Design And Simulation Research Of Electronic Throttle Control On LPG Engine

Electronic throttle control system is an important part of vehicle control technology,which is used in an increasing number of vehicles. Compared with the traditional throttle,electronic throttle can estimate torque on the actual operating conditions, which makesthe throttle opening be in the best position in order to improve the power, emissions andfuel economy of vehicles. As the research of domestic electronic throttle control systemwas started very late, the core technologies was monopolized by foreign, so the researchabout electronic throttle control system has important theoretical and practicalsignificance. This paper which based on the project " Research on LPG engine controlsystem " mainly has finished the research of simulation for electronic throttle controlsystem.Based on analyzing the electronic throttle control system structure and workingprinciple and selecting pulse width modulation to drive motor, nonlinear factors whichaffected accuracy of electronic throttle control system was Analyzed, including thenonlinear of the reduction gear backlash, the nonlinear of return spring and friction andthe nonlinear from changes of inlet gas flow. While, mathematical model of thenon-linear was established including mathematical model of DC motor, mathematicalmodel of the throttle valve, the mathematical model of the sensor. At the same time,control strategy of the electronic throttle was analyzed. Against the nonlinear,mathematical model of four ways which included the traditional incremental PID control,fuzzy adaptive PID control, synovial variable structure control, the global fuzzy synovialswitching gain control was established. And then, simulation model is built by usingmatlab/simulink/stateflow tools, the following characteristics under step was analyzedin four control strategy. According to control effectiveness, at the same time, taking intofactors of account cost, ultimately variable structure control was suggested as systemcontrol algorithms.