Machine Part Error Monitoring with Dynamic Threshold Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for error detection and monitoring in electronically closed-loop or open-loop controlled machine parts are inefficient due to the difficulty in determining initial monitoring configurations and the need for continuous adjustments over the machine's service life, as they do not account for factors like age and operating conditions.
Innovation Solution
A method and system that record and store operating parameters, determine a comparison group of comparable machine parts, and use statistical analysis to create dynamic threshold values, allowing for variance detection and assignment, thereby accounting for age, wear, and other factors without manual recalibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional static monitoring configuration is used at installation, then initial error detection is possible, but the monitoring system becomes obsolete over the machine's service life due to changing operating conditions and aging
Solution Approach 1:
The patent implements dynamic monitoring configurations that automatically adapt to changing operating conditions and machine aging. The system continuously updates threshold values and monitoring parameters based on actual operational data, transforming the static configuration into a dynamic one that evolves with the machine's service life without requiring manual intervention.
Solution Approach 2:
The monitoring system performs self-calibration and self-adjustment by automatically learning from operational data. The system uses machine learning algorithms to adapt threshold values and detection criteria autonomously, eliminating the need for manual recalibration by operators and enabling the system to maintain accuracy throughout its service life.
2Reliability
If multiple measurements are taken for comprehensive error detection, then more states can be monitored, but the quantity of measurements creates obstacles in effectively identifying the true cause of problems
Solution Approach 1:
The patent extracts and identifies the key measurement from among multiple measurements by using correlation analysis and machine learning algorithms. The system automatically determines which measurements are most relevant to specific error conditions and focuses analysis on those critical parameters, eliminating the need to manually examine all available measurements.
Solution Approach 2:
The system introduces an intelligent intermediary layer (machine learning model) that processes multiple measurements and translates them into meaningful error diagnoses. This intermediary automatically correlates measurements with potential causes and presents simplified error information to operators, reducing the complexity of analyzing multiple parameters.
3Ease of operation
If fixed threshold values are set for monitoring parameters, then error detection can be implemented, but the thresholds become inaccurate as machines age and operating conditions change
Solution Approach 1:
The patent implements automatic adjustment of monitoring parameter thresholds based on machine age, operating conditions, and historical data. The system dynamically modifies threshold values to account for normal aging effects and changing operational environments, maintaining measurement precision without requiring manual reconfiguration by operators.
Data Source
AI summary
In a method for error detection and monitoring an electronically closed-loop or open-loop controlled machine part, operating parameters and monitoring parameters of machine parts are recorded and stored. A comparison group of comparable machine parts and comparable operating parameters is determined based on the recorded and stored operating parameters and a machine part to be compared. A statistical analysis procedure is used for creating a threshold value based on the determined comparison group, and for detecting a variance of at least one state or at least one of the monitoring parameters based on the threshold value. The variance is assigned to the machine part.