Energy demand prediction method based on cuckoo algorithm optimized neural network

Abstract

The invention discloses an energy demand prediction method based on a cuckoo algorithm optimized neural network. The method comprises the steps: obtaining historical data of the total energy use amount of a to-be-predicted region, and carrying out the preprocessing to obtain a historical data matrix of the total energy consumption amount; solving each column of correlation coefficients of the matrix, and performing preliminary clustering on historical data in combination with a preset threshold value; adopting an improved K-means mean value clustering algorithm to carry out secondary clustering on the result of the preliminary clustering; optimizing the BP neural network by adopting a cuckoo algorithm until the BP neural network has an optimal weight and threshold value, and constructing an initial prediction model according to the BP neural network at the moment; and training the secondary clustering result according to the initial prediction model to obtain a target prediction model so as to predict the energy demand of the to-be-predicted region. According to the prediction method provided by the invention, the influence of regional difference on the energy demand can be comprehensively considered, the modeling number and complexity are effectively reduced, and meanwhile, the regional energy demand is accurately predicted.

Description

Technical field[0001]The present invention relates to the field of deep learning techniques, and more particularly to an energy demand prediction method based on a cumbersome algorithm optimized a neural network.Background technique[0002]Energy demand forecasts are of great significance for the development of national economies. At present, when performing energy demand forecasts, it is usually a historical data of the total consumption of energy consumption, combined with BP neural networks. However, this method usually has the following problems: First, due to the specific analysis of the influencing factors affecting the total amount of energy consumption in data modeling, the modeling process is complex; second, the second, using BP nerve When the network is predicted, it is easy to cause a local optimal solution due to the randomness of the initial weight and threshold, which in turn affects the accuracy of the prediction result. Therefore, how to provide a predictive method, i...

AI Technical Summary

Problems solved by technology

However, this method usually has the following problems: first, in data modeling, because there is no specific analysis of the factors that affect the total energy consumption, the number of models is redundant and the modeling process is complicated; second, the use of BP neural network When the network makes predictions, it is easy to fall into the local optimal solution due to the randomness of the initial weights and thresholds, which in turn affects the accuracy of the prediction results

Images