Fail-On Distributed Cooling System for Data Center Energy Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data center cooling systems consume excessive power due to their design, which exceeds the necessary cooling requirements, especially during low load conditions, leading to inefficient operation and increased energy costs.
Innovation Solution
A dynamic and distributed cooling system that adjusts the operation of multiple compressors based on various parameters such as temperature, humidity, and load, allowing for linear power consumption and efficient operation by maintaining each compressor within its efficient operating range, and includes redundant refrigeration circuits for fail-on functionality to prevent premature shutdowns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Temperature
If dedicated cooling units are sized to match maximum load on IT equipment, then cooling capacity is sufficient during high load conditions, but power consumption becomes excessive during low load conditions
Solution Approach 1:
The cooling system is divided into multiple independent cooling units, each capable of operating autonomously. This segmentation allows the system to activate only the necessary number of units based on current cooling load, rather than running a single oversized unit at partial capacity, thereby reducing overall power consumption during low load conditions.
Solution Approach 2:
The cooling system dynamically adjusts the number of active cooling units based on real-time monitoring of cooling load conditions. The control system enables flexible switching between one, two, or three active units depending on demand, allowing the system to adapt its capacity to match actual needs and minimize energy waste.
2Temperature
If cooling units switch on and off to maintain desired temperature, then cooling demand is met, but temperature control becomes less precise and causes cycling losses
Solution Approach 1:
Instead of completely switching cooling units on and off, the system maintains continuous operation of multiple units at reduced capacity. This approach eliminates the stop-start cycling behavior, maintaining steady cooling output that better matches fluctuating loads and reduces the energy losses associated with repeated startup and shutdown cycles.
3Reliability
If a cooling system is designed with redundant components for fail-safe operation, then reliability improves, but device complexity and initial cost increase
Solution Approach 1:
Multiple cooling units share common infrastructure including refrigerant piping, control systems, and mounting structures. This merging of common components reduces the overall complexity that would otherwise result from having fully independent redundant systems, while still providing fail-safe operation through the redundancy of the cooling units themselves.
4Power
If multiple compressors are operated in parallel, then cooling capacity increases, but power consumption increases non-linearly
Solution Approach 1:
The system operates multiple compressors at partial capacity rather than running a single compressor at full capacity. By distributing the cooling load across multiple units, each compressor operates in a more efficient portion of its performance curve, reducing the non-linear power consumption that occurs when compressors operate at extreme capacities.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces energy costs by optimizing cooling capacity with respect to power consumption and ensures continuous operation even in case of failures, maintaining efficient performance across varying loads and conditions.
Implementation Method 1
A first compressor of the plurality of compressors is activated at a first output level to meet a demand of a first cooling load
Implementation Method 2
multiple compressors coupled in parallel and configured to provide cooling to a confined space
Data Source
AI summary
One technique for improving cooling system efficiency is to operate cooling for a data center with a dynamic and distributed cooling system. A distributed cooling system may include multiple compressors operating cooperatively for cooling the data center. A dynamic cooling system may adjust operation of the compressors based on one or more parameters, such as inside temperature, outdoor temperature, inside humidity, outside humidity, and load on the data center. A dynamic cooling system may operate in a fail-on mode, adjusting the load placed on independent refrigeration circuits of an air handler system to compensate for a failure of one or more refrigeration circuits. By appropriately controlling speeds of the compressors, the power efficiency of the cooling system may be improved by ensuring that all activated compressors are operating within their efficient operating range.