Frequent readers of this blog know how much I dislike today's peer review system in the academic world. The experience this time was truly pleasant.
About two years ago, I submitted a paper to PMAPS2016, which later turned into a TSG paper. The quality of the review comments we received from PMAPS on the original submission was super high, way beyond my expectation. I managed to get the contact information from that reviewer, Antonio Bracale. I then reached out to him to express my appreciation. Later Antonio came back to me with a collaboration proposal on industrial load forecasting. This is the first paper from our collaboration.
The load forecasting literature has been so dominated by forecasting at high voltage level. The smart grid initiatives stimulated many papers at medium or low voltage level. Nevertheless, industrial load forecasting is still an important area that has not yet been extensively studied. This is certainly not the first industrial load forecasting paper, but our findings from the real-world data collected at an Italian factory may be helpful to the others dealing with similar problems.
Citation
Antonio Bracale, Guido Carpinelli, Pasquale De Falco and Tao Hong, "Short-term industrial load forecasting: a case study in an Italian factory," 2017 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), Torino, Italy, September 26-29, 2017
Abstract
Excellence in the planning and operations of power systems largely relies on accurate forecasts of loads. Although load forecasting has been extensively studied over the past several decades, the scientific community has not yet paid much attention to industrial load forecasting. The electricity demand of factories depends on many factors, of which some are uncommon or not as important in the classical load forecasting models. For instance, the scheduled processes and work shifts are very important to forecasting short-term industrial loads. In this paper, we offer some insights into modeling industrial loads. We develop a set of multiple linear regression models for an Italian factory that manufactures transformers. The proposed models outperform two other benchmark models for forecasting industrial loads 24 hours in advance.
About two years ago, I submitted a paper to PMAPS2016, which later turned into a TSG paper. The quality of the review comments we received from PMAPS on the original submission was super high, way beyond my expectation. I managed to get the contact information from that reviewer, Antonio Bracale. I then reached out to him to express my appreciation. Later Antonio came back to me with a collaboration proposal on industrial load forecasting. This is the first paper from our collaboration.
The load forecasting literature has been so dominated by forecasting at high voltage level. The smart grid initiatives stimulated many papers at medium or low voltage level. Nevertheless, industrial load forecasting is still an important area that has not yet been extensively studied. This is certainly not the first industrial load forecasting paper, but our findings from the real-world data collected at an Italian factory may be helpful to the others dealing with similar problems.
Citation
Antonio Bracale, Guido Carpinelli, Pasquale De Falco and Tao Hong, "Short-term industrial load forecasting: a case study in an Italian factory," 2017 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), Torino, Italy, September 26-29, 2017
Short-term Industrial Load Forecasting: a Case Study in an Italian Factory
Antonio Bracale, Guido Carpinelli, Pasquale De Falco and Tao Hong
Abstract
Excellence in the planning and operations of power systems largely relies on accurate forecasts of loads. Although load forecasting has been extensively studied over the past several decades, the scientific community has not yet paid much attention to industrial load forecasting. The electricity demand of factories depends on many factors, of which some are uncommon or not as important in the classical load forecasting models. For instance, the scheduled processes and work shifts are very important to forecasting short-term industrial loads. In this paper, we offer some insights into modeling industrial loads. We develop a set of multiple linear regression models for an Italian factory that manufactures transformers. The proposed models outperform two other benchmark models for forecasting industrial loads 24 hours in advance.
No comments:
Post a Comment
Note that you may link to your LinkedIn profile if you choose Name/URL option.