ABSTRACT
An aircraft is a highly intricate system that features numerous subsystems, assemblies, and individual components for which regular maintenance is inevitable. The operational efficiency of an aircraft can be maximised, and its maintenance needs can be reduced using an effective yet automatic AI-based health monitoring systems which are more efficient as compared to designing and constructing expensive and harder to operate engine testbeds. It has been observed that aircraft engine anomalies such as undergoing flameouts can occur due to the rapid change in the temperature of the engine. Engine oil temperature and cylinder head temperature, two measures connected to this issue, might be affected differently depending on flight modes and operational conditions which in turn hamper AI-based algorithms to yield accurate prediction on engine failures. In general, previous studies lack comprehensive analysis on anomaly prediction in piston engine aircraft using modern machine learning solutions. Furthermore, abrupt variation in aircraft sensors' data and noise result in either overfitting or unfavourable performance by such techniques. This work aims at studying conventional machine learning and deep learning models to foretell the possibility of engine flameout using engine oil and cylinder head temperatures of a widely used Textron Lycoming IO-540 six-cylinder piston engine. This is achieved through pre-processing the data extracted from the aircraft's real-time flight data recorder followed by prediction using specially designed multi-modal regularised Long Short-Term Memory network to enhance generalisation and avoid overfitting on highly variable data. The proposed architecture yields improved results with root mean square error of 0.55 and 3.20 on cylinder head and engine oil temperatures respectively averaged over three case studies of five different flights. These scores are significantly better i.e., up to 84% as compared to other popular machine learning predictive approaches including Random Forest, Decision Tree Regression, Artificial Neural Networks and vanilla Long Short-Term Memory networks. Through performance evaluation, it can be established that the proposed system is capable of predicting engine flameout 2 minutes ahead and is suitable for integration with the software system of aircraft's engine control unit.
ABSTRACT
Recent advancements in power electronic switches provide effective control and operational stability of power grid systems. Junction temperature is a crucial parameter of power-switching semiconductor devices, which needs monitoring to facilitate reliable operation and thermal control of power electronics circuits and ensure reliable performance. Over the years, various junction temperature measurement techniques have been developed, engaging both non-optical and optical-based methods, highlighting their advancements and challenges. This review focuses on several optical sensing-based junction temperature measuring techniques used for power-switching devices such as metal-oxide-semiconductor field-effect transistors (MOSFETs) and insulated-gate bipolar transistors (IGBTs). A comprehensive summary of recent developments in infrared camera (IRC), thermal sensitive optical parameter (TSOP), and fiber Bragg grating (FBG) temperature sensing techniques is provided, shedding light on their merits and challenges while providing a few possible future solutions. In addition, calibration methods and remedies for obtaining accurate measurements are discussed, thus providing better insight and directions for future research.
ABSTRACT
Broadband amplification in the O+E-band is very desirable nowadays as a way of coping with increasing bandwidth demands. The main issue with doped fiber amplifiers working in this band such as the bismuth-doped fiber amplifier is that they are costly and not widely available. Therefore, a wideband and flat-gain hybrid optical amplifier (HOA) covering the O+E-band based on a parallel combination of a praseodymium-doped fiber amplifier (PDFA) and a semiconductor optical amplifier (SOA) is proposed and demonstrated through numerical simulations. The praseodymium-doped fiber (PDF) core is pumped using a laser diode with a power of 500 mW that is centered at a wavelength of 1030 nm. The SOA is driven by an injection current of 60 mA. The performance of the HOA is analyzed by the optimization of various parameters such as the PDF length, Pr3+ concentration, pump wavelength, and injection current. A flat average gain of 24 dB with a flatness of 1 dB and an output power of 9.6 dBm is observed over a wavelength range of 1270-1450 nm. The noise figure (NF) varies from a minimum of 4 dB to a maximum of 5.9 dB for a signal power of 0 dBm. A gain reduction of around 4 dB is observed for an O-band signal at a wavelength of 1290 nm by considering the up-conversion effect. The transmission performance of the designed HOA as a pre-amplifier is evaluated based on the bit-error rate (BER) analysis for a coarse wavelength-division multiplexing (CWDM) system of eight on-off keying (OOK)-modulated channels, each having a data rate of 10 Gbps. An error-free transmission over 60 km of standard single-mode fiber (SMF) is achieved for different data rates of 5 Gbps, 7.5 Gbps, and 10 Gbps.
ABSTRACT
The performance of doped fiber amplifiers can be enhanced significantly with the help of multi-stage pumping technique provided that various critical parameters of pumps including their optical power and wavelength are optimized. We report the performance enhancement of a ytterbium doped fiber amplifier (YDFA) for a 1.02-1.08 µm spectral region with an optimized design based on a novel dual-stage in-band asymmetrical pumping scheme. By accurately adjusting the optical power and wavelength of pumps in both the stages, a record peak gain of around 62.5 dB and output power of 4.5 W are achieved for a signal wavelength of 1.0329 µm at an optimized length of Ytterbium-doped silica fiber and optimized doping concentration of Yb3+. Moreover, a minimum noise figure (NF) of 4 dB is observed for a signal wavelength of 1.0329 µm at the optimized parameters. Similarly, the effect of using high and low pump powers at the first and the second stage, respectively, on NF of the amplifier is also investigated at different values of signal powers. It is observed that the value of NF increases significantly by using high pump power at the first stage and low pump power at the second stage.
