Power distribution network disaster situation inference method and system based on tree-shaped Bayesian network

Abstract

The embodiment of the invention provides a power distribution network disaster situation inference method and system based on a tree-shaped Bayesian network. The method comprises the following steps:constructing the tree-shaped Bayesian network; determining a key node set and an outage combination set related to the evidence set of the key node set, and obtaining the joint probability of each outage combination in the outage combination set based on the tree-shaped Bayesian network; and determining a maximum value in the joint probability of each outage combination as a final result of disaster situation inference under the evidence set. According to the power distribution network disaster situation inference method and system based on the tree-shaped Bayesian network provided by the embodiment of the invention, the tree-shaped Bayesian network is built through the optimization of the structure of the Bayesian network and an inference algorithm; and based on the analysis of the jointprobability of the outage combination, the key load in the power distribution network is deduced, so that the complexity of probability calculation is effectively simplified, and the second-level real-time disaster situation deduction of the large-scale power distribution network is realized.

Description

technical field [0001] The invention relates to the technical field of power system detection, in particular to a distribution network disaster estimation method and system based on a tree-shaped Bayesian network. Background technique [0002] With the maturity of various technologies in the power grid system, many problems that threaten the security of the power grid in the traditional sense have been gradually resolved. However, extreme disasters have occurred frequently in recent years, which has become the main cause of large-scale power outages. In order to better cope with extreme disasters and ensure the continuous and effective power supply of key loads during disasters, a certain disaster inference system is required to determine the most likely outage of key loads based on limited communication capabilities and a large number of alarm information after the disaster, so as to provide Provide a basis for the recovery plan. [0003] Bayesian network is a commonly us...

AI Technical Summary

Problems solved by technology

But on the one hand, the inference complexity of the Bayesian network increases exponentially with the expansion of the network. Due to the large scale of the distribution network itself, its disaster inference is a task with extremely high timeliness and requires as short a time as possible. Inference results can be obtained in order to update the disaster situation and guide the emergency repair plan

Therefore, the usual Bayesian network is not up to the task

[0004] On the other hand, because the existing fast Bayesian inference methods have certain restrictions on the nodes that can be inferred, it is difficult to require the key load to meet this restriction in the actual distribution network, which makes these methods not practical in engineering

Images