Wind farm operators are facing unprecedented times. COVID-19 related lockdowns and supply chain disruptions have placed increased pressure on maintenance organizations across the industry to keep their wind turbines up and running to meet minimum production levels. Combined with dwindling PTCs and a growing number of service warranty contracts now coming to term, for many, asset maintenance is now taking center stage on the quarterly P&L. Visibility into asset health has never been more vital.
Adept managers realize it is no longer a decision of whether to implement predictive analytics, but rather, when. Having a systematic approach to reducing downtime and associated labor costs may ultimately be the determining factor of who thrives (or survives) in this new era.
To adapt to this new level of challenges, it is critical to implement an effective remote health monitoring system that is not only easy to deploy, but also has a track record of proven results.
Taking the guess work out of condition monitoring
Fischer Block, Inc., has created new asset health monitoring technology which combines advancements in electrical signature analysis (built over decades) with IoT, Machine Learning, and Big Data statistics as the backbone of their solution. Conforming to the strictest of industry standards (IEEE-519, IEC6100-4-7/30, and others), they pride themselves on being able to detect even the most subtle clue of a component health problem, deep within the system, and many months before actual failure.
This technology consists of SMART Block® intelligent sensors (one per turbine) combined with the wave iQ™ predictive analytics platform, providing new insights into the health of critical wind turbine system components.
Providing unprecedented visibility
The ability to detect and act on early detections of failure, before any impact on asset life, is the ultimate goal in a world-class preventative maintenance program. This was the guiding principle for Fischer Block, Inc. led by President & CEO, Greg Wolfe, as their team embarked on their groundbreaking technology. “Our team has spent many, many long nights and now, five patents later, we have produced technology which can extract even the most infinitesimal clue that something is beginning to go wrong, deep inside the drive-train system, and well in advance of actual failure”.
However, identifying a problem is only a portion of the challenge: incorporating this data into a platform with in-depth signature analysis, utilizing consistent and sound statistical and machine learning techniques, and actionable alerts are requisite in achieving efficient maintenance activity by minimizing both false positive and negative indications.
SMART Block® sensors and wave iQ™ provide a comprehensive solution in achieving such results.
How it works
For wind turbines, each element of the drive train has a specific purpose in helping the generator rotate in a purely uniform fashion.
As any particular element begins to degrade in health, no matter how small, its operating frequency begins to show up in the output signal of the generator, in the form of a slight anomaly to the overall composite signal (i.e., a distortion). Using proven frequency extraction techniques (mathematically based), these sub-frequency elements can be isolated and monitored over time, for changes in severity. Implementing IoT and Big Data technologies, comparisons are made in real-time, against historical norms of large populations of wind turbines, helping to further reduce the chances of false positive or negative indications.
With our patented IoT SMART Block® device and wave iQ™ predictive analytics engine, we are helping our customers achieve new levels in wind turbine availability, helping to avoid costly failures and extend the life of these critical assets
Significant benefits over traditional technologies
Vibration sensors: mechanical vibration sensors are designed to detect solely mechanical related problems. However, the generator’s electrical output signal contains health information for both mechanical and electrical system components, thus many critical component problems (such as rotor bar and wye ring problems) often can go undetected by vibration sensors.
Periodic/portable testing: Component signal distortion (within the generator output signal) has dependency on turbine operating conditions (such as rotor speed, load level, etc.). Therefore, repetitive scans (at each varying condition) are required for accurate baselining and ongoing component health assessments, far beyond what is viable with periodic testing.
Easy to deploy
Fischer Block, Inc. offers a suite of SMART Block® sensors which detect microscopic anomalies that cannot otherwise be detected through traditional monitoring systems.
Just one SMART Block® is all that is needed per turbine to provide high resolution monitoring the generator output signal; a very simple, nonintrusive installation, either up or down tower.
Sample of results from actual case studies:
• Spectral energy observed at increasing multiples of generator speed created alert indicating shaft alignment problems
• Side-bands on either side of fundamental frequency were detected, revealing rotor bar wye-ring degradation
• Excessive energy at Main Bearing Race Frequency detected, exposing a pre-fault condition.
• Pad mount transformer detected in saturation, creating a sixth harmonic pulsating torque, detrimental to drive train life, preventing potential catastrophic failure