Partial discharge (PD) is a maintenance challenge found in high-voltage equipment around the world, especially in older and aging infrastructure. Predictive maintenance workers are now beginning to use acoustic imaging to detect PDs by finding their distinct sound signature even before the equipment overheats. When used in conjunction with FLIR thermal imaging cameras, acoustic imaging cameras such as the FLIR Si124 are indispensable for effectively finding partial discharges before they cause equipment failure, costly damage, and unexpected downtime.
The electric current is always looking to escape when no one is looking, jumping out of its conductor and trying in vain to cross to a nearby electrode. In search of an escape route, it starts with a crack in a worn insulator. Or start on the surface of an overhead line insulator, dirty with years of contamination. Maybe it punches a small hole in the paper windings of high-voltage cables. Or it hides near a gas bubble that formed in aging liquid dielectric. It is relentless, trying again and again with every high and low peak of the voltage sine wave.
Utilities can reduce inspection time by up to 90% without extensive training with the FLIR Si124.
This type of partial discharge or PD remains hidden from view as the current attempts to cross to a neighboring conductor day after day. At some point, having deteriorated from the constant stress of high voltage, the nearby insulating material will fail and give way.
Eventually, the current breaks the division to another conductor and when that happens, the conductor will fail completely. This wreaks costly and destructive havoc on electrical equipment, switchgear, machinery or downstream facilities. Partial discharges can damage plant equipment or fry sensitive electronic components. Worse, partial discharges can cut off power supply to a community during idle plant hours or shifts, costing valuable productivity.
IEC 60270 more formally describes a PD as a "localized electrical discharge that only partially bonds the insulation between conductors and may or may not occur adjacent to a conductor". In general, a partial discharge is the consequence of localized electrical stress concentrations in the insulation or on the surface of the insulation and generally appear as pulses that have a duration of much less than 1 μs.
PARTIAL DISCHARGE DIAGNOSTICS: ESSENTIAL FOR PREDICTIVE MAINTENANCE
PD detection is an essential part of an effective condition-based monitoring (CBM) or predictive maintenance (PdM) program. The earlier they are detected, the less damage partial discharge can cause to insulators and the lower the risk of equipment failure and subsequent downtime.
The financial incentive to track partial discharges is simple: it is much less costly and disruptive to locate one, plan for scheduled downtime, and then repair and replace insulators and electrical connections at the PD site.
TOOLS FOR SUCCESS
To accurately locate a PD, there are numerous diagnostic technologies available to electrical contractors, inspectors and maintenance professionals. Insulation test meters provide numerical readings of how effective or resistive an insulator is. FLIR's thermal imaging cameras locate and identify resistive heat generated in electrical equipment, pinpointing it in a visual image with pixel-by-pixel temperature readings. Thermal images can be used in conjunction with acoustic images to determine the severity of the partial discharge. An increase in temperature along with an acoustic signature could indicate that the integrity of the insulating equipment is compromised.
LOCATE PARTIAL DOWNLOADS INSTANTLY
As part of a complete diagnostic ecosystem, FLIR also complements infrared thermal imaging diagnostics with acoustic imaging capabilities. Acoustic imaging cameras, such as FLIR Si124 offer advanced sound-based solutions for locating and analyzing industrial faults, deterioration and defects such as PD. It has been found that anomalies occur in the sound generated by PD before components begin to heat up and become visible to thermal imaging cameras. This provides an additional layer of early notification to detect potential future failures.
And while it is not unusual to hear humming near power lines, PDs are often inaudible to the human ear, making them particularly difficult to locate, especially in noisy workplaces with excessive background noise. With a portable acoustic imaging camera, much like a thermal imager, a user can scan an area and see the location of the ultrasonic sounds generated by PDs in a digital image of the components being inspected, either inaudible or hidden by background sound.
ONE-HANDED PORTABLE OPERATION
While there are numerous tools available for electricians to perform acoustic imaging, there are notable considerations to keep in mind, from portability to accuracy.
First, although most acoustic imaging tools are portable, select one that is easy to carry from place to place. Consider an acoustic imaging camera that is portable, ready to use, and easy enough to hold in one hand, making it useful for transportation, ergonomics, and improved aiming.
This illustration shows two signals of two sounds that can be lost if the sensitivity of the acoustic chamber is not good enough. The 16.5 kHz sound signal can be detected with a 32 microphone system and the 18.5 kHz sound signal with a 124 microphone system.
MORE MICROPHONES, BETTER RESULTS
The range of acoustic imaging tools available also reveals a wide mix in the number of microphones used to develop acoustic images. As a general rule of thumb in technology, more is better, so it goes without saying that employing more microphones is essential for creating acoustic images rich in detail. As with technology again, bigger is not always better when it comes to microphones. Look specifically for MEMS (microelectromechanical systems) type microphones. These can offer users a good balance of performance, stability in different environments, low power consumption for smaller batteries, and longer run time. In addition, the small size of the microphones means they are easier to organize compactly in a handheld tool.
Sensitivity: When examining the FLIR Si124 acoustic imaging camera, there is a carefully arranged array of 124 MEMS microphones that, working together, offer the highest level of sensitivity. The larger number of microphones also reduces the potential for "spatial aliasing," which is the improper positioning of the sound source in the image.
An acoustic imager such as the FLIR Si124 can help utilities analyze partial electrical discharge patterns, prioritize repairs with automated leak cost estimates and discharge type classification, and perform non-contact inspections quickly and safely.
Detection range and access: another advantage guaranteed by a large number of microphones is an extended detection range. Note that sound traveling through the air attenuates by 6 decibels each time the distance traveled is doubled. An average sized partial discharge might measure 40 dB(Z). The sound heard at 15 meters (approx. 50 feet) away from the source is 6 dB louder than at 30 meters (approx. 100 feet), and so on. To compensate for this, acoustic imaging camera manufacturers increase the number of microphones to increase the detection range. For FLIR, the result is a maximum detection range that is approximately doubled by using four times the number of microphones.
Many electrical components are difficult to access because they are fenced for security or are high off the ground. Access restrictions may also be time-based, limited to when a customer contact is on site to allow entry. Given these access barriers, it is critical to use tools that can find PD accurately, even from a distance. For example, the FLIR Si124 can be used to inspect overhead cables from the ground, as well as substation components that are secured behind fences, at a distance of up to 130 meters (426 feet).
FLIR Si124 can be used to inspect overhead cables from the ground, as well as substation components that are secured behind fences, at a distance of up to 130 meters (426 feet).
Processing power: FLIR Si124 produces 124 audio data streams that are processed and converted to a visual display. The camera has automatic audio frequency selection that simplifies use without sacrificing performance. Advances in data processing power and graphics have made it possible to integrate such large amounts of acoustic data instantly into an easy-to-understand image on the screen.
Users who compromise with cameras that have fewer microphones and/or older processors may end up with lower quality images, lower resolution, and potentially a slower refresh rate. In terms of productivity, a state-of-the-art camera like the FLIR Si124 can find problems up to 10 times faster compared to other available tools.
MICROPHONE FREQUENCIES CAN AFFECT THE INSPECTIONS
Inspection tools used by electrical contractors can themselves promote misconceptions about the best way to identify partial discharges. For example, PD constantly emits ultrasonic sound at a common frequency (40 kHz). Many acoustic imaging devices use or recommend almost exclusively this frequency. While that may be useful in some cases, in many others it can significantly affect detection sensitivity. Using a wider range of frequencies, from 10 kHz to 30 kHz, can generate better results when working at a distance, such as in an outdoor substation.
Intelligent noise cancellation: Partial discharges generate broadband sound, ranging from audible to inaudible or ultrasonic frequencies. In addition, inspections are rarely performed in quiet locations. Instead, the devices must deal with background noise from industrial facilities or outdoor sites, near the highway or aviation noise, for example. Smarter acoustic imaging cameras can make sense of interference and background noise, filtering them out to isolate the PD culprit.
PUTTING THE IA AND THE CLOUD TO WORK IN PARTIAL DISCHARGE DIAGNOSTICS
Categorizing PD is often a challenge. FLIR helps electrical contractors by applying AI algorithms to analyze partial discharges. A user can upload acoustic images to the FLIR Acoustic Camera Viewer cloud service and the image is automatically compared to thousands of images. The cloud service classifies the partial discharges found into three main categories: surface, floating and airborne discharges.
Reliance on an advanced AI service can help reduce errors, speed report preparation and serve as a key differentiator for inspection clients. The added ease of use also helps onboard more workers to perform acoustic imaging inspections as part of condition-based monitoring or predictive maintenance.
DECIDING ON THE RIGHT ACOUSTIC IMAGING TOOL
Acoustic imaging has quickly become a must-have technology for keeping power infrastructure up and running. More CBM managers are adding cameras like the FLIR Si124 to their toolbox. The return on investment comes quickly, as they detect problems quickly and easily while reducing repair costs and unplanned downtime.
PARTIAL DISCHARGE CLASSIFICATION CAPABILITIES TO BE CONSIDERED
Acoustic image analysis can require some training and learning, especially when it comes to understanding the different types of PD that have been located. Knowing what problems are evident and the severity can help formulate better reports, better repair recommendations and smarter next actions.
There are several types of PD, depending on where the discharge occurs and its pulse pattern. Surface discharges occur at the boundary of different insulation materials. Surface discharges can be found at various locations, including bushings, cable terminations or generator windings that overheat.
Example of the Partial Discharge pattern of a surface discharge.
Example of the Partial Discharge pattern of a floating discharge.
Example of the Partial Discharge pattern of negative and positive corona discharge. The positive corona is seen on the left side and the negative corona on the right side.
Floating discharges can occur when there is a floating conductor inside high voltage equipment, separated by a spacer, for example. Floating discharges are considered the most frequent form of partial discharges.
Finally, partial discharge into the air occurs when the air around a conductor, such as a power transmission line, which serves as an insulating material, may lose some of its insulating properties due to high humidity and/or contamination. This allows discharges into the air that further degrade the immediate quality of the air and conductor.
Knowing the type and severity of the discharge allows the facility to choose appropriate remediation measures and schedule maintenance to minimize failures and downtime.
Equipment areas to target for acoustic imaging of partial discharges
- Conductors and busbars
- Electric generators
- Transmission and distribution of electric power (T&D)
- Electrical substations
- Stators, motors and coils
- Control substation
- Transformers
Programs using partial discharge acoustic imaging
- Condition-based maintenance programs (CBM)
- Condition monitoring (CM) programs
- Predictive maintenance (PdM)