气体绝缘金属封闭开关设备运输振动风险评估算法设计与系统实现

Design and implementation of risk assessment algorithm and system for transportation vibration of gas insulated switchgear

气体绝缘金属封闭开关设备(GIS)在运输过程中受到的超标冲击可能会影响其未来的正常运行,造成安全隐患与经济损失。针对现有运输振动监测产品诊断规则单一、准确性与客观性不佳的现状,提取振动事件时域波形及其离散傅里叶变换的统计特征与形态学特征,进行基于大容量实际样本的神经网络训练优化,构造出面向GIS长途运输监测需求的振动风险评估算法,并以此为基础开发出配套的系统软件。测试结果表明,系统能够出具准确而详细的分析报告,有助于规范驾驶操作,并为责任界定提供有力依据。

The excessive impact on gas insulated switchgear(GIS)in the transportation process may affect its proper operation in the future,resulting in potential safety hazards and economic losses.Aiming at the current situation of“threshold-only”diagnostic rules,poor accuracy and objectivity of existing transportation vibration monitoring products,the statistical and morphological characteristics of time domain waveform and discrete Fourier transform(DFT)spectrum of vibration events are extracted for artificial neural networks(ANN)training and optimization based on large-capacity actual samples.On this basis,a vibration risk assessment algorithm for long-distance transportation monitoring needs of GIS and a supporting software are developed.The test results show that the monitoring system software can produce accurate and detailed analysis report,which is helpful to standardize driving operation and provide meaningful references for defining the responsibility.