Search

Bitumen in the oil sands industry is separated from sand using a water-based gravity separation process in a primary separation vessel (PSV). The interface between the froth and the middlings layer is an important parameter to control for optimal operation of the PSV unit. In this paper, a method using computer vision based on convolutional neural networks (CNNs) and Kalman filter (KF) is designed for the detection of the interface in PSV, with the CNN estimating both the interface level and the image quality. The proposed method consists of two parts: an offline and an online stage, wherein the parameters of CNN and KF are trained in the offline stage using the available data. The algorithm is made robust to any new type of occlusions, not present in the training dataset, in the online stage. Experimental results demonstrate that the proposed method is accurate and robust to different abnormalities in the process. [ABSTRACT FROM AUTHOR]

This article proposes a variational Bayesian (VB) inference based on the nonsynchronous array measurement (NAM) (VB-NAM) method in order to obtain the fast and robust localization of low-frequency acoustic sources. To enlarge the aperture size compared with the prototype array, the NAM is performed to measure the acoustic pressures with low-frequency based on the forward power propagation model. The implementation of the NAM can be reformulated into a cross-spectral matrix (CSM) completion problem. Then, to solve the inverse problem of the NAM power propagation model, the VB inference based on the Student-t priors and Kullback–Leibler (KL) divergence optimization is proposed. The advantages of the proposed VB-NAM benefit from the optimization of matrix inversion and adaptive estimation of regularization parameters. The contribution of the adaptive parameter evaluation is to reduce the impact of multiple interferences (such as additive noise and the matrix completion error) in NAM. Finally, both simulations at 800 Hz and experimental results at 1000 Hz are presented to show the validation of the proposed VB-NAM method, even under the anisotropic Gaussian noise conditions. Algorithm performance and iteration process are analyzed to demonstrate the efficiency and robustness. [ABSTRACT FROM AUTHOR]

Signals obtained from impact-echo techniques can be used to detect and classify the defects in damaged materials. The defects change the wave propagation between the impact and the sensors producing particular spectrum elements, which define the feature vector. We propose a hierarchical clustering method that models the feature vector as a mixture of Gaussians (MoG) for every class and then merges different clusters using as a distance measure the symmetric Kullback–Leibler (KL) divergence. Since there is no closed-form solution to the KL divergence between MoGs, some approximations are introduced. We apply the hierarchical clustering algorithms to the signals obtained from real specimens made of aluminum alloy. The samples are classified into four classes according to the state: homogeneous (no defect), one hole, one crack, and multiple defects. We compare the performance of different approximations and discuss the dendrograms that are obtained. Similar kinds of defects are clustered first, and more importantly, the high-level hierarchy is able to distinguish between the defective and nondefective materials. [ABSTRACT FROM AUTHOR]