Load prediction method for distribution transformer and distribution line

Abstract

The invention discloses a load prediction method for a distribution transformer and a distribution line, and the method comprises the steps: selecting a historical load, meteorological data and a workday type at a t moment, carrying out the initialization of a connection weight, and carrying out the normalization of a data sample; inputting data, calculating an input layer and a hidden layer of the Elam neural network, and outputting neurons of an output layer and a receiving layer; and performing neuron calculation and prediction based on an Elman neural network algorithm, and evaluating a prediction error. According to the method, an Elman neural network artificial intelligence algorithm is applied, ultra-short-term, short-term, medium-term and long-term power distribution line load prediction is achieved by training historical data and real-time data, accurate bases are provided for power distribution network planning construction and overhaul technical improvement project establishment of power grid enterprises, and the effectiveness and accuracy of investment are improved.

Description

technical field [0001] The invention relates to a distribution transformer and a load forecasting method of a distribution line, belonging to the technical field of electric power distribution. Background technique [0002] Distribution network load forecasting has strong guiding significance for power grid enterprises to adjust the distribution network operation mode, prevent distribution lines and distribution transformers from overloading, and even distribute network construction planning. However, due to the large amount of data and complex topology of the distribution network , it is difficult to achieve accurate prediction. [0003] Elman neural network is a typical dynamic neuron network. Based on the basic structure of BP artificial neural network, it has the function of mapping dynamic features by storing internal states, so that the system has the ability to adapt to time-varying characteristics. Compared with the BP neural network, the Elman neural network has an...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to prevent heavy overload of distribution lines and distribution transformers, solve the problems of large data volume and complex topological structure of distribution network, and it is difficult to realize accurate prediction. The present invention proposes a distribution transformer and distribution line load forecasting method

Images