Assessment of the performance of metaheuristic methods used for the inverse identification of effective heat capacity of phase change materials

journal article in Web of Science

en

Phase change materials (PCMs) are one of the promising technologies for the carbon neutral future, since they can be effectively used for thermal energy storage (TES), and therefore can mitigate the energy supply/demand mismatch characteristic for renewable energy resources. Mathematical models describing thermal behaviour of PCMs require various material properties, such as specific heat capacity, thermal conductivity or density, which are in many cases difficult to obtain experimentally. This paper introduces an inverse identification procedure which utilises heuristic optimisation methods to solve the inverse problem. The air-PCM heat exchanger was considered as a studied device. In order to avoid any uncertainties tied to the experimental measurement, the exact solution was generated by the numerical model directly (referred to as "pre-simulated"scenario). The cost function was constructed as a root mean square difference between the given calculated and pre simulated outlet air temperatures. In total seven different case studies (five with a base model and two with an extended model) were investigated for the inverse identification of the effective heat capacity of a PCM and seven different heuristic optimisation methods were utilised for each case study. The results have shown that the LSHADE and DE methods were the best performing ones and that one method outperformed the other depending on the problem settings.

Phase change materials (PCMs) are one of the promising technologies for the carbon neutral future, since they can be effectively used for thermal energy storage (TES), and therefore can mitigate the energy supply/demand mismatch characteristic for renewable energy resources. Mathematical models describing thermal behaviour of PCMs require various material properties, such as specific heat capacity, thermal conductivity or density, which are in many cases difficult to obtain experimentally. This paper introduces an inverse identification procedure which utilises heuristic optimisation methods to solve the inverse problem. The air-PCM heat exchanger was considered as a studied device. In order to avoid any uncertainties tied to the experimental measurement, the exact solution was generated by the numerical model directly (referred to as "pre-simulated"scenario). The cost function was constructed as a root mean square difference between the given calculated and pre simulated outlet air temperatures. In total seven different case studies (five with a base model and two with an extended model) were investigated for the inverse identification of the effective heat capacity of a PCM and seven different heuristic optimisation methods were utilised for each case study. The results have shown that the LSHADE and DE methods were the best performing ones and that one method outperformed the other depending on the problem settings.

Phase change materials; Metaheuristics; Inverse problem; Identification of thermophysical properties; Phase change hysteresis

15.03.2024

PERGAMON-ELSEVIER SCIENCE LTD

OXFORD

1873-6793

238

1

16