Research On Intelligent Gear Decisionmaking Of Dual Clutch Transmissions Based On Driving Intention And Driving Environment Identification

Gear decision-making of dual clutch transmissions(DCT)is the key to affecting the vehicle’s power,economy and comfort.It is necessary to comprehensively consider driving intention and driving environment to make intelligent gear decision-making rules.However,the current gear decision-making mostly depends on manual calibration,which cannot be reasonably optimized based on real-time driving intention and changes in the driving environment,that is,the degree of intelligence needs to be improved,which has a greater impact on the improvement of the overall performance of the vehicle.This article takes a manufacturer’s seven-speed DCT-fuelled vehicle as the research object,and conducts a study of an intelligent gear decision-making system that takes into account driving intention and driving environment.The main research contents are as follows:(1)Research on driving intention recognition based on long short term memory networks.The driving intention recognition problem is constructed as a classification problem in time series.After preprocessing steps such as the division of driving conditions,filtering of driving data,and K-means clustering,the driving intention is divided into five parts: acceleration,rapid acceleration,crurise,deceleration,and rapid deceleration.According to the continuity of driving behavior and the long-term dependence of long short-term memory networks,vehicle speed,accelerator pedal opening,and brake pedal force are selected as characteristic parameters to establish a driving intention recognition model based on long short term memory networks.The results of driving intention recognition are compared with those of traditional backpropagation neural network to verify the correctness of the results of driving intention recognition based on long short term memory networks.(2)Research on driving environment recognition based on vehicle driving data.The driving environment is divided into general form and special form,and real vehicle driving data is collected and pre-processing steps such as resampling and wavelet packet denoising of the selected driving condition characteristic parameters are performed.According to the vehicle longitudinal dynamics,principle of acceleration sensor and Kalman filter algorithm,the previous speed and measured acceleration value are used to identify the road slope,and the identified slope value is compare with the actual slope value using root mean square error to verify its effectiveness.Steering wheel angle and its angular velocity and vehicle lateral acceleration is used as characteristic parameters to establish a turning condition recognition model based on the Hidden Markov Model,and the accuracy of the model is verified using untrained driving data.(3)Research on intelligent gear decision-making system considering driving intention and driving environment.The vehicle dynamics model is established,and a traditional two-parameter shift rule based on vehicle speed and throttle opening as a comparison benchmark for the gear decision-making system is formulated.According to the dynamic characteristics of the vehicle,an optimal gear sequence based on dynamic programming algorithm is constructed to solve the problem,and the effects of different dynamic shift factors and economic shift factors on the vehicle’s dynamic index and economic index are analyzed.Then the optimal dynamic shifting rules and comprehensive shifting rules based on dynamic programming are extracted.Based on the influence of driving intention and driving environment on vehicle gear decision-making,different correction factors are selected to modify the shift speed,and an intelligent gear decision-making system considering driving intention and driving environment is formulated.Simulation results show that the gear decision-making system can obtain the comprehensive optimality of vehicle dynamics and economy,which is of great significance to promote the intelligent research level of the automatic transmission industry.