This work presents a new approach for detecting and isolating faults in nonlinear processes using independent neural network models. In this approach, an independent neural network is used to model the proton exchange membrane fuel cell nonlinear systems using a multi-input multi-output structure. This research proposed the use of radial basis function network and multilayer perceptron network to perform fault detection. After training, the neural network models can give accurate prediction of the system outputs, based on the system inputs. Using the residual generation concept developed in the model-based diagnosis, the difference between the actual and estimated outputs are used as residuals to detect faults. When the magnitude of these residuals exceed a predefined threshold, it is likely that the system is faulty. In order to isolate faults in the system, a second neural network is used to examine features in the residual. A specific feature would correspond to a specific fault. Based on features extracted and classification principles, the second neural network can isolate faults reliably and correctly. The developed method is applied to a benchmark simulation model of the proton exchange membrane fuel cell stacks developed at Michigan University. One component fault, one actuator fault and three sensor faults were simulated on the benchmark model. The simulation results show that the developed approach is able to detect and isolate the faults to a fault size of ±10% of nominal values. These results are promising and indicate the potential of the method to be applied to the real world of fuel cell stacks for dynamic monitoring and reliable operations.

The focus of the work in this thesis is concerned with the investigation and development of indirect measurement techniques. The methodology adopted is a combination of practical experimental, analytical deductive reasoning and simulation studies. This has led to proposals for a number of indirect tyre pressure monitoring systems, which are able to detect pressure loss under specific circumstances. The outcome overall is a proposal for a new supervisory system comprising of a modular framework, allowing various algorithms and techniques to be implemented in a complementary manner as they emerge and data sources become available. A number of contributions to the field have been made, which to the knowledge of the author, provide potential for further algorithm development and are imminently applicable given the above. The methods include a tyre pressure diagnosis via a wheel angular velocity comparator, the development of a model-based tyre pressure diagnosis via application of an unknown input observer and a parameter estimation scheme, a model-based tyre pressure diagnosis approach via an enhanced Kalman filter configured to estimate states including the input, a model-based tyre pressure diagnosis via cautious least squares, an investigation and critique of the effects of the choice of sampling interval on discrete-time models and estimation thereof. It is considered, that the extensive literature review provides a valuable historic insight into the tyre fault detection problem. It is clear, from the development and testing of the algorithms (and also the literature), that no single indirect pressure detection method is able to reliably detect changes in all driving scenarios which the regulations typically stipulate (depending on jurisdiction). In the absence of any information about the road input, the majority of the detection work must be shouldered by the wheel angular velocity comparator algorithm. As image recognition and sensor technology develops, it becomes possible to make estimates about the road surface and this removes some of the uncertainty on the input of the model-based parameter estimation approaches. Further work is detailed which goes some way towards realising the next steps in a development cycle suitable for a vehicle manufacturer to take through to the implementation stage.

Wrinkle ridges are among the most common tectonic structures on the terrestrial planets and provide important records of the history of planetary strain and geodynamics. The observed broad arches and superposed narrow wrinkles are thought to be the surface manifestation of blind thrust faults, which terminate in near-surface volcanic sequences and cause folding and layer-parallel shear. However, the subsurface tectonic architecture associated with the ridges remains a matter of debate. Here we present direct observations of a wrinkle ridge thrust fault where it has been exposed by erosion in the southern wall of Melas Chasma on Mars. The thrust fault has been made resistant to erosion, likely due to volcanic intrusion, such that later erosional widening of the trough exposed the fault plane as a 70km long ridge extending into the chasma. A plane fit to this ridge crest reveals a thrust fault with a dip of 13 degrees (+8 degrees, -7 degrees) between 1 and 3.5km depth below the plateau surface, with no evidence for listric character in this depth range. This dip is significantly lower than the commonly assumed value of 30 degrees, which, if representative of other wrinkle ridges, indicates that global contraction on Mars may have been previously underestimated.

The main topic of this licentiate thesis is fast faultdetection. The thesis summaries the work performed in theproject“Fast fault detection for distributionsystems”. In the first chapters of the thesis the term“fast”is used in a general manner. The term is laterdefined based upon considerations and conclusions made in thefirst chapters and then related to a specific time. To be able to understand and appreciate why fast faultdetection is necessary, power system faults and theirconsequences are briefly discussed. The consequences of a faultare dependent of a number of different factors, one of thefactors being the duration of the fault. The importance of the speed of the fault detection dependson the type of equipment used to clear the fault. A circuitbreaker which interrupt currents only when they pass through anatural zero crossing might be less dependent on the speed ofthe fault detection than a fault current limiter which limitsthe fault current before it has reached its first prospectivecurrent peak. In order to be able to detect a fault in a power system, thepower system must be observed, i.e., measurements of relevantquantities must be performed so that the fault detectionequipment can obtain information of the state of the system.The fault detection equipment and some general methods of faultdetection are briefly described. Some algorithms and their possible adaptation to fast faultdetection are described. A common principle of many algorithmsare that they assume that either a signal or the power systemobject can be described by a model. Sampled data values arethen fitted to the model so that an estimate of relevantparameters needed for fault detection is obtained. An algorithmwhich do not fit samples to a model but use instantaneouscurrent values for fault detection is also described andevaluated. Since the exact state of a power system never is known dueto variations in power production and load, a model of thepower system or of the signal can never be perfect, i.e., theestimated parameter can never be truly correct. Furthermore,errors from the data acquisition system contribute to the totalerror of the estimated parameter. Two case studies are used to study the performance of the(modified) algorithms. For those studies it has been shown thatthe algorithms can detect a fault within approximately 1msafter fault inception and that one of the algorithms candiscriminate between a fault and two types of common powersystem transients (capacitor and transformer energization). The second case study introduced a system with two sourceswhich required a directional algorithm to discriminate betweenfaults inside or outside the protection zone. It is concluded that under certain assumptions it ispossible to detect power system faults within approximately 1msand that it is possible to discriminate a power system faultfrom power system transient that regularly occurs within powersystems but which not are faults. / NR 20140805

Snehal Udar, for the Doctor of Philosophy degree in Electrical and Computer Engineering, presented on May 4, of 2011, at Southern Illinois University Carbondale. TITLE: BUILT-IN SCHEMES FOR TEST PATTERN GENERATION AND FAULT LOCATION MAJOR PROFESSOR: Dr. D. Kagaris In this dissertation, we studied the areas of test pattern generation and fault location for detecting and diagnosing the faults in today's complex chips. In the first problem, a novel reseeding based test pattern generation scheme is analyzed by proposing a hardware efficient technique that uses irreducible polynomial-primitive element pair to generate distinct subsequences of test patterns. It is shown that for the given characteristic polynomial the hardware cost remains the same irrespective of the number of seeds required to generate the test sequence of given length. This scheme is targeted at generating pseudo-random test patterns that detect easy-to-detect faults. A counter based reseeding scheme is further analyzed that embeds a given set of fully specified test patterns in minimum number of clock cycles. Second problem investigates the effectiveness of inserting observation points on the circuit lines that along with primary output lines distinguish a given set of faults. Three hardware based approaches are proposed that aim at inserting minimum observation points, and are compared with each other for different diagnostic resolutions.

No description available.

No description available.

No description available.

No description available.

No description available.