Cluster type electrical potential safety hazard pre-judgment method based on artificial neural network

Abstract

The invention relates to a cluster type electrical potential safety hazard pre-judgment method based on an artificial neural network. The method comprises steps of collecting monitoring data of the electrical safety monitoring points as training data; preprocessing to obtain sample training data, using electrical safety monitoring points as neurons of the artificial neural network; inputting the preprocessed sample training data into an artificial neural network to obtain an expected output vector and the actual output of the current artificial neural network, calculating the partial derivative of an error function to each neuron of an output layer, correcting a connection weight and a threshold value by using the partial derivative and the output of the neuron of an input layer, and judging whether to continue iteration or not by using a global error; and after training is completed, judging whether electrical potential safety hazards exist or not according to real-time acquired data.On the basis of relevance of an existing power grid, the electrical safety early warning effect under a cluster type monitoring mode is achieved, along with increase of the sample size and the self-learning time, the pre-judgment result and early warning are more accurate and effective, electrical safety hidden dangers are found out as soon as possible, and the life and property safety of peopleis effectively protected.

Description

technical field [0001] The present invention relates to the technical fields of calculation, estimation and counting, and in particular to a method for predicting hidden dangers of electrical safety clusters based on artificial neural networks. Background technique [0002] Due to the extensive application of electric energy, electrical safety is also extensive. No matter in the field of production or in the field of life, electricity is inseparable, and various electrical safety problems will be encountered. In order to better protect people's production and life Safety, the early warning system is used in a large number of electricity consumption occasions, and the electrical safety is guaranteed through the alarm. [0003] However, most of the existing methods on the safety of electricity use in the market are "false warnings" using sensor monitoring methods. The basic monitoring parameters of this type of method include current, voltage, residual current, temperature, et...

AI Technical Summary

Problems solved by technology

[0004] The invention solves the problem in the prior art that electrical safety monitoring starts to feed back information and give warnings only when an abnormality is found, and cannot achieve a real warning effect, and provides an optimized artificial neural network-based clustered electrical safety Hidden danger prediction method