<?xml version="1.0" encoding="UTF-8"?> <article key="pdf/9998341" mdate="2014-05-04 00:00:00"> <author>Tomohiro Hachino and Hitoshi Takata and Seiji Fukushima and Yasutaka Igarashi</author> <title>ShortTerm Electric Load Forecasting Using Multiple Gaussian Process Models</title> <pages>447 - 452</pages> <year>2014</year> <volume>8</volume> <number>2</number> <journal>International Journal of Electrical and Computer Engineering</journal> <ee>https://publications.waset.org/pdf/9998341</ee> <url>https://publications.waset.org/vol/86</url> <publisher>World Academy of Science, Engineering and Technology</publisher> <abstract>This paper presents a Gaussian process modelbased shortterm electric load forecasting. The Gaussian process model is a nonparametric model and the output of the model has Gaussian distribution with mean and variance. The multiple Gaussian process models as every hour ahead predictors are used to forecast future electric load demands up to 24 hours ahead in accordance with the direct forecasting approach. The separable leastsquares approach that combines the linear leastsquares method and genetic algorithm is applied to train these Gaussian process models. Simulation results are shown to demonstrate the effectiveness of the proposed electric load forecasting. </abstract> <index>Open Science Index 86, 2014</index> </article>