智能电网中电力电子设备故障预警与诊断技术研究

打开文本图片集

中图分类号：TM76 文献标识码：A 文章编号：1672-3791（2026）03-0195-03

Research on Fault Warningand Diagnosis Technology for Power Electronic Equipment in Smart Grid

LEI Ganqi

Guangdong Bangpu Recycling Technology Co.， Ltd.， Foshan， Guangdong Province， China Abstract：To address the fault isues of power electronic equipment in smart grids，this paper proposes a fault warningand diagnosis method that integrates inteligentsensors with deep learning-based feature extraction and Support Vector Machine（SVM） clasification.Real-time collectionof parameters such as voltage，current，temperature，and vibratioarecolected by inteligent sensors;local feature extractionof vibration signalsand spectrograms is conducted by means of deep learning; eficient fault classfication and diagnosis are realized using SVM.Test results show that the system has an accuracy rate of 90% infault prediction for high-characteristic intensity faults such as overloading and short circuits，with an overall accuracy rate of 87% .Additionally，the system response time is less than1 s，and the falsealarm and missed alarm are 2% and 3% ，respectively.

Keyords： Smart grid; Power electronic equipment; Fault warninganddiagnosis; Intellgentsensors; Machine learning

在能源数字化、智能化转型加速下，智能电网作为新型电力系统核心，正发生深刻变革[1]。（剩余3370字）