Component-based dual decomposition methods for the OPF problem

Mhanna, Sleiman; Chapman, Archie C.; Verbič, Gregor

doi:10.1016/j.segan.2018.04.003

Cited by 29 publications

(21 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, the number of market agents (consumers) is significantly larger than in the conventional OPF problem so that a centralized DOPF computation can be challenging if not intractable. To this end, distributed optimization approaches have been considered [20], [21] to ensure scalability. Next, the problem decomposition needs to be done at a household level to preserve consumers prerogative and privacy [22], [23].…”

Section: Local Energy Trading Under Network Constraintsmentioning

confidence: 99%

Decentralized P2P Energy Trading Under Network Constraints in a Low-Voltage Network

Guerrero

Chapman

Verbič

2019

IEEE Trans. Smart Grid

Self Cite

488

282

View full text Add to dashboard Cite

The increasing uptake of distributed energy resources (DERs) in distribution systems and the rapid advance of technology have established new scenarios in the operation of lowvoltage networks. In particular, recent trends in cryptocurrencies and blockchain have led to a proliferation of peer-to-peer (P2P) energy trading schemes, which allow the exchange of energy between the neighbors without any intervention of a conventional intermediary in the transactions. Nevertheless, far too little attention has been paid to the technical constraints of the network under this scenario. A major challenge to implementing P2P energy trading is that of ensuring that network constraints are not violated during the energy exchange. This paper proposes a methodology based on sensitivity analysis to assess the impact of P2P transactions on the network and to guarantee an exchange of energy that does not violate network constraints. The proposed method is tested on a typical UK low-voltage network. The results show that our method ensures that energy is exchanged between users under the P2P scheme without violating the network constraints, and that users can still capture the economic benefits of the P2P architecture.

show abstract

Section: Local Energy Trading Under Network Constraintsmentioning

confidence: 99%

Decentralized P2P Energy Trading Under Network Constraints in a Low-Voltage Network

Guerrero

Chapman

Verbič

2019

IEEE Trans. Smart Grid

Self Cite

488

282

View full text Add to dashboard Cite

show abstract

“…Furthermore, λ can trivially be computed in ADMM by (21), whereas in ALADIN it is a dual variable of the QP. In both algorithms, the penalty parameter ρ must be chosen carefully since it is widely known to be crucial for good convergence behavior [26]. In ALADIN, a weighting matrix Σ k is introduced for each region k to numerically cope with different value ranges or importance for convergence.…”

Section: Distributed Optimization Algorithmsmentioning

confidence: 99%

“…Furthermore, we neglect bus 67, which is a supplementary offshore wind park and the rated converter power is set to S VSC = 1800 MVA for all VSCs. Lastly, while in [30] converter losses are neglected, we use loss function (26). The centralized solution gives a total generated power of 9530.06 MW at costs of 209169.32 $/h.…”

mentioning

confidence: 99%

Distributed Optimal Power Flow in Hybrid AC–DC Grids

Meyer-Huebner

Suriyah

Leibfried

2019

IEEE Trans. Power Syst.

View full text Add to dashboard Cite

Distributed or multi-area optimal power flow (OPF) in alternating current (AC) grids is currently a subject undergoing intense study to cope with computational burdens in large-scale grids and to maintain self-control of a regional system operator. However, future power grids will most likely be hybrid grids consisting of the conventional AC transmission system combined with high voltage direct current (DC) technology. Thus, we reformulate the full AC-DC OPF problem such that it becomes separable and, therefore, accessible to distributed algorithms. Then, we show in detail two different approaches on the decomposition of a hybrid AC-DC grid. Finally, we implement an improved alternating direction of multipliers method (ADMM) as well as the most recently proposed augmented Lagrangian based alternating direction inexact Newton (ALADIN) method. Simulation results show that optimality gaps below 0.01% are reached with both decomposition approaches and algorithms for two different test systems (5-bus and 66-bus). Furthermore, convergence rates and wall clock times are reduced by around one order of magnitude from ADMM to ALADIN.Index Terms-Distributed optimal power flow, multi-area optimal power flow, AC-DC grid, ADMM, ALADIN.

show abstract

“…The values of r k and s k indicate how distant the iterates are from a solution. 2 If the sets X and Z are convex (which is the case for the SOCP-relaxed OPF) and problem (10) is feasible, ADMM is guaranteed to converge to an optimal point [24]. The main objective of this paper is to establish accelerated variants for the algorithm in (12) to ensure that these residuals decay quickly.…”

Section: Component-based Dual Decomposition and Admmmentioning

confidence: 99%

“…There is a plethora of existing works on distributed OPF. These can be broadly classified into three categories, dual decomposition methods, optimality conditions decomposition (OCD) methods and sparse semidefinite programming (SDP) decomposition methods (see [1] and [2] for a review). The dual decomposition techniques underlying the dual-decompositionbased distributed OPF methods in the literature can in turn be classified into two categories: region-based decompositions and component-based decompositions.…”

Section: Introductionmentioning

confidence: 99%

Accelerated Methods for the SOCP-Relaxed Component-Based Distributed Optimal Power Flow

Mhanna

Verbič

Chapman

2018

2018 Power Systems Computation Conference (PSCC)

Self Cite

View full text Add to dashboard Cite

In light of the increased focus on distributed methods, this paper proposes two accelerated subgradient methods and an adaptive penalty parameter scheme to speed-up the convergence of ADMM on the component-based dual decomposition of the second-order cone programming (SOCP) relaxation of the OPF. This work is the first to apply an adaptive penalty parameter method along with an accelerated subgradient method together in one scheme for distributed OPF. This accelerated scheme is demonstrated to reach substantial speed-ups, as high as 87%, on real-world test systems with more than 9000 buses, as well as on other difficult test cases.

show abstract

Component-based dual decomposition methods for the OPF problem

Cited by 29 publications

References 49 publications

Decentralized P2P Energy Trading Under Network Constraints in a Low-Voltage Network

Decentralized P2P Energy Trading Under Network Constraints in a Low-Voltage Network

Distributed Optimal Power Flow in Hybrid AC–DC Grids

Accelerated Methods for the SOCP-Relaxed Component-Based Distributed Optimal Power Flow

Contact Info

Product

Resources

About