Research On Networked Parallel Cascade Control Systems

The real-time communication network is inserted into the main and secondary closed-loop control loops of the parallel cascade control systems,so that the data between the sensor,controller and actuator nodes in the system is transmitted and exchanged through the real-time communication network,which constitutes networked parallel cascade control systems(NPCCS).NPCCS can realize remote real-time online control,realize data resource sharing between nodes,facilitate system fault diagnosis and maintenance,and have the advantages of parallel cascade control systems(PCCS)to overcome interference and improve the quality of control performance of main control loop.However,since NPCCS is used for data transmission and exchange through a real-time communication network,the introduction of the network inevitably brings a series of problems to be solved,such as network delay,data packet loss,and system stability.Network delay is an important factor affecting the stability of the system,which will deteriorate the quality of the control performance of the system and even lead to the system losing stability.Traditional control theory is difficult to directly use in the analysis and design of NPCCS.The introduction of network brings new opportunities and challenges for the analysis and design of NPCCS.The existence of network delay in NPCCS will reduce the stability of the system.The paper starts with the system structure of NPCCS,and proposes two kinds of new internal model control(IMC)combined with new Smith predictive control.The network delay compensation and control methods effectively reduce the influence of time-varying,random and uncertain network delay on system stability,improve the anti-interference ability of the system,and satisfy the requirements of system output for system input tracking performance.The robustness of the system to the parameter changes of the controlled object is enhanced,and the control performance quality and network service quality of the NPCCS are improved.The innovative research results of the paper are as follows:(1)Five basic structural forms of NPCCS are proposed for different location conditions of the network in the NPCCS structure,as well as sensor,controller and actuator independent nodes,or their shared nodes,and through the system configuration diagram,block diagram,the node device connection matrix,the network transmission matrix and the closed-loop transfer function of the system systematically describe and analyze the different structural forms and characteristics of these five NPCCS.Taking the most complex Type V NPCCS structure as an example,this paper analyzes and studies in detail the key technical problems,research ideas and research methods that need to be solved in order to realize the network delay compensation and control.(2)Aiming at the technical difficulties in the network delay compensation and control of Type V NPCCS,a delay compensation and control method based on the new IMC and Smith predictive control method(1)is proposed.The secondary control loop adopts the new Smith predictive control method(1),which reduces the influence of the network delay in the secondary control on the stability of the system,and combines the controller in the secondary control loop to further reduce the the influence of interference enter into the secondary control loop on the main control loop;the main control loop adopts the new IMC,in the case of realizing the dynamic compensation and control of the system for time-varying,random and uncertain network delay,the anti-interference ability of the system is further enhanced.The robustness of the system to the deviation between the predicted model of the controlled object and its real model is improved,and the system has good dynamic tracking performance and stability.(3)Aiming at the technical difficulties in the network delay compensation and control of Type V NPCCS,a delay compensation and control method based on the new IMC and Smith predictive control method(2)is proposed.The secondary control loop adopts the new Smith predictive control method(2),which reduces the influence of the network delay in the secondary control on the stability of the system,and combines the controller in the secondary control loop to further reduce the the influence of interference enter into the secondary control loop on the main control loop;the main control loop adopts the new IMC to achieve dynamic compensation and control of network delays greater than a few,or even dozens of sampling periods.At the same time,because the secondary control loop in the method adopts the new Smith predictive control method(2),eliminating the predictor compensation condition of the new Smith predictive control method(1)that the secondary controlled object prediction model is equal to its real model,which further enhances the robustness of the system and expands the scope of application of the method.Based on the research ideas and system design methods of the above two network delay compensation and control methods,the paper aims at network delays with time-varying,random and uncertainties,more than several,or even dozens of sampling periods,and the system of NPCCS also carries a certain amount of network data loss,and carries out systematic simulation research.The simulation results verify the effectiveness of the two methods presented in the paper,which show that the system has good tracking and stability,strong robustness and the ability to resist internal and external loop interference.The two methods proposed in the paper have the following characteristics:(1)Replace the network delay prediction compensation model with the real network data transmission process,and realize the network delay prediction compensation model from the system structure unconditionally equal to its real network data transmission process.(2)Elimination of estimation,measurement or identification of time-varying,random and uncertain network delay,avoiding delay estimation errors,and delay identification of node storage resources;reducing synchronization requirements for node clock signals;the compensation error caused by "empty sampling" or "multi-sampling" caused by delay is avoided.(3)The method adopted is based on the system structure and realizes the online,real-time and dynamic compensation methods for its network delay.The system structure is simple,and it is easy to be implemented in engineering.It can be combined with other intelligent algorithms to realize more complex and robust control.(4)Because it is dynamic compensation for network delay from the system structure,it can be used for wired networks,wireless networks,and even hybrid networks regardless of the choice of specific network communication protocols.(5)Since the new Smith predictive control is adopted for the secondary control loop,the implementation of the method is independent of the specific selection of the secondary controller in the secondary control loop.The secondary controller can adopt both the conventional control strategy and the intelligent control strategy.