Bandit-Based Solar Panel Control

Solar panels sustainably harvest energy from the sun. To improve performance, panels are often equipped with a tracking mechanism that computes the sun’s position in the sky throughout the day. Based on the tracker’s estimate of the sun’s location, a controller orients the panel to minimize the angle of incidence between solar radiant energy and the photovoltaic cells on the surface of the panel, increasing total energy harvested. Prior work has developed efficient tracking algorithms that accurately compute the sun’s location to facilitate solar tracking and control. However, always pointing a panel directly at the sun does not account for diffuse irradiance in the sky, reflected irradiance from the ground and surrounding surfaces, power required to reorient the panel, shading effects from neighboring panels and foliage, or changing weather conditions (such as clouds), all of which are contributing factors to the total energy harvested by a fleet of solar panels. In this work, we show that a bandit-based approach can increase the total energy harvested by solar panels by learning to dynamically account for such other factors. Our contribution is threefold: (1) the development of a test bed based on typical solar and irradiance models for experimenting with solar panel control using a variety of learning methods, (2) simulated validation that bandit algorithms can effectively learn to control solar panels, and (3) the design and construction of an intelligent solar panel prototype that learns to angle itself using bandit algorithms.

[1]  D. Yogi Goswami,et al.  Principles of Solar Engineering , 1978 .

[2]  J. Gittins Bandit processes and dynamic allocation indices , 1979 .

[3]  W. Peterson,et al.  The Ratio of Diffuse to Direct Solar Irradiance (Perpendicular to the Sun's Rays) with Clear Skies—A Conserved Quantity Throughout the Day , 1981 .

[4]  Dimitrios Passias,et al.  Shading effects in rows of solar cell panels , 1984 .

[5]  J. Michalsky The Astronomical Almanac's algorithm for approximate solar position (1950 - 2050). , 1988 .

[6]  Mahesan Niranjan,et al.  On-line Q-learning using connectionist systems , 1994 .

[7]  Martin L. Puterman,et al.  Markov Decision Processes: Discrete Stochastic Dynamic Programming , 1994 .

[8]  Andrew W. Moore,et al.  Reinforcement Learning: A Survey , 1996, J. Artif. Intell. Res..

[9]  S. Kalogirou Design and construction of a one-axis sun-tracking system , 1996 .

[10]  D. L. King,et al.  Analysis of factors influencing the annual energy production of photovoltaic systems , 2002, Conference Record of the Twenty-Ninth IEEE Photovoltaic Specialists Conference, 2002..

[11]  C. Long,et al.  Cloud Coverage Based on All-Sky Imaging and Its Impact on Surface Solar Irradiance , 2003 .

[12]  Tom Markvart,et al.  Practical handbook of photovoltaics : fundamentals and applications , 2003 .

[13]  I. Reda,et al.  Solar position algorithm for solar radiation applications , 2004 .

[14]  Peter Auer,et al.  Finite-time Analysis of the Multiarmed Bandit Problem , 2002, Machine Learning.

[15]  Chris C.S. Lau,et al.  Overcast sky conditions and luminance distribution in Hong Kong , 2004 .

[16]  M. Clifford,et al.  Design of a novel passive solar tracker , 2004 .

[17]  H. Vincent Poor,et al.  Bandit problems with side observations , 2005, IEEE Transactions on Automatic Control.

[18]  Yang Yang,et al.  High-efficiency solution processable polymer photovoltaic cells by self-organization of polymer blends , 2005 .

[19]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[20]  G. Kamali,et al.  Estimating solar radiation on tilted surfaces with various orientations: a study case in Karaj (Iran) , 2006 .

[21]  J. Rizk,et al.  Solar Tracking System: More Efficient Use of Solar Panels , 2008 .

[22]  Roberto Grena,et al.  An algorithm for the computation of the solar position , 2008 .

[23]  Karen Abrinia,et al.  A review of principle and sun-tracking methods for maximizing solar systems output , 2009 .

[24]  N. A. Kelly,et al.  Improved photovoltaic energy output for cloudy conditions with a solar tracking system , 2009 .

[25]  Wei Chu,et al.  A contextual-bandit approach to personalized news article recommendation , 2010, WWW '10.

[26]  M. Benghanem Optimization of tilt angle for solar panel: Case study for Madinah, Saudi Arabia , 2011 .

[27]  Roberto Grena,et al.  Five new algorithms for the computation of sun position from 2010 to 2110 , 2012 .

[28]  Manuel Berenguel,et al.  Control of Solar Energy Systems , 2012 .

[29]  Rustu Eke,et al.  Performance comparison of a double-axis sun tracking versus fixed PV system , 2012 .

[30]  Yongfang Li Molecular design of photovoltaic materials for polymer solar cells: toward suitable electronic energy levels and broad absorption. , 2012, Accounts of chemical research.

[31]  K. Mathieson,et al.  Performance of photovoltaic arrays in-vivo and characteristics of prosthetic vision in animals with retinal degeneration , 2015, Vision Research.

[32]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[33]  A. Bais,et al.  The effect of clouds on surface solar irradiance, based on data from an all-sky imaging system , 2016 .

[34]  M. Littman,et al.  Improving Solar Panel Efficiency Using Reinforcement Learning , 2017 .

[35]  M. Littman,et al.  Toward Improving Solar Panel Efficiency using Reinforcement Learning , 2017 .