Navegando por Assunto "Fault diagnosis"

Agora exibindo 1 - 4 de 4

Acesso aberto (Open Access)
Desenvolvimento de sistema de diagnóstico de falhas em roletes de transportadores de correia
(Universidade Federal do Pará, 2024-03-28) SOARES, João Lucas Lobato; MESQUITA, Alexandre Luiz Amarante; http://lattes.cnpq.br/3605920981600245; https://orcid.org/0000-0001-5605-8381
Belt conveyors are essential equipment in mining industry and require constant monitoring to maintain good reliability. In order to support the belt and the material being conveyed, rollers are components that constantly fail during operation, in which they present faults in bearings and surface wear in the shell as the most common failure modes. Thus, monitoring based on predictive maintenance is essential, and machine learning techniques can be used as an alternative for detecting equipment failures. In diagnostics using machine learning, the feature selection step is important to avoid loss of accuracy in the classification of the equipment's condition. The present study analyzes the performance of the decision tree algorithm and Analysis of Variance (ANOVA) as alternative methods for dimensionality reduction. Initially, the vibration signals were collected on the rollers of a belt conveyor bench and the Wavelet Packet Decomposition (WPD) was applied to the signals to obtain the energy ranges, which were used as features for classification. After the determination of the best features, two approaches were analyzed for the selection of features: one with the application of the method without dimensionality reduction and the other with the application of the decision tree. In addition, different classification algorithms were used: Support Vector Machine (SVM), k-Nearest Neighbor (kNN) and Artificial Neural Network (ANN). As a result, it was found a superior performance of diagnostic accuracy in all techniques with a reduction in the dimensionality of the characteristics selected by the decision tree. In addition, SVM, kNN and ANN showed increases in accuracy ranging among the fault diagnosis models approached.
Acesso aberto (Open Access)
Estudos de estratégias de identificação paramétrica para detecção e diagnóstico de faltas em um processo industrial do tipo tanques comunicantes
(Universidade Federal do Pará, 2012-04-22) SILVA, Raphael Diego Comesanha e; BARRA JUNIOR, Walter; http://lattes.cnpq.br/0492699174212608; COSTA JÚNIOR, Carlos Tavares da; http://lattes.cnpq.br/6328549183075122
This dissertation presents a technique for detection and diagnosis of incipient faults, which cause changes in behavior of the system under investigation and are reflected in the mathematical model’s parameters values variation. As a testbed, was constructed a model of an industrial system computing environment Matlab/Simulink, which consists of a dynamic plant composed of two tanks linked to each other. The modeling of this plant was carried out by physical equations that describe the dynamics of the system. The fault, which the system was submitted, represents a gradual clogging in the exit pipe of the tank 2. This bottleneck causes a gradual reduction, up to 20%, of the pipe section. The technique of fault detection was performed by real-time estimation of parameters Auto-regressive models with exogenous inputs (ARX) with fuzzy and Recursive Least Squares (RLS) estimators. Already, the percentage clogging diagnosis of the pipe was obtained by a fuzzy system parameter tracking, fed back by the integral of the residue detection. Using this methodology, it was possible to detect and diagnose the simulated fault in three differents operating points of the system. In both techniques tested, the RLS method perform well, only to detect fault. Otherwise, the fuzzy method performed better, in detect and diagnose the fault applied to the system, noting the work propose.
Acesso aberto (Open Access)
Metodologia baseada em sistema fuzzy intervalar do tipo-2 para detecção e identificação de faltas de incipientes em motores de indução
(Universidade Federal do Pará, 2013-02-27) ROCHA, Erick Melo; BARRA JUNIOR, Walter; http://lattes.cnpq.br/0492699174212608
Since the incorporation of automation in the production processes, aiming at order to improve productivity and quality of products and services, researches on more efficient methodologies for fault diagnosis became more intensive. Such techniques allow the early detection of faults, before then lead to failures. This work investigates techniques for detection and diagnosis of faults and its application to induction motors, limiting their study to two situations, namely: system free of faults and system under incipient partial short-circuit in the coils the stator winding. For faults detection, parametric analysis of fist order ARX (autoregressive with exogenous input) were applied. The parameters of identified ARX modes, which bring information about the dynamics of the dominant system, are recursively obtained by the techniques of recursive least squares (RLS). In order to evaluate the capability for early fault detection, a type-2 interval fuzzy system was developed. This kind of fuzzy system has capability to capture a larger set of uncertainties than conventional (type-1) fuzzy systems. The footprint of uncertainty (FOU), characteristic of type-2 fuzzy system, is a way to accounts for uncertainties coming from noise and numerical errors from the process of parameter estimation. The ARX model parameters are the inputs to the supervisor system. Genetic algorithms (GA’s) were used for optimization of SIF interval type-2, aiming at to reduce the diagnostic error. The results obtained in tests of computer simulation show the effectiveness of the proposed methodology.
Acesso aberto (Open Access)
Proposta de metodologia para diagnóstico de falha em rolamentos de baixa velocidade
(Universidade Federal do Pará, 2024-04-19) COSTA, Thiago Barroso; MESQUITA, Alexandre Luiz Amarante; http://lattes.cnpq.br/3605920981600245; https://orcid.org/0000-0001-5605-8381
Monitoring low-speed bearings with vibration analysis is more challenging due to the low energy level of the vibration signal that carries the failure data, making it more susceptible to interference from other sources, impairing the interpretation of information. Thus, an alternative is to calculate signal predictors that may be sensitive to pattern changes relative to failure onset and progression. Hence, the present work extracted different types of features, among them two nonlinear features and eleven extracted from the signal in the time domain. Those features were ranked and selected based on their sensibility to class differentiation, which was estimated using the t-Welch statistic value. Among them is the Largest Lyapunov Exponent, which, in this work, had a modification in one of its calculation steps, improving its sensitivity in some cases. In addition, the influence of the vibration signal window size on the class separability of the indicators was evaluated (which is a scarce content in low-speed bearing monitoring literature). After feature selection, the data were subjected to a linear transformation through PCA (Principal Component Analysis), aiming to reduce the data dimensionality to three dimensions and to minimize the redundancy effects of highly correlated features. In sequence, the data represented in the space of principal components were projected on a Hotelling T2 statistic control chart. The chart allowed the detection and rejection of potential outliers, which consisted of points above a limit line estimated based on F statistic distribution. Finally, binary and multiclass Support Vector Machine classification models were trained with experimental data acquired from normal conditions and three levels of incipient fault in bearing. The models performed well, mainly the binary model with test data obtained from belt conveyor pulley bearings in industrial operation.