Graduation Semester and Year
2018
Language
English
Document Type
Thesis
Degree Name
Master of Science in Mechanical Engineering
Department
Mechanical and Aerospace Engineering
First Advisor
Dereje Agonafer
Abstract
Acclimation is the process in which an IT equipment adjusts to a change in its environment such as a change in temperature and humidity, allowing it to maintain performance across a range of environmental conditions. In commissioning of a new cold shipped IT equipment in data centers, environmental acclimation is an important step. Proper acclimation also minimizes the risk of powering on a machine that may still be wet due to moisture condensation. An experimental study is carried out to find how the acclimation occurs and the factors influencing it. These experiments are designed with various temperature ranges and using the data from the experiment, a method is developed to predict the ACU output temperature that should be maintained during the acclimation of IT equipment to avoid condensation. A simple machine learning model is developed and trained with the data that was collected from the experiment, this model will predict the change in temperature of the server at each instance. With the predicted temperature and the relation that has been built in this paper, an ACU output temperature is calculated. In this ACU temperature the condensation does not happen since the server temperature is always higher than the dew point temperature of our predicted ACU output temperature. So, the waiting period that is required in the normal acclimation method for the water condensed on the server to get evaporated is not needed and the equipment can be powered immediately after the server acclimates to the room temperature without any delay.
Keywords
Data center optimization, Predictive analytics, Acclimation of data center, Acclimation of IT equipment, Machine learning for acclimation, Condensation, Transportation of servers, Cold shipping of servers, Acclimation
Disciplines
Aerospace Engineering | Engineering | Mechanical Engineering
License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Recommended Citation
Sadasivam, Premsanth, "Robust Predictive Analytics for Determining Acclimation parameter of IT Equipment" (2018). Mechanical and Aerospace Engineering Theses. 426.
https://mavmatrix.uta.edu/mechaerospace_theses/426
Comments
Degree granted by The University of Texas at Arlington