Generation method for the PV power time series combining the decomposition technique and Markov chain theory. Issue 13 (19th January 2018)

• Record Type: Journal Article
• Authors: Xu, Shenzhi; Ai, Xiaomeng; Fang, Jiakun; Wen, Jinyu; Li, Pai; Huang, Yuehui
• Is Part Of: Journal of engineering. Volume 2017:Issue 13(2017)
• Page Start: 2026
• Page End: 2031
• Publication Date: 2018-01-19
• Subjects: time series -- photovoltaic power systems -- power system security -- Markov processes -- Monte Carlo methods; PV power time series -- decomposition technique -- photovoltaic power generation -- power system security operation -- amplitude parameter series -- random fluctuating component -- Markov chain Monte Carlo method -- MCMC method -- random component -- random sampling -- real‐world PV plants -- Gansu -- China -- basic statistical characteristics -- distribution characteristics -- fluctuation characteristics; Engineering -- Periodicals; Engineering; Electronic journals; Periodicals; 620.005
• DOI: 10.1049/joe.2017.0685
• Languages: English
• ISSNs: 2051-3305

Abstract

Abstract : Photovoltaic (PV) power generation has made considerable developments in recent years. However, its intermittent and volatility of its output have seriously affected the security operation of the power system. In order to better understand the PV generation and provide sufficient data support for analysis the impacts, a novel generation method for PV power time series combining decomposition technique and Markov chain theory is presented here. It digs important factors from historical data from existing PV plants and then reproduce new data with similar patterns. In detail, the proposed method first decomposes the PV power time series into ideal output curve, amplitude parameter series, and random fluctuating component three parts. Then generating daily ideal output curve by the extraction of typical daily data, amplitude parameter series based on the Markov chain Monte Carlo (MCMC) method, and random component based on random sampling, respectively. Finally, the generated three parts are recombined into new PV power time series by the decomposition formula. Data obtained from real‐world PV plants in Gansu, China, validates the effectiveness of the proposed method. The generated series can simulate the basic statistical, distribution, and fluctuation characteristics of the measured series.