Power load prediction method

Abstract

The invention discloses a power load prediction method. The method comprises the steps of obtaining original input features of a model, wherein the original input features comprise power load data and meteorological factor data; inputting the original input features into a multilayer RBM network for training learning, and obtaining second input features through multiple times of nonlinear transformation of the RBM network and reconstruction and fine adjustment of parameters; obtaining third input features according to the second input features and a genetic algorithm; obtaining a second weight according to the initial weight of the BP neural network and the genetic algorithm; inputting the third input features and the second weight into the BP neural network, and according to a set error threshold value, performing reverse parameter fine tuning by using a BP algorithm until the error is smaller than or equal to the preset threshold value, thereby obtaining a model for predicting the power load; and inputting to-be-predicted time into the model for predicting the power load for prediction. According to the invention, the speed of training a power load prediction model and the accuracy of a prediction result can be improved.

Description

technical field [0001] The invention relates to the technical field of power load forecasting, in particular to a power load forecasting method. Background technique [0002] For decades, many people have been engaged in the research and application development of power system load forecasting and have achieved a lot of research results, and proposed many load forecasting methods. However, the change of power system load presents a high degree of nonlinearity, and it is still a difficult task to accurately predict the power system load. At present, there is no absolutely accurate power system load forecasting method that can be applied to any power system. A specific method has its own specific conditions for its own application. Only when certain specific power system operating conditions are met can it be achieve a satisfactory level of accuracy. [0003] The origin of short-term power load forecasting is earlier. A large number of domestic and foreign scholars have appl...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiments of the present invention is to provide a power load forecasting method to solve the technical problems that the training of the existing power load forecasting model takes a long time, and the accuracy of the model's power load forecasting is not high because the network model itself is easy to fall into a local minimum point.

Images