Power load fluctuation range prediction method and device

Abstract

The invention relates to a power load fluctuation range prediction method and device. The method comprises: obtaining a fuzzy matrix corresponding to a historical power load time sequence at a currentmoment and before the current moment; performing phase space reconstruction on each row of fuzzy data in the fuzzy matrix to obtain a phase space matrix corresponding to each row of fuzzy data; and determining a power load fluctuation range in a future time period after the current moment according to the phase space matrix corresponding to each row of fuzzy data. A phase space matrix corresponding to a historical power load time sequence is obtained; the load fluctuation range of the prediction moment is obtained by using the phase space matrix, the chaos and uncertainty of the power load time sequence are considered, the prediction result is more accurate than the result obtained in the prior art, and a powerful support can be provided for the formulation of a subsequent power distribution plan.

Description

technical field [0001] The invention relates to the technical field of electric power system load forecasting, in particular to a method and device for predicting electric load fluctuation range. Background technique [0002] With the development of big data and artificial intelligence technology, more and more load forecasting technologies have been applied. However, the chaos and variability of power load time series bring challenges to the application of accurate power load forecasting technology. How to solve the prediction problem of chaotic power load sequence has become a difficult research problem. In the existing research, this problem is solved by combining the linear model and the nonlinear model, that is, after the linear model is used for modeling, the prediction result of the linear model is subtracted from the original sequence to obtain the residual item containing the nonlinear component; then Use the nonlinear model to predict the residual item to obtain ...

AI Technical Summary

Problems solved by technology

In the existing research, this problem is solved by combining the linear model and the nonlinear model, that is, after the linear model is used for modeling, the prediction result of the linear model is subtracted from the original sequence to obtain the residual item containing the nonlinear component; then Use the nonlinear model to predict the residual item to obtain the final prediction result. This method subtracts the linear prediction component from the original time series to obtain the residual sequence. The noise component is often large and the signal-to-noise ratio is very low, resulting in nonlinear components. The features are also difficult to be learned by nonlinear methods, and the accuracy of prediction results is low

[0003] The prediction results obtained by the forecasting model in the prior art can only predict the fixed value of the load sequence at a certain moment, while the power load time series is affected by many external factors and has strong uncertainty. Consider this in load forecasting Uncertainty is reasonable and necessary

Images