Optimal Trajectory Planning of Electromagnetic Valve-Train Based on Multiple Algorithms for Improving Engine Performance

摘要

Electromagnetic valve-train (EMVT) is a camless variable valve technology that can adjust the valve movement state independently, continuously and in real time. It can adjust the engine charge efficiency according to engine working conditions. In this paper, in order to achieve the optimal inflation efficiency, a new-style moving-coil electromagnetic linear actuator (ELA) was designed and genetic algorithm (GA) was used to plan the valve motion trajectory in common working conditions. Furthermore, an improved BP neural network algorithm was used to optimize the valve operation mode in all working conditions. The new valve operation scheme was applied to the modified intake system of engine, and its simulation and test analysis were carried out. Compared with the prototype, the results showed that the charging efficiency was greatly improved in the range of full rotating speed, especially in the range of middle and low rotating speed up to 95.6%