Vibration Profile Evaluation for Ground Fault Reignition Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to efficiently and reliably detect anomalous events, such as ground faults and reignitions, in power supply networks with resonant neutral grounding, due to the challenges of identifying precise starting points and times of these events, which can lead to recurring faults and potential safety risks.
Innovation Solution
A method involving defining a time range with peak values, creating an envelope curve, comparing it to a theoretical vibration profile, and evaluating deviations to determine the anomalous event, using a data processing device to analyze electrical or mechanical vibration patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional vibration pattern evaluation methods are used, then the evaluation process is simple, but the precision of detecting anomalous events and determining their starting points is insufficient
Solution Approach 1:
The vibration profile is segmented into multiple peak value ranges, with each range containing at least three peak values. This segmentation allows for localized analysis of specific event windows, improving detection precision by focusing computational resources on relevant portions of the signal rather than analyzing the entire vibration profile at once.
Solution Approach 2:
The invention introduces a new evaluation dimension by comparing actual peak values against theoretically calculated peak values derived from envelope curves. This theoretical model provides an additional reference framework, enabling more precise anomaly detection by identifying deviations from expected vibration patterns rather than relying solely on absolute threshold comparisons.
2Measurement precision
If the evaluation method uses multiple peak values and theoretical comparisons, then the precision of anomaly detection improves, but the computational complexity and processing time increase
Solution Approach 1:
Envelope curves are pre-calculated based on the vibration profile before the actual anomaly detection process. This preliminary action creates a theoretical reference framework in advance, so that during actual evaluation, only straightforward comparisons between measured and theoretical peak values are needed, significantly reducing real-time computational burden while maintaining high precision.
3Reliability
If the system evaluates every peak value in detail, then the reliability of anomaly detection increases, but the productivity and efficiency of the evaluation process decreases
Solution Approach 1:
The evaluation applies different analysis depths to different portions of the vibration profile. By dividing the profile into peak value ranges and focusing detailed theoretical comparison only on ranges containing anomalies, the system maintains high reliability where needed while preserving overall evaluation efficiency. Normal portions of the signal require less computational scrutiny.
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
A method for evaluating an anomalous event (150) with respect to an oscillation profile (100) is described, comprising: i) defining a time domain of the oscillation profile (100) which has a peak value range (110) including: ia) a first peak value (111), ib) a second peak value (112) which follows the first peak value (111) in time and is decreasing compared to the first peak value, and ic) a third peak value (115) which follows the second peak value (112) in time and is increasing compared to the second peak value (112); ii) defining an envelope curve (120) with respect to the peak value range (110); iii) Determining a theoretical vibration profile (140) with respect to the third peak value (115) based on the envelope curve (120);iv) Comparing the theoretical vibration profile (140) with the actual vibration profile (130) with respect to the third peak value (115) to determine any deviation; and v) Evaluating the anomalous event (150) with respect to the vibration profile (100) based on the determined deviation.;