Design of an Intelligent Control System for a Stoker-Fired Boiler Based on Neuro-Fuzzy Modeling

摘要

A key issue of an industrial stoker-fired boiler is to design an efficient and robust controller for its combustion system, so that the boiler can provide a continuous supply of steam at the desired pressure conditions. However, it is impossible to use any model-based approaches to design such a controller due to highly nonlinear and uncertain dynamics of a boiler system. The most widely used design approach is to on-line tune control parameters during boiler operation. But, it is a hard and time consuming work because of strong load changes, large disturbances, large time lags, and so forth. In order to solve these problems, we adapt soft computing approaches to modeling of dynamic behavior based on real experimental data and off-line designing a hybrid FUZZY P+ID controller based on neuro-fuzzy modeling. This intelligent control system is applied to some industrial stoker-fired boilers and the experimental results demonstrate its effectiveness.