Diagnostic Robot Hammering Sound Analysis for Stator Wedge Loosening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional diagnostic methods require pre-existing diagnostic results from other inspection targets to accurately diagnose a single inspection target, which can lead to erroneous differentiation between normal and abnormal states when using a diagnostic robot for hammering sound diagnosis, especially in cases where the robot lacks the force to lift the entire abnormal portion.
Innovation Solution
A diagnostic robot equipped with a hammering sound signal acquisition unit, frequency characteristic conversion unit, and abnormality determination unit that analyzes and compares hammering sound signals from multiple struck positions on a single inspection target to determine abnormality based on matching degrees of waveforms around peaks, using linear predictive coding and prominence ratio analysis to calculate band energy ratios and determine the presence of loosening.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a diagnostic robot uses hammering sound diagnosis to inspect a single inspection target, then the robot can achieve efficient automatic diagnosis without relying on pre-existing diagnostic results from other targets, but the robot may erroneously discriminate between normal and abnormal portions due to insufficient force to lift the entire abnormal portion
Solution Approach 1:
The patent divides the inspection target into multiple struck positions and analyzes hammering sound signals from each position separately. By segmenting the diagnosis into multiple measurement points, the system can detect local abnormalities without requiring the robot to lift the entire abnormal portion, thus maintaining both efficiency and accuracy.
Solution Approach 2:
The patent introduces a new dimension of analysis by examining the matching degree of waveforms around peaks in frequency characteristics at different struck positions. This dimensional approach to signal analysis enables accurate discrimination between normal and abnormal portions without requiring additional mechanical force to lift the entire abnormal portion.
2Weight of moving object
If the diagnostic robot is designed to be small and light, then the robot can be more versatile and easier to deploy, but the robot lacks the force to lift the entire abnormal portion where a wedge has become loose
Solution Approach 1:
The patent replaces the mechanical approach of lifting the entire abnormal portion with an acoustic field-based diagnosis system. By using hammering sound signals and analyzing frequency characteristics, the system eliminates the need for mechanical lifting force, allowing lightweight robots to perform effective diagnosis without requiring significant lifting capability.
3Measurement precision
If the diagnostic robot analyzes hammering sound signals from multiple struck positions on a single inspection target, then the robot can accurately diagnose the state of the inspection target without relying on pre-existing diagnostic results from other targets, but the diagnosis process becomes more complex
Solution Approach 1:
The patent enables the inspection target to serve itself by analyzing its own hammering sound characteristics. The system uses the target's self-generated sound signals at multiple struck positions to diagnose its own state, eliminating the need for external reference data from other targets and simplifying the overall diagnostic process while maintaining high accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate and efficient automatic diagnosis of inspection targets without relying on pre-existing diagnostic results from other targets, reducing the risk of misclassification and allowing for smaller, lighter diagnostic robots capable of inspecting a single target.
Implementation Method 1
a sound collection unit 13 capable of collecting hammering sounds generated according to striking of the struck positions
Implementation Method 2
The frequency characteristic conversion unit 22 converts the N hammering sound signals acquired by the hammering sound signal acquisition unit 21 into N frequency characteristics
Implementation Method 3
a low-pitched sound is generated when the air flowing into a gap due to loosening of the surrounding support of the wedge is discharged according to piston vibration and thus a low-pitched sound is further excited at the abnormal portion
Data Source
AI summary
Diagnostic device including a hammering sound signal acquisition unit, a frequency characteristic conversion unit, and an abnormality determination unit. The hammering sound signal acquisition unit acquires an i-th hammering sound signal and a j-th hammering sound signal representing hammering sounds with respect to striking applied to an i-th and a j-th struck position of an inspection target. The frequency characteristic conversion unit respectively converts the i-th hammering sound signal and the j-th hammering sound signal into an i-th frequency characteristic and a j-th frequency characteristic. The abnormality determination unit determines presence or absence of an abnormality in the inspection target on the basis of a matching degree of waveforms around peaks in the i-th frequency characteristic and waveforms around peaks in the j-th frequency characteristic. The diagnostic device can be used with an inspection robot to automatically analyze for loose stator wedges.