Thermal Zone Modeling via Airflow Tracing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers face challenges in efficiently provisioning cooling due to varying heat densities within the facility, leading to suboptimal energy and cooling efficiency, as existing methods lack visibility into the cooling distribution across different physical areas.
Innovation Solution
The method involves creating a graphical representation of the data center, defining domains, subdividing them into finite element meshes, identifying air flow sources and sinks, measuring air flow velocities, and tracing these velocities to determine thermal zones, allowing for efficient cooling capacity assessment and optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If uniform cooling is provided across the entire data center, then cooling coverage is ensured, but energy efficiency deteriorates due to over-cooling low-density areas
Solution Approach 1:
The data center floor is divided into multiple thermal zones based on heat density measurements and air flow tracing. Each zone is independently defined and can be cooled according to its specific thermal requirements, replacing the uniform cooling approach with zone-specific cooling strategies.
Solution Approach 2:
Different cooling capacities and strategies are applied to different thermal zones based on their specific heat density requirements. High-density zones receive higher cooling capacity while low-density zones receive reduced cooling, optimizing energy efficiency while meeting local thermal demands.
2Temperature
If cooling capacity is increased to handle peak heat density areas, then hot spots are eliminated, but energy consumption increases due to over-cooling other areas
Solution Approach 1:
The cooling system is segmented into zone-specific cooling assignments, where each thermal zone is served by specific cooling sources. This allows independent temperature control in each zone, enabling peak heat density areas to receive adequate cooling while other areas receive only the cooling they actually need.
Solution Approach 2:
Temperature control is localized to each thermal zone with different temperature setpoints and cooling capacities assigned based on measured heat densities. This prevents the need to over-cool the entire facility to accommodate peak demands in localized areas.
3Loss of energy
If detailed thermal zone analysis is implemented, then cooling efficiency is improved, but system complexity increases due to measurement and modeling requirements
Solution Approach 1:
The system uses the existing air flow measurement infrastructure and thermal modeling capabilities already present in modern data centers. The thermal zone analysis leverages available sensor data and computational tools, minimizing the need for additional complex measurement equipment while still achieving detailed thermal characterization.
Data Source
AI summary
Techniques for using air flow analysis to model thermal zones are provided. In one aspect, a method for modeling thermal zones in a space, e.g., in a data center, includes the following steps. A graphical representation of the space is provided. At least one domain is defined in the space for modeling. A mesh is created in the domain by sub-dividing the domain into a set of discrete sub-domains that interconnect a plurality of nodes. Air flow sources and sinks are identified in the domain. Air flow measurements are obtained from one or more of the air flow sources and sinks. An air flow velocity vector at a center of each sub-domain is determined using the air flow measurements obtained from the air flow sources and sinks. Each velocity vector is traced to one of the air flow sources, wherein a combination of the traces to a given one of the air flow sources represents a thermal zone in the space.