基于NGO-SSA优化SVM算法的抽水蓄能水泵故障诊断预测模型研究

Research on a Fault Diagnosis and Prediction Model of Water Pump in Pumped Storage Based on NGO-SSA Optimized SVM Algorithm

为提高抽水蓄能水泵故障预测的准确性,提出一种基于北方苍鹰优化算法(Northern Goshawk Optimization,NGO)与麻雀搜索算法(Sparrow Search Algorithm,SSA)融合(NGO-SSA)优化支持向量机(Support Vector Machine,SVM)的水泵故障智能诊断模型。方法通过编码优化SVM模型参数,应用NGO-SSA算法全局搜索与局部优化,获得全局最优SVM参数组合。训练优化后的SVM分类器,构建故障诊断模型。仿真结果显示,相比SVM和单一优化算法,该方法可以显著提升准确率、降低误报率、增强正负样本识别能力。研究为抽水蓄能水泵健康监测提供了高效智能诊断思路,有助提高水泵故障预测性能。

To improve the accuracy of fault prediction for water pump in pumped storage,an intelligent diagnosis model for water pump faults is proposed based on Support Vector Machine(SVM)optimized with Northern Goshawk Optimization and Sparrow Search Algorithm(NGO-SSA).The parameters of the SVM model are encoded and then optimized by the NGO-SSA algorithm with performance of global search and local optimization,and finally obtains the global optimal combination of SVM parameters.The optimized SVM classifier is trained to construct the fault diagnosis model.The simulation results show that compared to standard SVM and a single optimization algorithm,the proposed method has the advantages of improved accuracy,reduced false alarm rate,and enhanced positive and negative sample recognition capability,which provides efficient and intelligent diagnostic ideas for health monitoring of water pump in pumped storage and helps improve the performance of pump fault prediction.