Interpreting Key Metrics for Partial Discharge Monitoring in Transformers: Sampling Rate, Number of Channels, and Sensitivity

• sampling rate: Determines the highest frequency range of partial discharge signals that the system can detect. The higher the sampling rate, the more complete the reconstructed pulse waveform will be, and the more accurate the identification of the discharge type will be.
• Number of channels: Determines how many sensors the system can connect to simultaneously. The greater the number of channels, the more comprehensive the coverage in terms of location and dimensions; however, this also increases the system’s complexity and cost.
• Sensitivity/Dynamic Range: Determines how faint a partial discharge signal the system can detect. High sensitivity allows for the detection of early, faint discharges, but may also introduce more environmental noise.
• anti-interference capability: The complexity of the on-site electromagnetic environment directly affects the accuracy of partial discharge monitoring. Effective digital filtering and pattern recognition algorithms are key to reducing the false alarm rate.

1. Key Performance Indicators (KPIs)

2. Sampling Rate—The Foundation of Signal Integrity

The rising edge of a partial discharge pulse typically occurs within nanoseconds; therefore, a sufficiently high sampling rate is required to capture the complete pulse waveform. According to the basic principles of signal processing, the sampling rate must be at least twice the highest frequency of the signal to avoid frequency aliasing. For high-frequency partial discharge signals, this requires the system to have a high sampling capability.

A high sampling rate not only allows for the capture of more complete waveform details, but also enables the system to more accurately distinguish between different partial discharge pulses over time—which is critical in complex scenarios where multiple discharge sources are present simultaneously. Pulse waveforms contain fingerprint information about the type of discharge; the more complete the waveform, the more reliable the type identification.

3. Number of channels—a key factor in coverage

3.1 4-Channel Configuration

The 4-channel configuration is the most common, allowing for the simultaneous connection of four sensors. A typical setup includes two ultrasonic sensors (positioned at different locations on the transformer for positioning) and two high-frequency current sensors (one each for the core grounding wire and the clamp grounding wire). This configuration meets the requirements of most 110 kV main transformers.

3.2 6-Channel Configuration

The 6-channel configuration offers higher monitoring density. By adding UHF sensors or additional ultrasonic sensors to the 4-channel setup, it enables more precise localization and multi-dimensional coverage. This configuration is suitable for large transformers or scenarios requiring high accuracy in partial discharge localization.

3.3 Channel Synchronization

The synchronization accuracy between multiple channels is an often-overlooked yet critical parameter. If there is a microsecond-level synchronization deviation between channels, discharge localization algorithms based on arrival time differences will produce significant errors. When selecting equipment, it is essential to verify that the acquisition clocks of all system channels are synchronized.

4. Dynamic Range and Sensitivity

Dynamic range determines a system’s ability to process both weak and strong signals simultaneously. Partial discharge monitoring deals with a wide range of signal intensities—from faint discharges of a few picocoulombs in the early stages of a fault to intense discharges in the later stages of a fault, often differing by several orders of magnitude. A wide dynamic range ensures that the system does not lose the faint signals present simultaneously due to saturation caused by strong signals.

In terms of sensitivity, high-frequency current sensors typically offer the highest sensitivity and can detect even faint pulse currents on ground wires; the sensitivity of ultrasonic sensors is significantly affected by installation location and propagation path, making them sensitive to discharges at close range but prone to significant signal attenuation at long distances.

5. Frequently Asked Questions FAQ

5.1 Q: Is a higher sampling rate always better? Is there an upper limit?

Answer: From a technical standpoint, a higher sampling rate is generally better; however, in practice, data processing capabilities and storage capacity must be taken into account. An excessively high sampling rate generates a massive amount of data, placing greater demands on communication bandwidth and storage space. As long as the system can capture partial discharge pulse waveforms, a moderate sampling rate is sufficient; there is no need to pursue the absolute maximum.

5.2 Q: Is there a significant difference between 4-channel and 6-channel systems in practical use?

A: For standard-sized 110 kV transformers, 4 channels are typically sufficient. For large transformers rated at 220 kV and above, 6 channels provide more comprehensive monitoring coverage and more accurate discharge localization. If you are unsure, we recommend consulting your supplier to assess channel requirements based on the specific transformer model and size.

5.3 Q: Does higher sensitivity lead to more false positives?

A: That is possible. Excessively high sensitivity can amplify ambient noise and external interference. The solution is to use multi-sensor cross-validation—triggering an alarm only when two or more sensors detect a signal simultaneously, which significantly reduces false alarms caused by the high sensitivity of a single sensor. A good system strikes a balance between sensitivity and interference resistance.

5.4 Q: Why is dynamic range important?

A: Different types of discharges may occur simultaneously inside a transformer—both weak internal discharges and strong corona discharges. If the dynamic range is insufficient, strong signals will saturate the detector, masking the weak signals that are occurring at the same time and leading to missed detections. A wide dynamic range ensures that both strong and weak signals are accurately captured.

5.5 Q: Why is there a discrepancy between the specifications in the parameter table and actual on-site performance?

A: The specifications listed in the parameter table are typically measured under ideal laboratory conditions. Factors such as electromagnetic interference, temperature fluctuations, and installation constraints in real-world environments can affect actual performance to varying degrees. When evaluating system performance, it is important to look beyond the specifications and focus on actual case studies from similar scenarios.

6. Summary of parametric evaluation

6.1 The sampling rate ensures that the entire waveform is captured; for high-frequency discharge scenarios, systems with a high sampling rate are preferred.

6.2 The number of channels corresponds to the number of sensors and the monitoring range; 4 channels meet mainstream needs.

6.3 Dynamic range and sensitivity together determine the ability to detect weak signals; a wide dynamic range is more valuable than high sensitivity alone.

Disclaimer: The content of this article is for technical exchanges and reference only, and does not constitute any form of procurement commitment or contract offer. Product technical parameters, configuration programs and prices are subject to the actual signed contracts and technical agreements. The technical data and cases involved in this article are from public information and engineering practice, if updated without notice.

Need help evaluating parameters or selecting a partial discharge monitoring system for transformers? Contact Innotongda for professional solutions. Service Hotline: 13959168359 (also on WeChat).