Direct Data-Driven Methods for Risk Limiting Dispatch

In the classical risk limiting dispatch (RLD) formulation, the system operator dispatches generators relying on information about the distribution of demand. In practice, such information is not readily available and therefore is estimated using historical demand and auxiliary information (or features) such as weather forecasts. In this paper, instead of using a separated estimation and optimization procedure, we propose learning methods that directly compute the RLD decision rule based on historical data. Using tools from statistical learning theory, we then develop generalization bounds and sample complexity results of the proposed methods. These algorithms and performance guarantees, developed for the single-bus network, are then extended to a general network setting for the uniform reserve case.

[1]  Ram Rajagopal,et al.  Risk-limiting dispatch for integrating renewable power , 2013 .

[2]  Jhi-Young Joo,et al.  Efficient Coordination of Wind Power and Price-Responsive Demand—Part I: Theoretical Foundations , 2011, IEEE Transactions on Power Systems.

[3]  Pierre Pinson,et al.  Non‐parametric probabilistic forecasts of wind power: required properties and evaluation , 2007 .

[4]  David Tse,et al.  Network Risk Limiting Dispatch: Optimal Control and Price of Uncertainty , 2012, IEEE Transactions on Automatic Control.

[5]  Haimonti Dutta,et al.  Machine Learning for the New York City Power Grid , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Ambuj Tewari,et al.  On the Complexity of Linear Prediction: Risk Bounds, Margin Bounds, and Regularization , 2008, NIPS.

[7]  Alex Smola,et al.  Kernel methods in machine learning , 2007, math/0701907.

[8]  R. Inman,et al.  Solar forecasting methods for renewable energy integration , 2013 .

[9]  Alexander Shapiro,et al.  The Sample Average Approximation Method for Stochastic Discrete Optimization , 2002, SIAM J. Optim..

[10]  Ram Rajagopal,et al.  Storage in Risk Limiting Dispatch: Control and approximation , 2013, 2013 American Control Conference.

[11]  O. Mangasarian,et al.  Lipschitz continuity of solutions of linear inequalities, programs and complementarity problems , 1987 .

[12]  Pravin Varaiya,et al.  Smart Operation of Smart Grid: Risk-Limiting Dispatch , 2011, Proceedings of the IEEE.

[13]  Michael T. Manry,et al.  Comparison of very short-term load forecasting techniques , 1996 .

[14]  Ruiwei Jiang,et al.  Robust Unit Commitment With Wind Power and Pumped Storage Hydro , 2012, IEEE Transactions on Power Systems.

[15]  C. Carmeli,et al.  Vector valued reproducing kernel Hilbert spaces and universality , 2008, 0807.1659.

[16]  Ram Rajagopal,et al.  Price of uncertainty in multistage stochastic power dispatch , 2014, 53rd IEEE Conference on Decision and Control.

[17]  Andrew R. Barron,et al.  Universal approximation bounds for superpositions of a sigmoidal function , 1993, IEEE Trans. Inf. Theory.

[18]  Yongpei Guan,et al.  Expected Value and Chance Constrained Stochastic Unit Commitment Ensuring Wind Power Utilization , 2014, IEEE Transactions on Power Systems.

[19]  Peter L. Bartlett,et al.  Rademacher and Gaussian Complexities: Risk Bounds and Structural Results , 2003, J. Mach. Learn. Res..

[20]  Kenji Kawaguchi,et al.  Deep Learning without Poor Local Minima , 2016, NIPS.

[21]  A. Papavasiliou,et al.  Reserve Requirements for Wind Power Integration: A Scenario-Based Stochastic Programming Framework , 2011, IEEE Transactions on Power Systems.

[22]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[23]  Paras Mandal,et al.  A review of wind power and wind speed forecasting methods with different time horizons , 2010, North American Power Symposium 2010.

[24]  Tao Hong,et al.  Probabilistic electric load forecasting: A tutorial review , 2016 .