Abstract:

In order to solve the problem of multiple estimation due to the complex characteristics of distribution network, a fault distance estimation method based on isolated forest algorithm and big data mining is proposed. Firstly, a new distribution Internet of things architecture of “cloud management side end” is constructed to realize efficient data management. Then, the isolated forest algorithm is introduced into the distribution edge agent to extract the abnormal voltage and current characteristic data. Finally, based on the distribution cloud master station to collect and obtain the abnormal feature data set extracted from a large number of devices, using the big data mining technology based on code base (BDMBCR) to search the association between fault events, classify the branches and groups according to the data similarity, and estimate the specific location of the fault. The IEEE 34 test simulation system is used for the experiment. The results show that the proposed method can mine the complex data characteristics. The multiple estimation length of two-phase and three-phase grounding fault is reduced by 24%-27% and 16%-20% respectively. The single-phase grounding fault influence distance is only 1.2 km, which is better than other comparison methods.

Key words: Distribution power Internet of Things, edge computing, isolated forest algorithm, fault distance estimation, big data mining