Application of machine learning to enhance the performance of a two-stage, two-temperature ejector cycle driven by the waste heat of exhaust gas

K. Megdouli,Bourhan Tashtoush,Paola Cinnella
DOI: https://doi.org/10.1016/j.enconman.2024.118091
IF: 10.4
2024-02-01
Energy Conversion and Management
Abstract:This article introduces an innovative two-stage, two-temperature ejector cycle (NTDERC) capable of simultaneously generating cooling and electricity. It harnesses waste heat from exhaust gases and is purposefully designed to operate with a neutral carbon footprint. To minimize global warming potential, we opted for a mixture of isobutane/propane (50 %/50 %) as the working fluid. In the NTDERC, the ejector is a crucial component, as it enhances refrigeration capacity and reduces electricity consumption by replacing the traditional compressor. Initially, we analyzed how geometric and operational parameters affect ejector efficiency using one-dimensional modeling and generated a database. An artificial neural networks (ANN) surrogate model is then developed to predict ejector performance, simplifying this otherwise laborious procedure. Finally, we conducted a comprehensive analysis of the NTDERC, considering energy, exergy, and exergoeconomic aspects through parameter evaluation. Results from the Multi-Objective Particle Swarm Optimization (MOPSO) indicate that the optimized total exergy destruction and the total unit cost of the product are 19.39 kW and 22.48 $/GJ, respectively. Implementation of this system in the refrigerated transportation industry has the potential to enhance efficiency, reliability, and sustainability, ultimately ensuring the safe and efficient transportation of temperature-sensitive goods.
energy & fuels,thermodynamics,mechanics
What problem does this paper attempt to address?