Modeling Method And Control Strategies Of Internet-Based Remote Control Systems

Along with the rapid development of the computer, communication and network technology, a new control system based on the Internet was produced. Networked control system is built on the Internet which exists in every corner of the world. It makes the control object can be controlled by the Internet in any position of the global. Networked control system is not limited to control distance and number of nodes, so it changes the control mode and the information transfer mode of the traditional control systems; Abundant information has accumulated on the Internet; multimedia storage, transfer and application technology become mature gradually; all kinds of database and related technology is developed rapidly. These provide networked control system an excellent environment and more control means. Networked control system does not need to build special-purpose communications network, and, it has the advantages of low investment costs. With mobile communications network access to the Internet, the networked control system can realize a variety of special control functions. In short, the networked control system with its unique advantage will get broad applications in national defense, industrial, medical and people’s daily life, and will promote the globalization of economic development, and significantly affect people’s life style. The networked control system becomes a new development direction of today’s control system and a hot research topic.Although the networked control system has many advantages, but there is also an urgent need of the theory and technology research on new problems. For example: from the controller to the actuators, and from the sensor to the controller, these components are connected through the Internet. As the network nodes reuse and bandwidth limited, network information transmission will inevitably produce delay. The length of time delay will be related with network protocol, transmission quantity, the transmission distance, transmission time and many other factors, has the remarkable uncertainty, time varying and randomness which not only has brought great difficulty to system modeling, analysis and control, but also affected the stability of system and control performance. So how to estimate and test network transmission delay accurately, put forward the reasonable modeling method and high performance control strategy of the networked control system, is in need of solution of some theoretical and technical problems. At a moment, some nodes and many information exchanges can cause the information transmission obstruction, the out-of-order phenomenon, and the loss of data packets, which will result in performance deterioration of control system, or even the whole control system collapse. How to prevent and compensate network information transmission jam and packet loss and improve the reliability of information transmission, become another urgent problem to solve for networked control system.In this paper, to analyze the networked control system with uncertain information transmission delay which constitute the lag loop, at first the factor what contain infinite dimension was studied using an approximate method for system modeling, with ignoring high-power terms which has a smaller influence on the system’s performance. Because the model predigestion of high power terms of the lag loop be bound to make certain error to the system, so as compensation a strong robust control strategy "ITAE criterion" was taken afterwards. Then, to restrain disturbance and noises the system quadratic optimal control was applied. Finally heating stove with long delay and brushless DC motor with fast dynamic response were selected for the controlled objects, and the simulation and experimental research were taken. After the prediction of network transmission time delay by using the stamped BP neural network, based on the dynamic planning theory and Lyapunov stability theory, the optimal state feedback matrices were derived for the heating stove networked control system, and Kalman filter is used for state estimation and disturbance rejection. By introducing adaptive tracking device to compensative input signal noises, simulation results were given.In order to realize accurate temperature control of the heating stove, the dead-beat response control strategy for networked based heating stove control system was studied. At first, the mathematical models of the networked based heating stove control system both in frequency domain and discrete state equation were established. Then the time-stamped BP neural network method to predict the delay time of network was used. This is followed the digital controller with dead-beat response design method has been gotten and the control algorithms of the digital controller were derived. Simulation results are illustrated that the accurete temperature control without steady-state error can be achieved. Besides, the digital controller without ripple of dead-beat response design method was discussed too.Multimodel control theory was introduced to the networked control system. The time-varying delay parameter of network information transmission was mapped for model set of system. The controller design method, switching method and stability were analyzed; after deriving the interactive multimodel control algorithms, the simulation was demonstrated.Finally, the work of this paper is summarized, the innovation points are presented and the future research work is prospected.