A simplified method of generating sequential meteorological parameters for uncertainty-based energy system design

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

View graph of relations



Original languageEnglish
Article number110780
Journal / PublicationEnergy and Buildings
Online published1 Feb 2021
Publication statusPublished - 15 Apr 2021


The random variation of meteorological parameters affects both building load and energy system design. Although long-term historical meteorological data can reflect this characteristic, a large amount of high-dimensional meteorological data can bring great difficulties to the design process. Therefore, a simplified method of generating meteorological parameters for an uncertainty-based energy system design is proposed in this study. To simplify the meteorological elements, a sensitivity analysis is first carried out based on the maximum information coefficient. Then, combined with the deterministic model and stochastic model of the meteorological parameters, a mixed time-series meteorological model is proposed to reflect the uncertainty. Finally, to generate random meteorological data, based on information entropy theory, Monte Carlo simulation method is adopted and improved by determining the minimum number of simulations. The efficiency and accuracy of this method are verified using a solar heating system design as an example. Results show that the characteristics of uncertainty and time sequence in the meteorological data generated by this method can be well retained. Moreover, the calculation amount can be reduced by 89.3% compared with the original design method.

Research Area(s)

  • Energy system design, Meteorological parameters, Monte Carlo, Sensitivity analysis, Uncertainty