考虑时空分布的配电网站房巡检策略

Inspection strategy of power distribution station considering temporal and spatial distribution

配电网站房巡检工作是全面摸排设备缺陷隐患,提供试验检验依据,提升配电网运行水平的重要手段。针对配电网站房点多面广、巡检工作量大、一线运维人员缺乏的突出问题,提出考虑时空分布的配电网站房巡检策略。该方法从地理信息系统(GIS)中导出运维辖区内站房坐标,经预处理形成数据源文件,综合考虑运维人员承载力、站房重要度和智能化程度、运维周期长短等实际应用场景特点,利用K均值聚类算法和遗传算法对预处理后的数据进行聚类优化分析,形成定期巡检模式下的站房巡检策略和考虑周期长短的站房差异化巡检策略。算例分析表明,所提方法能够有效优化巡检路径和指导运维人员开展巡检工作,提升巡视质量和效率。

The inspection of distribution network room is an important means to comprehensively find out the hidden dangers of equipment defects, provide the basis for test and inspection, and improve the operation level of distribution network. In view of the prominent problems of the wide range of distribution website rooms, large inspection workload and the lack of front-line operation and maintenance personnel, an inspection strategy of distribution website rooms considering temporal and spatial distribution is proposed. In this method, the coordinates of the station buildings within the operation and maintenance area are derived from the geographic information system(GIS) and preprocessed to form the data source file. Considering the characteristics of the actual application scenarios such as the carrying capacity of the operation and maintenance personnel, the importance and intelligence of the station buildings, and the length of the operation and maintenance cycle, K-means algorithm and genetic algorithm are used to cluster and optimize the preprocessed data, forming the station building patrol strategy under the regular patrol mode and the station building differential patrol strategy considering the length of the cycle. The example analysis shows that the proposed method can effectively optimize the inspection path, guide the operation and maintenance personnel to carry out the inspection work, and improve the inspection quality and efficiency.