基于风险关键特征量的输电线路运行环境风险评估

发布时间：2019-03-15 11:06

【摘要】：目前架空输电线路运行环境的整体风险评估主要基于复杂的危险源辨别和风险评估(HIRA)和累积扣分法,为了简化输电线路运行环境风险评估方法,建立了结合最小二乘支持向量机(LS-SVM)和主成分分析(PCA)的架空输电线路运行环境整体风险评估模型。首先,通过对故障跳闸情况进行统计分析,得到10种危害架空输电线路安全的运行环境技术因素,并提取出相应的表征风险特征量;考虑实际数据采集的工作量和风险评估的复杂性,使用PCA对特征量进行降维处理,得到权重较大的关键特征量;将这些风险关键特征量作为LS-SVM的输入,建立了风险评估预测模型。最后,以传统评估结果为样本给出评估模型训练和测试结果,结果表明模型预测分类的精度可以达到95%。
[Abstract]:At present, the overall risk assessment of overhead transmission line operation environment is mainly based on the complex hazard source identification and risk assessment (HIRA) and cumulative scoring method, in order to simplify the environmental risk assessment method of transmission line operation. A risk assessment model for overhead transmission line operation environment based on least square support vector machine (LS-SVM) and principal component analysis (PCA) is established. Firstly, through statistical analysis of fault tripping, 10 kinds of operational environment technical factors that endanger the safety of overhead transmission lines are obtained, and the corresponding characteristics of risk are extracted. Considering the workload of actual data collection and the complexity of risk assessment, we use PCA to deal with the dimension reduction of the feature quantity, and get the key feature quantity with large weight. Taking these key risk characteristics as the input of LS-SVM, a risk assessment and prediction model is established. Finally, the training and testing results of the evaluation model are given based on the traditional evaluation results. The results show that the accuracy of the model prediction classification can reach 95%.
【作者单位】：武汉大学电气工程学院;国网四川省电力公司德阳供电公司;国网河南省电力公司郑州供电公司;国家电网东北电力调控分中心;
【基金】：国家自然科学基金青年基金资助项目(51007066)~~
【分类号】：TM75

【相似文献】