基于GWKNN算法的风力发电机组内部短路故障辨识系统  

Internal Short Circuit Fault Identification System of Wind Turbine Based on GWKNN Algorithm

作　　者：赵海亮 ZHAO Hailiang(DCGN New Energy Anhui Co.,Ltd.,Hefei 230031,China)

机构地区：[1]中广核新能源安徽有限公司,合肥230031

出　　处：《自动化与仪表》2023年第9期36-40,共5页Automation & Instrumentation

摘　　要：风力发电机组内部短路故障特征模式多且复杂,导致故障辨识质量下降,该文研究基于GWKNN算法的风力发电机组内部短路故障辨识系统。由电源模块、显示模块、通信模块、控制模块、采集模块、存储模块搭建系统硬件架构,以保证系统控制、数据采集、供电、存储、显示、通信功能的实现。在系统硬件的支持下,采用GWKNN算法获取内部短路故障特征值的异常搜索因子以及特征模式,实现风力发电机组内部短路故障辨识算法设计。分析测试结果可知,该系统的主要模块功能均达到了预期设计目标,能快速、准确地辨识出短路故障信号与类型,平均辨识时间仅需3.99 s,证明所构建的系统应用效果好,能满足实际应用需求。There are many and complex characteristic modes of internal short circuit faults in wind turbines,which leads to a decrease in the quality of fault identification.Therefore,a wind turbine internal short circuit fault identification system based on the GWKNN algorithm is studied.The hardware architecture of the system is constructed by power module,display module,communication module,control module,acquisition module,and storage module to ensure the implementation of system control,data acquisition,power supply,storage,display,and communication functions.With the support of system hardware,the GWKNN algorithm is used to obtain the abnormal search factors and feature patterns of internal short-circuit fault eigenvalues,achieving the design of internal short-circuit fault identification algorithm for wind turbines.The analysis and testing results show that the main module functions of the system have achieved the expected design goals,and can quickly and accurately identify short circuit fault signals and types.The average identification time is only 3.99 s,proving that the constructed system has good application effects and can meet practical application needs.

关 键 词：GWKNN算法 风力发电机组 内部短路故障 故障辨识 异常搜索因子 特征模式 

分 类 号：TP206[自动化与计算机技术—检测技术与自动化装置] TM315[自动化与计算机技术—控制科学与工程]