Electrical load prediction method and system based on K-means clustering and BI-LSTM neural network

Abstract

The invention discloses an electrical load prediction method and system based on k-means clustering and a BI-LSTM neural network. The method comprises the following steps: preprocessing historical electrical load data through k-means clustering; taking historical electrical load data under the action of the same influence factor as original data through the clustering model to predict the electrical load in a certain time period in the future under the condition of the same influence factor, wherein the predicted data is closer to real data under the real condition, and the prediction reliability is high; on the other aspect, establishing the BI-LSTM neural network model is established to process the data, predicting the current state by using the historical data, and predicting the current state by considering the future condition, so that the basic LSTM neural network is considered, the influence of the hyper-parameter optimization model on the formation of each superposition layer is reduced, and the prediction precision is improved.

Description

technical field [0001] The invention belongs to the field of smart grids, and in particular relates to an electric load forecasting method and system based on k-means clustering and BI-LSTM neural network. Background technique [0002] In recent years, due to the rapid development of computer technology and information technology, the power transmission and distribution system of the power system has sufficient technical support to build a smart grid system. In the management of user electricity demand, the power system hopes to obtain a more accurate electric load for a certain period of time in the future, so as to coordinate the power generation of the unit, so as to avoid the waste of resources due to the grid load failing to meet the user's demand or the grid load being too high Case. [0003] The development of neural network and deep learning technology provides the possibility for big data processing. Before establishing the prediction model of electric load consump...

AI Technical Summary

Problems solved by technology

At present, in the electric load forecasting method, only the basic LSTM network is considered, that is, the one-way long short-term memory recurrent neural network, and the influence of the hyperparameter optimization model on the formation of each superposition layer is not explained, resulting in low electric load forecasting accuracy

Images