Multi-Data-Center Chiller Clustering for Energy Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing chiller energy efficiency in multiple data center setups are either inefficient due to compromised model performance or incur high development, deployment, and maintenance costs, especially when dealing with diverse chillers across different climate regions.
Innovation Solution
A method that integrates user-identified parameters with historical operations data from multiple data centers using a template-based framework, clusters these parameters into optimal groups with similar physical attributes, and generates analytical models for energy efficiency optimization, providing recommendations for parameter adjustments to enhance energy utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If one model is created across all data centers, then development and maintenance costs are reduced, but model performance is significantly compromised
Solution Approach 1:
The patent segments data centers into clusters based on geographical location, climate zones, and infrastructure characteristics. Instead of using a single universal model or individual models for each data center, the system creates specialized models for each cluster, achieving a balance between model performance and development cost.
Solution Approach 2:
The system dynamically adjusts model parameters based on cluster-specific characteristics such as climate conditions, geographical location, and infrastructure attributes. This allows models to adapt to different environments without requiring completely separate models for each data center.
2Reliability
If individual models are created for each chiller in each data center, then model performance is optimized, but development, deployment, and maintenance costs increase considerably
Solution Approach 1:
The patent merges multiple individual chiller models into cluster-level models that represent groups of chillers with similar characteristics. By combining data from multiple chillers within the same cluster, the system maintains optimized performance while reducing the total number of models required.
Solution Approach 2:
The created models serve multiple functions across different data centers within the same cluster. A single model can be applied to multiple chillers and data centers that share similar characteristics, making the model universally applicable and reducing redundant development efforts.
3Productivity
If numerous individual models are developed for diverse chillers across multiple data centers, then energy efficiency optimization is improved, but operational feasibility is reduced due to high maintenance costs
Solution Approach 1:
The system segments the large number of individual models into a smaller number of cluster-based models, making the system more operationally feasible while maintaining energy efficiency optimization capabilities through cluster-specific optimizations.
4Ease of manufacture
If a single model is used across all data centers, then maintenance costs are reduced, but the complexity and uniqueness of diverse chillers cannot be adequately addressed
Solution Approach 1:
The patent applies local quality by creating models that are specifically tailored to the characteristics of each cluster while maintaining a standardized approach across clusters. Each cluster model incorporates the unique properties of chillers in that region, ensuring adequate handling of chiller diversity.
Solution Approach 2:
The cluster-based models provide a universal solution that can be applied across multiple data centers while still addressing local characteristics. This multi-functional approach allows the same modeling framework to handle diverse chillers by adapting to cluster-specific parameters.
Data Source
AI summary
A method includes: integrating, by the computing device, a combination of user identified parameters in a template-based framework with historical operations data of a plurality of data centers in a multiple data center set-up to produce a feature set of parameters; clustering, by the computing device, the exclusive feature set of parameters into an optimal number of groups, each of which comprise similar physical attributes of chillers associated with any of the plurality of data centers; generating, by the computing device, analytical models for data center infrastructure component energy efficiency optimization of the multiple data center set-up based on the optimal number of groups; and providing recommendations from the analytical models as to which parameters are to be adjusted to have a more efficient energy utilization of the data center infrastructure components.