Short-term power load prediction method based on CNN-IPSO-GRU hybrid model

Abstract

The invention discloses a short-term power load prediction method based on a CNN-IPSO-GRU hybrid model. The method comprises the following steps: firstly, collecting historical load, meteorological factors, date information and other data of a power grid, performing data normalization processing, dividing a training set and a test set, extracting a multi-dimensional feature vector representing load change by using a convolutional neural network technology, and constructing a time sequence as input of a model; constructing a gating circulation unit network prediction model, optimizing the gating circulation unit network prediction model by utilizing the training set data through an improved particle swarm algorithm to obtain two optimal prediction model parameters, and reestablishing a gating circulation unit network model according to the obtained optimal prediction model parameters; and finally, realizing the short-term load prediction of the power grid by test set data. The method provided by the invention can accurately predict the short-term load change trend of the power grid, and further plays an important role in reducing the loss of the generator set and ensuring economic and reliable operation of the power grid.

Description

technical field [0001] The invention relates to a power load forecasting method, in particular to a short-term load forecasting method based on a CNN-IPSO-GRU hybrid model. Background technique [0002] With the rapid development of my country's electricity market, efficient and accurate short-term load forecasting is an important content of power grid research. Accurate short-term load forecasting plays an important role in reducing the loss of generating units and ensuring the economical and reliable operation of the power grid. Therefore, it is urgent to develop new methods to improve the accuracy of load forecasting and improve the economic benefits of power grids. [0003] Over the years, many scholars at home and abroad have done a lot of research on short-term load forecasting, which can be summarized into three categories: statistical methods, model combination methods and machine learning methods. Statistical methods mainly include time series models, fuzzy forecas...

AI Technical Summary

Problems solved by technology

GRU is improved on the basis of LSTM structure. Compared with LSTM, its own structure is simpler, faster convergence speed and higher prediction accuracy. Because some parameters usually need to rely on traditional experience when using GRU model for prediction. selection, with uncertainty, leading to a decrease in the prediction efficiency and accuracy of the model

Particle swarm optimization (PSO) has the advantages of fewer adjustment parameters, simple iterative optimization ideas, and fast convergence speed. It is widely used in determining model parameters. However, PSO algorithm is easy to fall into local optimum during the optimization process. Therefore, an improved particle swarm optimization (IPSO) algorithm is proposed to enhance the optimization ability of model parameters.

Images