Prediction Fuzzy Control Based On Neural Network Research And Applications

With the improvement of science and technology, the objects and processes that people want to control become more and more complex, but the aims for control are higher and higher. Intelligent control provides a effective way for solving the problem. Fuzzy logic and neural network are the important means of intelligent control. Fuzzy logic control does not need to set up the exact math model, and it can effectively solve the problem of time-delay systems. This paper introduces some theories about fuzzy logic and neural networks control and their advantages. The thesis brings up a fuzzy logic control system based on neural network prediction combining both two theories. Researches and simulations are also applied on the water level control of boiler.The last part of this paper introduces a way to design a fuzzy logic controller based on the MATLAB's fuzzy logic toolbox, and transform artificial neural networks designed into form of SIMULINK correspondingly by using functions of MATLAB. The results of simulation experiments of this control system illustrate the availability and feasibility of this means in theory.With the combination of automatic control theory and practical thermal process automation object in the power station, this paper provides an idea solving the problems of general large-time-delay systems in the industrial process. And this means has great reference value in engineering applications.