Generator Neutral Ground Monitoring via Waveform Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current protection techniques for electrical power generators rely solely on over-voltage trip relays, which may miss other fault events or impending faults, necessitating a more robust monitoring system to accurately detect and predict generator faults.
Innovation Solution
A system that monitors the voltage waveform across a neutral grounding transformer, compares present samples to stored historical samples, and generates diagnostic messages or control signals to identify and predict generator faults, including the component involved, using a voltage monitoring circuit, signal receiver, comparator, and fault analyzer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a simple over-voltage trip relay is used to monitor the neutral grounding circuit, then the device complexity is low and ease of operation is maintained, but the reliability of fault detection is insufficient and measurement precision is limited
Solution Approach 1:
The system performs preliminary action by continuously monitoring and storing historical voltage waveform data before faults occur. The waveform storage module saves normal operating waveforms as reference data, enabling early detection of abnormal patterns that precede actual faults, thus improving reliability without requiring complex real-time analysis hardware
Solution Approach 2:
The invention uses copying by creating digital replicas of voltage waveforms through analog-to-digital conversion. The waveform storage module stores copies of normal operating waveforms for comparison, and the fault detection module compares current waveform copies against historical copies to identify faults, enhancing detection reliability while keeping the physical system relatively simple
2Loss of information
If a simple over-voltage trip relay is used, then the device complexity is low, but the loss of information about fault patterns and predictive capabilities occurs
Solution Approach 1:
The waveform storage module performs preliminary action by continuously storing voltage waveform data during normal operation. This creates a historical database of normal patterns that can be compared against future abnormal patterns, preserving fault pattern information without requiring complex processing during actual fault events
Solution Approach 2:
The system uses copying by converting analog voltage waveforms into digital copies for storage and analysis. The analog-to-digital converter creates precise digital replicas of the continuous voltage signal, preserving all waveform information including subtle patterns that would indicate developing faults, thereby minimizing information loss
3Reliability
If a short time delay is used in the protection relay to prevent false trips, then the reliability of continuous operation is improved, but the speed of fault response is reduced
Solution Approach 1:
The system performs preliminary action by continuously monitoring and storing waveform data before faults occur. When a fault happens, the pre-stored normal waveforms are immediately available for comparison, eliminating the need for time delays to accumulate data. This enables instant fault detection while maintaining operational reliability through accurate pattern recognition
Solution Approach 2:
The fault detection module uses feedback by continuously comparing current voltage waveforms against stored historical waveforms. This real-time comparison provides immediate feedback when deviations occur, enabling fast fault response without time delays while maintaining system reliability through consistent pattern matching against known normal operation
Data Source
AI summary
Fault analysis and detection involving a generator neutral ground includes monitoring a voltage waveform across a neutral grounding transformer of a generator. During monitoring of the voltage waveform a sample of the voltage waveform is acquired. Previously, a number of similar past samples of the monitored waveform were acquired and stored as well. The present sample is compared to one or more of these stored, past samples so that, based on a similarity between the present sample to the past samples, a generator fault indicated by the present sample can be identified. Identifying the generator fault can include recognizing that it is presently occurring or that it may occur in a predictable timeframe. Additionally, identifying a component of the generator that caused, or will cause, the generator fault can be identified as well. A diagnostic message or control signal based on the identified generator fault can then be generated.