Reliable monitoring for detection of damage in epicyclic gearboxes is a serious concern for all industries in which these gearboxes operate in a harsh environment and in variable operational conditions. In this paper, autonomous multidimensional novelty detection algorithms are used to estimate the gearbox’ health state based on vectors of features calculated from the vibration signal. The authors examine various feature vectors, various sources of data and many different damage scenarios in order to compare novel detection algorithms based on three different principles of operation: a distance in the feature space, a probability distribution, and an ANN (artificial neural network)-based model reconstruction approach. In order to compensate for non-deterministic results of training of neural networks, which may lead to different network performance, the ensemble technique is used to combine responses from several networks. The methods are tested in a series of practical experiments involving implanting a damage in industrial epicyclic gearboxes, and acquisition of data at variable speed conditions.
The possibility of distinguishing and assessing the influences of defects in particular pump elements by registering vibration signals at characteristic points of the pump body would be a valuable way for obtaining diagnostic information. An effective tool facilitating this task could be a well designed and identified dynamic model of the pump. When applied for a specific type of the pump, such model could additionally help to improve its construction. This paper presents model of axial piston positive displacement pump worked out by the authors. After taking the simplifying assumptions and dividing the pump into three sets of elements, it was possible to build a discrete dynamic model with 13 degrees of freedom. According to the authors' intention, the developed dynamic model of the multi-piston pump should be used for damage simulation in its individual elements. By gradual change in values of selected construction parameters of the object (for example: stiffness coefficients, damping coefficients), it is possible to perform simulation of wear in the pump. Initial verification of performance of the created model was done to examine the effect of abrasive wear on the swash plate surface. The phase trajectory runs estimated at characteristics points of the pump body were used as a useful tool to determine wear of pump elements.