Control of a pneumatic actuator system using enhanced nonlinear proportional integral derivative controller algorithm

Abstract

Pneumatic actuators offer several advantages such as low cost, simple to maintain, high power to weight ratio, fast motion, free from overheating and reliable. Due to these advantages, this actuator continues to generate significant research interests and it has been promoted as an alternative to hydraulics and electric servo motors in many automated tasks. However, it exhibits high nonlinearities due to high friction forces, compressibility of air and dead band of the spool movement in the valve. These nonlinearities make an accurate position difficult to achieve and it requires an appropriate controller for better performance. This thesis presents a new approach to control the pneumatic positioning system. The mathematical modeling is a crucial part to be established before the control algorithm can be designed. Initially, the modeling of the system is presented based on physical derivation. The parameters of the system are identified, and comparison between system identification by means of the MATLAB software is performed. The result obtained from the modeling process is validated with experimental data. Subsequently, two new controller techniques are proposed based on the enhancement of the Nonlinear Proportional Integral Derivative (N-PID) controller. The first technique is called Multi-rate Nonlinear PID (MN-PID), which is performed by utilizing the characteristics of rate variation in nonlinear gain. Fuzzy logic is used to perform this task. Meanwhile, for the second technique, a Self-regulation Nonlinear (SN) function, is introduced to reprocess the error signal to continuously generate the values for rate variation. The proposed controllers are implemented to the system, and their performances are analyzed for both cases, with and without load. Simulation and experimental tests are conducted with different input, namely step, multistep, sinusoidal and S-curve waveforms to evaluate the performance of the proposed techniques. The existing techniques that include PID, N-PID, and Sliding Mode Control (SMC) are also tested to the system as a comparison. The results indicate that the system with MN-PID and SN-PID exhibits improvement of dynamic performance criterion exceeding 34% and 59%, respectively. In addition, both techniques succesfully provide fast response without overshoot where the rise time reduces more than 87%. It proves that the novel initiative is capable of examining and identifying the characteristics of rate variation based on a new controller that was derived from N-PID controller. Moreover, the system performance is successfully accomplished for each position and direction as well as under various loads.