Title: Shifting control optimisation of automatic transmission with congested conditions identification based on the support vector machine

Authors: Shang Peng; Guangqiang Wu; Kaixuan Chen; Minkai Jiang; Lijuan Ju

Addresses: Institute of Automotive Simulation Science, School of Automobile Studies, Tongji University, Shanghai, 201804, China ' Institute of Automotive Simulation Science, School of Automobile Studies, Tongji University, Shanghai, 201804, China ' Institute of Automotive Simulation Science, School of Automobile Studies, Tongji University, Shanghai, 201804, China ' Institute of Automotive Simulation Science, School of Automobile Studies, Tongji University, Shanghai, 201804, China ' Institute of Automotive Simulation Science, School of Automobile Studies, Tongji University, Shanghai, 201804, China

Abstract: Vehicles in congested conditions are often accompanied by frequent acceleration and deceleration, but also cause the phenomenon of frequent transmission shifting. In this paper, state parameters such as speed, accelerator pedal opening, and brake pedal opening are selected and preprocessed to obtain the relevant features of the vehicle state. Principal component analysis (PCA) is used to reduce the dimension of these features to simplify the computation. Then, a support vector machine (SVM) optimised by the genetic algorithm is used to identify congested conditions. Finally, based on the identification results and vehicle dynamics, the shift curve was corrected to reduce the frequent shifting phenomenon in congested conditions. Through the classification test of the training set and test set, the method used in this paper can effectively identify congested conditions, and the simulation test results show that use of the corrected shift curve can effectively reduce frequent shifting in congested conditions.

Keywords: automatic transmission; congested conditions; PCA; principal component analysis; SVM; support vector machine; shifting control optimisation.

DOI: 10.1504/IJVP.2023.130053

International Journal of Vehicle Performance, 2023 Vol.9 No.2, pp.204 - 224

Received: 24 Jun 2022
Accepted: 02 Sep 2022
Published online: 04 Apr 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article