A decoupling strategy for protecting sensitive process information in cooperative optimization of power flow. Issue 1 (12th October 2021)

• Record Type: Journal Article
• Authors: Varelmann, Tim; Otashu, Joannah I.; Seo, Kyeongjun; Lipow, Adrian W.; Mitsos, Alexander; Baldea, Michael
• Is Part Of: AIChE journal. Volume 68:Issue 1(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-10-12
• Subjects: congestion management -- cooperative optimal power flow -- demand‐side management -- load modeling; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; 660.28
• DOI: 10.1002/aic.17429
• Languages: English
• ISSNs: 0001-1541

Abstract

Abstract: Optimal power flow (OPF) with close cooperation between the power grid and flexible electricity‐intensive chemical processes can reduce costs for the grid and for electricity users. However, this would require sharing detailed chemical process models, which may reveal confidential information and potentially jeopardize competitive advantages for a chemical process operator. We propose an algorithm that enables economically advantageous cooperation without the need to exchange sensitive information. Low‐order linear models are used to represent the dynamic behavior of electricity‐intensive processes. We integrate these models into the OPF problem and solve the problem using a decoupling strategy based on Benders‐type cuts. The cuts introduce limited communication between the chemical processes and the grid without exchanging explicit information pertaining to the process dynamics and performance. Our results reproduce solutions obtained by sharing detailed process models up to a user‐defined optimality gap for several test cases that reflect both normal and congested grid states. We also investigate the trade‐off between the value of the optimality gap and computational effort. Finally, we study the scaling behavior of the iterative procedure with respect to flexible loads at multiple grid locations.