Search papers, labs, and topics across Lattice.
This paper introduces a decentralized framework for optimizing the joint operations of electrified ethane cracking plants and renewable-driven power grids while preserving the confidentiality of operational data. By employing the Alternating Direction Method of Multipliers, the framework allows for localized problem-solving and minimal data sharing, effectively addressing the Unit Commitment problem in a privacy-friendly manner. Numerical experiments on a synthetic Texas transmission network demonstrate that this approach maintains small optimality gaps and reveals complex, load-dependent emissions outcomes.
Data isolation in decentralized operations can optimize chemical plant scheduling while keeping operational secrets safe, revealing surprising emissions dynamics.
Electrification of ethane cracking offers a promising pathway to industrial decarbonization, provided that the electricity is sourced from renewable energy. However, integrating electrified chemical plant microgrids with a decarbonized power grid requires joint operations planning between Independent System Operators and chemical plants, which is hindered by the highly confidential nature of plant operational data. In this paper, we propose a privacy-friendly decentralized framework based on data isolation that jointly optimizes the Unit Commitment problem in the power system and microgrid scheduling in electrified ethane cracker plants. The framework employs the Alternating Direction Method of Multipliers, augmented with an auxiliary system-level penalty that accelerates convergence, allowing each subsystem to solve its local subproblem and share only minimal coordination signals. To reflect real-world conditions, numerical experiments are conducted on the ACTIVSg2000 test case, a synthetic model of the Texas transmission network, with 26 chemical plants identified from Texas mapped to their nearest grid connection points. In doing so, we characterize the cost of privacy-friendly decomposition on joint power and chemical system decisions, showing that data isolation results in consistently small optimality gaps, and that its emissions consequences are load-dependent and non-monotone.
