Predictive Fan Monitoring via Speed-Duty Cycle Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing cooling systems lack predictive monitoring capabilities to identify malfunctioning or deteriorating fans before complete failure, leading to unnecessary downtime and accelerated degradation of other fans when compensating for failed units.
Innovation Solution
A method that generates a model correlating fan rotational speed with the duty cycle of pulse-width modulation electrical current, allowing for the determination of effective rotational speed ranges and predictive failure alerts when operating parameters exceed these ranges, enabling proactive replacement of fans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional voltage-based monitoring methods are used, then the monitoring system is simple, but the system cannot predict fan failures before they occur
Solution Approach 1:
The patent changes the monitored parameter from voltage to rotational speed, which provides predictive information about fan degradation. By monitoring rotational speed across different duty cycles and comparing it to expected values from a model, the system can detect deviations indicating bearing degradation or grease dissipation, enabling failure prediction while maintaining manageable system complexity through software-based monitoring.
2Loss of time
If fans are monitored only after failure, then the monitoring system is simple, but the computing system experiences downtime and other fans accelerate deterioration
Solution Approach 1:
The patent implements preliminary action by establishing a model of expected rotational speed versus duty cycle relationships before fan failure occurs. The monitoring system continuously compares actual measurements against this model, enabling early detection of degradation trends. This allows proactive fan replacement before complete failure, preventing downtime and avoiding the need for other fans to compensate at elevated speeds, thereby preventing accelerated deterioration.
3Measurement precision
If rotational speed is monitored without considering duty cycle, then the monitoring method is simple, but the measurement precision is insufficient to predict failures
Solution Approach 1:
The patent introduces duty cycle as an additional parameter to create a two-dimensional monitoring approach (rotational speed versus duty cycle). By evaluating the relationship between these parameters and comparing against a model that defines expected relationships, the system achieves precise fan health assessment. The model captures the non-linear relationship between duty cycle and rotational speed, enabling accurate detection of degradation that would be invisible in single-parameter monitoring.
Data Source
AI summary
Predictive monitoring of computer cooling systems includes generating, at a computer system, a model representative of at least a relationship between a rotational speed of one or more fans configured to cool computing components and a duty cycle of the pulse-width modulation of electrical current supplied to the one or more fans. The computer system determines a first effective range for the rotational speed of the one or more fans as a function of the duty cycle based on the model and monitors data representative of the rotational speed of the one or more fans and data representative of the duty cycle. The computer system generates a predictive failure alert when the data representative of the rotational speed indicates that the rotational speed is outside of the first effective range for a particular value of the duty cycle.