基于布谷鸟搜索算法优化BP模糊Petri网的水电机组故障诊断

针对水电机组故障诊断的复杂性和传统算法存在的缺点,提出采用布谷鸟搜索算法优化BP模糊Petri网进行故障诊断。首先利用布谷鸟搜索算法的全局搜索功能对网络参数寻优,将得出的全局最优解作为BP模糊Petri网的最优初始参数,再用选取的故障样本数据对模糊Petri网进行学习训练,建立故障特征集到故障类型集的映射关系以实现故障分类。仿真试验表明,该故障诊断方法收敛速度快、准确率高,可应用于实际水电机组故障诊断。

Fault Diagnosis of Hydroelectric Set Based on Cuckoo Search Algorithm Evolving BP Fuzzy Petri Nets

Aiming at the complexity of fault diagnosis of hydroelectric set and the shortcomings of the traditional algorithm, a method based cuckoo search algorithm evolving BP fuzzy petri nets was presented in this paper. First, the cuckoo search algorithm was used to optimize the parameters of network, and the obtained global optimal solution was used as the initial optimal value of BP fuzzy Petri net. Then, the fuzzy Petri net was trained by the selected fault sample data. Thus, the mapping between fault feature set and fault type set was established and the fault of hydroelectric set could be diagnosed. Simulation experiment analysis shows that the proposed fault diagnosis method has a fast convergence rate and higher accurate. So, the method is effective for the fault diagnosis of hydroelectric set.