TCTA043 Data Center Cabinet Configuration Optimization

In recent years, with the vigorous development of science and technology, people gradually began to rely on computers and various electronic devices to process a large amount of information, and data centers process a large amount of data behind this information every day, accompanied by a large amount of power consumption and environmental impact. of destruction. In order to explore the energy-saving plan of the data center, this study uses the computational fluid dynamics software ANSYS Fluent to simulate the flow field in the data center, and uses the Taguchi method for analysis. Each factor has three levels of change, so an orthogonal table is selected, and nine groups of cases are simulated according to the orthogonal table.

In order to increase the return water temperature of the cooling system, three quality characteristics, the average temperature of the hot air duct, the temperature difference between the highest and the lowest temperature in the hot air duct, and the average temperature difference between the hot and cold air ducts, are selected as the optimization targets, and the grey relational theory is used to find out the quality characteristics under various quality characteristics. , and finally simulate and verify the optimization results. The results show that the factor levels of the optimized case are that the server position is concentrated at the bottom, the air is supplied from the bottom, the air supply flow rate is 800CMH, and the air supply temperature is 288.15K. The temperature difference between the highest and lowest temperature = 2.52K and the average temperature difference between the hot and cold air ducts = 11.06K.

Keywords: data center, computational fluid dynamics, Taguchi method, grey relational analysis

(Read the full text please subscribe to e-journals)

Dear friends~ The 43rd issue of electronic journal has been sent!

Subscribers who have not completed the subscription on the publication date will be sent to your personal mailbox 3-5 days after the subscription.

★ If you do not receive the journal, please write to: tcta.info@gmail.com, we will deal with you as soon as possible, thank you!

Leave a Reply