A least squares support vector machine model optimized by moth-flame optimization algorithm for annual power load forecasting

Cunbin Li

doi:10.1007/s10489-016-0810-2

Return

A least squares support vector machine model optimized by moth-flame optimization algorithm for annual power load forecasting

2016-07-08

123

PRE

Cunbin Li

Shuke Li

Yunqi Liu *

DOI：10.1007/s10489-016-0810-2

Original

Original request for help

Save

Abstract

En 中文

Annual power load forecasting is essential for the planning, operation and maintenance of an electric power system, which can also mirror the economic development of a country to some extent. Accurate annual power load forecasting can provide valuable references for electric power system operators and economic managers. With the development of Energy Internet and further reformation of electric power market, power load forecasting has become a more difficult and challenging task. In this paper, a new hybrid annual power load forecasting model based on LSSVM (least squares support vector machine) and MFO (Moth-Flame Optimization algorithm) is proposed, which the parameters of LSSVM model are optimally determined by the latest nature-inspired metaheuristic algorithm MFO. Meanwhile, the rolling mechanism is also employed. The forecasting results of China's annual electricity consumption indicate the proposed MFO-LSSVM model shows much better forecasting performance than single LSSVM, FOA-LSSVM (LSSVM optimized by fruit fly optimization), and PSO-LSSVM (LSSVM optimized by particle swarm optimization). MFO, as a new intelligence optimization algorithm, is attractive and promising. The LSSVM model optimized by MFO can significantly improve annual power load forecasting accuracy.

Keywords:

Annual power load forecasting

LSSVM (least squares support vector machine)

MFO (Moth-Flame Optimization algorithm)

Rolling mechanism

Parameter optimization

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.