首页|Data on Computational Intelligence Detailed by Researchers at Changsha Universit y of Science and Technology (A Collaborative Multi-component Optimization Model Based On Pattern Sequence Similarity for Electricity Demand Prediction)
Data on Computational Intelligence Detailed by Researchers at Changsha Universit y of Science and Technology (A Collaborative Multi-component Optimization Model Based On Pattern Sequence Similarity for Electricity Demand Prediction)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Fresh data on Machine Learning - Compu tational Intelligence are presented in a new report. According to news originati ng from Changsha, People's Republic of China, by NewsRx correspondents, research stated, "In the new electricity market, the accurate electricity demand predict ion can make high possible profit. However, electricity consumption data exhibit s nonlinearity, high volatility, and susceptibility to various factors." Financial support for this research came from National Natural Science Foundatio n of China (NSFC). Our news journalists obtained a quote from the research from the Changsha Univer sity of Science and Technology, "Most existing prediction schemes inadequately a ccount for these traits, resulting in weak performance. In view of this, we prop ose a collaborative multi-component optimization model (MCOBHPSF) to achieve hi gh accuracy electricity demand prediction. For this model, the original data is first decomposed into linear trend components and nonlinear residual components using the Moving Average filter. Then, the enhanced Pattern Sequence-based Forec asting (PSF) algorithm that can effectively capture data patterns with obvious c hanges is used to accurately forecast the trend component and the embedded Light GBM for residual components. We further optimize the prediction results by using an error optimization scheme based on online sequence extreme learning machines to reduce prediction errors. The results of extensive experiments on four real- world datasets demonstrate that our proposed MCO-BHPSF model outperforms four ad vanced baseline models. In day-ahead prediction, our model is on average 31 % better than PSF baselines."
ChangshaPeople's Republic of ChinaAs iaComputational IntelligenceMachine LearningChangsha University of Science and Technology
NETL
NSTL
