Multi-objective optimization of supply air jet enhancing airflow uniformity in data center using Taguchi/CRITIC/TOPSIS triple method

Abstract

Jet fans have been introduced into under-floor air distribution systems as an active local measure to reduce hot spot temperatures and improve airflow distribution uniformity in data centers. However, as the rack load increases, the two jet fans are insufficient to balance the airflow distribution in the hotspot area, more jet fans and the corresponding parameters need setting and optimizing. A couple of methods of criteria importance through intercriteria correlation/technique for order preference by similarity to ideal solution were introduced as a multi-criteria decision making method to solve this problem. The optimal design parameters of multiple jet fans were determined by combining the Taguchi method with Computational Fluid Dynamics. The contribution of each parameter to the evaluation index was calculated by signal-to-noise ratio analysis and analysis of variance using Taguchi method, and the nozzle angle was determined to be the most important parameter. Then, Taguchi method was combined with criteria importance through intercriteria correlation/technique for order preference by similarity to ideal solution to transform the multi-objective into a single-objective to determine the optimal combination of design parameters for multiple jet fans. Finally, the best combination case was verified by evaluation indexes including supply heat index, return temperature index, ratio of temperature differentials and index of mixing. In the optimal combination, the highest server temperature was reduced by 5.3 °C compared with two jet fans. The multi-criteria decision making method is suitable for choosing the optimal parameter combination of airflow organization when faced with complex decisions involving several dimensions.