Short-term load prediction method for optimizing SVM based on MWOA algorithm

Abstract

The invention discloses a short-term load prediction method for optimizing SVM based on MWOA algorithm in an electric power system. The method comprises the steps of firstly obtaining the historical load data and weather data of an electric power system, and carrying out the preprocessing of the obtained data; secondly, constructing a similar day set through grey relational degree analysis, and generating a training sample and a test sample; establishing a multi-input single-output support vector machine model (SVM) for feature learning, and searching an optimal kernel parameter p and a regularization parameter C of the vector machine model by adopting an improved whale algorithm (MWOA) in the training process; and finally, inputting the test sample into the prediction model to obtain a short-term load prediction result of the power system. The load prediction model algorithm provided by the invention is high in convergence rate, has relatively good balanced traversal, is not liable tofall into a local extreme value, and effectively obtains a relatively high-precision short-term load prediction result of the power system.

Description

Technical field: [0001] The invention relates to the field of power system dispatching communication, in particular to a short-term load forecasting method of a power system based on MWOA algorithm optimization SVM. Background technique: [0002] Electric energy is currently the most important source of energy in the world, but the shortcoming of difficult storage of electric energy has not been effectively resolved, which requires a dynamic balance between power generation planning and load demand. An important guarantee for high power supply quality. Accurate prediction results are the basis for scheduling and refined management, and are the premise for carrying out work such as safety and stability analysis, dynamic state estimation, and maintenance plan preparation. Short-term load forecasting is an important part of the user-side microgrid energy management system and the basis for optimal dispatching of the microgrid. The forecast results can affect the microgrid's op...

AI Technical Summary

Problems solved by technology

[0006] The main purpose of the present invention is to provide a short-term load forecasting method for electric power system to overcome certain defects in the existing load forecasting methods, including the defects of slow learning speed, weak generalization ability and low forecasting accuracy

Images