Cylinder Failure Prediction Using Statistical Degradation Indexes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to predict cylinder failures in actuators before they occur, leading to potential breakdowns in manufacturing systems, and require inefficient post-failure diagnosis and replacement.
Innovation Solution
A cylinder failure prediction system that uses statistical degradation indexes based on operation time data to identify potential failures, reducing the need for extensive data processing and enabling proactive replacement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional failure diagnosis methods are used (dividing operation time into items and comparing with normal range), then the diagnosis process is simple, but the ability to predict cylinder failure in advance is insufficient
Solution Approach 1:
The system performs preliminary actions by continuously monitoring operation time data and calculating statistical degradation indexes before failure occurs. The determination part ranks degradation index values and identifies failure-predicted cylinders in advance, enabling proactive maintenance before actual failure happens.
Solution Approach 2:
The patent replaces conventional mechanical/time-based diagnosis methods with statistical analysis methods. Instead of simply dividing operation time into items and comparing with ranges, the system uses statistical degradation indexes (standard deviation, variance, skewness, kurtosis) to objectively quantify and predict failure trends.
2Reliability
If all cylinders are inspected to ensure reliability, then failure prediction accuracy improves, but the workload and time required for inspection increases significantly
Solution Approach 1:
The system extracts only the cylinders that are predicted to fail from the entire population. The determination part ranks degradation index values and identifies failure-predicted cylinders based on predetermined ranges, so only these specific cylinders need inspection, extracting the critical subset from the whole group.
Solution Approach 2:
The patent changes the inspection approach from universal to targeted by using statistical parameters (degradation index values and their rankings) to identify which cylinders require inspection. This parameter-based selection method reduces inspection workload while maintaining high reliability by focusing on cylinders with abnormal degradation patterns.
Data Source
Figure 1
Figure 2(a)~2(b)
Figure 3(a)~3(b)
AI summary
A cylinder failure prediction system of the present technology includes a data acquisition part configured to acquire a plurality of pieces of operation time data for each of a plurality of cylinders by repeatedly detecting operation times of the plurality of cylinders over an equipment operation elapsed time, a degradation index value calculation part configured to calculate degradation index values of a plurality of statistical degradation indexes that can represent a failure of the cylinder on the basis of the operation time data, and a determination part configured to rank the degradation index values of the statistical degradation indexes calculated for each cylinder according to a magnitude, and assign a predetermined score to each of the cylinders having an upper value, which is greater than or equal to a predetermined range, in the ranked degradation index values and determine a failure-predicted cylinder on the basis of a sum of the predetermined scores.