船用锅炉水位控制系统的补偿模糊神经网络仿真研究

The Compensatory Fuzzy Neural Networks Control and Simulation of Water Level in Marine Boiler

针对船用锅炉这种具有非线性、参数不稳定、难以建立精确的数学模型的特性,结合模糊控制理论和人工神经元网络理论,提出用补偿模糊神经网络构造自适应控制器实现对锅炉的水位控制。通过补偿模糊推理和快速学习算法的引入,使其在性能上优于一般的模糊神经网络,并且利用MATLAB仿真工具对这种自适应控制器的性能作初步研究。

In this paper, the control system of marine boiler is studied. Aiming at the nonlinear object of industrial boiler with instability parameter and difficulty of modelling, combining fuzzy control theory and neural networks theory, a adaptive controller built with compensatory fuzzy neural networks to control boiler is proposed. With the compensatory fuzzy inference and quick arithmetic, the compensatory fuzzy neural networks is superior to that of common fuzzy neutral networks. Simulation by Matlab software is also performed to study the controller＇s performance.