Abstract
To supply power to customers, a power company procures fuel from suppliers while also holding an inventory of such fuel and/or storing a very limited amount of excess power in a battery. Additionally, a power company may directly sell renewable energy sources to customers and exchange power with a neighboring grid to ameliorate shortages. The standard techniques of optimizing profit entail a tedious, human-driven decision-making procedure that results only in local optima. This disclosure optimizes the profit of a power company by automatically making intelligent procurement and selling decisions using machine learning. The decisions are treated as an end-to-end supply-chain problem and jointly optimized, such that optimal trade-offs are achieved amongst supply, demand, system restrictions, and environmental constraints. In particular, the techniques jointly optimize over the end-to-end supply chain, including fuel procurement, fuel and electricity selling, fuel stocking, etc.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Wang, Yixin and Li, Xin, "Optimized Decision Making in the Power Industry Using Machine Learning", Technical Disclosure Commons, (December 24, 2020)
https://www.tdcommons.org/dpubs_series/3924