Power distribution network line fault prediction method based on deep learning

Abstract

The invention discloses a power distribution network line fault prediction method based on deep learning. The power distribution network line fault prediction method comprises the steps of constructing a neural network training model; acquiring historical data of the operation state of the power distribution network, and obtaining power grid inherent attribute data related to the fault cause of the power distribution network; obtaining external weather state data corresponding to the historical data of the operation state of the power distribution network; fusing the power grid inherent attribute data and the external weather state data to form sample data beneficial to power distribution network fault cause and fault prediction; integrating the sample data, and modeling by adopting a long-term and short-term memory neural network to obtain a neural network training model; performing fault prediction by using the neural network training model to obtain a power distribution network fault prediction model; and analyzing the historical data of the power distribution network fault and the current operation state data, weather state data, inherent attributes and other data, and constructing a power distribution network fault prediction model. According to the power distribution network line fault prediction method, the operation states of all lines can be predicted, and the probability of power failure events can be effectively reduced.

Description

technical field [0001] The invention belongs to the technical field of electric power fault diagnosis, and in particular relates to a method for predicting faults of distribution network lines based on deep learning. Background technique [0002] The distribution network is located at the end of the power system and is closely connected with users. It directly distributes and supplies power to users. The operating status of the distribution network will directly affect the quality of power consumption of users. Users require the power supply system to meet the power supply requirements at any time. Demand, the reliability of power supply is high, once a power outage occurs, it will affect the user's experience and cause adverse effects. The statistical results show that the vast majority of customer failures and power outages are caused by distribution network failures. If distribution network failures can be effectively predicted and operation and maintenance performed in a...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, the proportion of cable lines is increasing and the cable lines are buried underground. It is not suitable to find problems intuitively. Once a problem occurs, it will often affect multiple lines, and its influence is relatively large.

At the same time, the existing distribution network structure is becoming more and more complex, the scale is getting larger and larger, and the number of equipment is increasing. However, the limited human and material resources can only be placed on the monitoring of key lines, and all lines cannot be maintained in time. As a result, some lines can only be operated and maintained after a fault occurs. How to effectively monitor all lines has become an urgent problem to be solved