A novel proportional-integral-derivative-based fuzzy neural network (PID-based FNN) controller is proposed in this study to control the speed of a vane-type air motor (VAM) servo system for tracking periodic speed command. First, the structure and operating principles of the VAM servo system are introduced. Then, the dynamics of the VAM servo system is analyzed to derive the second-order state equation of the VAM. Moreover, due to the dynamic characteristics and system parameters of the VAM servo system are highly nonlinear and time-varying, a PID-based FNN controller, which integrates conventional proportional-integral-derivative neural network (PIDNN) control with fuzzy rules, is proposed to achieve precise speed control of VAM servo system under the occurrences of the inherent nonlinearities and external disturbances. The network structure and its on-line learning algorithm using delta adaptation law are described in detail. Meanwhile, the convergence analysis of the speed tracking error is given using the discrete-type Lyapunov function. To enhance the control performance of the proposed intelligent control approach, a 32-bit floating-point digital signal processor (DSP), TMS320F28335, is adopted for the implementation of the proposed control system. Finally, experimental results are illustrated to show the validity and advantages of the proposed PID-based FNN controller for VAM servo system.
ASJC Scopus subject areas
- Theoretical Computer Science
- Computational Theory and Mathematics
- Artificial Intelligence