CDU Anomaly Detection Using Correlated Virtual Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increased complexity of coolant distribution units (CDUs) due to extensive sensor instrumentation leads to higher failure rates, as the added sensors and firmware complexity result in increased failure rates and costs, particularly in non-redundant deployments where a single CDU failure can shut down expensive IT racks.
Innovation Solution
Implementing multi-metric artificial intelligence/machine learning models for anomaly detection in CDUs, which reduces the need for redundant sensors and simplifies firmware, enabling scalable, automatic, and high-speed anomaly detection across multiple deployments, thereby reducing system failure rates and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If extensive sensor instrumentation is added to CDUs to improve monitoring capability, then measurement precision is improved, but device complexity increases leading to higher failure rates
Solution Approach 1:
The patent creates a virtual copy of the physical sensor system through software-based sensors implemented in firmware. These virtual sensors replicate the functionality of physical sensors but exist as software constructs, allowing the system to monitor CDU parameters without adding physical hardware. This resolves the contradiction by providing measurement precision through software simulation rather than physical instrumentation.
Solution Approach 2:
The patent replaces the mechanical/physical sensor system with a software-based monitoring system. Instead of using physical sensors that require hardware installation and maintenance, the system uses firmware-based virtual sensors that compute sensor readings from existing operational data. This substitution eliminates the need for extensive physical sensor instrumentation while maintaining monitoring capability.
2Reliability
If redundant sensors are deployed to improve reliability, then reliability is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates virtual redundant sensors through software implementation. Instead of deploying multiple physical redundant sensors that would increase hardware complexity, the system implements virtual sensors in firmware that can be replicated and configured software-defined. This provides redundancy at no additional hardware cost and with minimal complexity increase.
Solution Approach 2:
The patent makes the sensor system universal by implementing a unified firmware-based virtual sensor framework that can serve multiple monitoring functions simultaneously. The same virtual sensor infrastructure supports various sensing operations and redundancy configurations without requiring separate dedicated hardware for each function, thereby improving reliability without proportionally increasing complexity.
3Measurement precision
If multiple correlated sensors are used to monitor CDU operations, then measurement precision is improved, but anomaly detection accuracy deteriorates due to correlated failures
Solution Approach 1:
The patent introduces an intermediary layer between the physical sensors and the anomaly detection system. This intermediary is the virtual sensor framework implemented in firmware, which acts as a mediator that processes, validates, and correlates sensor data before presenting it to anomaly detection algorithms. This intermediary layer can identify and filter out correlated sensor failures, preventing them from misleading the anomaly detection system while preserving the benefits of multiple correlated sensors for comprehensive monitoring.
Data Source
AI summary
Multi-metric artificial intelligence (AI)/machine learning (ML) models for detection of anomalous behavior of a machine/system are disclosed. The multi-metric AI/ML models are configured to detect anomalous behavior of systems having multiple sensors that measure correlated sensor metrics such as coolant distribution units (CDUs). The multi-metric AI/ML models perform the anomalous system behavior detection in a manner that enables both a reduction in the amount of sensor instrumentation needed to monitor the system's operational behavior as well as a corresponding reduction in the complexity of the firmware that controls the sensor instrumentation. As such, AI-enabled systems and corresponding methods for anomalous behavior detection disclosed herein offer a technical solution to the technical problem of increased failure rates of existing multi-sensor systems, which is caused by the presence of redundant sensor instrumentation that necessitates complex firmware for controlling the sensor instrumentation.