Proximal ADMM for Nonconvex and Nonsmooth Optimization

By enabling the nodes or agents to solve small-sized subproblems to achieve coordination, distributed algorithms are favored by many networked systems for efficient and scalable computation. While for convex problems, substantial distributed algorithms are available, the results for the more broad nonconvex counterparts are extremely lacking. This paper develops a distributed algorithm for a class of nonconvex and nonsmooth problems featured by i) a nonconvex objective formed by both separate and composite objective components regarding the decision components of interconnected agents, ii) local bounded convex constraints, and iii) coupled linear constraints. This problem is directly originated from smart buildings and is also broad in other domains. To provide a distributed algorithm with convergence guarantee, we revise the powerful tool of alternating direction method of multiplier (ADMM) and proposed a proximal ADMM. Specifically, noting that the main difficulty to establish the convergence for the nonconvex and nonsmooth optimization within the ADMM framework is to assume the boundness of dual updates, we propose to update the dual variables in a discounted manner. This leads to the establishment of a so-called sufficiently decreasing and lower bounded Lyapunov function, which is critical to establish the convergence. We prove that the method converges to some approximate stationary points. We besides showcase the efficacy and performance of the method by a numerical example and the concrete application to multi-zone heating, ventilation, and air-conditioning (HVAC) control in smart buildings.

[1]  C. Spanos,et al.  A Survey of ADMM Variants for Distributed Optimization: Problems, Algorithms and Features , 2022, ArXiv.

[2]  C. Spanos,et al.  Optimal Network Charge for Peer-to-Peer Energy Trading: A Grid Perspective , 2022, IEEE Transactions on Power Systems.

[3]  Yue-Chun Chen,et al.  Towards Transactive Energy: An Analysis of Information-related Practical Issues , 2021, Energy Conversion and Economics.

[4]  W. Hager,et al.  An inexact accelerated stochastic ADMM for separable convex optimization , 2020, Computational Optimization and Applications.

[5]  Ahmad Afshar,et al.  Distributed Nonconvex Optimization for Energy Efficiency in Mobile Ad Hoc Networks , 2021, IEEE Systems Journal.

[6]  Qing-Shan Jia,et al.  Efficient Real-Time EV Charging Scheduling via Ordinal Optimization , 2021, IEEE Transactions on Smart Grid.

[7]  Pierluigi Siano,et al.  A Comprehensive and Efficient Decentralized Framework for Coordinated Multiperiod Economic Dispatch of Transmission and Distribution Systems , 2021, IEEE Systems Journal.

[8]  Guoqiang Hu,et al.  Distributed Control of Multizone HVAC Systems Considering Indoor Air Quality , 2021, IEEE Transactions on Control Systems Technology.

[9]  Guoqiang Hu,et al.  Optimal Sharing and Fair Cost Allocation of Community Energy Storage , 2020, IEEE Transactions on Smart Grid.

[10]  Xu Andy Sun,et al.  A Two-Level ADMM Algorithm for AC OPF With Global Convergence Guarantees , 2020, IEEE Transactions on Power Systems.

[11]  Gang Feng,et al.  Distributed Proximal Algorithms for Multiagent Optimization With Coupled Inequality Constraints , 2019, IEEE Transactions on Automatic Control.

[12]  Meisam Razaviyayn,et al.  Linearized ADMM Converges to Second-Order Stationary Points for Non-Convex Problems , 2021, IEEE Transactions on Signal Processing.

[13]  Costas J. Spanos,et al.  Selling Renewable Utilization Service to Consumers via Cloud Energy Storage , 2020, ArXiv.

[14]  Yan Peng,et al.  Optimal Resource Allocation in Energy-Efficient Internet-of-Things Networks With Imperfect CSI , 2020, IEEE Internet of Things Journal.

[15]  C. Spanos,et al.  A Proximal Linearization-based Decentralized Method for Nonconvex Problems with Nonlinear Constraints , 2020, ArXiv.

[16]  Giuseppe Notarstefano,et al.  Tracking-ADMM for Distributed Constraint-Coupled Optimization , 2019, Autom..

[17]  Costas J. Spanos,et al.  HVAC Energy Cost Optimization for a Multizone Building via a Decentralized Approach , 2019, IEEE Transactions on Automation Science and Engineering.

[18]  Xuan Zhang,et al.  Decentralized EV-Based Charging Optimization With Building Integrated Wind Energy , 2019, IEEE Transactions on Automation Science and Engineering.

[19]  X. A. Sun,et al.  A two-level distributed algorithm for general constrained non-convex optimization with global convergence , 2019 .

[20]  Qinghua Liu,et al.  Linearized ADMM for Nonconvex Nonsmooth Optimization With Convergence Analysis , 2017, IEEE Access.

[21]  Wotao Yin,et al.  Global Convergence of ADMM in Nonconvex Nonsmooth Optimization , 2015, Journal of Scientific Computing.

[22]  Hongchao Zhang,et al.  Generalized symmetric ADMM for separable convex optimization , 2017, Computational Optimization and Applications.

[23]  Zechun Hu,et al.  Distributed Coordination of EV Charging With Renewable Energy in a Microgrid of Buildings , 2018, IEEE Transactions on Smart Grid.

[24]  N. S. Aybat,et al.  Distributed Linearized Alternating Direction Method of Multipliers for Composite Convex Consensus Optimization , 2015, IEEE Transactions on Automatic Control.

[25]  Prateek Jain,et al.  Non-convex Optimization for Machine Learning , 2017, Found. Trends Mach. Learn..

[26]  Yu Yang,et al.  Stochastic coordination of aggregated electric vehicle charging with on-site wind power at multiple buildings , 2017, 2017 IEEE 56th Annual Conference on Decision and Control (CDC).

[27]  Deren Han,et al.  Convergence of alternating direction method for minimizing sum of two nonconvex functions with linear constraints , 2017, Int. J. Comput. Math..

[28]  Michael M. Zavlanos,et al.  On the Convergence of a Distributed Augmented Lagrangian Method for Nonconvex Optimization , 2017, IEEE Transactions on Automatic Control.

[29]  Maria Prandini,et al.  Dual decomposition for multi-agent distributed optimization with coupling constraints , 2016, Autom..

[30]  Georgios B. Giannakis,et al.  Scalable Electric Vehicle Charging Protocols , 2015, IEEE Transactions on Power Systems.

[31]  Xiaojun Chen,et al.  Alternating Direction Method of Multipliers for a Class of Nonconvex and Nonsmooth Problems with Applications to Background/Foreground Extraction , 2015, SIAM J. Imaging Sci..

[32]  Wotao Yin,et al.  Parallel Multi-Block ADMM with o(1 / k) Convergence , 2013, Journal of Scientific Computing.

[33]  Xiaoming Yuan,et al.  On the convergence of the direct extension of ADMM for three-block separable convex minimization models with one strongly convex function , 2017, Comput. Optim. Appl..

[34]  Qing-Shan Jia,et al.  The joint scheduling of EV charging load with building mounted wind power using simulation-based policy improvement , 2016, 2016 International Symposium on Flexible Automation (ISFA).

[35]  Moritz Diehl,et al.  An Augmented Lagrangian Based Algorithm for Distributed NonConvex Optimization , 2016, SIAM J. Optim..

[36]  Huan Li,et al.  Accelerated Proximal Gradient Methods for Nonconvex Programming , 2015, NIPS.

[37]  Zhi-Quan Luo,et al.  Convergence analysis of alternating direction method of multipliers for a family of nonconvex problems , 2014, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[38]  Guoyin Li,et al.  Global Convergence of Splitting Methods for Nonconvex Composite Optimization , 2014, SIAM J. Optim..

[39]  Ion Necoara,et al.  On linear convergence of a distributed dual gradient algorithm for linearly constrained separable convex problems , 2014, Autom..

[40]  Xiangfeng Wang,et al.  Multi-Agent Distributed Optimization via Inexact Consensus ADMM , 2014, IEEE Transactions on Signal Processing.

[41]  Tianyi Lin,et al.  On the Convergence Rate of Multi-Block ADMM , 2014, 1408.4265.

[42]  Qing Ling,et al.  On the Linear Convergence of the ADMM in Decentralized Consensus Optimization , 2013, IEEE Transactions on Signal Processing.

[43]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..