Papers by Keyword: Multivariate Statistic

Abstract: In this paper, a time-domain analysis method based on multivariate statistic is presented for wind power generation fault diagnosis. Generally, the sound and vibration signals obtained from wind power generation are time-variant since they are strongly related to the rotational speed which is not constant even in the macro steady state. Since the mostly used signal processing method, the Fourier analysis, is only suitable for stationary signals, the development of the joint time-frequency analysis is demanded. Here, Q statistic (also referred as squared prediction error, SPE) is introduced, it is used to monitor the vibration signals and three-phase currents. The control limit of the Q statistics is calculated to decide the state of the rotating machine, and the contribution plot of SPE is used to find the fault source. The method can efficiently detect faint change and the validity of the method is proved by experiments.

Abstract: Rock cuttability is expressed by specific energy (SE) that is defined as the energy required for cutting unit volume of rock. Direct determination of SE requires a rock cutting rig and is expensive and time-consuming. Therefore, empirical models have been alternative methods for predicting SE from rock properties. Two different predictive models of SE have been developed in this study using regression tree and artificial neural network (ANN) methods. Both empirical models employed the uniaxial compressive strength (UCS) and Mode I fracture toughness (KIC), being derived from tensile strength (t), as predictors. Data from four different studies have been used to develop the models. Statistical analyses on the data set have shown that both UCS and KIC are closely related to SE in a nonlinear form. Numerical and graphical measures of the goodness of the fit and ANOVA test have shown that regression tree and ANN models have performed similarly.