In EDS ansehen

Approximate dynamic programming with feasibility guarantees

Gespeichert in:
Bibliographische Detailangaben
Titel: Approximate dynamic programming with feasibility guarantees
Autoren: Engelmann, Alexander, Faulwasser, Timm
Quelle: IEEE transactions on control of network systems. 12(2):1565-1576
Verlagsinformationen: 2025.
Publikationsjahr: 2025
Schlagwörter: ADP | flexibility aggregation | hierarchical optimization | large-scale optimization | tree structures | TSO-DSO coordination
Beschreibung: Safe and economic operation of networked systems is challenging. Optimization-based schemes are frequently considered, since they achieve near-optimality while ensuring safety via the explicit consideration of constraints. In applications, these schemes often require solving large-scale optimization problems. Iterative techniques from distributed optimization are frequently proposed for complexity reduction. Yet, they achieve feasibility only asymptotically, which induces a substantial computational burden. This work presents an approximate dynamic programming scheme, which is guaranteed to deliver a feasible solution in "one shot", i.e., in one backward-forward iteration over all subproblems provided they are coupled by a tree structure. Our approach generalizes methods from seemingly disconnected domains such as power systems and optimal control. We demonstrate its efficacy for problems with nonconvex constraints via numerical examples from both domains.
Publikationsart: Article
Sprache: English
ISSN: 2325-5870
Dokumentencode: edsair.dris...01170..4672f29b39ae6f5b53d1a459ceb7d294
Datenbank: OpenAIRE
Beschreibung
Abstract:Safe and economic operation of networked systems is challenging. Optimization-based schemes are frequently considered, since they achieve near-optimality while ensuring safety via the explicit consideration of constraints. In applications, these schemes often require solving large-scale optimization problems. Iterative techniques from distributed optimization are frequently proposed for complexity reduction. Yet, they achieve feasibility only asymptotically, which induces a substantial computational burden. This work presents an approximate dynamic programming scheme, which is guaranteed to deliver a feasible solution in "one shot", i.e., in one backward-forward iteration over all subproblems provided they are coupled by a tree structure. Our approach generalizes methods from seemingly disconnected domains such as power systems and optimal control. We demonstrate its efficacy for problems with nonconvex constraints via numerical examples from both domains.
ISSN:23255870