The Open Automation and Control Systems Journal

2015, 7 : 1356-1363
Published online 2015 September 17. DOI: 10.2174/1874444301507011356
Publisher ID: TOAUTOCJ-7-1356

Design and Optimization of Feedforward-PID Integrated Controller for Pre-Injection Numerical Control Air-Powered Engine

Zhao Caihong , Chen Shian and Zhang Shijin
School of Mechanical Engineering, Nanhang Jincheng College, Nanjing, Jiangsu, 211156, P.R. China.

ABSTRACT

A feedforward-PID integrated controller is developed and optimized for the pre-injection numerical control (NC) air-powered engine (PINCAPE) to track the expected speed quickly and accurately. The proposed integrated controller comprises two parts, namely a feedforward controller and a PID one. According to the real-time workload and the expected speed, the feedforward controller uses a numerical fitting function to determine the main part of the supply air pressure to obtain good stable accuracy. The function is developed with the output of the supply air pressure, and the inputs of the engine speed and the output power, which are based on the different steady-state operations of the engine. To improve the engine response performance, the PID controller gives the fine-tuning of the supply air pressure with the input of the difference between the expected speed and the real-time speed. The parameters of the PID controller are optimized according to the genetic algorithm, with an objective function that minimizes the mean square of the speed error. The results show that the optimized feedforward-PID integrated controller shortens the response time of the PINCAPE by more than 47% and 64.5% under varied set point speed and engine load variations respectively, compared with the single feedforward controller. Furthermore, the response steady state error of the feedforward-PID integrated controller is less than 1%, which perfectly satisfies the control specification.

Keywords:

Air-powered engine, feedforward-PID integrated controller, genetic algorithm, NC, pre-injection.