can send execution information for the target operating mode to the motion platform. The execution information is used to instruct the vehicle to be identified to dynamically adjust its energy control strategy for the future time period to the energy control strategy corresponding to the target operating mode. The cloud platform receives and stores the execution information.

[0109] S502, Receive the working condition query request for the vehicle to be identified sent by the query device.

[0110] S503: Respond to the operating condition query request and send the target operating condition mode to the querying device based on the execution information.

[0111] If it is necessary to query the operating condition mode of the vehicle to be identified, the querying device can send an operating condition query request for the vehicle to be identified to the cloud platform. The operating condition query request includes: the identification information of the vehicle to be identified. The cloud platform receives the operating condition query request and responds to the operating condition query request. According to the execution information, it sends the target operating condition mode to the querying device.

[0112] In this embodiment, the vehicle-cloud collaborative operation mode dispatch and status tracking mechanism allows the cloud platform to actively send signals after vehicle startup based on vehicle operation recognition results. When the vehicle's operating mode execution status changes, real-time feedback is sent to the cloud platform, enabling the cloud platform to track and monitor the vehicle's status throughout the entire process. This mechanism supports vehicle-cloud collaborative analysis, which not only verifies the algorithm's recognition accuracy and execution effect but also provides a data foundation for subsequent algorithm iteration and optimization, thereby continuously improving recognition accuracy and overall vehicle energy consumption management effectiveness.

[0113] Figure 8 This is an overall framework diagram of vehicle operation mode recognition provided in the embodiments of this application, such as... Figure 8 As shown, the system includes several modules: data acquisition and preprocessing, key indicator calculation, operating condition identification, and operating condition distribution. Data acquisition and preprocessing refers to the cloud platform's preprocessing of data collected during vehicle operation, including handling missing values, identifying outliers, and standardizing formats, to ensure the accuracy and consistency of subsequent analysis. Specifically, data can be collected at a frequency of 1Hz during vehicle operation and synchronized to the cloud platform in real time via Flink.

[0114] Key performance indicator (KPI) calculation refers to the process by which the cloud platform, after preprocessing, aggregates and performs preliminary calculations on the data collected during vehicle operation using the Spark distributed computing framework to obtain vehicle data, providing a high-quality data foundation for operation mode recognition and energy consumption optimization.

[0115] Operating condition mode recognition refers to the process by which the cloud platform identifies the operating condition mode of a vehicle using a pre-trained vehicle operating condition mode recognition model based on vehicle data.

[0116] Operating condition data transmission refers to the cloud platform transmitting the identified vehicle operating condition mode to the vehicle, enabling the vehicle to dynamically adjust its energy control strategy, thereby achieving efficient utilization of fuel cells and optimization of overall vehicle energy consumption.

[0117] In this embodiment, by fully combining the powerful data computing and model training capabilities of the cloud with the flexible control execution capabilities of the vehicle, a closed-loop optimization mechanism of "cloud recognition - vehicle execution - cloud feedback" is formed to achieve accurate identification of commercial vehicle operating modes and dynamic energy management, ultimately achieving the goal of improving energy efficiency and reducing energy consumption.

[0118] Figure 9 This is a schematic diagram of the overall process of vehicle operation mode recognition provided in the embodiments of this application, such as... Figure 9 As shown, it includes the following steps: Step 1: Wait for the vehicle to be powered on and send vehicle data to the cloud platform.

[0119] Step 2: The cloud platform acquires operating condition modes based on vehicle data.

[0120] Step 3: The cloud platform sends the operating mode to the vehicle.

[0121] Step 4: The vehicle reports the execution information of the operating mode to the cloud platform.

[0122] Step 5: The cloud platform sends the vehicle's operating status mode to the query device and displays it through the webpage or application (APP) provided by the query device.

[0123] Figure 10 This is a schematic diagram of the vehicle operating condition pattern recognition device provided in an embodiment of this application. The device can be integrated into a computer device.

[0124] like Figure 10 As shown, the device may include: The acquisition module 601 is used to acquire the first vehicle data of the vehicle to be identified within the first time period. The acquisition module 601 is also used to acquire the historical operating condition pattern of the vehicle to be identified in the first time period based on the first vehicle data and a pre-trained vehicle operating condition pattern recognition model. The determination module 602 is used to determine the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle in the first time period. The sending module 603 is used to send the target operating mode to the vehicle to be identified, so that the vehicle to be identified will dynamically adjust its energy control strategy for future time periods to the energy control strategy corresponding to the target operating mode.

[0125] In an optional implementation, the acquisition module 601 is further configured to: Obtain the target driving route of the vehicle to be identified within the first time period; Acquire second vehicle data for multiple other vehicles on the target driving route during the second time period prior to the first time period; Based on the second vehicle data of multiple other vehicles, a vehicle operating condition pattern recognition model is used to obtain the historical operating condition patterns of multiple other vehicles in the second time period. Module 602 is specifically used for: The target operating mode is determined based on the historical operating mode of the vehicle to be identified in the first time period and the historical operating mode of multiple other vehicles in the second time period.

[0126] In an optional implementation, the determining module 602 is specifically used for: The operating mode with the most occurrences is selected as the candidate operating mode from among multiple operating modes of other vehicles. If the number of other vehicles exceeds a preset threshold, the candidate operating condition mode is not the historical operating condition mode of the vehicle to be identified in the first time period, and the proportion of the number of candidate operating condition modes exceeds a preset proportion threshold, then the candidate operating condition mode is determined as the target operating condition mode. The proportion of the number of candidate operating condition modes is the ratio of the number of candidate operating condition modes to the number of operating condition modes of all other vehicles.

[0127] In an optional implementation, the determining module 602 is specifically used for: If the number of other vehicles does not exceed the preset threshold, the historical operating mode of the vehicle to be identified in the first time period is determined as the target operating mode.

[0128] In an optional implementation, the determining module 602 is specifically used for: Obtain the confidence level of the historical operating condition patterns of the vehicle to be identified in the first time period; If the confidence level exceeds the preset confidence threshold, the historical operating condition mode of the vehicle to be identified in the first time period is determined as the target operating condition mode.

[0129] In an optional implementation, the vehicle operating condition pattern recognition model is trained using the following steps: Acquire historical vehicle data for multiple historical vehicles within a historical time period; Based on the historical vehicle data of each historical vehicle, the labeled operating condition mode of each historical vehicle in the historical time period is obtained respectively. A training dataset is generated based on historical vehicle data from multiple historical vehicles and the labeled operating conditions of multiple historical vehicles within historical time periods. Based on the training dataset, the initial random forest model is trained to obtain a vehicle condition pattern recognition model.

[0130] In an optional embodiment, the device further includes: Receiver module 604 is used for: Receive and store the execution information sent by the vehicle to be identified for the target operating mode. The execution information is used to instruct the vehicle to be identified to dynamically adjust the energy control strategy for future time periods to the energy control strategy corresponding to the target operating mode.

[0131] In an optional implementation, the receiving module 604 is further configured to receive a status query request for the vehicle to be identified sent by the query device. The sending module 603 is also used to respond to the operating condition query request and send the target operating condition mode to the querying device based on the execution information.

[0132] The processing flow of each module in the device and the interaction flow between each module can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0133] Figure 11 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. A cloud platform may be deployed on this device.

[0134] like Figure 11 As shown, the device may include a processor 701, a memory 702, and a bus 703. The memory 702 stores machine-readable instructions that can be executed by the processor 701. When the computer device is running, the processor 701 communicates with the memory 702 through the bus 703, and the processor executes the machine-readable instructions to perform the above-described method.

[0135] This application also provides a computer-readable storage medium storing a computer program, which is executed by a processor to perform the above-described method.

[0136] In this embodiment, the computer program, when run by the processor, can also execute other machine-readable instructions to perform other methods as described in the embodiments. For details on the specific execution steps and principles, please refer to the description of the embodiments, which will not be repeated here.

[0137] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0138] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0139] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0140] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0141] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0142] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the technical scope disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for recognizing vehicle operating conditions, characterized in that, Applied to a cloud platform, the method includes: Acquire the first vehicle data of the vehicle to be identified within a first time period, where the first time period is a historical time period prior to the current moment; Based on the first vehicle data, a pre-trained vehicle operating condition pattern recognition model is used to obtain the historical operating condition pattern of the vehicle to be identified in the first time period. Based on the historical operating condition mode of the vehicle to be identified in the first time period, the target operating condition mode of the vehicle to be identified in the future time period is determined, wherein the future time period is the time period after the current time. The target operating condition mode is sent to the vehicle to be identified, so that the vehicle to be identified dynamically adjusts its energy control strategy for the future time period to the energy control strategy corresponding to the target operating condition mode. The method further includes: Obtain the target driving route of the vehicle to be identified within the first time period; Acquire second vehicle data of multiple other vehicles on the target driving route during the second time period prior to the first time period; Based on the second vehicle data of multiple other vehicles, the vehicle operating condition pattern recognition model is used to obtain the historical operating condition patterns of multiple other vehicles in the second time period. The step of determining the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle to be identified in the first time period includes: The target operating condition mode is determined based on the historical operating condition mode of the vehicle to be identified in the first time period and the historical operating condition modes of multiple other vehicles in the second time period.

2. The method according to claim 1, characterized in that, The step of determining the target operating condition mode based on the historical operating condition mode of the vehicle to be identified in the first time period and the historical operating condition modes of multiple other vehicles in the second time period includes: The operating mode with the most occurrences is selected as the candidate operating mode from among the operating modes of the other vehicles. If the number of other vehicles exceeds a preset threshold, the candidate operating condition mode is not the historical operating condition mode of the vehicle to be identified in the first time period, and the proportion of the number of candidate operating condition modes exceeds a preset proportion threshold, then the candidate operating condition mode is determined as the target operating condition mode, and the proportion of the number of candidate operating condition modes is the ratio of the number of candidate operating condition modes to the number of operating condition modes of all other vehicles.

3. The method according to claim 1, characterized in that, The step of determining the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle to be identified in the first time period includes: If the number of other vehicles does not exceed a preset threshold, then the historical operating mode of the vehicle to be identified in the first time period is determined to be the target operating mode.

4. The method according to claim 1, characterized in that, The step of determining the target operating mode of the vehicle to be identified in a future time period based on the historical operating mode of the vehicle to be identified in the first time period includes: Obtain the confidence level of the vehicle to be identified in the historical operating condition mode during the first time period; If the confidence level exceeds a preset confidence threshold, then the historical operating condition mode of the vehicle to be identified in the first time period is determined as the target operating condition mode.

5. The method according to any one of claims 1-4, characterized in that, The vehicle operating condition pattern recognition model is trained using the following steps: Acquire historical vehicle data for multiple historical vehicles within a historical time period; Based on the historical vehicle data of each historical vehicle, the labeled operating condition mode of each historical vehicle in the historical time period is obtained respectively. A training dataset is generated based on the historical vehicle data of multiple historical vehicles and the labeled operating condition patterns of multiple historical vehicles during the historical time period. Based on the training dataset, the initial random forest model is trained to obtain the vehicle operating condition pattern recognition model.

6. The method according to claim 1, characterized in that, After sending the target operating mode to the vehicle to be identified, the method further includes: The system receives and stores execution information sent by the vehicle to be identified for the target operating mode. The execution information is used to instruct the vehicle to be identified to dynamically adjust the energy control strategy for the future time period to the energy control strategy corresponding to the target operating mode.

7. The method according to claim 6, characterized in that, The method further includes: Receive the operational status query request for the vehicle to be identified sent by the query device; In response to the operating condition query request, the target operating condition mode is sent to the querying device based on the execution information.

8. A computer device, characterized in that, include: The device includes a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the method according to any one of claims 1 to 7.

Images