Sub-cycle Fault Detection Using Transient Waveform Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fault detection methods in electrical networks are unable to locate sub-cycle faults, which are transient and often precursors to permanent faults, as they require a steady-state condition that sub-cycle faults do not reach, leading to missed detection and potential loss of electric service.
Innovation Solution
The approach calculates the source impedance and line inductance under fault conditions using time domain differential equations, allowing for the determination of fault distance and type without prior knowledge of the circuit's impedance, and can differentiate between upstream and downstream faults, accommodating various capacitor bank configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fault detection methods using reactance algorithms are used, then fault location can be determined for steady-state faults, but sub-cycle transient faults cannot be detected because they do not reach steady state
Solution Approach 1:
The patent changes the detection parameter from steady-state voltage and current magnitudes to transient parameters including rate of change of current (di/dt), second derivative of current (d²i/dt²), and phase angle differences during the transient period. This allows fault detection to occur during the transient phase before steady state is reached, enabling detection of sub-cycle faults that conventional methods miss.
Solution Approach 2:
The system performs preliminary detection and classification of faults during the transient period immediately after fault occurrence, before the system reaches steady state. By calculating di/dt and d²i/dt² in real-time during the transient phase, the system can identify and locate faults immediately, preventing the need to wait for steady-state conditions that sub-cycle faults never achieve.
2Measurement precision
If steady-state voltage and current data are used for fault location, then accurate fault distance can be calculated, but sub-cycle faults lasting less than one cycle cannot be located
Solution Approach 1:
The patent transforms the measurement approach by using transient parameters (di/dt, d²i/dt², and phase angle differences) instead of steady-state magnitudes. This allows accurate fault distance calculation to be achieved during the transient period itself, making the measurement precision independent of fault duration and enabling accurate location of sub-cycle faults.
Solution Approach 2:
The patent replaces the conventional mechanical approach of waiting for steady-state electrical conditions with a mathematical transformation approach using differential equations and phase angle analysis. This substitution allows fault location to be determined from transient waveforms through mathematical processing rather than requiring steady-state electrical parameters.
3Reliability
If conventional fault location methods are used, then permanent faults can be detected, but intermittent and transitory faults are ignored and lead to loss of electric service
Solution Approach 1:
The system continuously monitors transient parameters (di/dt, d²i/dt², phase angles) and provides immediate feedback when a fault is detected during the transient period. This real-time feedback mechanism allows the system to identify intermittent and transitory faults as they occur, enabling preventive action before they develop into permanent faults that would cause service loss.
Solution Approach 2:
The system performs preliminary detection and classification of all fault types during the transient period, including intermittent and transitory faults that conventional methods ignore. By identifying these faults early in their development phase, the system can trigger preventive maintenance or isolation actions before the faults progress to permanent conditions causing service disruption.
Data Source
Figure 1~2
Figure 3~4
Figure 5A~6
AI summary
An electrical waveform is received over an electrical power line. A plurality of nominal electrical parameters are determined for the electrical power network and the plurality of nominal electrical parameters are associated with a state of the electrical power network in the absence of at least one transitory electrical fault in the network. Subsequently, a plurality of electrical parameters of the electrical waveform are sampled when the at least one transitory electrical fault exists in the network. A plurality of inductances are determined based at least in part upon a comparison of the nominal electrical parameters and the plurality of sampled electrical parameters. The plurality of inductances are representative of inductances present in the network when the at least one transitory electrical fault exists in the network. The plurality of inductances are analyzed to determine a distance and/or direction to the at least one electrical fault.