基于遗传算法的故障诊断方法研究

Research on Fault Diagnosis Method Based on Genetic Algorithm

现代工业系统的大型化、复杂化、自动化的发展趋势使得工业系统的故障诊断方法遇到一系列的考验,误差小、精度高的故障诊断方法的研究成为重中之重。为此本文对故障诊断方法的发展状况进行简介,并对当下研究最热门、效果较好的遗传算法进行详细分析。遗传算法是基于自适应启发式全局搜索概率算法,在故障诊断中主要用于算法的优化,提高故障诊断准确率。本文最后列举了遗传算法优化Bp神经网络的方法用于风机齿轮箱故障诊断的案例。结果显示,在遗传算法的优化下,风机齿轮箱故障诊断的误差得到明显改善,预测精度也提高了。

The large-scale,complicated,and automated development trend of modern industrial systems makes the fault diagnosis methods of industrial systems encounter a series of tests,the research of fault diagnosis methods with small errors and high precision has become the top priority.For this reason,this paper introduces the development status of fault diagnosis methods.And analyzes the most popular and eff ective genetic algorithm in detail.Genetic algorithm is based on an adaptive heuristic global search probability algorithm,which is mainly used for algorithm optimization in fault diagnosis to improve the accuracy of fault diagnosis.In the end,this article lists the case of using genetic algorithm to optimize Bp neural network for fault diagnosis of wind turbine gearbox.The results show that under the optimization of genetic algorithm,the error of wind turbine gearbox fault diagnosis has been signifi cantly improved,and the prediction accuracy has also been improved.