Multi-Sensor Signal Fusion for Non-Stationary Machine Health Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional machine monitoring and diagnostic techniques are poorly suited for analyzing non-stationary machines due to their time-varying signal characteristics, which makes it difficult to accurately extract features related to machine health and performance.
Innovation Solution
The method involves using multiple sensors to acquire non-stationary signals from machines operating in a non-stationary manner, fusing these signals to produce a combined output, and employing techniques like wavelet transforms and deep learning to extract features that are insensitive to the level of stationarity, allowing for accurate analysis of machine health and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional machine monitoring techniques are used, then the system is simple and easy to operate, but the analysis accuracy for non-stationary machines deteriorates due to time-varying signal characteristics
Solution Approach 1:
The patent segments the non-stationary signal analysis into multiple stationary segments using windowing functions. Each segment is analyzed independently using conventional techniques, and the results are combined to achieve accurate analysis of the entire non-stationary signal. This allows conventional simple techniques to be applied while maintaining high accuracy for non-stationary machines.
Solution Approach 2:
The patent employs adaptive windowing and segmentation strategies that dynamically adjust to the characteristics of non-stationary signals. The system automatically identifies stationary segments within non-stationary signals and applies appropriate analysis methods to each, enabling accurate analysis without requiring completely new complex systems.
2Reliability
If multiple sensors and signal fusion techniques are employed, then the measurement precision and reliability improve, but the device complexity and data processing requirements increase
Solution Approach 1:
The patent combines signals from multiple sensors (vibration, acoustic, temperature, etc.) using signal fusion techniques. The signals are merged in the time domain, frequency domain, or transformed domain to create a comprehensive view of machine health. This merging approach improves reliability by cross-validating measurements from different sensor types while using systematic methods to manage the complexity.
Solution Approach 2:
The patent introduces signal processing intermediaries such as wavelet transforms, Hilbert transforms, and envelope analysis as mediator techniques. These intermediaries process and reconcile data from multiple sensors before final analysis, making the fusion process more manageable and reducing the direct complexity of integrating multiple sensor outputs.
3Measurement precision
If advanced signal processing techniques like wavelet transforms and deep learning are used, then the feature extraction accuracy improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies preliminary signal processing steps such as filtering, detrending, and segmentation before applying advanced techniques like wavelet transforms or deep learning. This preliminary action reduces the complexity and size of the data that requires computationally intensive processing, thereby reducing processing time while maintaining feature extraction accuracy.
Solution Approach 2:
The patent selectively applies advanced signal processing techniques only to specific segments or portions of signals where they are most needed, rather than processing entire datasets with computationally intensive methods. This partial application approach maintains high feature extraction accuracy for critical regions while significantly reducing overall processing time.
Data Source
AI summary
A method for monitoring at least one machine including causing at least a first sensor to acquire at least a first non-stationary signal from at least one machine operating in a non-stationary manner during at least one operational time frame, the at least first sensor providing at least a first non-stationary output, causing at least a second sensor to acquire at least a second non-stationary signal from the at least one machine during the operational time frame, the at least second sensor providing at least a second non-stationary output, fusing the at least first non-stationary output with the at least second non-stationary output to produce a fused output, extracting at least one feature of at least one of the first and second non-stationary signals based on the fused output, analyzing the at least one feature to ascertain a state of health of the at least one machine and performing at least one of a repair operation, maintenance operation and modification of operating parameters of the at least one machine based on the state of health as found by the analyzing.