Partial Discharge Recognition and Localization Method of High Voltage Power Supply for Insulation Material Testing
Partial discharge testing is a fundamental diagnostic technique for assessing the quality and integrity of insulation materials. These small electrical discharges occur at localized defects within the insulation and are precursors to eventual failure. The high voltage power supply used for testing must provide clean, stable voltage while enabling detection and analysis of partial discharge signals. Recognition and localization of partial discharges provide valuable information about the nature and location of defects.
Insulation materials in high voltage equipment are subjected to intense electric fields during operation. Manufacturing defects such as voids, inclusions, or delaminations create regions of enhanced electric field where partial discharges can initiate. Over time, the discharges erode the insulation, eventually leading to complete breakdown. Early detection of partial discharges enables intervention before failure.
The high voltage power supply for partial discharge testing must provide extremely low noise output. Any noise or ripple on the supply can mask the small partial discharge signals or be mistaken for discharges. The supply must also provide stable voltage to maintain consistent test conditions. Voltage fluctuations can cause variations in partial discharge activity that complicate interpretation.
Partial discharge detection circuits measure the electrical signals produced by the discharges. Each discharge creates a small current pulse that can be detected by a coupling circuit. The coupling capacitor provides a low impedance path for the high frequency discharge signals while blocking the power frequency voltage. The measurement impedance converts the current pulse to a voltage signal for analysis.
Recognition of partial discharge types enables diagnosis of the defect mechanism. Different types of defects produce characteristic discharge patterns. Internal voids produce discharges that occur around the peaks of the applied voltage, with similar patterns on positive and negative half cycles. Surface discharges occur preferentially on one polarity. Corona from sharp points produces a distinctive pattern with discharges on one half cycle only.
Phase resolved partial discharge analysis displays the discharge activity as a function of the phase of the applied voltage. The phase pattern provides information about the discharge type. Statistical parameters extracted from the pattern, such as the skewness and kurtosis, can be used for automated recognition. Pattern recognition algorithms can classify the discharge type based on these parameters.
Pulse sequence analysis examines the sequence of discharge pulses rather than just the phase pattern. The time intervals between successive pulses contain information about the discharge physics. Different discharge mechanisms produce different pulse sequences. This analysis can distinguish between multiple discharge sources that might be present simultaneously.
Localization of partial discharges determines the position of the discharge source within the test object. For simple geometries, the position can be determined from the apparent charge magnitude and the electric field distribution. For complex geometries, more sophisticated techniques are required.
Time of flight localization uses the propagation time of discharge pulses through the test object. Sensors at multiple positions detect the arrival times of the pulses. The differences in arrival times determine the distances to the source. Triangulation from multiple sensors locates the source in three dimensions. This technique requires knowledge of the propagation velocity in the test object.
Acoustic localization detects the sound waves produced by the discharges. Each discharge creates a small acoustic emission that propagates through the material. Acoustic sensors at multiple positions detect the arrival times. The time differences enable localization similar to time of flight electrical methods. Acoustic localization is particularly useful for locating discharges in large equipment such as transformers.
Ultra high frequency detection captures the electromagnetic waves radiated by partial discharges. The discharges produce electromagnetic energy in the ultra high frequency range, hundreds of megahertz to gigahertz. UHF sensors detect this energy, and the signal strength at multiple sensors enables localization. UHF detection is less susceptible to external noise than conventional electrical detection.
Calibration of the partial discharge measurement system establishes the relationship between the measured signal and the actual discharge magnitude. A calibrator injects a known charge into the measurement circuit, and the system response is recorded. The calibration enables reporting of discharge magnitudes in picocoulombs, the standard unit for partial discharge measurement. Regular calibration ensures the accuracy of the measurements over time.
