Abstract
With the rapid advancement of information technology, the energy consumption of data centers—critical facilities for data storage, processing, and transmission—has increased significantly. To address this challenge, this study proposes two off-grid distributed systems capable of simultaneously providing cooling and power to data centers. Two different energy supply systems were evaluated based on total present value (TPV), the proportion of renewable energy sources (RES), and carbon dioxide emissions (CE), differing primarily in the integration of waste heat recovery subsystems. Furthermore, a multi-objective optimization methodology combining the Non-dominated Sorting Genetic Algorithm II (NSGA-II) with the entropy-weighted Technique for Order Preference by Similarity to the Ideal Solution (EW-TOPSIS) was employed for decision-making to determine the optimal system configuration. The analysis indicates that System 2, which includes the heat recovery subsystem, is the superior option. The optimized configuration consists of an internal combustion engine (ICE) with a power output of 4,255 kW, photovoltaic (PV) modules with a capacity of 1,490 kW, a wind turbine (WT) rated at 100 kW, a waste heat-driven absorption refrigeration (WHAR) unit with a cooling capacity of 668 kW, and an electric chiller (EC) rated at 3,731 kW. This approach enhances data center sustainability by reducing reliance on traditional energy sources and promoting a more resilient energy infrastructure.

This publication has 27 references indexed in Scilit: