Energy-Efficient Placement Optimization of the HVAC System for 5G Base Station with Redundant Configuration

The HVAC system is a major energy consumer in a base station (BS), accounting for about 40% of the total energy, and its efficiency is greatly influenced by its placement. Due to the necessity of high reliability and availability in 5G BS, HVAC redundancy is required. Therefore, optimizing the placement of the HVAC system is crucial yet challenging for improving energy efficiency in BS. Existing simulation-based analysis methods are infeasible due to their high computational costs, and traditional blind solution search methods are impractical due to the complexities and uncertainties involved. This paper addresses these challenges by proposing an ordinal optimization (OO)-based approach considering practical engineering constraints and HVAC redundancy. The proposed approach aims to find near-optimal solutions with a high probability, facilitating better decision-making in practical design, it involves developing a hybrid physical-based and data-driven parallel simulation model to efficiently analyze the performance of candidate designs. Additionally, a detailed CFD model is developed to make accurate evaluations and generate the best design. Numerical test results demonstrate the effectiveness of the proposed approach in terms of thermal performance, energy efficiency, and search speed.