Adaptive Maintenance Timing from Sensor Threshold Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Condition-based maintenance (CBM) faces challenges in accurately scheduling maintenance due to the vast amount of data generated by sensors, making it difficult to determine the precise time of maintenance need for equipment, especially when multiple sensors are used to monitor complex devices.
Innovation Solution
A method that adapts the estimated time of maintenance need by using changes in device usage over time, involving sensors to monitor operational parameters and adjust threshold levels, allowing for more accurate scheduling of maintenance operations before the need arises, utilizing a network interface and controller to process data and provide adaptive maintenance estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensors are used to monitor equipment for condition-based maintenance, then equipment failure can be detected, but the amount of data generated becomes vast and difficult to manage
Solution Approach 1:
The patent extracts only the essential features and parameters from the vast sensor data that are relevant to predicting maintenance needs. Instead of processing all raw sensor data, the system identifies and extracts key indicators that signal equipment degradation, thereby reducing data volume while maintaining reliability.
Solution Approach 2:
The system performs preliminary analysis of sensor data to identify trends and patterns that indicate future maintenance needs. By analyzing historical sensor data and operational parameters in advance, the system predicts when maintenance will be needed, allowing proactive scheduling before actual failure occurs.
2Measurement precision
If multiple sensors are installed to monitor complex equipment, then characterization accuracy improves, but data amount and processing complexity increase
Solution Approach 1:
The patent segments the monitoring system into distinct functional modules: sensor data acquisition, data processing, predictive analysis, and maintenance scheduling. Each module handles specific tasks independently, making the complex system more manageable and easier to maintain while preserving measurement precision through specialized processing for each sensor type.
Solution Approach 2:
The system employs a universal data processing framework that can handle multiple sensor types and equipment configurations through a single integrated platform. This multi-functional approach allows the same system to process data from various sensors (vibration, temperature, pressure) and apply consistent analytical methods, reducing overall system complexity.
3Ease of operation
If maintenance is scheduled based on fixed intervals, then planning is simple, but it does not accurately reflect actual equipment maintenance needs
Solution Approach 1:
The patent transitions from static fixed-interval scheduling to dynamic condition-based scheduling. The system continuously monitors equipment condition through sensor data and adjusts maintenance timing dynamically based on actual equipment state. This allows maintenance to be scheduled neither too early (wasting resources) nor too late (risking failure), optimizing both simplicity and accuracy.
Solution Approach 2:
The system incorporates feedback loops where sensor data continuously informs maintenance scheduling decisions. Actual equipment performance and degradation trends feed back into the predictive model, which then adjusts maintenance recommendations. This feedback mechanism ensures maintenance timing accurately reflects real equipment needs while maintaining operational simplicity through automated decision support.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
There is provided condition-based maintenance of a device. One or more time instants associated with a change of the maintenance need of the device are estimated on the basis of measurements from a sensor and threshold levels indicating a maintenance need of the device. At least one of the previous estimated time instants is adapted on the basis of the obtained measurements in a subsequent time period.