Conference Paper

Simultaneous faults diagnosis for automotive ignition patterns.

DOI: 10.1109/ICMLC.2011.6016890 Conference: International Conference on Machine Learning and Cybernetics, ICMLC 2011, Guilin, China, July 10-13, 2011, Proceedings
Source: DBLP

ABSTRACT Many practical applications have the property that different possible faults may appear at one time. Simultaneous faults diagnosis is referred to accurately diagnose the possible faults based on the symptoms from an observed pattern. There are two key challenges in this kind of problem: 1) the symptoms of different faults are mixed or combined into one (input) pattern which makes accurate diagnosis difficult, 2) the preparation of a large amount of training patterns because there are many different combinations of faults. We proposed a framework to effectively resolve these challenges using feature extraction and multi-label probabilistic classification. This framework has been applied and verified in the domain of automotive ignition patterns.

  • [Show abstract] [Hide abstract]
    ABSTRACT: One of the main limitations of current plant supervisory control systems is the reliability and the correct management of simultaneous faults, which is crucial for supporting the plant operators' decision making. In this work, a MultiLabel approach that makes use of support vector machines as the learning algorithm is employed to arrange a novel fault diagnosis system (FDS). The FDS is trained to address a difficult control case study from industry widely studied in the literature, the Tennessee Eastman process. Successful results have been obtained when diagnosing up to four simultaneous faults. These results are very promising since they have been obtained by just using simple training sets consisting of single faults, thus proving a very high learning capacity. © 2007 American Institute of Chemical Engineers AIChE J, 2007
    AIChE Journal 09/2007; 53(11):2871 - 2884. · 2.58 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Degradation parameter or deviation parameter from normal to failure condition of machine part or system is needed as an object of prediction in prognostics method. This study proposes the combination between relevance vector machine (RVM) and logistic regression (LR) in order to assess the failure degradation and prediction from incipient failure until final failure occurred. LR is used to estimate failure degradation of bearing based on run-to-failure datasets and the results are then regarded as target vectors of failure probability. RVM is selected as intelligent system then trained by using run-to-failure bearing data and target vectors of failure probability estimated by LR. After the training process, RVM is employed to predict failure probability of individual units of machine component. The performance of the proposed method is validated by applying the system to predict failure time of individual bearing based on simulation and experimental data. The result shows the plausibility and effectiveness of the proposed method, which can be considered as the machine degradation assessment model.
    Mechanical Systems and Signal Processing 05/2010; · 2.47 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Engine ignition pattern analysis is one of the trouble-diagnosis methods for automotive gasoline engines. Based on the waveform of the ignition pattern, the mechanic guesses what may be the potential malfunctioning parts of an engine with his/her experience and handbooks. However, this manual diagnostic method is imprecise because many ignition patterns are very similar. Therefore, a diagnosis may need many trials to identify the malfunctioning parts. Meanwhile the mechanic needs to disassemble and assemble the engine parts for verification. To tackle this problem, Wavelet Packet Transform (WPT) is firstly employed to extract the features of the ignition pattern. With the extracted features, a statistics over the frequency subbands of the pattern can then be produced, which can be used by Multi-classLeast Squares Support Vector Machines (MCLS-SVM) for engine fault classification. With the newly proposed classification system, the number of diagnostic trials can be reduced. Besides, MCLS-SVM is also compared with a typical classification method, Multi-layer Perceptron (MLP). Experimental results show that MCLS-SVM produces higher diagnostic accuracy than MLP.
    Expert Systems with Applications 07/2011; 38:8563-8570. · 1.97 Impact Factor