2020/10/18 by Paritosh Ramanan, Ramanan, Paritosh, Murat Yildirim +5
Engineering · #Distributed #Electric Power System Optimization #FOS: Computer and information sciences #FOS: Mathematics #Islanding Detection in Power Systems #Optimization and Control (math.OC) #Parallel #Power System Reliability and Maintenance #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.2010.09055
openalex publication_date 2020/10/18 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Unit Commitment (UC) is a fundamental problem in power system operations.\nWhen coupled with generation maintenance, the joint optimization problem poses\nsignificant computational challenges due to coupling constraints linking\nmaintenance and UC decisions. Obviously, these challenges grow with the size of\nthe network. With the introduction of sensors for monitoring generator health\nand condition-based maintenance(CBM), these challenges have been magnified.\nADMM-based decentralized methods have shown promise in solving large-scale UC\nproblems, especially in vertically integrated power systems. However, in their\ncurrent form, these methods fail to deliver similar computational performance\nand scalability when considering the joint UC and CBM problem.\n This paper provides a novel decentralized optimization framework for solving\nlarge-scale, joint UC and CBM problems. Our approach relies on the novel use of\nthe subgradient method to temporally decouple various subproblems of the\nADMM-based formulation of the joint problem along the maintenance horizon. By\neffectively utilizing multithreading, our decentralized subgradient approach\ndelivers superior computational performance and eliminates the need to move\nsensor data thereby alleviating privacy and security concerns. Using\nexperiments on large scale test cases, we show that our framework can provide a\nspeedup of upto 50x as compared to various state of the art benchmarks without\ncompromising on solution quality.\n