Driving condition construction method and device, vehicle and storage medium

Abstract

The invention relates to the technical field of vehicle engineering, in particular to a driving condition construction method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining real vehicle operation data, and extracting speed data and slope data in the real vehicle operation data; dividing the vehicle travel into a plurality of speed and gradient segments according to the speed data and the gradient data; and classifying the plurality of speed and gradient segments, selecting a target segment from each type of speed and gradient segments by using a genetic algorithm, and generating the speed and gradient driving condition of the vehicle according to each type of target segment. Therefore, the problems that strong coupling of speed and gradient is neglected, gradient characteristics are difficult to consider and the like in related technologies are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle engineering, and in particular relates to a driving condition construction method and device, a vehicle and a storage medium. BACKGROUND

[0002] The construction of typical driving conditions plays an important role in the design, testing, optimization and certification process of vehicles. At present, most of the commonly used typical driving conditions are constructed based on speed-time curves, ignoring the significant influence of slope on vehicle energy consumption and performance. With the increasing demand for the restoration of actual driving scenarios, composite conditions that integrate speed and slope information have gradually become a research and application hotspot.

[0003] In related technologies, one type of method generates multiple condition sequences through a speed state transition probability matrix, but does not consider the slope factor. Another type of method introduces slope information, but decouples the speed and slope characteristics for processing, making it difficult to reflect the coupling relationship between the two in actual driving. In summary, related technologies generally ignore the strong coupling between speed and slope, making it difficult to consider slope characteristics and resulting in insufficient condition representation. SUMMARY

[0004] The present application provides a driving condition construction method, device, vehicle and storage medium to solve the problem of related technologies ignoring the strong coupling between speed and slope and making it difficult to consider slope characteristics.

[0005] The first aspect embodiment of the present application provides a driving condition construction method, including the following steps: obtaining real vehicle running data, extracting speed data and slope data in the real vehicle running data; dividing the vehicle trip into multiple speed and slope segments according to the speed data and the slope data; classifying the multiple speed and slope segments, selecting target segments in each type of speed and slope segment using a genetic algorithm, and generating speed and slope driving conditions for the vehicle according to each type of target segment.

[0006] Optionally, in an embodiment of the present application, dividing the vehicle trip into multiple speed and slope segments according to the speed data and the slope data includes: identifying the idle time of the vehicle in the real vehicle running data; dividing the vehicle trip into multiple sub-trips according to the idle time of the vehicle; and matching the multiple sub-trips with the speed data and the slope data to obtain the multiple speed and slope segments.

[0007] Optionally, in an embodiment of the present application, classifying the multiple speed and slope segments includes: extracting speed characteristics and slope characteristics in the speed and slope segments; clustering the speed and slope segments according to the speed characteristics to obtain a first clustering result; and clustering the segments in the first clustering result according to the slope characteristics to obtain a second clustering result.

[0008] Optionally, in an embodiment of the present application, the target segment is selected in each speed and slope segment by using a genetic algorithm, comprising: defining a multi-objective optimization problem of the genetic algorithm; solving the multi-objective optimization problem based on a multi-objective optimization algorithm of the genetic algorithm; and determining the target segment according to a solution of the multi-objective optimization problem.

[0009] Optionally, in an embodiment of the present application, the multi-objective optimization problem comprises a control action, a control target and a control constraint, wherein, The formula of the control action is:

[0010] wherein, a represents the control action, and the selection of each kinematic segment is defined as 1, and the non-selection is defined as 0.

[0011] The formula of the control target is:

[0012] wherein, J is the control target, p v is a Pearson coefficient of the speed distribution, p s is a Pearson coefficient of the slope distribution.

[0013] The formula of the control constraint is:

[0014] wherein, is a selection constraint of the i-th short-range segment, i is a minimum time of the constructed typical working condition, is a maximum time of the constructed typical working condition, is a total time of the i-th short-range segment, is a total time of all short-range segments. i

[0015] Optionally, in an embodiment of the present application, the multi-objective optimization algorithm based on the genetic algorithm solves the multi-objective optimization problem, comprising: randomly generating an initial population, each individual representing a solution; evaluating the fitness value of each individual in the population and performing non-dominated sorting on the individuals in the population; based on the fitness value and the non-dominated sorting result, using a roulette method to select a target individual for breeding, generating a new individual through a crossover operation, and performing mutation on the new individual; adding the newly generated individual to the population, and performing fitness evaluation and non-dominated sorting, until a preset termination condition is met, then stopping the iteration solution of the multi-objective optimization algorithm.

[0016] ​​Optionally, in an embodiment of the present application, the selection formula of the target segment is:

[0017] wherein, is the final segment selection solution, is the minimum limit value of the Pearson coefficient of the speed distribution, is the minimum limit value of the Pearson coefficient of the slope distribution.

[0018] The second aspect embodiment of the present application provides a driving condition construction device, comprising: an acquisition module, configured to acquire real vehicle running data, and extract speed data and slope data in the real vehicle running data; a division module, configured to divide a vehicle trip into a plurality of speed and slope segments according to the speed data and the slope data; and a generation module, configured to classify the plurality of speed and slope segments, select a target segment in each type of speed and slope segment by using a genetic algorithm, and generate a speed and slope driving condition of the vehicle according to each type of target segment.

[0019] Optionally, in an embodiment of the present application, the division module is further configured to identify an idle time of the vehicle in the real vehicle running data; divide the vehicle trip into a plurality of sub-trips according to the idle time of the vehicle; and match the plurality of sub-trips with the speed data and the slope data to obtain the plurality of speed and slope segments.

[0020] Optionally, in an embodiment of the present application, the generation module is further configured to extract speed features and slope features in the speed and slope segments; cluster the speed and slope segments according to the speed features to obtain a first clustering result; and cluster the segments in the first clustering result according to the slope features to obtain a second clustering result.

[0021] Optionally, in an embodiment of the present application, the generation module is further configured to define a multi-objective optimization problem of the genetic algorithm; solve the multi-objective optimization problem based on a multi-objective optimization algorithm of the genetic algorithm; and determine the target segment according to a solution result of the multi-objective optimization problem.

[0022] Optionally, in an embodiment of the present application, the multi-objective optimization problem comprises a control action, a control target, and a control constraint, wherein, The formula of the control action is:

[0023] wherein, a represents the control action, and the selection of each kinematic segment is defined as 1, and the non-selection is defined as 0.

[0024] The formula of the control target is:

[0025] wherein,J is a control target, p v is a Pearson coefficient of the speed distribution, p s is a Pearson coefficient of the slope distribution.

[0026] The formula of the control constraint is:

[0027] wherein, is a minimum time of the i class short-trip segment selection constraint, is a minimum time of the constructed typical working condition, is a maximum time of the constructed typical working condition, is a total time of the i class short-trip segment, is a total time of all short-trip segments.

[0028] Optionally, in an embodiment of the present application, the generating module is further configured to randomly generate an initial population, each individual representing a solution; evaluate the fitness value of each individual in the population and perform non-dominated sorting on the individuals in the population; based on the fitness value and the non-dominated sorting result, select a target individual for reproduction by using a roulette method, generate a new individual by using a crossover operation, and perform mutation on the new individual; add the newly generated individual to the population, and perform fitness evaluation and non-dominated sorting, until a preset termination condition is met, and then stop the iteration of the multi-objective optimization algorithm.

[0029] Optionally, in an embodiment of the present application, the selection formula of the target segment is:

[0030] wherein, is a final segment selection solution set, is a minimum limit value of the Pearson coefficient of the speed distribution, is a minimum limit value of the Pearson coefficient of the slope distribution.

[0031] The third aspect embodiment of the present application provides a vehicle, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the driving working condition construction method of the above-described embodiments.

[0032] The fourth aspect embodiment of the present application provides a computer readable storage medium, which stores a computer program executable by a processor to implement the driving working condition construction method of the above-described embodiments.

[0033] Therefore, the present application includes the following beneficial effects: First, raw data from the actual vehicle's operation on real roads is collected, and the required speed and corresponding slope data are extracted from it. Then, based on the combined characteristics of speed and slope, the entire trip is divided into several speed-slope combined segments. These segments are then classified, with segments with similar characteristics grouped into the same category. Finally, within each category, a genetic algorithm is used to screen the segments, selecting representative target segments. Finally, based on these target segments, a composite driving cycle containing speed and slope information is generated to more accurately reflect the vehicle's operating characteristics under real-world road conditions. This overcomes the problem that related technologies ignore the strong coupling between speed and slope, making it difficult to take slope characteristics into account.

[0034] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A flowchart of a driving condition construction method provided according to an embodiment of the present application; Figure 2 A flowchart of the specific steps of the driving condition construction method according to an embodiment of the present application; Figure 3 This is an example diagram of a driving condition construction device according to an embodiment of the present application; Figure 4 Schematic diagram of the structure of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0037] The driving condition construction method, device, vehicle and storage medium of the embodiments of the present application are described below with reference to the accompanying drawings. In view of the problems mentioned in the above background art, the present application provides a driving condition construction method. In the method, first, the original data of the actual road running process of the real vehicle is collected, and the required speed data and corresponding slope data are extracted therefrom. Then, according to the joint characteristics of speed and slope, the entire trip is divided into a plurality of speed-slope combination segments. Next, the segments are classified by features, and segments with similar features are classified into the same category. Finally, in each category, a genetic algorithm is used to select representative target segments, and finally a composite driving condition containing speed and slope information is generated based on these target segments, which is used to more accurately reflect the running characteristics of the vehicle under real road conditions. Thus, the problem that related technologies ignore the strong coupling of speed and slope and are difficult to consider slope characteristics is solved.

[0038] Specifically, Figure 1 A flowchart of a driving condition construction method provided by an embodiment of the present application is shown.

[0039] As Figure 1 shown, the driving condition construction method includes the following steps: In step S101, real vehicle running data is obtained, and speed data and slope data in the real vehicle running data are extracted.

[0040] The real vehicle running data is the real dynamic data collected by vehicle-mounted sensors, Internet of Vehicles devices, etc. during the vehicle running on the actual road, which commonly includes speed, acceleration, engine speed, throttle opening, brake state, slope, geographic position, time stamp, etc. The speed data represents the running speed of the vehicle at different times, usually expressed in kilometers per hour (km / h). The slope data represents the longitudinal inclination of the vehicle running section, i.e. the slope of the road relative to the horizontal plane, usually expressed in percentage (%), for example, a slope of 10% means that the road rises 10 meters in height for every 100 meters of forward progress.

[0041] It can be understood that by obtaining the real vehicle running data and extracting the speed data and slope data therefrom, the running state of the vehicle in the actual road environment can be truly reflected, providing accurate data support for subsequent construction of driving conditions. At the same time, real vehicle data has higher authenticity and representativeness, which can effectively capture key features such as different driving behaviors and road slope changes, improving the accuracy and applicability of the condition modeling. The present application realizes continuous collection of the running state of the vehicle in the actual road environment through the Internet of Vehicles function, and the data obtained includes but is not limited to the speed information and slope information of the vehicle.

[0042] In step S102, the vehicle trip is divided into multiple speed and slope segments according to the speed data and the slope data.

[0043] It can be understood that, according to the speed data and the slope data, the vehicle trip is divided into multiple speed and slope segments, which can divide the continuous and complex vehicle running data into a plurality of short time intervals with time sequence continuity and relatively stable motion characteristics, thereby finely characterizing the dynamic behavior of the vehicle under different road conditions and driving states, and avoiding the loss of key information caused by overall data smoothing processing. At the same time, the structured segmentation simplifies the data analysis process, and improves the efficiency and accuracy of abnormal data elimination, feature extraction and classification clustering.

[0044] In an embodiment of the present application, the vehicle trip is divided into multiple speed and slope segments according to the speed data and the slope data, comprising: identifying an idle time of the vehicle in the real vehicle running data; dividing the vehicle trip into multiple sub-trips according to the idle time of the vehicle; and matching the multiple sub-trips with the speed data and the slope data to obtain multiple speed and slope segments.

[0045] The idle time is a time point at which the vehicle is stationary while the engine is running, i.e., a state in which the vehicle speed is zero but the engine is not turned off, which is usually used to identify the beginning or end of a driving period.

[0046] It can be understood that, according to the time node from one idle time to the next idle time, the vehicle trip is divided into multiple short trips, and the speed-slope segment corresponding to each short trip is stored at the same time. Then, the stored short trip segments are preprocessed, and the segments with abnormal values are eliminated or corrected. For example, the kinematic segment with a speed exceeding 120 km / h is considered as an overspeed abnormality and is eliminated; the segment with a speed not exceeding 10 km / h for 180 seconds is considered as a long-time congestion abnormality and is eliminated; the segment with an acceleration exceeding the range of [-8, 4] m / s 2 The segment with a slope exceeding the range of [-30%, 30%] is also considered as an abnormality and is eliminated. Through the above steps, the complex long-time driving data is structured into independent and representative short-time segment, while ensuring the accuracy and effectiveness of the data used for subsequent analysis.

[0047] In step S103, the multiple speed and slope segments are classified, a genetic algorithm is used to select a target segment in each class of speed and slope segments, and a speed and slope driving cycle of the vehicle is generated according to each class of target segments.

[0048] The genetic algorithm is a global optimization algorithm based on evolutionary ideas, and multiple candidate segments are iteratively optimized by simulating natural selection and genetic mechanisms. After classifying multiple speed and slope segments, the genetic algorithm is used to select representative target segments from each type of segment. The driving cycle refers to the description of the running state of the vehicle changing with time during actual driving, usually including the change curve or discrete data sequence of parameters such as speed, acceleration, and slope with time. In this embodiment, a joint driving cycle of speed and slope is constructed to describe the running characteristics of the vehicle under specific conditions.

[0049] It can be understood that through the classification processing, a large amount of real vehicle data can be divided into a more regular segment set according to the characteristic segments of the vehicle under different running states, and data redundancy is reduced. The introduction of the genetic algorithm for target segment selection can automatically select the sample that best reflects the characteristics of each type of segment under the premise of ensuring representativeness.

[0050] In an embodiment of the present application, classifying multiple speed and slope segments includes: extracting speed features and slope features in the speed and slope segments; clustering the speed and slope segments according to the speed features to obtain a first clustering result; and clustering the segments in the first clustering result according to the slope features to obtain a second clustering result.

[0051] The clustering is a data analysis method that divides similar data into the same category to reveal the structural characteristics of the data, facilitating effective classification and processing of the data. The first clustering result is obtained by clustering the speed and slope segments according to speed features such as average speed and acceleration, resulting in categories such as high speed, medium speed, and low speed. The second clustering result is obtained by further clustering within each speed category according to the slope features, finally subdividing the segments into nine categories such as high-speed high-slope and high-speed medium-slope, which helps to more accurately reflect the typical driving cycles of the vehicle under different speeds and slopes.

[0052] It can be understood that extracting speed features and slope features in the speed and slope segments can accurately capture key change information during vehicle driving. Clustering based on speed features effectively divides different speed categories and clearly defines the speed levels of vehicle operation. Further clustering within each speed category based on slope features refines the road condition differences and improves the representativeness and precision of the driving cycle.

[0053] The embodiments of the present application extract speed features of the preprocessed short-range segments, including average speed, average travel speed, speed standard deviation, average acceleration, acceleration standard deviation, average deceleration, deceleration standard deviation, idling time ratio, acceleration time ratio, deceleration time ratio, and uniform speed time ratio, and perform principal component analysis on the above speed features to select principal components with a cumulative contribution rate of more than 90% to reduce dimensions and retain main information.

[0054] Subsequently, K-means clustering is performed based on the principal components, where K-means clustering is a commonly used unsupervised machine learning algorithm, mainly used to automatically divide data into several categories, so that the data in the same category has higher similarity, and the data between different categories has greater difference. All segments are divided into high-speed, medium-speed, and low-speed three categories.

[0055] On this basis, the slope features of the segments in each speed category are further calculated, the slope features are further subjected to principal component analysis, the principal components with a cumulative contribution rate of more than 90% are selected, and the K-means clustering method is used again to further subdivide all segments into high-speed high-slope, high-speed medium-slope, high-speed low-slope, medium-speed high-slope, medium-speed medium-slope, medium-speed low-slope, low-speed high-slope, low-speed medium-slope, and low-speed low-slope, a total of nine categories, realizing double-layer clustering and classification of speed and slope.

[0056] In an embodiment of the present application, a genetic algorithm is used to select target segments in each speed and slope segment, including: defining a multi-objective optimization problem of the genetic algorithm; solving the multi-objective optimization problem based on the multi-objective optimization algorithm of the genetic algorithm; and determining the target segments according to the solution of the multi-objective optimization problem.

[0057] Wherein, multi-objective optimization refers to considering two or more conflicting or related objectives simultaneously when solving a problem, seeking one or more compromise solutions, and making each objective as optimal as possible, which is widely used in engineering design, resource allocation, scheduling, etc.

[0058] It can be understood that the embodiments of the present application explicitly define multiple optimization objectives to ensure that multiple key performance indicators are considered when selecting speed and slope segments. The global search capability and multi-objective optimization characteristics of the genetic algorithm are used to effectively explore the complex solution space and avoid falling into local optimum. The optimal speed and slope segments are selected in combination with the optimization results to ensure that the selected segments are representative and high quality.

[0059] In an embodiment of the present application, the multi-objective optimization problem includes control actions, control targets, and control constraints, wherein, The formula of the control action is:

[0060] Wherein,a The control action is represented by defining the selection of each kinematic segment as 1, and non-selection as 0.

[0061] The formula of the control target is:

[0062] wherein, J is the control target, p v is the Pearson coefficient of the speed distribution, p s is the Pearson coefficient of the slope distribution.

[0063] The formula of the control constraint is:

[0064] wherein, is the selection constraint of the i short-range segment of the first type, is the minimum time of the constructed typical working condition, is the maximum time of the constructed typical working condition, is the total time of the i short-range segment of the second type, is the total time of all short-range segments.

[0065] It can be understood that the Pearson coefficient of the speed distribution is calculated as follows:

[0066] wherein, V 1 is the speed distribution within the current short-range type, V 2 is the speed distribution of the short-range segment selected by the control action, wherein , t'_sum is the total driving time, , is the cumulative time of the speed distribution within the range of 0-10 km / h, , is the cumulative time of the speed distribution within the range of 10-20 km / h, , is the cumulative time of the speed distribution within the range of 20-30 km / h, , is the cumulative time of the speed distribution within the range of 30-40 km / h, , is the cumulative time of the speed distribution within the range of 40-50 km / h, , is the cumulative time of the speed distribution within the range of 50-60 km / h, , Cumulative time for speed distribution in the range 60-70 km / h, Cumulative time for speed distribution in the range 70-80 km / h, Cumulative time for speed distribution in the range 80 km / h and above, The Pearson coefficient for the slope distribution is calculated as follows: V 1 and V 2 covariance, and standard deviation of V 1 and V 2 respectively.

[0067] The Pearson coefficient for the slope distribution is calculated as follows:

[0068] wherein S 1 is the slope distribution within the current short trip class, S 2 is the slope distribution of the short trip segment selected by the control action, wherein , is the total slope time, , is the cumulative time for slope distribution in the range -30% -20%, , is the cumulative time for slope distribution in the range -20% -10%, , is the cumulative time for slope distribution in the range -10% -5%, , is the cumulative time for slope distribution in the range -50% 0%, , is the cumulative time for slope distribution in the range 0% 5%, , is the cumulative time for slope distribution in the range 5% 10%, , is the cumulative time for slope distribution in the range 10% 20%, , is the cumulative time for slope distribution in the range 20% 30%. The Pearson coefficient for the slope distribution is calculated as follows: S 1 and S 2 covariance, and standard deviation of S 1 and S 2 respectively.

[0069] ​​In an embodiment of the present application, a multi-objective optimization algorithm based on genetic algorithm is used to solve the multi-objective optimization problem, including: randomly generating an initial population, each individual representing a solution; evaluating the fitness value of each individual in the population and non-dominant sorting of individuals in the population; based on the fitness value and the non-dominant sorting result, using the roulette method to select target individuals for breeding, generating new individuals through crossover operation, and mutating the new individuals; adding the newly generated individuals to the population, and performing fitness evaluation and non-dominant sorting until the preset termination condition is met, then stopping the iteration of the multi-objective optimization algorithm.

[0070] wherein the initial population is a set of candidate solutions randomly generated at the beginning of the multi-objective optimization algorithm. Non-dominant sorting is a method of stratifying individuals in the population according to their advantages and disadvantages in multiple objectives, and the individuals are divided into multiple levels according to the "domination" relationship, the first layer of individuals is not dominated by any other individual; the second layer of individuals is only dominated by the first layer of individuals, and so on. The fitness value is a numerical indicator for measuring the quality or degree of excellence of an individual. The roulette method is a probability selection mechanism based on the fitness value, each individual is assigned a selection probability proportional to its fitness value, similar to occupying a corresponding arc length on the roulette wheel, by randomly rotating the roulette wheel to select individuals, the population diversity can be maintained while excellent individuals are preferentially selected, avoiding early convergence. The crossover operation is a recombination mechanism simulating the natural genetic process, extracting part of the gene fragments from two or more parent individuals, and generating new individuals through exchange or combination. Common crossover methods include single-point crossover, multi-point crossover, etc.

[0071] It can be understood that, in order to solve the multi-objective optimization problem, the multi-objective optimization method of genetic algorithm is used to select the short-travel construction condition, and the algorithm process is as follows: (1) Initialize the population: randomly generate an initial population, each individual representing a solution.

[0072] (2) Calculate the fitness: evaluate the fitness value of each individual in the population. For multi-objective optimization, the fitness function needs to consider multiple objectives, i.e. control objectives J .

[0073] (3) Non-dominant sorting: sort the individuals in the population to identify the Pareto frontier, wherein Pareto is used to describe the solution set that achieves the "optimal balance" between multiple objectives. The individuals on the frontier are non-dominated solutions, i.e. no other individual is better than these solutions in all objectives.

[0074] (4) Selection: according to the fitness value and the non-dominant sorting result, using the roulette method to select excellent individuals for breeding.

[0075] (5) Crossover and mutation: new individuals are generated by crossover operation, and the new individuals are mutated to increase the diversity of the population.

[0076] (6) Update population: new generated individuals are added to the population, and fitness evaluation and non-dominated sorting are performed to update the Pareto front.

[0077] (7) Termination condition: determine whether to end the algorithm according to the preset termination condition (such as the maximum number of iterations). If the termination condition is not reached, return to step (2) to continue iteration.

[0078] In an embodiment of the present application, the selection formula of the target segment is:

[0079] wherein, is the final segment selection solution set, is the minimum limit value of the Pearson coefficient of the speed distribution, is the minimum limit value of the Pearson coefficient of the slope distribution.

[0080] It can be understood that the Pareto front solution set X={ x 1 ,x 2 ....x n} constructed by each type of short trip segment is obtained, and logical judgment is performed using the selection formula of the target segment to select segments that represent both speed characteristics and slope characteristics and are selected from each type of short trip segment.

[0081] The embodiments of the present application will be further described below in conjunction with Figure 2 , as shown in Figure 2 , including: First, the embodiments of the present application use the Internet of Vehicles function to collect real vehicle running data, and the data contains speed information and slope information; second, short trips are divided according to from one idle speed to the next idle speed, and the corresponding speed-slope segments of the short trips are stored at the same time; third, the data is preprocessed, and the short trip segments with abnormal data values are removed or corrected; then, through the steps of abnormal data removal, key feature extraction, and hierarchical clustering, the data is divided into several representative short trip types, such as high-speed high-slope, medium-speed low-slope, etc. Nine types of driving states. Each type of segment reflects typical road conditions and driving behavior characteristics; In each type of segment, the embodiments of the present application use a multi-objective optimization method based on genetic algorithm to screen representative segments. The specific process is as follows: First, an initial population is randomly generated, each individual representing a piecewise combination scheme; the fitness value of the individual is calculated to measure its representativeness of the speed and slope characteristics of the class of fragments. The fitness evaluation is based on the Pearson correlation coefficient, which measures the correlation between the speed distribution and the slope distribution of the individual and the overall distribution of the class of fragments, respectively; then the population is non-dominantly sorted to identify non-dominant solutions on the Pareto frontier.

[0082] Based on the fitness and non-dominant sorting results, high-quality individuals are selected for breeding using the roulette method, and new individuals are generated through crossover and mutation to increase population diversity and explore better solutions; after the new individuals are added to the population, the fitness is re-evaluated and the Pareto frontier is updated. The process continues to iterate until the preset termination condition is met.

[0083] Finally, from the Pareto frontier solution set, select the individual with both speed and slope Pearson coefficients not lower than the specified threshold and the largest sum, as the representative working condition of the class of fragments, for subsequent typical working condition construction, to ensure that the selected fragments are highly representative in multiple dimensions.

[0084] According to the driving cycle construction method proposed in the embodiments of the present application, first, the original data of the actual road operation of the real vehicle is collected, and the required speed data and corresponding slope data are extracted therefrom; then, according to the joint characteristics of speed and slope, the entire journey is divided into a plurality of speed-slope combination fragments; then, the characteristics of these fragments are classified, and fragments with similar characteristics are classified into the same category; finally, in each category, a genetic algorithm is used to select representative target fragments, and finally a composite driving cycle containing speed and slope information is generated based on these target fragments, which can more accurately reflect the operating characteristics of the vehicle under real road conditions. Thus, the problem of ignoring the strong coupling of speed and slope in related technologies and being difficult to consider slope characteristics is solved.

[0085] Second, the driving cycle construction device according to the embodiments of the present application is described with reference to the accompanying drawings.

[0086] Figure 3 is a block schematic diagram of the driving cycle construction device according to the embodiments of the present application.

[0087] As shown in the figure, the driving cycle construction device 10 includes an acquisition module 101, a division module 102, and a generation module 103.

[0088] The obtaining module 101 is configured to obtain real vehicle running data, and extract speed data and gradient data in the real vehicle running data.

[0089] In an embodiment of the present application, the dividing module 102 is further configured to identify an idling time of the vehicle in the real vehicle running data, divide the vehicle trip into a plurality of sub-trips according to the idling time of the vehicle, and match the plurality of sub-trips with the speed data and the gradient data to obtain the plurality of speed and gradient segments.

[0090] In an embodiment of the present application, the generating module 103 is further configured to extract speed features and gradient features in the speed and gradient segments, cluster the speed and gradient segments according to the speed features to obtain a first clustering result, and cluster the segments in the first clustering result according to the gradient features to obtain a second clustering result.

[0091] In an embodiment of the present application, the generating module 103 is further configured to define a multi-objective optimization problem of a genetic algorithm, solve the multi-objective optimization problem based on a multi-objective optimization algorithm of the genetic algorithm, and determine the target segment according to a solution of the multi-objective optimization problem.

[0092] In an embodiment of the present application, the multi-objective optimization problem includes a control action, a control target, and a control constraint, wherein, The formula of the control action is:

[0093] wherein, a represents the control action, and the selection of each kinematic segment is defined as 1, and the non-selection is defined as 0.

[0094] The formula of the control target is:

[0095] wherein, J is the control target, p v is a Pearson coefficient of the speed distribution, p s is a Pearson coefficient of the gradient distribution.

[0096] The formula of the control constraint is:

[0097] wherein, is the i th control constraint, iClass short-stroke segment selection constraints, is the minimum time of the typical working condition constructed, is the maximum time of the typical working condition constructed, For the i The total time of the short-trip segment, is the total time of all short trip segments.

[0098] In one embodiment of the present application, the generation module 103 is further used to randomly generate an initial population, where each individual represents a solution; evaluate the fitness value of each individual in the population and perform non-dominated sorting on the individuals in the population: based on the fitness value and non-dominated sorting results, use the roulette method to select target individuals for reproduction, generate new individuals through crossover operations, and mutate the new individuals; add the newly generated individuals to the population, and perform fitness evaluation and non-dominated sorting until the preset termination condition is met, then stop the iterative solution of the multi-objective optimization algorithm.

[0099] In one embodiment of the present application, the target segment selection formula is:

[0100] in, Select the solution set for the final segment, is the minimum limit value of the Pearson coefficient of velocity distribution, is the minimum limiting value of the Pearson coefficient of the slope distribution.

[0101] It should be noted that the above explanation of the embodiment of the driving condition construction method is also applicable to the driving condition construction device of this embodiment, and will not be repeated here.

[0102] According to the driving condition construction device proposed in the embodiment of this application, first, raw data of the actual vehicle during its operation on the actual road is collected, and the required speed data and corresponding slope data are extracted from it. Then, based on the joint characteristics of speed and slope, the entire journey is divided into several speed-slope combination segments. Next, these segments are feature-classified, and segments with similar characteristics are classified into the same category. Finally, within each category, a genetic algorithm is used to screen the segments, from which representative target segments are selected. Finally, based on these target segments, a composite driving condition containing speed and slope information is generated to more accurately reflect the operating characteristics of the vehicle under real road conditions. This solves the problem that related technologies ignore the strong coupling between speed and slope and have difficulty in taking into account slope characteristics.

[0103] Figure 4 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include: The memory 401, the processor 402 and the computer program stored in the memory 401 and executable on the processor 402.

[0104] The processor 402 implements the driving condition construction method provided in the above embodiments when executing the program.

[0105] Further, the vehicle further comprises: The communication interface 403 is used for communication between the memory 401 and the processor 402.

[0106] The memory 401 is used for storing the computer program executable on the processor 402.

[0107] The memory 401 can include a high-speed RAM (Random Access Memory) memory, and can also include a non-volatile memory, for example, at least one disk memory.

[0108] If the memory 401, the processor 402 and the communication interface 403 are independently implemented, the communication interface 403, the memory 401 and the processor 402 can be connected to each other through a bus and complete communication between each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 4 In the figure, only one thick line is used to represent, but it does not mean that there is only one bus or one type of bus.

[0109] Optionally, in specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can complete communication between each other through an internal interface.

[0110] The processor 402 can be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0111] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the program is executed by the processor to implement the driving condition construction method as above.

[0112] The embodiment of the present application also provides a computer program, which, when executed by a processor, implements the driving condition construction method.

[0113] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms is not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0114] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "N" is at least two, for example, two, three, etc., unless otherwise explicitly specified.

[0115] Any process or method descriptions in flow charts or described elsewhere herein can be understood as representing code modules, segments, or portions of code which include one or more executable instructions for implementing the specified logical functions or steps, and the preferred embodiments of the present application also include the possibility that the functions described can be implemented using hardware, software, firmware, or any combination thereof. The preferred embodiments of the present application should be understood to include the possibility that the functions described can be implemented using hardware, software, firmware, or any combination thereof.

[0116] It should be understood that parts of the present application can be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. As in another embodiment, if implemented in hardware, any of the following technologies known in the art or their combinations can be used: discrete logic circuit with logic gate circuit for implementing logical functions on data signals, application specific integrated circuit with suitable combination logic gate circuit, programmable gate array, field programmable gate array, etc.

[0117] Those skilled in the art can understand that all or part of the steps of the method for implementing the above-mentioned embodiments can be instructed by a program to complete the relevant hardware, and the above-mentioned program can be stored in a computer readable storage medium. When the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0118] Although the embodiments of the present application have been shown and described above, it should be understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application. Those skilled in the art can make changes, modifications, replacements and variations to the above-mentioned embodiments within the scope of the present application.

Claims

1. A driving condition construction method, characterized in that: The following steps are involved: Acquiring actual vehicle operation data, and extracting speed data and slope data from the actual vehicle operation data; dividing the vehicle travel into a plurality of speed and gradient segments according to the speed data and the gradient data; The plurality of speed and slope segments are classified, a target segment is selected from each type of speed and slope segments using a genetic algorithm, and a speed and slope driving condition of the vehicle is generated according to each type of target segment.

2. The driving condition construction method according to claim 1, characterized in that: The step of dividing the vehicle travel into a plurality of speed and gradient segments according to the speed data and the gradient data includes: Identifying the idling moment of the vehicle in the actual vehicle operation data; Dividing the vehicle trip into a plurality of sub-trips according to the idle time of the vehicle; The plurality of sub-trips are matched with the speed data and the grade data to obtain a plurality of speed and grade segments.

3. The driving condition construction method according to claim 1, characterized in that: The classifying the plurality of speed and slope segments comprises: extracting speed features and slope features from the speed and slope segments; Clustering the speed and slope segments according to the speed feature to obtain a first clustering result; The segments in the first clustering result are clustered according to the slope feature to obtain a second clustering result.

4. The driving condition construction method according to claim 1, characterized in that: The method of selecting a target segment from each type of speed and gradient segments using a genetic algorithm includes: defining a multi-objective optimization problem of the genetic algorithm; Solving the multi-objective optimization problem using a multi-objective optimization algorithm based on a genetic algorithm; The target segment is determined according to the solution result of the multi-objective optimization problem.

5. The driving condition construction method according to claim 4, characterized in that: The multi-objective optimization problem includes control actions, control objectives and control constraints, wherein: The formula for the control action is: in, a Represents control action; The formula of the control objective is: in, J is the control target, p v is the Pearson coefficient of velocity distribution, p s is the Pearson coefficient of the slope distribution; The formula for the control constraint is: in, For the i Class short-stroke segment selection constraints, is the minimum time of the typical working condition constructed, is the maximum time of the typical working condition constructed, For the i The total time of the short-trip segment, is the total time of all short trip segments.

6. The driving condition construction method according to claim 4, characterized in that: The multi-objective optimization algorithm based on genetic algorithm solves the multi-objective optimization problem, including: Randomly generate the initial population, each individual represents a solution; Evaluate the fitness value of each individual in the population and perform non-dominated sorting on the individuals in the population: Based on the fitness value and non-dominated sorting results, the roulette wheel method is used to select the target individuals for reproduction, new individuals are generated through crossover operation, and the new individuals are mutated; The newly generated individuals are added to the population, and fitness evaluation and non-dominated sorting are performed until a preset termination condition is met, and then the iterative solution of the multi-objective optimization algorithm is stopped.

7. The driving condition construction method according to claim 4, characterized in that: The selection formula of the target fragment is: in, Select the solution set for the final segment, is the minimum limit value of the Pearson coefficient of velocity distribution, is the minimum limiting value of the Pearson coefficient of the slope distribution.

8. A driving condition construction device, characterized in that: include: An acquisition module is used to acquire actual vehicle operation data and extract speed data and slope data from the actual vehicle operation data; a dividing module, dividing the vehicle travel into a plurality of speed and slope segments according to the speed data and the slope data; The generating module classifies the plurality of speed and slope segments, selects a target segment in each type of speed and slope segment using a genetic algorithm, and generates a speed and slope driving condition of the vehicle according to each type of target segment.

9. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the driving condition construction method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed, the driving condition construction method according to any one of claims 1 to 7 is implemented.

Images