An effective and novel roller bearing fault diagnosis technique based on empirical mode decomposition (EMD) energy entropy and support vector machine (SVM) is put forward in this article. The vibration signal of roller bearing is decomposed by EMD and the first 5 intrinsic mode function (IMF) components are obtained. SVM served as a fault diagnosis classifier and the extracted energy features of the first 5 IMFs are taken as network input vectors, and then the fault bearing and the normal bearing can be distinguished. An technique for fault of roller bearing by SVM is evaluated against a series of fault diagnosis methods that are widely used in machinery, with particular regard to the effect of training set size on fault diagnosis accuracy. We trained the SVM using RBF kernel function. We compare our experimental results with the existing results given by SMO and SVM-light algorithms. It can be seen that the fault diagnosis method based on SVM-light is superior to that based on SMO in diagnosis accuracy of roller bearing. In addition to the SVM, the same datasets were classified using RBF NN and Hopfield NN. The experimental results show that the technique of support vector machine based on EMD energy entropy has higher fault diagnosis ability.
Feed-forward artificial neural networks (ANNs) have been applied to the diagnosis of mixed-mode electronic circuit. In order to tackle the circuit complexity and to reduce the number of test points, hierarchical approach to the diagnosis generation was implemented with two levels of decision: the system level and the circuit level. For every level, using the simulation-before-test (SBT) approach, fault dictionary was created first, containing data relating to the fault code and the circuit response for a given input signal. ANNs were used to model the fault dictionaries. During the learning phase, the ANNs were considered as an approximation algorithm to capture the mapping enclosed within the fault dictionary. Later on, in the diagnostic phase, the ANNs were used as an algorithm for mapping the measured data into fault code, which is equivalent to searching the fault dictionary performed by some other diagnostic procedures. At the topmost level, the fault dictionary was split into parts simplifying the implementation of the concept. A voting system was created at the topmost level in order to distinguish which ANN's output is to be accepted as the final diagnostic statement. The approach was tested on an example of an analog-to-digital converter, and only one test point was used, i.e. the digital output. Full diversity of faults was considered in both digital (stuck-at and delay faults) and analog (parametric and catastrophic faults) parts of the diagnosed system. Special attention was paid to the faults related to the A/D and D/A interfaces within the circuit.