Study And Application On The Adaptive Control Of Nonlinear Dynamic Plants Based On U-Model

The research on adaptive control is an important field based on the nonlinear dynamic plants. However, the appropriate or general mathematic model for studying the nonlinear dynamic system of adaptive control method is important. But in the studies, the main difficulty is no general model structural frame, which is able to accurately express nonlinear dynamic system structure, or nonlinear dynamic system model structure and parameter is too complex. All above are giving rise to no have standard to track, because each controller have itself designing methods.The topic research objectives are: for a class of unknown nonlinear stochastic dynamic system, based on control mathematical original model U-Model (composed by controlled plants past input data formed model), design an adaptive control criteria standard. Consider U-Model coefficients and there is completely unknown uncertainty factors influencing system. Design two different adaptive control algorithms, respectively adaptive poles placement control method and RBF neural network method. In each control algorithm for U-Model represent different nonlinear simulation model and the model of double welding pipe based on the U-Model is simulated.Based on the control standard U-Model, respectively for carrying on the theory is simulation and the double welded pipe of welding process on temperature control is experiment. Because in theory simulation, consider all sorts of unknown factors of the system of influence. Welding pipe temperature control, U-Model can be convenient, concise representation of the unknown welding pipe model. Through theoretical simulation and the double welded pipe welding process of temperature control experiment, we can get a conclusion that: 1) U-Model can be widely express category (smooth polynomial) unknown nonlinear dynamic system. 2) Based on U-Model design standard controls in theory and practical application is completely feasible. 3) U-Model can easily establish analysis, research and design of nonlinear dynamic system model.