Optimal sizing of small wind/battery systems considering the DC bus voltage stability effect on energy capture, wind speed variability, and load uncertainty

• Published in: Applied energy 2012-05, Vol.93, p.404-412
• Main Authors: Lujano-Rojas, Juan M., Dufo-López, Rodolfo, Bernal-Agustín, José L.
• Format: Article
• Language: English
• Subjects: ambient temperature; Applied sciences; Banks; batteries; Battery; Charge; Charge controller; Coulombic efficiency; Electric batteries; Energy; Exact sciences and technology; Mathematical models; Optimal sizing; Optimization; stochastic processes; Uncertainty; uncertainty analysis; Wind energy systems; wind power; Wind speed
• ISSN: 0306-2619; 1872-9118
• DOI: 10.1016/j.apenergy.2011.12.035

► We propose a mathematical model for optimal sizing of small wind energy systems. ► No other previous work has considered all the aspects included in this paper. ► The model considers several parameters about batteries. ► Wind speed variability is considered by means of ARMA model. ► The results show how to minimize the expected energy that is not supplied. In this paper, a mathematical model for stochastic simulation and optimization of small wind energy systems is presented. This model is able to consider the operation of the charge controller, the coulombic efficiency during charge and discharge processes, the influence of temperature on the battery bank capacity, the wind speed variability, and load uncertainty. The joint effect of charge controller operation, ambient temperature, and coulombic efficiency is analyzed in a system installed in Zaragoza (Spain), concluding that if the analysis without considering these factors is carried out, the reliability level of the physical system could be lower than expected, and an increment of 25% in the battery bank capacity would be required to reach a reliability level of 90% in the analyzed case. Also, the effect of the wind speed variability and load uncertainty in the system reliability is analyzed. Finally, the uncertainty in the battery bank lifetime and its effect on the net present cost are discussed. The results showed that, considering uncertainty of 17.5% in the battery bank lifetime calculated using the Ah throughput model, about 12% of uncertainty in the net present cost is expected. The model presented in this research could be a useful stochastic simulation and optimization tool that allows the consideration of important uncertainty factors in techno-economic analysis.