基于人工智能算法的线圈电流故障自动诊断方法  

Automatic fault diagnosis method of coil current based on artificial intelligence algorithm

作　　者：许丹[1] XU Dan(Xianyang Vocational Technical College,Xianyang Shanxi 712000,China)

机构地区：[1]咸阳职业技术学院,陕西咸阳712000

出　　处：《自动化与仪器仪表》2021年第7期96-99,共4页Automation & Instrumentation

基　　金：陕西省教育厅科学研究项目:移动电源双温双控集成电路的设计与应用(No.19JK0936)。

摘　　要：传统线圈电流故障诊断一般以电流局部特征值作为特征向量来分析,这种分析内容的局限性导致特征分类准确度较低,因此,设计一种基于人工智能算法的线圈电流故障自动诊断方法。首先利用开口式霍尔电流传感器采集线圈电流信号,为了消除信号中的噪声,采用五点三次平滑法进行简单去噪滤波,利用多项式最小二乘逼近,建立三次多项式曲线方程,然后引入人工智能算法,采用典型优化动态时间规整算法,通过最优路径来寻找采集信号与参考信号,使用支持向量机将样本映射到高维特征空间中实现结构风险的最小化,将线圈电流数据通过支持向量确定构造成最优超平面完成故障的分类和诊断。通过模拟实验结果表明,设计方法中的故障识别与分类的准确率与传统方法相比高出39.8%,验证了方法的有效性。The traditional coil current fault diagnosis usually takes the current local eigenvalue as the eigenvector to analyze.The limitation of this analysis content leads to the low accuracy of feature classification.Therefore, an automatic coil current fault diagnosis method based on artificial intelligence algorithm is designed.First using open type hall current sensor coil current signal is collected, in order to eliminate the noise in the signal, using simple denoising of five point three times smoothing filter, using the multinomial least square approximation, cubic polynomial curve equation, and then introduce artificial intelligence algorithms, dynamic time neat algorithm was optimized by using typical, through to find the optimal path to collect signal and reference signal, using support vector machine(SVM) samples will be mapped to high-dimensional feature space to realize structural risk minimization, the coil current data by support vector to determine the structure into the optimal hyperplane complete fault classification and diagnosis.The simulation results show that the accuracy of fault identification and classification in the design method is 39.8% higher than that of the traditional method, which verifies the effectiveness of the proposed method.

关 键 词：人工智能算法 线圈电流 故障自动诊断 支持向量机 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TM401[自动化与计算机技术—控制科学与工程]