High operations and maintenance costs for wind turbines reduce their overall cost effectiveness. One of the biggest drivers of maintenance cost is unscheduled maintenance due to unexpected failures. Continuous monitoring of wind turbine health using automated failure detection algorithms can improve turbine reliability and reduce maintenance costs by detecting failures before they reach a catastrophic stage and by eliminating unnecessary scheduled maintenance. A SCADA (Supervisory Control and Data Acquisition System) -data based condition monitoring system uses data already collected at the wind turbine controller. It is a cost-effective way to monitor wind turbines for early warning of failures and performance issues. In this paper, we describe our exploration of existing wind turbine SCADA data for development of fault detection and diagnostic techniques for wind turbines. We used a number of measurements to develop anomaly detection algorithms and investigated classification techniques using clustering algorithms and principal components analysis for capturing fault signatures. Anomalous signatures due to a reported gearbox failure are identified from a set of original measurements including rotor speeds and produced power.
Incipient fault detection and diagnosis in turbine engines is key to effective maintenance and improved availability of systems dependent on these engines. In this paper, we present a novel method for incipient fault detection and diagnosis using Hidden Markov Models (HMMs). In particular, we focus on engine faults that are manifest in transient operating conditions such as engine startup and acceleration. HMMs are stochastic signal models that are effective in modeling transient signals. They are developed with engine data collected under nominal operating conditions. Engine data representing different fault conditions are used to develop the fault HMMs; a separate model is developed for each of the faults. Once the nominal and fault HMMs are developed, new engine data collected from the engine are evaluated against the HMMs and a determination is made whether a fault is indicated. Here, we demonstrate our HMM-based fault detection and diagnosis approach on engine speed profiles taken from a real engine. Further, the effectiveness of the HMM-based approach is compared with a neural-network-based approach and a method based on using principal component analysis in conjunction with a neural network approach.
A fault diagnosis and prognosis method is developed for the fuel supply system in gas turbine engines. The engine startup profiles of the core speed (N2) and the exhaust gas temperature (EGT) collected with high speed sampling rate are extracted and processed into a more compact data set. The fuzzy clustering method is applied to the smaller number of parameters and the fault is detected by differentiating the clusters matching the failures. In this work, the actual flight data collected in the field is used to develop and validate the system, and the results are shown for the test on nine engines that experienced fuel supply system failure. The developed fault diagnosis system detects the failure successfully for all nine cases. For the earliest detection cases, the alarms start to trigger 26 days before the system completely fails and 7 days in advance for the last detection.
Within the context of preventive health maintenance in complex engineering systems, novel sensor fault detection methodologies are developed for an aircraft auxiliary power unit. Promising results at operational and sensor failure conditions are obtained for temperature and pressure sensors. In the methodology proposed, first covariance and noise analyses of sensor data are performed. Next, auto-associative and heteroassociative neural networks for sensor validation are designed and trained. These neural networks are used together to provide validation for pressure and temperature sensors. The last step consists of development of detection and identification logic for sensor faults. In spite of high noise levels, the methodology is shown to be very robust. More than 90% correct sensor failure detection is achieved when noise on the order of noise inherently present in sensor readings is added.
Bayesian Networks has been proven to be successful tool for fault diagnosis. There are a variety of approaches for learning the structure of Bayesian Networks from data. This learning problem has been proven to be NP-hard hence none of the approaches are exact when no prior knowledge about the domain of the variables exists. Our approach is based on searching the best network by using particle swarm optimization (PSO) technique. PSO is inherently parallel, works for large domains and does not trap into local maxima. This paper is an application of this technique to a real world problem; fault diagnosis of an airplane engine for oil related variables. It is implemented by our improved software written in C/C++ by using MPI on Linux. Our implementation has the advantages of being general, robust and scalable. Moreover neither expert knowledge, nor node ordering is necessary prior to the optimization. The datasets are generated by preprocessing oil related sensor readings of airplane engines taken during the approach phase of flights. Using this datasets and our software, we constructed Bayesian Networks of the oil related variables in an airplane engine for diagnostics and predictive purposes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.