Engine Health Monitoring Using Fuzzy Signatures for Deterioration Order
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing engine health monitoring (EHM) methods struggle to accurately determine the order of events leading to engine deterioration in turbo engines, particularly between the compressor and turbine, which affects maintenance scheduling and cost estimation.
Innovation Solution
The proposed method employs an Engine Simulation Unit (ESU), Possibilistic Drift Computation Unit (PDCU), Fuzzy String Generator Unit (FSGU), and Information Fusion and Prognosis Unit (IFPU) to analyze engine data, transforming it into fuzzy signatures and predicting Remaining Useful Life (RUL) by comparing engine conditions to a database of known engines, accounting for different filter cut-off frequencies and confidence levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing EHM methods are used to monitor engine deterioration, then maintenance scheduling can be performed, but the accuracy of determining the order of events (compressor vs turbine deterioration) is insufficient
Solution Approach 1:
The patent segments the engine monitoring into separate analyses for compressor and turbine deterioration. It divides the deterioration assessment into distinct components (compressor deterioration indicator and turbine deterioration indicator) that can be independently evaluated and ordered, allowing the system to determine the sequence of deterioration events between these two engine components.
Solution Approach 2:
The patent introduces intermediary indicators (compressor deterioration indicator and turbine deterioration indicator) that mediate between raw engine parameters and the final assessment of deterioration order. These indicators serve as intermediate representations that capture the state of each component, enabling the system to compare and determine the chronological order of deterioration events.
2Measurement precision
If detailed engine parameter monitoring is implemented to improve RUL prediction accuracy, then maintenance planning precision increases, but system complexity and data processing requirements increase
Solution Approach 1:
The patent extracts only the essential features needed for RUL prediction from the full set of engine parameters. It identifies and extracts key parameters related to compressor and turbine deterioration, separating them from other less relevant data. This extraction reduces the complexity of data processing while maintaining the accuracy needed for reliable RUL prediction and maintenance planning.
Solution Approach 2:
The patent applies local quality by focusing monitoring resources on specific critical areas (compressor and turbine deterioration) rather than uniformly monitoring all engine parameters. It concentrates data collection and analysis on the local regions (specific engine components) where deterioration most significantly impacts RUL, thereby reducing overall system complexity while preserving prediction accuracy.
Data Source
Figure 1
Figure 2
Figure 3A~3C
AI summary
The invention relates to an Equipment Health Monitoring method for an engine (100) with the following steps a) a possibilistic drift computation unit (52) automatically allots an upper probability distribution to drift rates of the differences between measured and predicted values generated by an engine simulation unit (51), b) a fuzzy string generator unit (53) transforms the numerical sequence of upper probabilities of the drift rates generated by the possibilistic drift computation unit (52) into a sequence of quantified terms in a fuzzy term set, c) an experience-based string matching unit (54) compares the string of terms generated by the fuzzy string generator unit (53) with at least one other sequence or portion of sequence of previously obtained fuzzy terms in order to determine the degree of similarity to each of the database set, and d) an information fusion and prognosis unit (55) determining in dependence of the matching patterns or portions of the patterns resulting from the comparisons carried out in the experience-based string matching unit (54), providing a rate of engine deterioration indicating the current level of deterioration, the rate of deterioration change and the remaining useful life for a given level of deterioration or requirement or otherwise for engine maintenance or the most likely level of deterioration and the likelihood for the requirement or otherwise of engine maintenance of the engine (100) under test.