Data Center Dynamic Cooling System for Linear Power Consumption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data center cooling systems consume excessive power due to being designed to operate at maximum cooling rates, leading to inefficiencies when IT equipment is at low load, resulting in significant energy waste.
Innovation Solution
A dynamic and distributed cooling system with multiple compressors operating in parallel, controlled by a controller to adjust speeds based on temperature, humidity, and load, providing a linear power consumption profile that matches cooling capacity to demand.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If dedicated cooling units are sized to match maximum load on IT equipment, then cooling capacity is sufficient during peak demand, but power consumption becomes excessive during low load conditions
Solution Approach 1:
The cooling system dynamically adjusts compressor operation based on real-time cooling demand. Multiple compressors can operate at variable speeds rather than fixed on/off states, allowing the system to match cooling output precisely to the thermal load from IT equipment, thereby reducing energy waste during low-demand periods while maintaining adequate capacity during peaks
Solution Approach 2:
The system changes operational parameters of compressors (such as speed, capacity modulation) in response to varying cooling loads. By continuously adjusting these parameters rather than maintaining fixed operation, the system optimizes the balance between cooling capacity and power consumption across different load conditions
2Device complexity
If cooling units operate at fixed cooling rates, then system design is simplified, but efficiency deteriorates when IT equipment load varies
Solution Approach 1:
Rather than fixed cooling rates, the system implements dynamic control where compressor speeds and operational states are continuously adjusted based on sensor feedback from the data center environment. This dynamic approach improves energy efficiency across varying loads while the control system manages the added complexity through automated algorithms
3Ease of operation
If cooling units switch on and off to maintain temperature, then operational control is simplified, but power efficiency worsens due to frequent cycling and oversized operation
Solution Approach 1:
The system replaces simple on/off cycling with continuous variable-speed operation of compressors. This allows for smooth modulation of cooling output to precisely match demand, eliminating the energy waste associated with frequent start-stop cycling and oversized operation, while control algorithms manage the increased operational complexity
Solution Approach 2:
Rather than periodic on/off cycling, the system employs continuous adjustment of compressor operation. This eliminates the inefficiencies of repeated startup and shutdown cycles, maintaining steady operation that better matches the continuous thermal load of IT equipment
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. 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. By doing so, the cooling system may be controlled to obtain an approximately linear power consumption as a function of cooling load.