A hybrid forecasting models using least square support vector machine (LSSVM) and genetic algorithm (GA)

Thian, Swee Liang (2021) A hybrid forecasting models using least square support vector machine (LSSVM) and genetic algorithm (GA). Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) A hybrid forecasting models using least square support vector machine (LSSVM) and genetic algorithm (GA).pdf - Submitted Version Download (301kB)
	Text (Full text) A hybrid forecasting models using least square support vector machine (LSSVM) and genetic algorithm (GA).pdf - Submitted Version Restricted to Repository staff only Download (890kB)

Abstract

The project is to develop electricity price forecasting models. Forecast price of electricity become an important issue in the electricity power generation field. The prediction price of electricity price can help the electricity provider decide their electricity price strategy. It can also help the electricity consumer know their electricity price more accurately and estimate their daily electricity usage a day ahead of 24-hour cost. The challenge face on forecasting prices is more on the accuracy and the efficiency of the forecasting model. Therefore, some researchers have developed a forecasting model by using the algorithm to optimize the forecast output. Hence, a combination of the Least Square Support Vector Machine (LSSVM) and Genetic Algorithm (GA) forecast model was developed in this project. The parameter used in the project was the hourly Ontario electricity price (HOEP) and Ontario demand in years 2010 provided by The Independent Electricity System Operator (IESO) website.This type of hybrid forecasting model will be tested on the Ontario power market, which is report as the most volatile market worldwide, to compare with the existing model and provide better accuracy of forecast price and performance. The hybrid forecasting model has completed developed come out with the forcasting results and performance on the hybrid forecasting model itself.

Actions (login required)

View Item