Hybrid CMAC-PID Controller in Heating Ventilating and Air-Conditioning System

The controlled objects are modeled in heating, ventilating and air-conditioning (HVAC) system, mainly including the heat exchanger and the air-conditioning space. The HVAC system has large inertia, pure lag and nonlinear characteristic. The uncertain disturbance factors affect the control performance. To obtain better performance in HVAC system, this study proposes a hybrid Cerebellar model articulation controller (CMAC)-PID control system, which combines the CMAC neural network and general PID control. This method realized the feedback control by using traditional PID controller to enhance the stability and reject the disturbance and realized the feed forward control by using CMAC neural network to increase the response speed and control precision in HVAC system. The simulation results show the hybrid CMA-PID control system possesses the advantages of high precision, realtime standard, strong ability of anti-interfere and robustness.

[1]  Aníbal R. Figueiras-Vidal,et al.  Generalizing CMAC architecture and training , 1998, IEEE Trans. Neural Networks.

[2]  Shih-Lin Hung,et al.  High-order MS CMAC neural network , 2001, IEEE Trans. Neural Networks.

[3]  Li Hui The implementation of CMAC Neural Network in measuring thermal-comfort-degree , 2003 .

[4]  Wang Jin Simulation Research on Application of Single Neuron Adaptive Control in High Precision Air-conditioning System , 2004 .

[5]  T.J. Stonham,et al.  Associative memory neural networks for time series prediction , 1996, Proceedings of Third International Conference on Electronics, Circuits, and Systems.

[6]  Li Jie-jia Application of Adopting Fuzzy Neural Network Adaptive Control in VAV Temperature Control System Based on Wavelet Neural Network , 2004 .

[7]  C Ongfang The Fuzzy-Self-Adaptive Control of Room Temperature in VAV systems , 2005 .

[8]  D.A. Handelman,et al.  Theory and development of higher-order CMAC neural networks , 1992, IEEE Control Systems.

[9]  Zuo Xin A Comparative Study of PID Turning Methods , 2008 .

[10]  James S. Albus,et al.  New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC)1 , 1975 .

[11]  Frank L. Lewis,et al.  Optimal design of CMAC neural-network controller for robot manipulators , 2000, IEEE Trans. Syst. Man Cybern. Part C.

[12]  Wang Jiangjiang THE SIMULATION AND RESEARCH OF THE FUZZY NEURAL NETWORK CONTROL IN THE TEMPERATURE OF AIR - CONDITIONING SYSTEM , 2004 .