AI Current Waveform Diagnosis for Abnormal Part Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predictive maintenance methods for devices struggle to accurately determine the specific abnormal part causing a symptom, requiring expert intervention and resulting in inefficient and unreliable diagnosis and repair processes.
Innovation Solution
A method utilizing artificial intelligence that collects waveform data from devices before faults occur, constructs reference models via deep learning, and compares real-time waveforms to detect abnormal parts, providing probability values for rapid and accurate identification of issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If expert intervention is used to diagnose abnormal parts, then diagnostic reliability is improved, but diagnostic time and operational complexity increase
Solution Approach 1:
The system enables self-diagnosis by automatically comparing real-time waveform data with reference models to identify abnormal parts, eliminating the need for expert intervention while maintaining high diagnostic reliability and reducing diagnostic time
Solution Approach 2:
The patent replaces the mechanical expert diagnosis process with an automated AI-based waveform analysis system that uses deep learning models to detect and diagnose abnormal parts, thereby reducing both time and operational complexity
2Measurement precision
If expert intervention is used to diagnose abnormal parts, then diagnostic accuracy is improved, but device complexity and operational requirements increase
Solution Approach 1:
The system performs automated diagnosis by comparing real-time waveforms with pre-stored reference models, achieving high diagnostic accuracy without requiring expert operators or complex manual diagnostic procedures
Solution Approach 2:
Reference models for various abnormal parts are pre-collected and stored during normal operation, enabling rapid and accurate comparison and diagnosis when abnormalities occur, thereby maintaining high accuracy while simplifying the diagnostic process
3Reliability
If conventional predictive maintenance methods are used, then abnormal symptoms are detected, but specific abnormal part identification is insufficient
Solution Approach 1:
The patent segments the diagnosis process into waveform data collection, reference model construction for specific parts, and comparative analysis, enabling identification of specific abnormal parts rather than just detecting general abnormal symptoms
Solution Approach 2:
The patent introduces waveform data as an intermediary that carries information about the state of specific device parts, allowing the system to detect both abnormal symptoms and their specific locations by comparing real-time waveforms with part-specific reference models
Data Source
AI summary
A method for learning and detecting an abnormal part of a device through artificial intelligence comprises: an information collection step for collecting a current waveform of a current value that changes over time in a driving state of at least one device and collecting information about a faulty part of the device, together with current waveform information before a fault occurs in the device; a model setting step for learning, by a control unit, information collected in the information collection step and setting a reference model of a current waveform for each faulty part of the device; and a detection step for, when an abnormal symptom of the device is detected in a real-time driving state, comparing, by the control unit, a real-time current waveform of the device and the reference model, and detecting and providing an abnormal part regarding the abnormal symptom of the device.