基于边缘计算的低压配电网健康状态评价方法

张明泽; 栾文鹏; 艾 欣; 胡博伟; 刘博; 刘文彬

张明泽,栾文鹏,艾欣,胡博伟,刘博,刘文彬.基于边缘计算的低压配电网健康状态评价方法[J].电测与仪表,2024,61(5):91-97.
zhangmingze,luanwenpeng,aixin,hubowei,liubo,liuwenbin.A health status assessment method for low-voltage distribution networksbased on edge computing[J].Electrical Measurement & Instrumentation,2024,61(5):91-97.

基于边缘计算的低压配电网健康状态评价方法

A health status assessment method for low-voltage distribution networksbased on edge computing

DOI：10.19753/j.issn1001-1390.2024.05.014

中文关键词: 低压配网碳中和分布式新能源边缘计算健康状态评价

英文关键词:low-voltage distribution network, carbon neutrality, distributed new energy, edge computing health status assessment

基金项目:国家电网公司科技资助项目（520600230011）

作者中文名	作者英文名	单位
张明泽	zhangmingze	华北电力大学
栾文鹏	luanwenpeng	天津大学
艾欣	aixin	华北电力大学
胡博伟	hubowei	天津大学
刘博	liubo	天津大学
刘文彬	liuwenbin	国网山东电力科学研究院

摘要点击次数: 343

中文摘要:

分布式新能源在配网中的大规模接入是实现碳中和目标的必要过程，但由此引发的若干配网稳定性隐患问题需要重视，从庞大的配网数据中快速、有效的筛选出低压配电台区的薄弱环节，从而进行针对性的优化改造具有重要意义。本文提出了一种基于边缘计算的低压配电网健康状态评价方法，首先在众多指标中选取5+1项重要指标，包括5项基础指标以及1项数据合格率指标，随后对指标进行标准化处理；为提高评价方法的合理性，分别采用基于专家经验的序关系分析法和基于数据差异性的变异系数法，得到各个指标的主观权重和客观权重；最后通过拉格朗日最优乘子法对主观权重和客观权重进行优化，得到各个指标的综合权重与评价函数，并根据函数值得到低压配电网的健康状况进行最终评价结果。所提方法采用边缘计算的方式为通信和主站系统缓解计算压力，通过对淄博市的8个低压台区进行评价分析，结果表明本文所提方法更具有效性和客观性，有望在实际系统中实施。

英文摘要:

The large-scale integration of renewable distributed generations into the distribution network is a necessary process to achieve carbon neutrality, but the potential stability hazards brought in need to be treated seriously. It is of great significance to effectively screen the weak links in the low-voltage distribution network via data analysis, and carry out targeted optimization and improvement. Therefore, a health status assessment method for low-voltage distribution networks based on edge computing is proposed in this paper. In the method, 5+1 indexes are selected for the assessment of low-voltage distribution network, which are performed normalization processing before application. Then the subjective and objective weights of each evaluation index are obtained by the order relation analysis method and the variation coefficient method which simulate the expert knowledge respectively. The subjective weight and objective weight are optimized by Lagrange optimal multiplier method, and the comprehensive weight and evaluation function of each index are obtained. The health status of low-voltage distribution network is reasonably quantified the value of evaluation function. The proposed method is performed by edge computing in the smart terminals on site to relieve computing pressure for communication and main station systems. The analysis results of 8 low-voltage transformer service areas in Zibo City, prove the effectiveness of the proposed method, which can be implemented in practical power systems.

查看全文查看/发表评论下载PDF阅读器

关闭