基于监督机器学习的配电网电力智能运维检修研究

Research on Intelligent Operation and Maintenance of Power Distribution Network Based on Supervised Machine learning

配电变压器是电力配电系统的核心组件,为确保配电系统达到预期的可靠性,必须对配电系统进行定期的检修和提前的风险筛选。为了对配电变压器组进行有效的风险筛选,提出一种基于监督机器学习的配电网配电变压器风险筛选策略。首先介绍实物资产风险管理的行业要求,然后采用案例研究和行动研究相结合的方法,将资产风险管理方法和有监督的机器学习算法有机结合,对待筛选的配电变压器组进行风险筛选,将其分为低、中、高危组,最后在位于高度密集地区的配电网络中使用40个配电变压器的单元作为验证对象,案例结果表明所提方案可以有效实现配电变压器组的风险筛选,增强配电系统中的资产管理能力。

Distribution transformer is the core component of power distribution system.In order to ensure the expected reliability of power distribution system,regular maintenance and risk screening must be carried out in advance.In order to effectively screen the risk of distribution transformer banks,a risk screening strategy for distribution transformers based on supervised machine learning is proposed.First,the industry requirements of real asset risk management are introduced.Then,the method of case study and action study is adopted to combine asset risk management method with supervised machine learning algorithm.The screened distribution transformer groups are screened for risk and divided into low,medium and high risk groups.Finally,40 units of distribution transformers are used as verification objects in a distribution network located in a highly dense area.The case results show that the proposed scheme can effectively realize the risk screening of distribution transformer groups and enhance the asset management ability in the distribution system.