Research on Shifting Strategy of Hybrid Electric Vehicle Based on Dynamic Programming Algorithm

In this paper, for a hybrid power transmission system composed of a planetary gear mechanism and an improved DCT transmission, the vehicle dynamics model, energy management strategy and torque distribution strategy are established through MATLAB, and two-parameter optimal economical shifting law is obtained by analytical method. By constructing a global optimization problem and solving it through a dynamic programming algorithm, the shifting sequence under the WLTC driving cycle is also obtained. The optimal economic shifting law and the shifting law obtained by the dynamic programming algorithm are verified in MATLAB by running the WLTC driving cycle. The shifting law obtained by the dynamic programming algorithm can make the engine work in a more efficient working range, so that the hybrid vehicle achieves a better fuel economy.

I. Introduction

Hybrid vehicles have the advantages of low energy consumption and low pollution, and are often equipped with automatic transmissions. The automatic transmission shift rule, also known as the shifting strategy, refers to the relationship between the shift points of two adjacent gears and shift control parameters. According to the number of shift control parameters, the classic shifting law can be divided into: single parameter, double parameter, three parameter and so on. It is difficult to obtain better economy and power with single-parameter shifting, and it has been used less now. The two-parameter shifting law is established under the condition of steady-state driving, and its parameters are usually selected as vehicle speed and throttle. Li Hongcai [1] considered the influence of the battery state of charge, divided the vehicle working mode, and proposed an improved dual-parameter shifting strategy, which improved both power performance and fuel economy. Chengsheng Miao [2] added the terrain coefficient as the third parameter and used dynamic programming and moving least squares to optimize the gear sequence, which improved fuel economy by 3.5% compared with the traditional shifting strategy. The classic shifting law only considers the steady-state operation of the vehicle, and the hybrid electric vehicle transmission system has more control parameters and more controlled components than traditional internal combustion engine vehicle, as a global optimal method, dynamic programming has been widely used in gear shifting. Jacobson [3] applied the trajectory optimization method of dynamic programming to optimize the shifting strategy of automatic transmission AMT with the minimum acceleration time as the optimization criterion, and achieved good power performance. Erik [4] extracted the information of the road with on-board road gradient database and GPS unit, and used the dynamic programming algorithm in the model predictive control, reduced the fuel consumption of the vehicle. VD Ngo [5] proposed a method based on stochastic dynamic programming to find the optimal shifting strategy in the average sense over several driving cycles, thus taking the drivability of the vehicle into account.