摘要
配电网拓扑结构和设备规模日益复杂,报警信息在配电网主站中汇集,采用智能算法能够高效地进行识别和评估。为此,采用差分进化改进鲸鱼优化算法以生成更合适的初始种群;之后,融合支持向量机构建适应动态条件下配电网风险等级预测的评估模型。结果表明:所提算法的预测准确率达到了95%以上;此外,所提算法的耗时虽然较高,但相比其他四种算法没有超过3s,可以满足实际需求。因此,所提算法在配电网风险评估中具备一定的实用性,可以为配电网风险分级评估提供可靠的算法支撑。
The topology structure and equipment scale of the distribution network are becoming increasingly complex.Alarm information is collected in the main station of the distribution network,and intelligent algorithms can be used to efficiently identify and evaluate it.To this end,the differential evolution was adopted to improve the whale optimization algorithm,in order to generate a more suitable initial population;afterwards,a support vector mechanism was integrated to establish an evaluation model for predicting the risk level of distribution networks under dynamic conditions.The results show that the prediction accuracy of the proposed algorithm reaches over 95%;in addition,although the time consumption of the proposed algorithm is relatively high,it does not exceed 3 seconds compared to the other four algorithms,which can meet practical needs.Therefore,the proposed algorithm has certain practicality in risk assessment of distribution networks and can provide reliable algorithm support for risk classification assessment of distribution networks.
作者
胡晓燕
史静
李冰洁
李泽森
李琥
谈健
Hu Xiaoyan;Shi Jing;Li Bingjie;Li Zesen;Li Hu;Tan Jian(Economic and'Technological Research lnstitute,State Grid Jiangsu Electric Power Go.,Ld.,Nanjing Jiangsu 210000,China)
出处
《电气自动化》
2025年第3期94-96,101,共4页
Electrical Automation
基金
国网江苏经研院项目“基于典型气象场景及重点行业发展的江苏负荷曲线推演技术研究”(SGISJY00GHJS2400158)。
关键词
配电网
风险分级
支持向量机
鲸鱼优化算法
差分进化
distribution network
risk classification
support vector machine
whale optimization algorithm
differential evolution
