Multi-Objective Energy, Exergy and Economic Optimization of a Liquid Air Energy Storage System Using Artificial Neural Network and Genetic Algorithm

Multi-Objective Energy, Exergy and Economic Optimization of a Liquid Air Energy Storage System Using Artificial Neural Network and Genetic Algorithm

Mahziyar Ghaedi1 Arman Galedari2 Mahdi Moghimi3

1) Ph.D. Student, School of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran Energy Conversion Department Email:
2) M.Sc. Student, School of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran Energy Conversion Department Email:
3) Associate Professor, School of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran Energy Conversion Department Email:

IOI Code : ABEB-ABKCDC
Publication : 4th International Conference on Recent Advances in Engineering, Innovation and Technology - Belgium(eitconf.com)
Abstract :
    With the increasing penetration of renewable energy sources, the development of high-efficiency and economically viable energy storage systems has become one of the major challenges facing modern power grids. In this study, a conventional Liquid Air Energy Storage (LAES) system was thermodynamically modeled and optimized from energy, exergy, and economic perspectives. First, a comprehensive thermodynamic model was developed in MATLAB, and the thermophysical properties of the working fluids were calculated using REFPROP. The developed model was then validated against the results reported by Zhang et al., yielding a maximum relative error of 6.67%. To reduce computational cost, 2,000 datasets were generated using the thermodynamic model and employed to train an Artificial Neural Network (ANN). The trained ANN was subsequently integrated with a multi-objective Genetic Algorithm (GA) to maximize the Standalone Round-Trip Efficiency (S-RTE) and Standalone Total Exergy Efficiency (S-TEE) while minimizing the Dynamic Payback Period (DPP). The optimization results indicated that, at the optimal operating point, the S-RTE increased from 56.37% to 66.78%, the S-TEE improved from 58.18% to 69.43%, and the DPP decreased from 12.89 to 4.84 years, corresponding to improvements of 18.5% and 19.3% in the energy and exergy performance indices, respectively, together with a 62.4% reduction in the payback period compared with the baseline system. The results demonstrate that integrating thermodynamic modeling with Artificial Neural Networks and Genetic Algorithms provides an effective framework for the simultaneous thermodynamic and economic optimization of Liquid Air Energy Storage systems and facilitates their deployment for large-scale grid energy storage applications.

Keywords : energy storage liquid air energy storage genetic algorithm multi objective optimization energy analysis