Engine Speed Based On Second Order Sliding Mode Control Study

Engine, as the power source of the vehicle, its performance determines the overall performance of the vehicle. With the continuous improvement of performance of the modern vehicle, a higher demand is proposed in the engine speed control of the whole conditions. With the increasing requirement of control, the more advanced control theory and control methods are needed. Thus, this paper presents a second-order sliding mode control algorithm to design the engine speed controller, and also verifies the control effect from the software and hardware simulition.Based on Crossley and Cook's engine model for four-cylinder four-stroke gasoline engine in low-frequency range, we summarized the approximate nonlinear model between the engine torque output, engine speed and throttle position. Based on this nonlinear model and the research data of the engine test bench, the gasoline engine torque model by multiple stepwise regression method is build. We simulated the engine torque model's step response in the Matlab/Simulink environment, and compared it with the actual engine exprimental data. The results show that the model can better reflect the actual characteristics of the engine, which laid the foundation for the design and simulation of speed controller.This paper also describes the development and basic theory of high-order sliding mode control, focusing on the principles of second-order sliding mode controller and several basic common algorithms, and illustrates the design theory and related theorems of first-order robust exact differentiator. Based on the nonlinear engine model, we designed a traditional one-order sliding controller and a second-order super twisting controller for speed control system. A first-order robust exact differentiator for the requirements of the differential values in the sliding mode algorithm was designed, which can calculate the system error differential values in real-time. In Matlab/Simulink simulation environment, we established the controller's simulation model and validated its control effect based on the performance evaluation. From the smooth continuous signal tracking capability, discontinuous square wave signal tracking capability, load disturbance robustness, parameter change robustness and other aspects, we simulated the control effect of the two sliding mode controllers, and compared with traditional PID controller. The analysis results showed that the second-order sliding mode controller has good performance compared to other algorithms, and the deficiency of the traditional first-order sliding mode algorithm in the engine speed control system.Introduction the Matlab/xPC Target hardware in the loop simulation platform briefly, we designed and built the engine speed control hardware in the loop system based on this platform. According to the xPC Target environment device driver development process, we designed speed acquisition driver module, throttle control drive module and so on. We also established the second-order super twisting controller's hardware in the loop model in xPC Target environment, which is installed in the industrial computer to control the actual engine, whose control effect is tested, and which is compared with the engine test bench's PID controller. Simulation results showe that compared with the PID algorithm the second-order sliding mode super twisting algorithm has following advantages, such as good tracking performance, small overshoot, fast response, strong robustness and so on.