Multi-Model Based Operational Feedback Adaptive Dual-Rate Decoupling Control And Its Application In Dual-Tank Cascade Control System

Operational feedback control of practical industral processes is always one of the hot spots and difficulties in the field of control.Although the existing operational feedback control methods have considerd the influence of the errors in the underlying-loop on the performance of the whole control system,they supposed the model in the operational layer is static and did not consider the couplings among different variables.Moreover,they did not consider the fact that the input and output variables may have different sampling rates.To overcome these limitations,relying on the National Natural Science Fund Project "Nonlinear multivariable adaptive optimal decoupling control and its application in steel ball mill coal pulverizing system(61573090)",this article studies the multi-model based operational feedback adaptive dual-rate decoupling control method,and applies it to a dual-tank cascade control system.The main contributions of this article are as follows:(1)By combining the pole assignment control system in the underlying loop with the dynamic model in the operational loop,the operational controller design model is established.When the model parameters are known,the multiple model operational feedback control method composed of a one-step ahead linear controller,a one-step ahead nonlinear controller with neural network compensation and a switching mechanism is proposed.When the model parameters are unknown,by combining the above control method with the recursive least square identification method with dead zone,the multi-model based operational feedback adaptive control method is proposed.Simulations are conducted by using the above two methods,the results show the effectiveness of the proposed methods.(2)By using the model transformation technique,the single-rate model is converted into a dual-rate model.When the model parameters are known,a one-step ahead weighting dual-rate decoupling control method based on multiple models is proposed.When the model parameters are unknown,by combining the above control method with the recursive least square identification method with dead zone,the multi-model based operational feedback dual-rate adaptive decoupling control method is proposed.Simulations are conducted by using the above two methods,the results show the effectiveness of the proposed methods.(3)The flow-level cascade control system of a dual-tank is a typical abstraction of the industrial process operational feedback control process,so it has broadly representative.Firstly,by using the control variable method,the objectives in the flow loop and the liquid level loop are established.Then the controller model of the cascate system is obtained.Finally,based on this model,the multi-model based operational feedback dual-rate adaptive decoupling controller is designed and applied to a dual-tank.The application results show the effectiveness and practicability.