[论文解读] A Fully-Distributed Asynchronous Approach for Multi-Area Coordinated Network-Constrained Unit Commitment

Yamin Wang, Lei Wu+1 more|arXiv (Cornell University)|Jan 19, 2018

Optimal Power Flow Distribution参考文献 37被引用 4

一句话总结

本文提出了一种用于多区域网络约束经济调度（NCUC）的全分布式异步ADMM方法，通过有限的区域间信息交换实现独立的本地优化。通过引入自适应惩罚参数、联络线协调机制以及单位启停的启发式固定策略，该方法提升了收敛性与可行性，在存在非凸性的情况下仍表现出良好的数值性能。

ABSTRACT

This paper discusses a consensus-based alternating direction method of multipliers (ADMM) approach to solve the multi-area coordinated network-constrained unit commitment (NCUC) problem in a distributed manner. Due to political and technical difficulties, it is neither practical nor feasible to solve the multi-area coordination problem in a centralized fashion, which requires full access to all data of individual areas. In comparison, in the proposed fully-distributed approach, local NCUC problems of individual areas can be solved independently, and only limited information is exchanged among adjacent areas for facilitating multi-area coordination. Furthermore, as traditional ADMM can only guarantee convergence for convex problems, this paper discusses several strategies to mitigate oscillations, enhance convergence performance, and derive good-enough feasible solutions, including: (i) A tie-line power flow based area coordination strategy is designed to reduce the number of global consensus variables; (ii) Different penalty parameters {ho} are assigned to individual consensus variables and are updated via certain rules during the iterative procedure, which would reduce the impact of initial values of {ho} on convergence performance; (iii) Heuristic rules are adopted to fix certain unit commitment variables for avoiding oscillations during the iterative procedure; and (iv) An asynchronous distributed strategy is studied, which solves NCUC subproblems of small areas multiple times and exchanges information with adjacent areas more frequently within one complete run of slower NCUC subproblems of large areas. Numerical cases illustrate effectiveness of the proposed asynchronous fully-distributed NCUC approach, and investigate key factors that would affect its convergence performance.

研究动机与目标

• 为解决集中式多区域NCUC的局限性，实现去中心化、数据隐私保护的协调机制。
• 克服传统ADMM在非凸NCUC问题中面临的收敛性问题。
• 减少对全局一致性变量的依赖，通过区域特定协调提升可扩展性。
• 开发一种异步策略，通过优先更新较快的子问题以加速收敛。

提出的方法

• 采用基于一致性ADMM框架，将多区域NCUC问题分解为本地子问题。
• 将联络线功率潮流用作协调变量，以最小化全局一致性变量的数量。
• 对每个一致性变量动态更新自适应惩罚参数，以降低对初始值的敏感性。
• 在迭代过程中通过启发式规则固定部分单位启停变量，以抑制振荡。
• 采用异步更新策略，使小型区域子问题可在大型区域子问题周期内多次迭代。
• 通过仅交换联络线潮流和对偶变量等有限的区域间信息，确保数据隐私。

实验结果

研究问题

• RQ1 如何在无需所有区域完整数据访问的前提下，以全分布式方式求解多区域NCUC？
• RQ2 在ADMM框架下，采用何种策略可改善非凸NCUC问题的收敛性并减少振荡？
• RQ3 自适应惩罚参数调整如何影响分布式NCUC中的收敛性能？
• RQ4 异步子问题更新在多大程度上提升了计算效率与收敛性？
• RQ5 联络线功率潮流在减少全局一致性变量数量方面发挥何种作用？

主要发现

• 所提方法即使在存在非凸性的情况下，也能收敛至可行解，优于标准ADMM。
• 自适应惩罚参数更新显著降低了对初始值的敏感性，提升了鲁棒性。
• 基于联络线的协调机制减少了全局一致性变量的数量，增强了可扩展性。
• 对单位启停变量的启发式固定策略能有效抑制迭代更新过程中的振荡。
• 异步策略通过允许较快的子问题独立推进，加速了收敛。
• 数值算例验证了该方法的有效性，表明惩罚参数自适应调整与协调策略是影响收敛性能的关键因素。