It is widely documented that breakdowns of the electrical insulation system of generators can lead to system failures and in more extreme cases, result in catastrophic failure of critical equipment.
Analysing Partial Discharges (PD) occurring in generators is in order to understand the condition off the stator insulation and help to avoid potential failures occurring. Despite the availability of permanent PD continuous monitoring solutions, only a minority of electrical generation companies have integrated them into their operations, with the majority relying on intermittent testing.
Those engaged in intermittent or periodic testing of PD tend to experience mixed results in the analysis of the occurring PD which can lead to a lack of confidence in the results. Sporadic acquisition of PD data is one of the main barriers to widespread deployment of permanent monitoring systems, and the clear understanding of results is another drawback.
Numerous stator failure mechanisms have been found to exhibit a close correlation with PD activity. Some of the most common defects generating PD are:
- Thermal deterioration
- Thermal cycling
- Loose stator bars: due to the vibrations
- Semicon coating
- Semicon/Stress-grading junction
- Inadequate end-winding spacing:
- Contamination
It is now well established that PD trending analysis is an effective method to assess the insulation degradation rate. It is to be expected that an increase in a defect size will typically result in an increase in the intensity of PD activity in respect to amplitude and/or quantity of discharges.
Are you limiting your analysis with offline testing?
PD measurements obtained in an offline test where the machine is not running are not effective and can prove to be inadequate. In offline testing they fail to capture all potential operating conditions which impact PD such as loose-bars defects, deterioration of the voltage stress coatings and the ability to detect discharges between phases.
Enhance the precision of you analysis through online monitoring
Online measurements prove to be the most effective approach for identifying all potential sources of PD, as they involve continuous monitoring of the machine under real operational conditions.
However, the challenge with online tests which are performed periodically is they are limited in their ability to replicate the PD measurement at the same identical operating conditions. This leads to concerns regarding the reliability of the data being compared, PD intensity is known to significantly vary within short time periods, even at normal operation conditions, due to the load and temperature changes.
Example1: High PD activity can occur during generator start-up, due to open gaps in the ground-wall insulation, following the start-up it would be expected that this activity would no longer exist as the coil temperature begins to rise during normal operation.
Example 2: Sudden increases in slot discharges can occur in conditions where there is high load or high temperatures. When the generator is operating at low load condition or at low temperatures, these slot discharges would disappear.
Conclusion
PD trending evaluations need to consider the impact of both the operating condition but also the environmental conditions. The critical point is the comparison of PD activity/intensity at the same, load, voltage, temperature and where applicable humidity conditions.
Asset managers and decision makers need to understand the status of their assets, manage operational conditions across their fleet in real time enabling them to device a real condition-based maintenance plan they can trust.
The INTEGO from Camlin Energy is a unique online monitoring solution, that provides a comprehensive picture of PD events under different operating conditions including load, temperature, humidity and other parameters. INTEGO uses powerful algorithms that can detect subtle anomalies in PD data, giving a clear picture of actual generator condition with lower risk of false alarms.
