多层递阶方法在故障预报中的应用

The application of multi-level recursive method in fault prediction

针对连续搅拌釜式反应器(CSTR)系统动态时变的特点,仅用系统观测数据,分别应用基本的多层递阶方法和综合的多层递阶方法对CSTR系统进行了故障预报.通过仿真研究选择并确定了最佳匹配的模型族和算法族.采用最佳匹配的多层递阶方法进行全数字仿真实验,从而证明了这种方法对于CSTR系统故障预报的有效性.该方法能够对一类带时变参数的非线性系统进行故障预报,并具有工程实用价值.

The fault prediction for continuous stirred tank reactor(CSTR) system has been studied based on the integrated multi-level recursive forecasting method which considering the CSTR system′s dynamic and time-variable characteristics.The optimal match of models and the algorithm of multi-level recursive method have been investigated through simulation.Through used in digital simulation experiments,the proposed method which is specific for CSTR system fault prediction has been validated and proved to be effective.This method can be used to predict the faults in such a class of nonlinear time-varying systems.Hence applying the proposed method in engineering and industry is proved to be feasible.