Research On Multivariable Control Methods And Applications Based On Information Theory

Multivariable control method has always been a hot issue in the field of control. With the modern industrial systems becoming more and more large and complex, most of actual systems are multivariable systems and random phenomenon is inevitable. Therefore, this article has theoretical and practical value. The birth of information theory provides a new angle for research of control theory, and enriches the control methods. In addition, information measures can characterize the stochastic uncertainty effectively and further facilitate us to deal with practical problems.In this paper, we focus on the multivariable control system based on the relevant knowledge of information theory, and put forward new multivariable control methods with considering the non-Gaussian noises in industrial systems. The main work and contribution are as follows:First of all, based on the information theory, this paper studies the core issue of decentralized control system, how to measure the multivariable coupling degree accurately. In the frequency domain, we use the mutual information rate to describe the internal coupling of the system and propose the relevant loop pairing criteria.Secondly, this paper focuses on the multivariable networked control system. With the development of computer and communication technology, the birth of networked control system broadens the application fields of the traditional control system. For non-Gaussian random time delays in networked control systems, we propose the improved minimum entropy control algorithm to design the equivalent multivariable nonlinear non-Gaussian system and use the linearization method for local stability analysis.Finally, this paper uses the survival information potential to describe the uncertainty of the closed-loop system and constitute a new performance index. Based on the above, we present a new multivariable control strategy and then analyze the stability of the closed-loop system by using the stochastic linearization model. And we do the relevant simulation experiments to verify the effectiveness of the proposed strategy in the final. This part makes an improvement on the minimum entropy control algorithm and put forward relevant control strategies, which lays a solid foundation for the further theoretical research and engineering practice.