Method and device for generating test working condition of energy consumption of whole automobile and electronic equipment
By using multi-source traffic data clustering and particle swarm optimization algorithms to generate test conditions in hybrid vehicle testing, the problem of existing technologies being unable to accurately simulate the real driving behavior of hybrid vehicles is solved, achieving more efficient test coverage and accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CAS NEW ENERGY TECH CO LTD
- Filing Date
- 2026-02-11
- Publication Date
- 2026-05-15
AI Technical Summary
Existing standard test conditions cannot fully reflect the real driving behavior and road conditions of hybrid vehicles, especially the impact on their working mode switching and energy recovery, resulting in a lack of targeted test conditions.
By acquiring multi-source traffic data from real traffic environments, clustering and optimization processes are performed to generate more accurate test conditions that simulate the actual conditions of hybrid vehicles. Particle swarm optimization algorithm is used to optimize the test condition parameters, and a target test condition parameter vector is constructed to cover high-energy-consumption sensitive areas.
The generated test conditions can more accurately reflect actual driving characteristics, improve the accuracy and coverage of hybrid vehicle testing, and simulate more realistic vehicle operation.
Smart Images
Figure CN122048154A_ABST
Abstract
Description
Technical Field
[0001] The embodiments in this specification belong to the field of test condition simulation, and specifically relate to a method, apparatus and electronic equipment for generating test conditions for the energy consumption of a complete vehicle. Background Technology
[0002] When testing the energy consumption of hybrid vehicles, existing standard test cycles, such as the New European Driving Cycle (NEDC) and the Worldwide Harmonized Light Vehicles Test Cycle (WLTC), cannot fully reflect real-world driving behavior and road conditions. In particular, they have a significant impact on the switching of operating modes and energy recovery of hybrid vehicles, resulting in a lack of test cycles that can be used to specifically test hybrid vehicles. Summary of the Invention
[0003] The embodiments of this disclosure provide a method, apparatus, and electronic device for generating test conditions for vehicle energy consumption, which are intended to solve one or more of the above-mentioned problems and other potential problems.
[0004] According to a first aspect of this disclosure, a method for generating test conditions for vehicle energy consumption is provided. The method includes acquiring multi-source traffic data under real traffic conditions and slicing the multi-source traffic data into feature vectors for each window; clustering the feature vectors for each window to construct a basic test condition library, where each basic test condition corresponds to a cluster; for each basic test condition, outputting the equivalent energy consumption per unit distance from an energy consumption model based on the test condition parameter vector, ranking the sensitivity of each test condition parameter based on the equivalent energy consumption per unit distance, and assigning weights to each test condition parameter according to the sensitivity ranking; and optimizing the test condition parameter vector for each basic test condition using a particle swarm optimization algorithm based on the weights to obtain target test condition parameter vectors, and updating the basic test condition library based on each target test condition parameter vector, wherein the target test condition parameter vectors have the largest energy consumption sensitivity coverage and the smallest representative error.
[0005] According to a second aspect of this disclosure, a test condition generation device for vehicle energy consumption is provided. The device includes a data acquisition module configured to acquire multi-source traffic data under real traffic conditions and slice the multi-source traffic data into feature vectors for each window; a clustering module configured to cluster the feature vectors for each window to construct a basic test condition library, wherein each basic test condition corresponds to a cluster in the basic test condition library; a sorting module configured to, for each basic test condition, output the equivalent energy consumption per unit distance from the energy consumption model based on the test condition parameter vector, and sort the test condition parameters by sensitivity based on the equivalent energy consumption per unit distance, so as to assign weights to each test condition parameter according to the sensitivity sort; and an optimization module configured to optimize the test condition parameter vector for each basic test condition based on the weights using a particle swarm optimization algorithm to obtain a target test condition parameter vector, and update the basic test condition library based on each target test condition parameter vector, wherein the target test condition parameter vector has the largest energy consumption sensitivity coverage and the smallest representative error.
[0006] According to a third aspect of this disclosure, an electronic device is provided, including one or more processors and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform a method provided according to a first scheme.
[0007] According to a fourth aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method provided according to the first aspect.
[0008] The solutions provided in the embodiments of this specification can optimize and adjust the operating parameters of each basic operating condition in the basic operating condition library, so that the basic operating conditions can better cover high energy consumption sensitive areas, more accurately reflect actual driving characteristics, and thus more accurately simulate the actual situation of automobiles, especially hybrid vehicles, when using the basic operating conditions for testing. Attached Figure Description
[0009] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:
[0010] Figure 1 A flowchart illustrating a method for generating test conditions for vehicle energy consumption according to some embodiments of this disclosure is shown.
[0011] Figure 2 A schematic diagram of the structure of a test condition generation apparatus for vehicle energy consumption according to some embodiments of the present disclosure is shown.
[0012] Figure 3 A schematic block diagram of an electronic device according to some embodiments of the present disclosure is shown. Detailed Implementation
[0013] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0014] The terms “comprising” and “having”, and any variations thereof, in this specification, claims, and the foregoing drawings are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. Depending on the context, the word “if” as it applies herein may be interpreted as “when”, “in response to determination”, or “in response to detection”.
[0015] Figure 1 A flowchart illustrating a method 100 for generating test conditions for vehicle energy consumption according to some embodiments of this disclosure is shown. Method 100 can be executed by a terminal, which may include, but is not limited to, mobile phones, tablets, desktop computers, servers, etc. Figure 1 As shown, in method 100, step 102 can obtain multi-source traffic data in a real traffic environment and slice the multi-source traffic data into feature vectors for each window.
[0016] In this embodiment, firstly, multi-source traffic data from both vehicle and road perspectives is acquired under real-world traffic conditions. This multi-source traffic data specifically includes vehicle speed, wheel speed, generator speed and torque, and road data such as traffic flow, congestion index, and traffic light timing. Next, the multi-source traffic data is sliced according to pre-set windows (e.g., 1km as a road segment window, or 60s as a time window), resulting in several window feature vectors. Each window feature vector can be a feature vector composed of all types of multi-source traffic data within that window. As an example, a window feature vector can be represented as {average speed, speed variance, mean acceleration, acceleration volatility, idling time percentage, average gradient, braking event rate, road type label, and traffic density}.
[0017] In method 100, step 104 can cluster the feature vectors of each window to construct a basic working condition library, in which each basic working condition corresponds to a cluster.
[0018] In this embodiment, the clustering process of the window feature vector can be a two-level clustering. First, DBSCAN is used for denoising and identification of clustered outliers, and then K-means is used for refinement to obtain several well-clustered clusters. Each cluster is considered to correspond to a basic working condition. In K-means clustering, the K value can be adaptively determined using the silhouette coefficient or the Calinski-Harabasz exponent. Within each cluster, the parameter values corresponding to various working condition parameters can be the average of all values in the cluster, or all parameter values can be stored and randomly selected each time they are generated. By integrating the working condition parameters corresponding to all clusters, a basic working condition library can be constructed. Each basic working condition in the basic working condition library preferably includes a speed-time curve, acceleration / deceleration segment distribution, gradient segment, and classic signal events (such as parking / red light).
[0019] In method 100, step 106 can be used to output the equivalent energy consumption per unit distance from the energy consumption model for each basic working condition based on the working condition parameter vector, and to sort the sensitivity of each working condition parameter based on the equivalent energy consumption per unit distance, so as to assign weights to each working condition parameter according to the sensitivity sort.
[0020] In this embodiment, for each basic operating condition in the basic operating condition library, an operating condition parameter vector is constructed based on the corresponding operating condition parameters. Then, the energy consumption model outputs the equivalent energy consumption per unit distance. The energy consumption model can be an Equivalent Consumption Minimization Strategy (ECMS) model, a physical mechanism model. This model can be calculated based on preset physical formulas (such as driving resistance equations, engine universal characteristic diagrams, motor efficiency diagrams, and battery models) by inputting operating condition parameters such as vehicle speed, acceleration, road gradient, and pedal opening, and outputs instantaneous fuel consumption, electricity consumption, and equivalent energy consumption per unit distance. Through a global sensitivity analysis method (such as SOBOL index), the sensitivity of each operating condition parameter can be ranked to determine the contribution of each parameter to the equivalent energy consumption per unit distance. Different weights are assigned to different operating condition parameters based on the final sensitivity ranking; the higher the ranking, the greater the contribution of the operating condition parameter, and the higher its weight.
[0021] In method 100, step 108 can optimize the working condition parameter vector of each basic working condition based on the weights using the particle swarm optimization algorithm to obtain the target working condition parameter vector, and update the basic working condition library based on each target working condition parameter vector. The target working condition parameter vector has the largest energy consumption sensitivity coverage and the smallest representative error.
[0022] In this embodiment, the optimization objective is to maximize energy-sensitive coverage and minimize representative error. An evaluation function for the particle swarm optimization algorithm is constructed to optimize and update the parameter vectors corresponding to each basic operating condition until the evaluation function meets a preset convergence value. Then, based on the target operating condition parameter vector, the parameters corresponding to each basic operating condition in the basic operating condition library are updated again. This ensures that the updated operating condition parameters in the basic operating condition library can more accurately simulate the real road conditions required by hybrid vehicles, thus enabling better vehicle testing. Specifically, based on preliminary experiments, the energy-sensitive interval corresponding to each operating condition parameter can be determined. The coverage area of the operating condition parameter within the energy-sensitive interval can be determined based on the numerical range of the operating condition parameter, and the energy-sensitive coverage can be determined by weighting the coverage area with weights. The representative error can be determined based on the average error between the generated values of the basic operating conditions and the actually collected traffic data.
[0023] In one possible implementation, multi-source traffic data under real traffic conditions is acquired, and the multi-source traffic data is sliced into feature vectors for each window, including:
[0024] Acquire multi-source traffic data in real-world traffic environments, including vehicle-side and roadside data; and
[0025] After unifying the timestamps and preprocessing the multi-source traffic data, the multi-source traffic data is sliced according to the preset window to obtain the feature data of each window, and the window feature vector corresponding to each slice is constructed according to the feature data of each window.
[0026] In this embodiment, vehicle-side data may include CAN data, GNSS trajectories, and OBD / CAN logs (including speed, wheel speed, acceleration, pedal opening, gear position, engine / motor speed and torque, instantaneous fuel flow, battery voltage / current / SOC, braking event indicators, etc.) from real fleets / shared mobility / test vehicles. Roadside data may include traffic perception data (intersection cameras / traffic flow), navigation history trajectories, traffic light timing, road longitudinal slope and curvature (high-precision map), road segment speed limits and congestion indices, and meteorological data (temperature, wind speed), etc. After unifying the timestamps of each multi-source traffic data using UTC or PPS clocks to ensure a network time synchronization error of less than 10ms, the multi-source traffic data will be preprocessed. Preprocessing may include data cleaning (removing GPS jumps and missing CAN frame segments), padding, and low-pass filtering (e.g., 0.5 Hz filtering of speed signals to remove glitches, retaining transient acceleration data using a high-sampling channel). After preprocessing, the preprocessed multi-source traffic data will be sliced according to the preset window to obtain the window feature data in each slice window. Then, all the window feature data in the slice window will be integrated to construct the window feature vector.
[0027] In one possible implementation, the feature vectors of each window are clustered to construct a basic operating condition library, including:
[0028] Cluster the feature vectors of each window and fit the mapping curve between speed and time for each cluster. Determine the working parameters for each cluster based on the mapping curve and clustering data, and build a basic working condition library based on the working parameters.
[0029] In this embodiment, after clustering the feature vectors of each window, a speed-time curve can be fitted by selecting several central samples in the cluster or by using Dynamic Time Warping (DTW). This allows for the identification of statistical distribution data such as speed quantile curves, acceleration distribution, idling distribution, and gradient distribution based on the speed-time curve. In addition to the clustered data obtained from clustering, the operating parameters also include statistical distribution data obtained from the mapping curve.
[0030] In one possible implementation, the sensitivity ranking of various operating parameters is based on the equivalent energy consumption per unit distance, including:
[0031] Candidate influencing factors are determined from various operating parameters, and the target factor is obtained by screening from these candidates using the Morris method; and
[0032] Based on SOBOL global sensitivity analysis, the contribution of each target factor to the equivalent energy consumption per unit distance is determined, and the sensitivity of each target factor is ranked according to its contribution.
[0033] In this embodiment, several candidate influencing factors can be pre-selected from the operating parameters to characterize the high correlation between these candidate influencing factors and the equivalent energy consumption per unit distance. Next, the Morris method will be used to preliminarily screen each candidate influencing factor to obtain the target factors. The specific process can be as follows:
[0034] First, calculate using the elementary effect:
[0035]
[0036] Where f(·) is the equivalent energy consumption per unit distance output based on the values of each factor. Let be the factor value of the i-th factor in a k-dimensional vector. Let be the perturbation step size of the i-th factor.
[0037] Next, multiple calculations will be performed on r random trajectories. Calculate the mean and standard deviation of these values. A larger mean indicates a greater impact of the factor on the output, while a larger standard deviation indicates a non-linear factor or strong interaction with other factors. Finally, appropriate thresholds for the mean and standard deviation can be set to filter candidate influencing factors and retain target factors that meet the threshold requirements.
[0038] Next, a Sobol global sensitivity analysis will be performed on the target factors. Sobol global sensitivity analysis is a sensitivity analysis method based on variance decomposition, used to quantify the contribution of model input parameters to output uncertainty. It decomposes the total variance of the model output into components caused by each input parameter and their interactions, thereby determining the contribution of each target factor to the equivalent energy consumption per unit distance. Finally, the sensitivity can be ranked in descending order of contribution.
[0039] In one possible implementation, based on weights, the particle swarm optimization algorithm optimizes the parameter vector for each basic working condition to obtain the target working condition parameter vector, including:
[0040] For each operating condition parameter, the parameter range of the operating condition parameter is divided into various parameter intervals, and the sensitivity of feature change corresponding to each parameter interval is calculated. The parameter interval with the highest sensitivity of feature change is taken as the high sensitivity interval of the operating condition parameter.
[0041] An evaluation function is constructed based on the difference between energy consumption sensitive coverage and representative error. The particle swarm optimization algorithm optimizes the operating parameter vector of each basic operating condition according to the evaluation function until the convergence condition is met, and then the target operating parameter vector corresponding to each basic operating condition is obtained. The energy consumption sensitive coverage is determined based on the product between the coverage of the operating parameters in the high-sensitivity interval and the weight.
[0042] In this embodiment, for any operating condition parameter, the high-sensitivity interval is the region where energy consumption is most sensitive to changes in driving characteristics, not the region where the absolute energy consumption value is high when the vehicle is driving. Therefore, it is first necessary to determine the high-sensitivity interval for each operating condition parameter. Specifically, the parameter range of each operating condition parameter can be divided into several parameter intervals according to a preset number of divisions, and the standard deviation of energy consumption corresponding to each parameter interval can be calculated. The ratio of the standard deviation of energy consumption to the grid width (i.e., the width of the divided parameter interval) is used as the sensitivity to feature changes, and the parameter interval with the highest sensitivity to feature changes is determined as the high-sensitivity interval of that operating condition parameter.
[0043] After identifying the high-sensitivity intervals, for a specific basic operating condition, the parameter range of the operating condition is determined within that condition. Then, the coverage of each operating condition parameter within the high-sensitivity interval is determined. The coverage is then weighted according to its weight to obtain the energy consumption sensitivity coverage. The calculation formula is:
[0044]
[0045] in, The weight of the i-th operating condition parameter, Let represent the coverage of the i-th operating condition parameter.
[0046] Furthermore, the formula for calculating the representative error is:
[0047]
[0048] in, This is the generated value of the i-th operating condition parameter in the basic operating condition library. Let be the actual value of the i-th working condition parameter in the real scenario, and n be the total number of categories of working condition parameters.
[0049] The expression for the evaluation function is:
[0050]
[0051] in, , The weights can be adjusted as needed.
[0052] Under the constraints of the evaluation function, the target working condition parameter vector that satisfies the convergence condition can be obtained by iterative optimization using the particle swarm optimization algorithm.
[0053] In one possible implementation, the method further includes:
[0054] Acquire test data of the test bench under the updated basic working conditions, and verify the target working condition parameter vector based on the test data;
[0055] In response to the verification results indicating that the test error exceeds the error threshold, the adjustment weights of energy-sensitive coverage and representative error in the evaluation function are updated.
[0056] In this embodiment, the test vehicle can also be placed on a roller test bench, and the target operating condition parameter vector corresponding to the basic operating condition can be used as the input to the bench to simulate and test the actual test data corresponding to the test vehicle. The test data includes the energy consumption data obtained from the test, which is compared with the equivalent energy consumption per unit distance corresponding to the basic operating condition to determine the test error between the two. If the test error is not greater than the error threshold, the optimized target operating condition parameters are considered reliable. If the test error is greater than the error threshold, the optimized result is considered unsatisfactory, and the adjustment weight of the evaluation function will be adjusted, and optimization training will be carried out again until the test error is not greater than the error threshold.
[0057] Figure 2 This document illustrates a schematic diagram of a vehicle energy consumption test condition generation apparatus 200 according to some embodiments of the present disclosure. The various embodiments in this specification are described in a progressive manner; similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. Figure 2 As shown, the device 200 includes a data acquisition module 201, configured to acquire multi-source traffic data under real traffic conditions and slice the multi-source traffic data into feature vectors for each window; a clustering module 202, configured to cluster the feature vectors for each window to construct a basic operating condition library, in which each basic operating condition corresponds to a cluster; a sorting module 203, configured to, for each basic operating condition, output the equivalent energy consumption per unit distance from the energy consumption model based on the operating condition parameter vector, and sort the operating condition parameters by sensitivity based on the equivalent energy consumption per unit distance, so as to assign weights to each operating condition parameter according to the sensitivity sort; and an optimization module 204, configured to, based on the weights, optimize the operating condition parameter vectors for each basic operating condition using a particle swarm optimization algorithm to obtain target operating condition parameter vectors, and update the basic operating condition library based on each target operating condition parameter vector, wherein the target operating condition parameter vectors have the largest energy consumption sensitivity coverage and the smallest representative error.
[0058] In one possible implementation, the data acquisition module 201 is further configured to acquire multi-source traffic data under real traffic conditions, including vehicle-side data and road-side data; and after unifying and preprocessing the multi-source traffic data with timestamps, slice the multi-source traffic data according to a preset window to obtain feature data for each window, and construct a window feature vector corresponding to each slice according to the feature data of each window.
[0059] In one possible implementation, the clustering module 202 is further configured to cluster the feature vectors of each window and fit the mapping curve between speed and time corresponding to each cluster, and determine the working condition parameters corresponding to each cluster based on the mapping curve and the clustering data, so as to construct a basic working condition library based on the working condition parameters.
[0060] In one possible implementation, the sorting module 203 is further configured to determine candidate influencing factors among various operating parameters, select target factors from among the candidate influencing factors based on the Morris method, and determine the contribution of each target factor to the equivalent energy consumption per unit distance based on SOBOL global sensitivity analysis, and sort the target factors by sensitivity according to their contribution.
[0061] In one possible implementation, the optimization module 204 is further configured to divide the parameter range of each operating condition parameter into parameter intervals, and calculate the sensitivity of feature changes corresponding to each parameter interval. The parameter interval with the highest sensitivity of feature changes is taken as the high-sensitivity interval of the operating condition parameter. An evaluation function is constructed based on the difference between the energy consumption sensitivity coverage and the representative error. The particle swarm optimization algorithm optimizes the operating condition parameter vector of each basic operating condition according to the evaluation function until the convergence condition is met, and the target operating condition parameter vector corresponding to each basic operating condition is obtained. The energy consumption sensitivity coverage is determined based on the product between the coverage of the operating condition parameter in the high-sensitivity interval and the weight.
[0062] In one possible implementation, the apparatus further includes a verification module configured to acquire test data of the test bench under updated baseline conditions to verify the target operating condition parameter vector based on the test data; and in response to the verification result indicating that the test error is greater than an error threshold, to update the adjustment weights of energy-sensitive coverage and representative error in the evaluation function.
[0063] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this specification is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).
[0064] Figure 3 A block diagram of an electronic device 300 that can implement various embodiments of the present disclosure is shown. For example... Figure 3 As shown, the electronic device 300 includes a processor 310, a disk drive 320, an input / output interface 330, a network interface 340, and a memory 350. The processor 310, disk drive 320, input / output interface 330, network interface 340, and memory 350 can communicate with each other via a communication bus 360.
[0065] The processor 310 can be implemented using a general-purpose CPU, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs in order to implement the technical solution provided in this application.
[0066] The memory 350 can be implemented in the form of ROM (Read Only Memory), RAM (Read Access Memory), static memory, dynamic storage devices, etc. The memory 350 can store the operating system 351 used to control the operation of the electronic device 300, and the basic input / output system (BIOS) 352 used to control the low-level operations of the electronic device 300. Additionally, it can store a web browser 353, a data storage management system 354, etc. In summary, when the technical solution provided in this application is implemented through software or firmware, the relevant program code is stored in the memory 350 and is called and executed by the processor 310.
[0067] Input / output interface 330 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.
[0068] Network interface 340 is used to connect a communication module (not shown in the figure) to enable communication and interaction between the device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).
[0069] Bus 360 includes a pathway for transmitting information between various components of the device, such as processor 310, disk drive 320, input / output interface 330, network interface 340, and memory 350.
[0070] It should be noted that although the above-described device only shows the processor 310, disk drive 320, input / output interface 330, network interface 340, memory 350, bus 360, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the method of this application, and does not necessarily include all the components shown in the figures.
[0071] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0072] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. Furthermore, although operations are depicted in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the foregoing discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.
[0073] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.
Claims
1. A method for generating test conditions for vehicle energy consumption, characterized in that, The method includes: Acquire multi-source traffic data under real traffic conditions, and slice the multi-source traffic data into feature vectors for each window. Cluster the feature vectors of each window to construct a basic working condition library, in which each basic working condition corresponds to a cluster. For each basic operating condition, based on the operating condition parameter vector, the energy consumption model outputs the equivalent energy consumption per unit distance. The sensitivity of each operating condition parameter is ranked based on the equivalent energy consumption per unit distance, and weights are assigned to each operating condition parameter according to this sensitivity ranking. Based on the weights, the particle swarm optimization algorithm optimizes the operating condition parameter vector for each of the basic operating conditions to obtain the target operating condition parameter vector. The basic operating condition library is then updated based on each target operating condition parameter vector. The target operating condition parameter vector corresponds to the maximum energy consumption sensitivity coverage and the minimum representativeness error.
2. The method for generating test conditions for vehicle energy consumption according to claim 1, characterized in that, The process of acquiring multi-source traffic data under real-world traffic conditions, and slicing the multi-source traffic data into feature vectors for each window, includes: Acquire multi-source traffic data under real-world traffic conditions, including vehicle-side data and roadside data; and After unifying the timestamps and preprocessing the multi-source traffic data, the multi-source traffic data is sliced according to a preset window to obtain feature data for each window, and a window feature vector corresponding to each slice is constructed based on the feature data of each window.
3. The method for generating test conditions for vehicle energy consumption according to claim 1, characterized in that, The step of clustering the feature vectors of each window to construct a basic operating condition library includes: Cluster the feature vectors of each window and fit the mapping curve between speed and time for each cluster. Determine the working condition parameters for each cluster based on the mapping curve and the clustering data, so as to construct a basic working condition library based on the working condition parameters.
4. The method for generating test conditions for vehicle energy consumption according to claim 1, characterized in that, The sensitivity ranking of various operating parameters based on equivalent energy consumption per unit distance includes: Candidate influencing factors are determined from various operating parameters, and the target factor is obtained by screening from these candidates using the Morris method; and Based on SOBOL global sensitivity analysis, the contribution of each target factor to the equivalent energy consumption per unit distance is determined, and the sensitivity of each target factor is ranked according to its contribution.
5. The method for generating test conditions for vehicle energy consumption according to claim 1, characterized in that, The step of optimizing the working condition parameter vector for each of the basic working conditions based on the weights using a particle swarm optimization algorithm to obtain the target working condition parameter vector includes: For each operating condition parameter, the parameter range of the operating condition parameter is divided into various parameter intervals, and the sensitivity of feature change corresponding to each parameter interval is calculated. The parameter interval with the highest sensitivity of feature change is taken as the high sensitivity interval of the operating condition parameter. An evaluation function is constructed based on the difference between energy consumption sensitive coverage and representative error. The particle swarm optimization algorithm optimizes the operating condition parameter vector for each basic operating condition according to the evaluation function until the convergence condition is met, and then obtains the target operating condition parameter vector for each basic operating condition. The energy consumption sensitive coverage is determined based on the product between the coverage of the operating condition parameter in the high sensitivity interval and the weight.
6. The method for generating test conditions for vehicle energy consumption according to claim 5, characterized in that, The method further includes: Acquire test data of the test bench under the updated basic working conditions, and verify the target working condition parameter vector based on the test data; In response to the verification results indicating that the test error is greater than the error threshold, the adjustment weights of energy-sensitive coverage and representative error in the evaluation function are updated.
7. A device for generating test conditions for the energy consumption of a complete vehicle, characterized in that, The device includes: The data acquisition module is configured to acquire multi-source traffic data under real traffic conditions and slice the multi-source traffic data into feature vectors for each window. The clustering module is configured to cluster the feature vectors of each window to build a basic working condition library, wherein each basic working condition in the basic working condition library corresponds to a cluster. The sorting module is configured to, for each basic operating condition, based on the operating condition parameter vector, output the equivalent energy consumption per unit distance from the energy consumption model, and sort the operating condition parameters by sensitivity based on the equivalent energy consumption per unit distance, thereby assigning weights to each operating condition parameter according to the sensitivity sort; and The optimization module is configured to optimize the operating condition parameter vector of each of the basic operating conditions based on the weights using a particle swarm optimization algorithm to obtain a target operating condition parameter vector, and update the basic operating condition library based on each target operating condition parameter vector. The target operating condition parameter vector corresponds to the maximum energy consumption sensitivity coverage and the minimum representativeness error.
8. An electronic device, characterized in that, include: One or more processors, and A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method for generating test conditions for the energy consumption of a vehicle as described in any one of claims 1-6.
9. A computer program product, characterized in that, The system includes a computer program that, when executed by a processor, implements a method for generating test conditions for the energy consumption of a vehicle according to any one of claims 1-6.