Optimization Online


A Hierarchical Alternating Direction Method of Multipliers for Fully Distributed Unit Commitment

Jian Jinbao(2268752262***at***qq.com)
Zhang Chen(2268752262***at***qq.com)
Yang Linfeng(2268752262***at***qq.com)

Abstract: Abstract—This paper discusses a hierarchical alternating direction method of multipliers (ADMM) approach for the unit commitment (UC) problem in a fully distributed manner. Decentralized unit commitment operation schemes have several advantages when compared with the traditional centralized management system for smart grid. Specifically, decentralized management is more flexible, less computationally intensive, and easier to implement without relying on communication infrastructure. In this paper, the fully distributed UC approach is used to solve the UC problem of each individual unit in parallel. This approach consists of two layers of ADMM. In outer ADMM, x-update steps and v-update steps can be decoupled for each unit and executed in parallel after we rearranging and grouping the variables and constraints of the UC mixed integer quadratic programming (MIQP) model according to each unit. The z-update steps, which couple all units because of system constraints, were decoupled for each unit by using an inner consensus ADMM to solve their Lagrangian dual problems. The proposed method can be implemented in master-slave distributed and parallel schema. The master node can be deployed in regional center; other computing nodes can be deployed (or installed) on each unit. Each unit keeps its information secret and does its computations in parallel. The simulation results show that the proposed hierarchical ADMM can obtain high-quality solutions in reasonable times and keep the information of units secret. The results in parallel environment show that the proposed method can obtain good speedup and is suit for solving large-scale D-UC problems.

Keywords: ADMM, full-distributed unit commitment, hierarchical.

Category 1: Applications -- Science and Engineering

Citation: The University of Guangxi. June 1, 2017

Download: [PDF]

Entry Submitted: 08/05/2017
Entry Accepted: 08/06/2017
Entry Last Modified: 08/05/2017

Modify/Update this entry

  Visitors Authors More about us Links
  Subscribe, Unsubscribe
Digest Archive
Search, Browse the Repository


Coordinator's Board
Classification Scheme
Give us feedback
Optimization Journals, Sites, Societies
Mathematical Optimization Society