Optimization of immersion cooling performance using the Taguchi Method

Data centers have become one of the most important elements of the Industrial Revolution 4.0. However, conventional data centers face the problem of electricity consumption, with 40% being used for the cooling systems. This led to the proposed development of immersion cooling as one of the methods to solve the current data center cooling system issues. This paper, therefore, was aimed at determining the factors affecting the immersion cooling performance using mineral oil. The experiment was further conducted by creating an immersion cooling box and running the server by loading the software. The contribution and effect of each factor were also determined through the use of the Taguchi Method with Orthogonal Array L8. The performances measured include the CPU temperature and Energy Consumption using the effect of the inlet-outlet location, flowrate and cooling fan speed as the Taguchi factors. The results showed that the CPU temperature was affected by cooling fans with 71.278% and energy consumption by 75.034%. [Display omitted]