Exergy-based optimisation of a phase change materials integrated hybrid renewable system for active cooling applications using supervised machine learning method

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

10 Scopus Citations
View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Pages (from-to)514-526
Journal / PublicationSolar Energy
Volume195
Online published2 Dec 2019
Publication statusPublished - 1 Jan 2020

Abstract

The active cooling, latent thermal storage and advanced energy conversions are effective solutions to high-efficiently utilise renewable energy for building applications, whereas the electricity consumption of active facilities for the thermal performance enhancement needs to be considered. In this study, the exergy analysis of a hybrid renewable system, with on-site thermal and electric energy forms, sensible and latent heat storages, was investigated, in terms of technical feasibility of proposed active cooling solutions. The contradiction between the increased electricity consumption of active cooling facilities and the enhancement of renewable generations has been presented, discussed, together with effective solutions, from the perspective of exergy. A machine-learning based optimisation methodology was proposed and used, to address the contradiction and to maximise the overall exergy, with the integration of an advanced optimisation algorithm. The results showed that, in regard to the contradiction, effective solutions include the active water-based cooling and the optimal design of the geometric and operating parameters. Furthermore, with the adoption of optimal parameters through the machine-learning based optimisation, the overall exergy of the hybrid renewable system is 872.06 kWh, which is 2.6% higher than the maximum overall exergy through the Taguchi standard orthogonal array (849.9 kWh). This study demonstrates an effective solution to the contradiction of an active renewable system, together with a machine-learning based optimisation methodology, which can promote the practical feasibility and applicability of active renewable systems in renewable and sustainable buildings.

Research Area(s)

  • Exergy analysis, Machine learning, On-site thermal and electric energy, Phase Change Materials (PCMs), Sensible and latent storage, Teaching-learning-based optimisation

Citation Format(s)

Exergy-based optimisation of a phase change materials integrated hybrid renewable system for active cooling applications using supervised machine learning method. / Tang, Liang; Zhou, Yuekuan; Zheng, Siqian; Zhang, Guoqiang.

In: Solar Energy, Vol. 195, 01.01.2020, p. 514-526.

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