Learning Warm-Start Points For Ac Optimal Power Flow

A large amount of data has been generated by grid operators solving AC optimal power flow (ACOPF) throughout the years, and we explore how leveraging this data can be used to help solve future ACOPF problems. We use this data to train a Random Forest to predict solutions of future ACOPF problems. To preserve correlations and relationships between predicted variables, we utilize a multi-target approach to learn approximate voltage and generation solutions to ACOPF problems directly by only using network loads, without the knowledge of other network parameters or the system topology. We explore the benefits of using the learned solution as an intelligent warm start point for solving the ACOPF, and the proposed framework is evaluated numerically using multiple IEEE test networks. The benefit of using learned ACOPF solutions is shown to be solver and network dependent, but shows promise for quickly finding approximate solutions to the ACOPF problem.

[1]  Sairaj V. Dhople,et al.  Linear approximations to AC power flow in rectangular coordinates , 2015, 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[2]  Thomas J. Overbye,et al.  A comparison of the AC and DC power flow models for LMP calculations , 2004, 37th Annual Hawaii International Conference on System Sciences, 2004. Proceedings of the.

[3]  Russell Bent,et al.  Tight Piecewise Convex Relaxations for Global Optimization of Optimal Power Flow , 2018, 2018 Power Systems Computation Conference (PSCC).

[4]  R D Zimmerman,et al.  MATPOWER: Steady-State Operations, Planning, and Analysis Tools for Power Systems Research and Education , 2011, IEEE Transactions on Power Systems.

[5]  Krishnamurthy Dvijotham,et al.  Optimal Adaptive Linearizations of the AC Power Flow Equations , 2018, 2018 Power Systems Computation Conference (PSCC).

[6]  Felix F. Wu,et al.  Power system state estimation: a survey , 1990 .

[7]  Concha Bielza,et al.  A survey on multi‐output regression , 2015, WIREs Data Mining Knowl. Discov..

[8]  S. Cvijic,et al.  Applications of Homotopy for solving AC Power Flow and AC Optimal Power Flow , 2012, 2012 IEEE Power and Energy Society General Meeting.

[9]  Daniela M. Witten,et al.  An Introduction to Statistical Learning: with Applications in R , 2013 .

[10]  Kory W. Hedman,et al.  Though a number of algorithms were proposed in literature to improve the convergence performance of the AC model , 2018 .

[11]  Thomas Navidi,et al.  Predicting Solutions to the Optimal Power Flow Problem , 2016 .

[12]  S. Low,et al.  Zero Duality Gap in Optimal Power Flow Problem , 2012, IEEE Transactions on Power Systems.

[13]  F. Milano.  Continuous Newton's Method for Power Flow Analysis , 2009, IEEE Transactions on Power Systems.

[14]  Venkataramana Ajjarapu,et al.  The continuation power flow: a tool for steady state voltage stability analysis , 1991 .

[15]  Ahmed S. Zamzam,et al.  Data-Driven Learning-Based Optimization for Distribution System State Estimation , 2018, IEEE Transactions on Power Systems.

[16]  Fred C. Schweppe,et al.  Power System Static-State Estimation, Part I: Exact Model , 1970 .

[17]  O. Alsac,et al.  Fast Decoupled Load Flow , 1974 .

[18]  Andrey Bernstein,et al.  Joint Chance Constraints in AC Optimal Power Flow: Improving Bounds Through Learning , 2018, IEEE Transactions on Smart Grid.

[19]  Pascal Van Hentenryck,et al.  A Linear-Programming Approximation of AC Power Flows , 2012, INFORMS J. Comput..

[20]  S. Zampieri,et al.  On the Existence and Linear Approximation of the Power Flow Solution in Power Distribution Networks , 2014, IEEE Transactions on Power Systems.

[21]  Line Roald,et al.  Learning for Constrained Optimization: Identifying Optimal Active Constraint Sets , 2018, INFORMS J. Comput..

[22]  Daniel K. Molzahn,et al.  Inexact convex relaxations for AC optimal power flow: Towards AC feasibility , 2019, Electric Power Systems Research.