Neural Network Control and Optimization

This book presents the development of intelligent control and optimization technologies for Heating, Ventilating, and Air-Conditioning (HVAC) systems. The HVAC industry expects the automation and control systems to perform well through the years without the requirement to re-tune the system. A novel neural network assisted control scheme is proposed. By taking advantage of neural network technology, the deficiency of traditional fixed controller (PID type) can be compensated to certain degree and better performance is achieved in term of set point tracking performance and disturbance rejection ability. Another important contribution of this book is that two robust neural network training algorithms are proposed. The neural network is trained in an online fashion and convergence of the network weights is guaranteed. For a neural network based control system, the stability issue is always the main concern in term of the closed control loop.

Qing Song is with the School of Electrical and Electronic Engineering, Nanyang Technological University an associate Professor. He held short visiting research positions at Johns Hopkins University and Tsinghua University, respectively.