A nonparametric Bayesian model for forecasting residential solar generation

This paper presents a framework for generating synthetic residential solar generation profiles. Using the Dirichlet process, characteristics can be clustered and assigned to unobservable connections in a network. This approach retains the variance of this assignment introduced by sparse data, and allows for profiles to be generated specific to individual characteristics. It is also demonstrated that a Markov process modelling changing solar irradiance can be defined from existing solar generation data, rather than specific solar irradiance data in the event that it is not available. This model was applied to data sourced from Ausgrid's Smart Grid, Smart City Program, and a limited initial application found that profiles specific to assigned characteristics could be generated successfully.

[1]  Korbinian Breinl,et al.  Simulating daily precipitation and temperature: a weather generation framework for assessing hydrometeorological hazards , 2015 .

[2]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[3]  Michael I. Jordan,et al.  Hierarchical Dirichlet Processes , 2006 .

[4]  J. Munda,et al.  MODELLING OF CLOUDLESS SOLAR RADIATION FOR PV MODULE PERFORMANCE ANALYSIS , 2009 .

[5]  Zhong-Ping Jiang,et al.  Analysis of Voltage Profile Problems Due to the Penetration of Distributed Generation in Low-Voltage Secondary Distribution Networks , 2012, IEEE Transactions on Power Delivery.

[6]  Alejandro Navarro-Espinosa,et al.  Probabilistic Impact Assessment of Low Carbon Technologies in LV Distribution Systems , 2016, IEEE Transactions on Power Systems.

[7]  T. Ferguson.  A Bayesian Analysis of Some Nonparametric Problems , 1973 .

[8]  Ian Richardson,et al.  Integrated simulation of photovoltaic micro-generation and domestic electricity demand: a one-minute resolution open-source model , 2013 .

[9]  Christopher J. Smith,et al.  Stochastic generation of synthetic minutely irradiance time series derived from mean hourly weather observation data , 2015 .

[10]  J. Farmer,et al.  How Predictable is Technological Progress? , 2015, 1502.05274.

[11]  Gregor Verbic,et al.  Impact study of prosumers on loadability and voltage stability of future grids , 2016, 2016 IEEE International Conference on Power System Technology (POWERCON).

[12]  Gunther Seckmeyer,et al.  Improved Synthesis of Global Irradiance with One-Minute Resolution for PV System Simulations , 2014 .

[13]  Alexis Boukouvalas,et al.  What to Do When K-Means Clustering Fails: A Simple yet Principled Alternative Algorithm , 2016, PloS one.

[14]  Murray Thomson,et al.  High-resolution stochastic integrated thermal–electrical domestic demand model , 2016 .

[15]  Jukka Paatero,et al.  Impacts of distributed photovoltaics on network voltages: Stochastic simulations of three Swedish low-voltage distribution grids , 2010 .

[16]  Maya R. Gupta,et al.  Introduction to the Dirichlet Distribution and Related Processes , 2010 .

[17]  Michael I. Jordan,et al.  Variational inference for Dirichlet process mixtures , 2006 .

[18]  Murray Thomson,et al.  Four-state domestic building occupancy model for energy demand simulations , 2015 .

[19]  Yee Whye Teh,et al.  Stick-breaking Construction for the Indian Buffet Process , 2007, AISTATS.

[20]  D. Blackwell,et al.  Ferguson Distributions Via Polya Urn Schemes , 1973 .

[21]  Yee Whye Teh,et al.  Dirichlet Process , 2017, Encyclopedia of Machine Learning and Data Mining.

[22]  Yann Ménière,et al.  Predicting the costs of photovoltaic solar modules in 2020 using experience curve models , 2013 .

[23]  Gregor Verbic,et al.  Aggregated Demand Response Modelling for Future Grid Scenarios , 2015, 1502.05480.