Eran holds a Ph.D. in Lasers (Physical Electronics) from Tel-Aviv University with many patents relating to novel laser design.
He was the CEO and founder (2001-2016) of V-Gen: among the early and dominant players in the Fiber-Laser field. V-Gen was acquired by Newport/Spectra-Physics in 2014.
Since 2017 he is the founder and CEO of Prisma-Photonics: focusing on Smart Physical Infrastructure based on novel Fiber Sensing technology. Over the recent years Prisma Photonics won multiple leading industry awards and successfully demonstrated and deployed its unique solutions in the Pipeline, Power, Communications and other infrastructure fields.
Virtual Pipeline Summit 7 Oct
Novel classification of leak detection and third-party intrusion enabling best-in-class false alarm rate
CEO & Founder, Prisma Photonics, Israel
In recent years there has been a growing interest in distributed fiber-optic sensing for monitoring oil and gas pipelines. Pipeline leakage typically generates both pipeline mechanical vibrations and acoustic signals, originate from the interaction between the leaking substance and the soil surrounding the pipeline, that can be sensed by an adjacent optical fiber. Digging activity, that might threaten the pipeline, also creates strong acoustic signals in the ground, that are readily detected by optical interrogators.
However, in a real-life scenario, there are many other sources of these types of signals along the pipeline, such as traffic, farming, and construction as a few examples. The signals from these sources, that are also detected by the optical interrogator, must be filtered out by the system and not lead to alerts.
In fact, the high False Alarm Rate (FAR) or Nuisance Alarm Rate (NAR) of fiber-optic sensing systems is the limiting factor of the existing technology and is mainly originating from insufficient classification capabilities.
Prisma Photonics pipeline monitoring system, PrismaFlow™, uses its novel Hyper-Scan Fiber-Sensing™, an ultra-sensitive sensing capability to generate data with a very high signal to noise ratio (SNR), even when using pre-deployed fiber-optic cables. Advanced machine learning capabilities harnessed together with the high data SNR enables extraction of a large set of features that assist, for example, in differentiating background noises from digging activities that endanger the pipeline. These features enable the construction of novel real-time algorithms for classification of different activities, leading to a dramatic decrease in the false and nuisance alarm rate.
We demonstrate PrismaFlow™’s high events detection rate, accompanied by a negligible false alarm rate in a 30 km long pre-deployed optical cable inside a conduit, running along infrastructure placed along highly noisy environment such as highways, rural roads, railway, and through farm fields.