Real-Time Optical Fiber-Based Distributed Temperature Monitoring of Insulation Oil-Immersed Commercial Distribution Power Transformer

Badar, Mudabbir; Lü, Ping; Wang, Qirui; Boyer, Thomas; Chen, Kevin P.; Ohodnicki, Paul R.

doi:10.1109/jsen.2020.3024943

Cited by 35 publications

(13 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The calculated spectral shift can be translated into the strain or temperature using the corresponding coefficients of the optical fiber. [33,34] The data generated by the sensors vary in modalities (functional data associated with spectra, time-series data, and spatiotemporal data as shown in Figure 9), requiring the development of a wide variety of ML modeling techniques.…”

Section: Dfos Systemsmentioning

confidence: 99%

Recent Advances in Machine Learning for Fiber Optic Sensor Applications

Venketeswaran

Lalam

Wuenschell

et al. 2021

Advanced Intelligent Systems

Self Cite

View full text Add to dashboard Cite

The latest digital revolution involves the rise of smart devices composed of sensor hardware and artificial intelligence (AI) software for performing intelligent tasks. Smart sensors have become ubiquitous in our lives with varied applications ranging from voice-enabled home devices (Google Home, Alexa, etc.) to the Industrial Internet of Things (IIoT). This revolution has been fueled by 1) miniaturization of sensing hardware, 2) easy access to cloud and high-performance computing, 3) development of big data storage and analytics technologies, and 4) the latest breakthroughs in machine learning (ML) and AI technologies. The emergence of AI since 2012 and its major breakthroughs can be attributed to the research and development (R&D) in deep learning, a subfield of ML that uses biologically inspired neural networks to perform learning tasks. [1] The performance of conventional ML algorithms depends on the individual selection of specific features, while deep neural networks (DNN) automatically generate features as part of the learning process. Deep learningbased AI technologies are increasingly showing performance

show abstract

Section: Dfos Systemsmentioning

confidence: 99%

Recent Advances in Machine Learning for Fiber Optic Sensor Applications

Venketeswaran

Lalam

Wuenschell

et al. 2021

Advanced Intelligent Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…The stable result in this work eliminates the potential error in point sensing with pre-defined locations and highlights the possibility of locating abnormal cracks and temperature hotspots due to cell aging. Rayleigh-based OFDR systems have also been applied to the temperature monitoring of the magnetic core and insulation oil of a power transformer by Lu et al [ 109 ] and Badar et al [ 110 , 111 ]. The above cases have confirmed the unique advantage of immunity to EMI of fiber optic sensors in high power electrical components and demonstrated their distributed sensing capability, which is highly critical to monitor the health of these electrical assets.…”

Section: Operating Principle and Recent Advances In Fiber Optic Sementioning

confidence: 99%

Fiber Optic Sensing Technologies for Battery Management Systems and Energy Storage Applications

Preger

Burroughs

et al. 2021

Sensors

View full text Add to dashboard Cite

Applications of fiber optic sensors to battery monitoring have been increasing due to the growing need of enhanced battery management systems with accurate state estimations. The goal of this review is to discuss the advancements enabling the practical implementation of battery internal parameter measurements including local temperature, strain, pressure, and refractive index for general operation, as well as the external measurements such as temperature gradients and vent gas sensing for thermal runaway imminent detection. A reasonable matching is discussed between fiber optic sensors of different range capabilities with battery systems of three levels of scales, namely electric vehicle and heavy-duty electric truck battery packs, and grid-scale battery systems. The advantages of fiber optic sensors over electrical sensors are discussed, while electrochemical stability issues of fiber-implanted batteries are critically assessed. This review also includes the estimated sensing system costs for typical fiber optic sensors and identifies the high interrogation cost as one of the limitations in their practical deployment into batteries. Finally, future perspectives are considered in the implementation of fiber optics into high-value battery applications such as grid-scale energy storage fault detection and prediction systems.

show abstract

“…Optical fiber-based sensors offer a variety of unique advantages over traditional sensing technology that are relevant to transformer monitoring, including robust operation under harsh chemical conditions, immunity to electromagnetic interference, and compatibility with tools for distributed and quasi-distributed sensing schemes. Optical fiber sensing has been demonstrated for transformer monitoring, primarily focused on detection of thermal and acoustic signatures of fault conditions [3][4][5]. Techniques such as Raman [6] and tunable diode laser absorption spectroscopy (TDLAS) [7] can be used to uniquely identify spectroscopic signatures of relevant gas species, for example, in conjunction with hollow core waveguides or photonic crystal fiber [6,8].…”

Section: Introductionmentioning

confidence: 99%

Multi-parametric gas sensing for transformer monitoring using an optical fiber sensor array

Wuenschell

Kim

Lander

et al. 2023

Optical Waveguide and Laser Sensors II

View full text Add to dashboard Cite

The maintenance and inspection of power transformers can be a time-consuming task for electric utilities, but it is a necessity for maintaining electrical grid reliability. A standard strategy for diagnosis of fault conditions in an oil-filled transformer is to periodically acquire oil samples and perform dissolved gas analysis (DGA). Aging and temperature variation can induce varying concentrations of hydrogen, methane, and other hydrocarbons, which all form as the oil degrades. Acetylene (C2H2) is generated only during localized, high-temperature events such as partial discharge, and its presence is a key marker for identifying these conditions..The development of optical fiber-based sensors to fill the role of DGA offers several advantages, including the option to implement real-time, in-situ, or even spatially resolved (distributed or quasi-distributed) sensing schemes. The evanescent field approach, in conjunction with tailored sensing materials, provides a cheap and scalable solution to this problem. However, this solution is oftentimes hampered by long-term stability and cross-sensitivity issues. One solution is to gather data from multiple optical fiber sensors designed to eliminate cross-sensitivity and calibrate drift. In this work, a multi-sensor array is developed to target multiple gas species relevant to transformer monitoring (C2H2, CH4, H2). This approach, combined with machine learning models such as support vector machines (SVM), can be used to identify the gas species present at concentrations relevant to DGA (ppm levels) with dramatically increased accuracy.

show abstract

Real-Time Optical Fiber-Based Distributed Temperature Monitoring of Insulation Oil-Immersed Commercial Distribution Power Transformer

Cited by 35 publications

References 27 publications

Recent Advances in Machine Learning for Fiber Optic Sensor Applications

Recent Advances in Machine Learning for Fiber Optic Sensor Applications

Fiber Optic Sensing Technologies for Battery Management Systems and Energy Storage Applications

Multi-parametric gas sensing for transformer monitoring using an optical fiber sensor array

Contact Info

Product

Resources

About