Method for testing overall performance of electric tricycle

Abstract

This invention discloses a method for testing the overall performance of an electric tricycle, relating to the field of vehicle performance testing technology. The method includes: data mining based on multiple operating scenario factors of the electric tricycle to construct Q test scenarios; performing multi-dimensional performance expectation prediction to obtain the expected power performance, expected electrical performance, and expected comfort performance for the Q scenarios; performing power characteristic detection on the electric tricycle to obtain power characteristic detection results; performing electrical characteristic detection on the electric tricycle to obtain electrical characteristic detection results; performing comfort characteristic detection on the electric tricycle to obtain comfort characteristic detection results; and constructing a visual report on the overall vehicle performance of the electric tricycle based on the power characteristic detection results, electrical characteristic detection results, and comfort characteristic detection results. This invention solves the technical problem that existing technologies cannot comprehensively evaluate the multi-dimensional performance of electric tricycles, achieving the technical effect of comprehensively evaluating the multi-dimensional performance of electric tricycles.

Description

Technical Field

[0001] This invention relates to the field of vehicle performance testing technology, specifically to a method for testing the overall performance of an electric tricycle. Background Technology

[0002] With the widespread use of electric tricycles, especially in urban transportation, logistics, and short-distance travel, the requirements for their overall performance are becoming increasingly stringent. The performance of electric tricycles includes not only traditional power performance but also electrical performance and comfort performance. Currently, the performance evaluation of electric tricycles mainly relies on testing methods under single scenarios, lacking a comprehensive evaluation system for various complex operating scenarios. Summary of the Invention

[0003] This application provides a method for testing the overall performance of electric tricycles, which addresses the technical problem that existing technologies cannot comprehensively evaluate the multi-dimensional performance of electric tricycles.

[0004] In view of the above problems, this application provides a method for testing the overall performance of electric tricycles.

[0005] This application provides a method for testing the overall performance of an electric tricycle, the method comprising:

[0006] Data mining is performed on the electric tricycle based on multiple operating scenario factors to construct Q test scenarios. These factors include road conditions, environmental conditions, and load conditions, with Q being a positive integer greater than 1. Multidimensional performance expectation prediction is then performed on the electric tricycle based on these Q test scenarios to obtain Q scenarios of expected power performance, Q scenarios of expected electrical performance, and Q scenarios of expected comfort performance. Based on these Q test scenarios, power feature detection is performed on the electric tricycle according to the expected power performance, yielding power feature detection results. Similarly, electrical feature detection is performed on the electric tricycle according to the expected electrical performance, yielding electrical feature detection results. Finally, comfort feature detection is performed on the electric tricycle based on the expected comfort performance, yielding comfort feature detection results. A visual report of the electric tricycle's overall vehicle performance is then constructed based on the power feature detection results, the electrical feature detection results, and the comfort feature detection results.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] This application utilizes data mining based on multiple operating scenario factors of electric tricycles to construct Q test scenarios. These multiple operating scenario factors include road conditions, environmental conditions, and load conditions, where Q is a positive integer greater than 1. Based on these Q test scenarios, multi-dimensional performance expectations of the electric tricycle are predicted, resulting in Q expected power performance, Q expected electrical performance, and Q expected comfort performance. Based on these Q test scenarios, power feature detection is performed on the electric tricycle according to the expected power performance, yielding power feature detection results. Similarly, electrical feature detection is performed on the electric tricycle according to the expected electrical performance, yielding electrical feature detection results. Finally, based on the power feature detection results, electrical feature detection results, and comfort feature detection results, a visual report on the overall vehicle performance of the electric tricycle is constructed. This invention addresses the technical problem that existing technologies cannot comprehensively evaluate the multi-dimensional performance of electric tricycles. By constructing Q test scenarios based on multiple operating scenario factors, and combining data mining technology to predict the power, electrical, and comfort performance of electric tricycles in different scenarios, a feature detection model is used to detect power, electrical, and comfort features respectively to obtain various performance data. The detection results are summarized and visualized to form a vehicle performance report, achieving the technical effect of comprehensively evaluating the multi-dimensional performance of electric tricycles. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of the process for testing the overall performance of an electric tricycle, as provided in an embodiment of this application.

[0011] Figure 2 This is a schematic diagram illustrating the process of obtaining the power characteristic detection results in the whole vehicle performance testing method for electric tricycles provided in the embodiments of this application. Detailed Implementation

[0012] This application provides a method for testing the overall performance of electric tricycles, addressing the technical problem that existing technologies cannot comprehensively evaluate the multi-dimensional performance of electric tricycles. By constructing Q test scenarios based on multiple operating scenario factors, and combining data mining techniques to predict the power, electrical, and comfort performance of electric tricycles in different scenarios, a feature detection model is used to detect power, electrical, and comfort features respectively, obtaining various performance data. The detection results are summarized and visualized to form a vehicle performance report, achieving the technical effect of comprehensively evaluating the multi-dimensional performance of electric tricycles.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only 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] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0015] Examples, such as Figure 1 As shown, this application provides a method for testing the overall performance of an electric tricycle, the method comprising:

[0016] Step S100: Perform data mining based on the multi-factor operation of the electric tricycle to construct Q test scenarios, wherein the multi-factor operation includes the road condition, the environment condition, and the load condition, and Q is a positive integer greater than 1.

[0017] In this embodiment, the multiple operating scenario factors include road conditions, environmental conditions, and load conditions. Road conditions refer to the type of road, road conditions, gradient, and traffic conditions on which the tricycle travels. Environmental conditions refer to external environmental conditions, such as temperature, humidity, wind speed, and other weather factors. Load conditions refer to the tricycle's performance under different load conditions.

[0018] When performing data mining based on the diverse operating scenario factors of electric tricycles, data mining techniques, such as cluster analysis, are employed to identify and construct multiple test scenarios. Specifically, historical data on the diverse operating scenario factors are first obtained from a historical database, and representative scenario patterns are identified using data mining techniques, such as cluster analysis.

[0019] Next, based on the mined multivariate factor data, multiple possible test scenario schemes were constructed. Each test scenario includes different combinations of roads, environments, and loads. Subsequently, pairwise similarity assessments were performed on these test scenarios, and similarity coefficients were calculated to determine the similarity between different scenarios.

[0020] Finally, based on a pre-set similarity threshold, Q test scenarios with similarity below the threshold are selected.

[0021] Furthermore, the method provided in the application embodiment, which constructs Q test scenarios by performing data mining based on multiple operating scenario factors of the electric tricycle, also includes:

[0022] Test scenarios are set for the electric tricycle based on the multi-factor operating scenarios to obtain a set of test scenario schemes; pairwise similarity evaluation is performed on the set of test scenario schemes to obtain a set of test scenario similarity coefficients; based on the set of test scenario similarity coefficients, the set of test scenario schemes is filtered according to a test scenario similarity threshold to obtain the Q test scenarios that are less than the test scenario similarity threshold.

[0023] In this embodiment, test scenarios are first set based on the multiple operating scenario factors of the electric tricycle. Each test scenario includes different combinations of these three factors; that is, each test scenario is determined by multiple factors (such as road conditions, environmental conditions, and load conditions). Specifically, firstly, operating data of the electric tricycle under different conditions is collected from a historical database, including multi-dimensional indicators such as vehicle speed, acceleration, power consumption, and battery status. Subsequently, data mining techniques, such as cluster analysis (e.g., K-means clustering), are used to analyze this collected data. Through cluster analysis, test scenarios with similar characteristics are grouped into the same category based on factors such as road conditions, environmental conditions, and load conditions, with each category representing a specific test scenario. Therefore, multiple sets of test scenario schemes containing different combinations of road, environmental, and load conditions are ultimately obtained.

[0024] Next, pairwise similarity assessments are performed based on the test scenario set. Cosine similarity is used to calculate the similarity between scenarios. Specifically, cosine similarity measures the similarity between the factor vectors of the test scenarios. Higher similarity indicates greater similarity in road, environmental, or load conditions between the two test scenarios; conversely, lower similarity indicates significant differences. A set of test scenario similarity coefficients is obtained through this calculation.

[0025] Next, to remove redundancy and ensure the representativeness of the test scenarios, a selection process is performed based on the obtained test scenario similarity coefficient set. The test scenario similarity coefficient set is compared with a preset test scenario similarity threshold to filter the test scenario scheme set, resulting in Q test scenarios with similarity coefficients lower than the threshold. Specifically, for each test scenario scheme, its similarity coefficient with other scenarios is compared with the preset test scenario similarity threshold. If the similarity coefficient with all other scenarios is lower than the test scenario similarity threshold, the scenario is considered independent and representative. Finally, Q test scenarios are selected.

[0026] Step S200: Perform multi-dimensional performance expectation prediction on the electric tricycle based on the Q test scenarios to obtain the expected power performance, expected electrical performance, and expected comfort performance for the Q scenarios.

[0027] In this embodiment, the process of multi-dimensional performance expectation prediction for an electric tricycle includes predictions of power performance, electrical performance, and comfort performance. Specifically, firstly, vehicle manufacturing process information of the electric tricycle is obtained, such as motor parameters, drive system configuration, and battery technology. Then, based on Q test scenarios and the vehicle manufacturing process information, expected predictions are made for power performance, electrical performance, and comfort performance. In the power performance expectation prediction, the expected maximum speed, expected maximum climbing angle, and expected braking distance for each scenario are predicted according to the road, environmental, and load conditions of the test scenarios. In the electrical performance expectation prediction, the expected driving range, motor efficiency, and motor temperature rise are predicted under the same scenario conditions. Finally, in the comfort performance expectation prediction, the expected driving noise level, driving vibration amplitude, and vibration frequency are predicted for each scenario.

[0028] Through these steps, we can ultimately obtain the expected power performance, expected electrical performance, and expected comfort performance for Q scenarios.

[0029] Furthermore, in the method provided in the application embodiments, the method further includes performing multi-dimensional performance expectation prediction on the electric tricycle based on the Q test scenarios to obtain the expected power performance, expected electrical performance, and expected comfort performance in the Q scenarios, and also includes:

[0030] Obtain the vehicle manufacturing process information of the electric tricycle; based on the Q test scenarios and the vehicle manufacturing process information, predict the expected power performance of the electric tricycle to generate the Q expected power performance for each scenario, wherein the expected power performance for each scenario includes the expected maximum speed, expected maximum climbing angle, and expected braking distance corresponding to each test scenario; based on the Q test scenarios and the vehicle manufacturing process information, predict the expected electrical performance of the electric tricycle to generate the Q expected electrical performance for each scenario, wherein the expected electrical performance for each scenario includes the expected driving range, expected motor efficiency, and expected motor temperature rise corresponding to each test scenario; based on the Q test scenarios and the vehicle manufacturing process information, predict the expected comfort performance of the electric tricycle to generate the Q expected comfort performance for each scenario, wherein the expected comfort performance for each scenario includes the expected driving noise level, expected driving vibration amplitude, and expected driving vibration frequency corresponding to each test scenario.

[0031] In this embodiment of the application, the vehicle manufacturing process information of the electric tricycle that has been uploaded in advance is first obtained, including the key technical parameters of the electric tricycle, such as motor power, drive system configuration, battery capacity, battery management system, braking system and frame design.

[0032] Next, based on Q test scenarios and vehicle manufacturing process information, expected power performance is predicted to generate the expected power performance for each test scenario. Specifically, first, the test scenario records, vehicle manufacturing process records, and scenario-standard power performance records for the electric tricycle are loaded. The test scenario records include road type, environmental conditions, and load conditions, while the scenario-standard power performance records provide the standard power performance of the electric tricycle under different test scenarios, such as expected maximum speed, maximum climbing angle, and braking distance. Then, supervised learning methods, such as neural networks, are used to train the system, taking the test scenario records and vehicle manufacturing process information as input data and the scenario-standard power performance records as the target output. The training objective is to minimize the power performance expectation prediction loss coefficient, i.e., to reduce the error between the predicted result and the actual standard power performance. After training, if the prediction loss coefficient is less than the preset power performance expectation prediction loss threshold, a power performance expectation prediction model is generated. At this point, the q-th test scenario and vehicle manufacturing process information are input into this model to obtain the expected power performance for that test scenario, which is then added to the list of Q scenario expected power performances. By inputting one test scenario from each of the Q test scenarios and vehicle manufacturing process information into the model, the expected power performance for each of the Q scenarios is obtained. The expected power performance for each scenario includes the expected maximum speed, expected maximum climbing angle, and expected braking distance for that specific test scenario.

[0033] Similarly, based on Q test scenarios and vehicle manufacturing process information, expected electrical performance is predicted, generating expected electrical performance for Q scenarios. The specific process is similar to the power performance prediction: first, the test scenario records and vehicle manufacturing process information of the electric tricycle are loaded; then, a supervised learning method is used to train the electrical performance prediction model. The model aims to predict indicators such as expected driving range, motor efficiency, and motor temperature rise for each test scenario. Finally, the trained electrical performance prediction model is applied to each test scenario to obtain Q expected electrical performance values ​​for each scenario, which are then added to the corresponding lists. The expected electrical performance for each scenario includes the expected driving range, expected motor efficiency, and expected motor temperature rise for each test scenario.

[0034] Finally, based on Q test scenarios and vehicle manufacturing process information, the expected comfort performance of the electric tricycle is predicted, generating expected comfort performance for Q scenarios. This prediction process is similar to that of power performance and electrical performance prediction. After loading relevant data, a supervised learning method is used to train the expected comfort performance prediction model. The model aims to predict the expected driving noise level, vibration amplitude, and vibration frequency for each test scenario. The trained expected comfort performance prediction model is applied to each test scenario to obtain the expected comfort performance for Q scenarios. The expected comfort performance for each scenario includes the expected driving noise level, expected driving vibration amplitude, and expected driving vibration frequency corresponding to each test scenario.

[0035] Through the above steps, the expected power performance, expected electrical performance, and expected comfort performance are generated for each test scenario.

[0036] Furthermore, in the method provided in the application embodiment, the expected power performance of the electric tricycle is predicted based on the Q test scenarios and the vehicle manufacturing process information to generate the expected power performance for the Q scenarios, and the method further includes:

[0037] Load electric tricycle test scenario records, electric tricycle production process records, and scenario standard power performance records; use the electric tricycle test scenario records and electric tricycle production process records as input information, and the scenario standard power performance records as output information to perform supervised training on a predetermined learning model to obtain the power performance expectation prediction loss coefficient; if the power performance expectation prediction loss coefficient is less than the power performance expectation prediction loss threshold, generate a power performance expectation prediction model; input the q-th test scenario and the vehicle production process information into the power performance expectation prediction model to obtain the q-th scenario expectation power performance, and add the q-th scenario expectation power performance to the Q scenario expectation power performances.

[0038] In this embodiment, the test scenario records, manufacturing process records, and standard power performance records for electric tricycles are first loaded from a historical database. The test scenario records include different test conditions, such as road type, ambient temperature, humidity, and vehicle load status, which directly affect the vehicle's dynamic performance. The manufacturing process records provide manufacturing parameters, including motor power, drive system configuration, battery type and capacity, and braking system design, which significantly affect the electric tricycle's power performance. The standard power performance records contain known standard power performance data for the electric tricycle under different test scenarios, such as the expected maximum speed, maximum climbing angle, and braking distance.

[0039] Once the input data is ready, the next step is to train a supervised learning model. Specifically, using records of electric tricycle test scenarios and production processes as input features, and records of standard power performance under different scenarios as the target output, the predetermined learning model is trained using methods such as support vector machine regression. The goal of supervised learning is to minimize the error between predicted and actual values; that is, the model learns how to predict power performance under a given test scenario and production process through the input features.

[0040] During training, the expected prediction loss coefficient for dynamic performance is calculated. This is a metric that evaluates model performance by comparing the predicted results with the actual standard dynamic performance. The expected prediction loss coefficient for dynamic performance is obtained by calculating the error between the predicted value and the true value using the mean squared error, and measures the difference between the model's prediction and the actual data.

[0041] If the predicted loss coefficient for the expected power performance after training is less than the predicted loss threshold for the expected power performance preset by technical experts, then the model is considered to have reached the required accuracy and can be used for actual prediction. At this point, using this trained expected power performance prediction model, the q-th test scenario and the production process information of the electric tricycle are input into the model to obtain the expected power performance for the q-th scenario, including the expected maximum speed, maximum climbing angle, and braking distance. Finally, the expected power performance for the q-th scenario is added to the expected power performance for all Q scenarios.

[0042] Step S300: Based on the Q test scenarios, perform power characteristic detection on the electric tricycle according to the expected power performance of the Q scenarios, and obtain the power characteristic detection results.

[0043] In this embodiment of the application, when performing power characteristic detection on an electric tricycle, a power performance confidence test is first performed to obtain the actual test power performance under each test scenario. Then, the test power performance for each scenario is compared with the corresponding expected power performance to identify the deviations, and a power performance deviation matrix containing all scenarios is generated.

[0044] Subsequently, the power performance deviation matrix is ​​input into a power performance testing model to further calculate the power performance testing coefficients. These coefficients are used to quantify the differences in power performance for each test scenario. By calculating the reciprocal of the variance of these testing coefficients, the power performance stability is obtained, which is an indicator measuring the consistency of power output of the electric tricycle under different test scenarios. Finally, the power performance stability, the tested power performance, and the power performance testing coefficients are combined and output to form the power characteristic testing results.

[0045] Furthermore, such as Figure 2 As shown, the method provided in the application embodiment, which performs power characteristic detection on the electric tricycle based on the Q expected power performance scenarios to obtain power characteristic detection results, further includes:

[0046] Based on the Q test scenarios, a power performance confidence test is performed on the electric tricycle to obtain the test power performance of Q scenarios. Based on the expected power performance of the Q scenarios, deviations in the test power performance of the Q scenarios are identified to generate a power performance deviation matrix for the Q scenarios. The power performance deviation matrix for the Q scenarios is input into a power performance detection model to obtain power performance detection coefficients for the Q scenarios. The reciprocal of the variance of the power performance detection coefficients for the Q scenarios is used as the power performance stability. The power performance stability, the test power performance of the Q scenarios, and the power performance detection coefficients for the Q scenarios are output as the power feature detection result.

[0047] In this embodiment, the electric tricycle is first subjected to a power performance confidence test based on Q test scenarios. During this stage, the electric tricycle undergoes actual testing under different test scenario conditions (such as load, road conditions, and environment) to obtain the test power performance for each scenario. Through this process, the test power performance for Q scenarios is obtained.

[0048] Next, based on the expected power performance of Q scenarios, deviation identification is performed on the tested power performance of Q scenarios. That is, the expected power performance of Q scenarios is compared with the tested power performance of Q scenarios. By comparing the difference between the actual and the expected, the deviation in each scenario is identified, and a power performance deviation matrix of Q scenarios is generated.

[0049] Subsequently, the Q scenario dynamic performance deviation matrices are input into the dynamic performance detection model to obtain Q scenario dynamic performance detection coefficients. The dynamic performance detection model is pre-trained. During training, historical test data, including the dynamic performance deviation matrices and corresponding dynamic performance detection coefficients for historical scenarios, are first obtained from a historical database. Using the historical scenario dynamic performance deviation matrices as input data and the dynamic performance detection coefficients for each test scenario as output data, a multilayer perceptron neural network is trained to obtain the dynamic performance detection model.

[0050] Then, using the power performance detection coefficients of Q scenarios, the variance of the detection coefficients is calculated, and the reciprocal of the variance of the power performance detection coefficients of Q scenarios is taken as the power performance stability.

[0051] Finally, the dynamic performance stability, the dynamic performance tested in Q scenarios, and the dynamic performance detection coefficients in Q scenarios are summarized as the final dynamic characteristic detection result.

[0052] Furthermore, in the method provided in the application embodiment, the power performance confidence test is performed on the electric tricycle according to the Q test scenarios to obtain the power performance of the Q scenarios, and the method further includes:

[0053] Based on the Q test scenarios, the q-th test scenario is extracted, where q is a positive integer, 1 ≤ q ≤ Q. Multiple maximum speed tests are performed on the electric tricycle according to the q-th test scenario to obtain a maximum speed test dataset. Multiple maximum climbing angle tests are performed on the electric tricycle according to the q-th test scenario to obtain a maximum climbing angle test dataset. Braking distance tests are performed on the electric tricycle according to the q-th test scenario to obtain a braking distance test dataset. The mean values ​​of the maximum speed test dataset, the maximum climbing angle test dataset, and the braking distance test dataset are calculated to obtain the confidence maximum speed, confidence maximum climbing angle, and confidence braking distance for the q-th scenario. The confidence maximum speed, confidence maximum climbing angle, and confidence braking distance for the q-th scenario are output as the test power performance for the q-th scenario, and this test power performance for the q-th scenario is added to the test power performance of the Q scenarios.

[0054] In this embodiment, the q-th test scenario is first randomly selected from Q test scenarios, where q is a positive integer and 1 ≤ q ≤ Q. Each test scenario includes different environmental conditions, road conditions, and load conditions, and the performance of the electric tricycle will change according to these factors.

[0055] Next, based on the q-th test scenario, a series of specific performance tests are conducted on the electric tricycle. First, a maximum speed test is performed, with multiple tests conducted to record the maximum speed of the electric tricycle in each scenario. This data forms the maximum speed test dataset. Then, in the same test scenario, a maximum climbing angle test is performed. Multiple tests are conducted to measure the maximum climbing angle that the electric tricycle can climb. Finally, the maximum climbing angles measured in multiple tests are rounded to obtain the maximum climbing angle test dataset. Next, based on the q-th test scenario, a braking distance test is performed on the electric tricycle. Multiple braking tests are conducted, and the braking distance required from maximum speed to stopping in each test is recorded. The longest braking distance is taken as the result of each test. Finally, these longest braking distance values ​​are integrated to obtain the braking distance test dataset.

[0056] Then, the mean values ​​of the maximum vehicle speed test dataset, the maximum climb angle test dataset, and the braking distance test dataset were calculated respectively. The mean value of the maximum vehicle speed test dataset was calculated to obtain the confidence maximum vehicle speed for the q-th scenario; the mean value of the maximum climb angle test dataset was calculated to obtain the confidence maximum climb angle for the q-th scenario; and the mean value of the braking distance test dataset was calculated to obtain the confidence braking distance for the q-th scenario.

[0057] Finally, the calculated confidence maximum vehicle speed, confidence maximum climbing angle, and confidence braking distance for the q-th scenario are output as the test power performance for the q-th scenario, and the test power performance for the q-th scenario is added to the test power performance of the Q scenarios.

[0058] To obtain the global test results, we iterate through all Q test scenarios and repeat the above steps for the test results of each scenario, ultimately obtaining the test dynamic performance of Q scenarios.

[0059] Step S400: Based on the Q test scenarios, perform electrical feature detection on the electric tricycle according to the expected electrical performance of the Q scenarios, and obtain the electrical feature detection results.

[0060] In this embodiment, based on Q test scenarios, electrical performance factors for the electric tricycle are first set, including key electrical performance indicators such as range, motor efficiency, and motor temperature rise. Then, for each test scenario, an electrical performance confidence test is performed to obtain test electrical performance data for each scenario. Next, by comparing the actual electrical performance with the expected electrical performance for each test scenario, deviation identification is performed to generate an electrical performance deviation matrix. This deviation matrix is ​​then input into a pre-trained electrical performance detection model to calculate the electrical performance detection coefficients for each scenario. Next, by calculating the reciprocal of the variance of these coefficients, the electrical performance stability is obtained. Finally, the electrical performance stability, test electrical performance, and electrical performance detection coefficients are integrated and output to generate the electrical feature detection results.

[0061] Furthermore, in the method provided in the application embodiment, based on the Q test scenarios, electrical feature detection is performed on the electric tricycle according to the expected electrical performance of the Q scenarios to obtain electrical feature detection results, which further includes:

[0062] The electric tricycle is configured with electrical performance factors, including driving range, motor efficiency, and motor temperature rise. Based on these factors, electrical performance confidence tests are performed on the electric tricycle under Q test scenarios to obtain Q scenario test electrical performances. Deviations in the Q scenario test electrical performances are identified based on the expected electrical performances under the Q scenarios, generating a Q scenario electrical performance deviation matrix. The Q scenario electrical performance deviation matrix is ​​input into an electrical performance detection model to obtain Q scenario electrical performance detection coefficients. The reciprocal of the variance of the Q scenario electrical performance detection coefficients is calculated to obtain the electrical performance stability. The electrical performance stability, the Q scenario test electrical performances, and the Q scenario electrical performance detection coefficients are output as the electrical feature detection results.

[0063] In this embodiment, the electrical performance factors of the electric tricycle are first set, including driving range, motor efficiency, and motor temperature rise. Next, based on these electrical performance factors, electrical performance confidence tests are conducted on the electric tricycle under Q test scenarios. Specifically, the process of conducting the electrical performance confidence test is similar to the previous power performance confidence test. The electrical performance confidence test first randomly extracts the q-th test scenario from the Q test scenarios. Then, the electric tricycle is tested multiple times in the randomly extracted scenario to obtain electrical performance data under different scenarios, including indicators such as maximum driving range, maximum motor efficiency, and maximum motor temperature rise. These data form the maximum driving range test dataset, the maximum motor efficiency test dataset, and the maximum motor temperature rise test dataset, respectively. Subsequently, the mean values ​​of the maximum driving range test dataset, the maximum motor efficiency test dataset, and the maximum motor temperature rise test dataset are calculated to obtain the maximum driving range, maximum motor efficiency, and maximum motor temperature rise of the q-th scenario. The calculated maximum driving range, maximum motor efficiency, and maximum motor temperature rise for scenario q are then output as the electrical performance test results for scenario q. These results are then added to the total electrical performance test results for all Q scenarios. To obtain the global test results, all Q test scenarios are iterated over, and the above steps are repeated for each scenario's test results, ultimately yielding the electrical performance test results for all Q scenarios.

[0064] Next, based on the expected electrical performance of Q scenarios, deviation identification is performed on the tested electrical performance of Q scenarios. That is, the difference between the actual test data and the expected value in each scenario is calculated, and based on these deviations, an electrical performance deviation matrix for Q scenarios is generated.

[0065] Subsequently, the Q scene electrical performance deviation matrices are input into the electrical performance detection model to obtain the Q scene electrical performance detection coefficients. The electrical performance detection model is pre-trained. During training, a set of historical electrical performance deviation matrices and corresponding electrical performance detection coefficient sets are first obtained from a historical database. The multilayer perceptron is trained using the historical electrical performance deviation matrix set as input data and the electrical performance detection coefficient set as output data to obtain the electrical performance detection model.

[0066] Then, the reciprocal of the variance of the electrical performance detection coefficients for the Q scenarios is calculated to obtain the electrical performance stability. The electrical performance stability, the electrical performance of the Q scenarios, and the electrical performance detection coefficients of the Q scenarios are output as the electrical feature detection results.

[0067] Step S500: Based on the Q test scenarios, perform comfort feature detection on the electric tricycle according to the expected comfort performance of the Q scenarios, and obtain the comfort feature detection results.

[0068] In this embodiment, a comfort performance factor for the electric tricycle is first set, including driving noise level, driving vibration amplitude, and driving vibration frequency. Then, in each test scenario, a comfort performance confidence test is performed based on the comfort performance factor to obtain test comfort performance data for Q scenarios. Next, the test comfort performance for each scenario is compared with the expected value to identify deviations and generate a comfort performance deviation matrix. The deviation matrix is ​​input into a pre-trained comfort performance detection model to calculate and obtain the comfort performance detection coefficient for each scenario. Finally, the comfort performance stability is obtained by calculating the reciprocal of the variance of the detection coefficients, and the stability, test comfort performance, and detection coefficients are output as the final comfort feature detection result.

[0069] Furthermore, in the method provided in the application embodiment, based on the Q test scenarios, comfort feature detection is performed on the electric tricycle according to the expected comfort performance of the Q scenarios to obtain comfort feature detection results, which further includes:

[0070] A comfort performance factor is set for the electric tricycle, wherein the comfort performance factor includes driving noise level, driving vibration amplitude, and driving vibration frequency; based on the comfort performance factor, a comfort performance confidence test is performed on the electric tricycle according to Q test scenarios to obtain Q scenario test comfort performance; based on the expected comfort performance of the Q scenarios, deviations are identified in the Q scenario test comfort performance to generate a Q scenario comfort performance deviation matrix; the Q scenario comfort performance deviation matrix is ​​input into a comfort performance detection model to obtain Q scenario comfort performance detection coefficients; the reciprocal of the variance of the Q scenario comfort performance detection coefficients is used as the comfort performance stability; the comfort performance stability, the Q scenario test comfort performance, and the Q scenario comfort detection coefficients are output as the comfort feature detection result.

[0071] In this embodiment of the application, the comfort performance factor of the electric tricycle is first set, wherein the comfort performance factor includes driving noise level, driving vibration amplitude and driving vibration frequency.

[0072] Next, based on the comfort performance factor, a confidence test of the electric tricycle's comfort performance is conducted under Q test scenarios. Specifically, firstly, the q-th scenario is randomly selected from the Q test scenarios, and the electric tricycle is tested multiple times under this scenario. Data such as driving noise level, driving vibration amplitude, and driving vibration frequency are recorded for each test. The maximum value from each test is taken to form the test dataset for each indicator. Then, the mean of the test dataset for each indicator is calculated to obtain the test data for the maximum driving noise level, maximum driving vibration amplitude, and maximum driving vibration frequency in the q-th scenario. The calculated maximum driving noise level, maximum driving vibration amplitude, and maximum driving vibration frequency in the q-th scenario are then output as the test comfort performance for the q-th scenario, and this test comfort performance is added to the total Q scenario test comfort performance. To obtain the global test results, all Q test scenarios are iterated through, and the above steps are repeated for the test results of each scenario, ultimately obtaining the test comfort performance for the Q scenarios.

[0073] The obtained comfort performance data from the Q scenarios are then compared with the expected comfort performance value for each scenario to identify the comfort performance deviation for each scenario. This deviation reflects the difference between the actual test results and the expected values, and a comfort performance deviation matrix for the Q scenarios is generated through comparison.

[0074] Next, the comfort performance deviation matrices of Q scenarios are input into the comfort performance detection model. The model analyzes the patterns in the deviation matrices to calculate the comfort performance detection coefficients for each scenario, resulting in Q scenario comfort performance detection coefficients. The comfort performance detection model is pre-trained. During training, a set of historical electrical performance deviation matrices and corresponding electrical performance detection coefficient sets are first obtained from a historical database. The multilayer perceptron is trained using the historical comfort performance deviation matrix set as input data and the comfort performance detection coefficient set as output data. After training, the comfort performance detection model can predict the corresponding comfort performance detection coefficients based on the comfort performance deviation matrix of each test scenario.

[0075] Next, the variance of the comfort performance detection coefficients for the Q scenarios is calculated, and the reciprocal of the variance of the comfort performance detection coefficients for the Q scenarios is taken as the comfort performance stability.

[0076] Finally, the comfort performance stability, the comfort performance tested in Q scenarios, and the comfort performance detection coefficients in Q scenarios are output as the comfort feature detection results.

[0077] Step S600: Based on the power characteristic detection results, the electrical characteristic detection results, and the comfort characteristic detection results, construct a visual report on the overall performance of the electric tricycle.

[0078] In this embodiment, when constructing a visual report on the overall performance of the electric tricycle based on the results of power characteristic detection, electrical characteristic detection, and comfort characteristic detection, the power performance, electrical performance, and comfort performance data obtained from various test scenarios are first directly compiled and displayed. By presenting these test results in chart form, the report can clearly show the actual performance of each aspect under different test scenarios. Furthermore, through deviation analysis and the display of detection coefficients, the report demonstrates the difference between each performance indicator and the expected value. Finally, all this information is aggregated to form a comprehensive report, completing the construction of the visual report on the overall performance of the electric tricycle.

[0079] In summary, the embodiments of this application have at least the following technical effects:

[0080] This application utilizes data mining based on multiple operating scenario factors of electric tricycles to construct Q test scenarios. These multiple operating scenario factors include road conditions, environmental conditions, and load conditions, where Q is a positive integer greater than 1. Based on these Q test scenarios, multi-dimensional performance expectations of the electric tricycle are predicted, resulting in Q expected power performance, Q expected electrical performance, and Q expected comfort performance. Based on these Q test scenarios, power feature detection is performed on the electric tricycle according to the expected power performance, yielding power feature detection results. Similarly, electrical feature detection is performed on the electric tricycle according to the expected electrical performance, yielding electrical feature detection results. Finally, based on the power feature detection results, electrical feature detection results, and comfort feature detection results, a visual report on the overall vehicle performance of the electric tricycle is constructed. This invention addresses the technical problem that existing technologies cannot comprehensively evaluate the multi-dimensional performance of electric tricycles. By constructing Q test scenarios based on multiple operating scenario factors, and combining data mining technology to predict the power, electrical, and comfort performance of electric tricycles in different scenarios, a feature detection model is used to detect power, electrical, and comfort features respectively to obtain various performance data. The detection results are summarized and visualized to form a vehicle performance report, achieving the technical effect of comprehensively evaluating the multi-dimensional performance of electric tricycles.

[0081] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0082] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0083] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for testing the overall performance of an electric tricycle, characterized in that, The method includes: Data mining is performed on the multiple operating scenario factors of electric tricycles to construct Q test scenarios. The multiple operating scenario factors include the operating road condition, the operating environment condition, and the operating load condition, and Q is a positive integer greater than 1. Based on the Q test scenarios, multidimensional performance expectation predictions are made for the electric tricycle to obtain the expected power performance, expected electrical performance, and expected comfort performance in the Q scenarios. Based on the Q test scenarios, the electric tricycle is subjected to power characteristic detection according to the expected power performance of the Q scenarios, and the power characteristic detection results are obtained. Based on the Q test scenarios, the electric tricycle is subjected to electrical feature detection according to the expected electrical performance of the Q scenarios, and the electrical feature detection results are obtained. Based on the Q test scenarios, the electric tricycle is subjected to comfort feature detection according to the expected comfort performance of the Q scenarios, and the comfort feature detection results are obtained. Based on the power characteristic detection results, electrical characteristic detection results, and comfort characteristic detection results, a visual report on the overall vehicle performance of the electric tricycle is constructed. Based on the Q test scenarios, the electric tricycle is subjected to power characteristic detection according to the expected power performance of the Q scenarios, and the power characteristic detection results are obtained, including: Based on the Q test scenarios, a power performance confidence test is conducted on the electric tricycle to obtain the power performance of the Q scenarios. Based on the expected dynamic performance of the Q scenarios, deviations in the tested dynamic performance of the Q scenarios are identified, and a dynamic performance deviation matrix for the Q scenarios is generated. Input the Q scenario power performance deviation matrices into the power performance detection model to obtain the Q scenario power performance detection coefficients; The inverse of the variance of the Q scenario dynamic performance detection coefficients is taken as the dynamic performance stability. The dynamic performance stability, the dynamic performance tested in the Q scenarios, and the dynamic performance detection coefficients in the Q scenarios are output as the dynamic characteristic detection results. The electric tricycle is subjected to electrical feature detection based on the expected electrical performance of the Q scenarios, and the electrical feature detection results are obtained, including: The electric tricycle is configured with electrical performance factors, which include driving range, motor efficiency, and motor temperature rise. Based on the electrical performance factor, electrical performance confidence tests are conducted on the electric tricycle according to the Q test scenarios to obtain the electrical performance of the Q scenarios. Based on the expected electrical performance of the Q scenarios, deviations in the tested electrical performance of the Q scenarios are identified, and a deviation matrix of electrical performance of the Q scenarios is generated. Input the Q scenario electrical performance deviation matrices into the electrical performance detection model to obtain the Q scenario electrical performance detection coefficients; The electrical performance stability is obtained by calculating the reciprocal of the variance of the electrical performance detection coefficients for the Q scenarios. The electrical performance stability, the electrical performance of the Q scenarios, and the electrical performance detection coefficients of the Q scenarios are output as the electrical feature detection results.

2. The method as described in claim 1, characterized in that, Based on the Q test scenarios, a power performance confidence test is conducted on the electric tricycle to obtain the power performance in the Q scenarios, including: Based on the Q test scenarios, extract the q-th test scenario, where q is a positive integer, 1≤q≤Q; The electric tricycle is subjected to multiple maximum speed tests based on the qth test scenario to obtain the maximum speed test dataset. Based on the qth test scenario, the electric tricycle is subjected to multiple maximum climbing angle tests to obtain the maximum climbing angle test dataset. The braking distance of the electric tricycle is tested according to the qth test scenario to obtain a braking distance test dataset. The mean values ​​of the maximum vehicle speed test dataset, the maximum climbing angle test dataset, and the braking distance test dataset are calculated respectively to obtain the confidence maximum vehicle speed, the confidence maximum climbing angle, and the confidence braking distance for the qth scenario. The confidence maximum vehicle speed, the confidence maximum climbing angle, and the confidence braking distance of the qth scenario are output as the test power performance of the qth scenario, and the test power performance of the qth scenario is added to the test power performance of the Q scenarios.

3. The method as described in claim 1, characterized in that, Based on the Q test scenarios, the electric tricycle is subjected to comfort feature detection according to the expected comfort performance of the Q scenarios, and the comfort feature detection results are obtained, including: The comfort performance factor of the electric tricycle is set, wherein the comfort performance factor includes driving noise level, driving vibration amplitude and driving vibration frequency; Based on the comfort performance factor, the electric tricycle is subjected to comfort performance confidence tests according to the Q test scenarios to obtain the comfort performance of the Q scenarios. Based on the expected comfort performance of the Q scenarios, deviations in the tested comfort performance of the Q scenarios are identified, and a comfort performance deviation matrix for the Q scenarios is generated. Input the Q scenario comfort performance deviation matrices into the comfort performance detection model to obtain the Q scenario comfort performance detection coefficients; The inverse of the variance of the Q scenario comfort performance detection coefficients is taken as the comfort performance stability. The comfort performance stability, the comfort performance tested in the Q scenarios, and the comfort performance detection coefficients in the Q scenarios are output as the comfort feature detection results.

4. The method as described in claim 1, characterized in that, Based on the Q test scenarios, multi-dimensional performance expectation predictions are performed on the electric tricycle to obtain the expected power performance, expected electrical performance, and expected comfort performance for the Q scenarios, including: Obtain the vehicle manufacturing process information of the electric tricycle; Based on the Q test scenarios and the vehicle manufacturing process information, the expected power performance of the electric tricycle is predicted, generating the expected power performance of the Q scenarios. The expected power performance of each scenario includes the expected maximum speed, expected maximum climbing angle and expected braking distance corresponding to each test scenario. Based on the Q test scenarios and the vehicle manufacturing process information, the electrical performance expectation of the electric tricycle is predicted to generate the Q scenario expected electrical performance, wherein the expected electrical performance of each scenario includes the expected driving range, expected motor efficiency and expected motor temperature rise corresponding to each test scenario; Based on the Q test scenarios and the vehicle manufacturing process information, the expected comfort performance of the electric tricycle is predicted, generating the expected comfort performance of the Q scenarios. The expected comfort performance of each scenario includes the expected driving noise level, expected driving vibration amplitude, and expected driving vibration frequency corresponding to each test scenario.

5. The method as described in claim 4, characterized in that, Based on the Q test scenarios and the vehicle manufacturing process information, the expected power performance of the electric tricycle is predicted, generating the expected power performance for the Q scenarios, including: Load test scenario records, production process records, and standard power performance records for electric tricycles; Using the electric tricycle test scenario record and the electric tricycle production process record as input information, and the scenario standard power performance record as output information, supervised training is performed on the predetermined learning model to obtain the power performance expected prediction loss coefficient. If the expected prediction loss coefficient of dynamic performance is less than the expected prediction loss threshold of dynamic performance, a expected prediction model of dynamic performance is generated. The q-th test scenario and the vehicle production process information are input into the power performance expectation prediction model to obtain the expected power performance of the q-th scenario, and the expected power performance of the q-th scenario is added to the expected power performance of the Q scenarios.

6. The method as described in claim 1, characterized in that, Data mining was conducted based on the diverse operating scenario factors of electric tricycles to construct Q test scenarios, including: Based on the aforementioned multi-operational scenario factors, test scenarios are set for the electric tricycle to obtain a set of test scenario schemes. Based on the set of test scenario schemes, pairwise similarity evaluation is performed to obtain a set of test scenario similarity coefficients; Based on the set of test scenario similarity coefficients, the set of test scenario schemes is filtered according to the test scenario similarity threshold to obtain the Q test scenarios that are less than the test scenario similarity threshold.

Images