期刊
ENERGY
卷 227, 期 -, 页码 -出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.energy.2021.120436
关键词
Energy modelling; Manufacturing energy analysis; Energy prediction; HVAC; Building energy analysis; Random forest; Peak shaving
资金
- Engineering and Physical Sciences Research Council (EPSRC)
- University of Strathclyde
This study utilized simulation and machine learning to analyze thermal energy flows in and around a manufacturing facility, predicting spikes in energy consumption and improving energy efficiency, resulting in energy consumption reduction.
Manufacturing companies are subjected to peak-load-dependent energy prices/tariffs, and are faced with high costs for the peak consumption utilised. Complex interactions between HVAC control management, manufacturing schedules, required facility conditions and thermal energy flows in and around the building are difficult to analyse, thus little knowledge exists regarding the interactions between machine, building and HVAC level thermal energy flows. This study utilised simulation for the analysis of thermal energy flows in and around a manufacturing facility, with machine learning adopted for the prediction of spikes in energy consumption based upon weather conditions, occupancy and manufacturing schedules, thus identifying potential energy inefficiencies. The asynchronous optimisation of manufacturing and HVAC schedules allowed for a 15.1% reduction in peak energy demand whilst maintaining levels of manufacturing productivity, and the provision of a more energy efficient management methodology within a manufacturing facility. The resulting model was able to predict energy consumption with an accuracy of 96.5%, and accurately identify patterns in energy profiles. Such a methodology holds the potential to allow for peak consumption reduction and can provide companies with an incentive to monitor and reduce their energy consumption patterns. (c) 2021 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据