A Distributed Multi-Energy Optimal Coordination Scheme Exploiting Neighborhood Flexibility

The optimal coordination of distributed energy resources in large residential neighborhoods presents a critical and computationally challenging problem in demand-side management. To fully exploit demand-side flexibility, we propose a decentralized scheme to address this challenge, in which multiple energy sources and social interactions among residents are taken into account. Local operations are cast as mixed-integer subproblems, organized into blocks to integrate social factors, while several linking constraints are enforced at the neighborhood level. Exploiting this structure, we propose to tackle the dual problem in a distributed manner, followed by primal feasibility recovery. The cutting plane algorithm presented is novel in its asynchronous concurrent mechanism, and its efficiency is validated through multiple tests across various problem configurations. In a one-device-per-block setting with 443 households, our algorithm is able to find relative gap < 0.1% solutions in 2 minutes. In a large-scale test with millions of decision variables out of 28336 households, our algorithm managed to find relative gap < 0.01% solutions within 310 seconds. Unlike existing studies using mainly synchronized algorithms in simulations, ours reflects realistic operating conditions better. Our results demonstrate the effectiveness and applicability of the proposed methodology in addressing complex real-world coordination problems.