| description abstract | When evaluating the shift quality of a manual transmission automotive, the subjective evaluation method is fuzzy, and the objective evaluation method with dynamic parameters as indexes is complex. To solve the problem of evaluating the quality of automobile gear-shifting simply and accurately, first the data of vehicle gear-free clearance was collected by setting up a gear data acquisition platform. Then, the collected data were displayed and stored in real time by the LabVIEW data-processing platform. Next, the DBSCAN algorithm was used to delete outliers, and the K-means clustering algorithm was used to classify the gear data. Then, taking the automobile gear data as the object, the evaluation significance groups were extracted from the three dimensions of gear selection travel, shift travel, and gear-free clearance; and the function factors were given to each evaluation significance group to establish the evaluation model of the manual transmission vehicle gear selection quality. The weight of each parameter index in the evaluation method of automobile gear shifting quality as obtained by using the entropy weight method from objective data, and a comprehensive evaluation method for shift selection quality of a manual transmission vehicle is established. The results show that for three different types of vehicles, the shift performance can be tested by using this evaluation method. The test results are consistent with the subjective testers, which verifies the effectiveness of the objective evaluation index selected for the quality evaluation of gear-shifting. The evaluation method can be used to evaluate the quality of automobile gear selection and shift simply and conveniently, which improves the simplicity of test, the reliability of analysis, and provides a reference for the evaluation of automobile gear selection quality. | |
