A novel intelligent fault diagnosis method for motor roller bearings which operate under unsteady rotating speed and load is proposed in this paper. The pseudo Wigner-Ville distribution (PWVD) the relative crossing information (RCI) methods are used extracting feature spectra from non-stationary vibration signal measured condition diagnosis. RCI to automatically extract spectrum time-frequency of signal. extracted instantaneous, not correlated with rotation load. By using ant colony...

Detecting and cleaning grain caking on the inner walls of silos is an important task to ensure food safety in storage facilities. However, response challenges such as insufficient lighting conditions, small diverse forms caking, this paper proposes development evaluation a convolutional neural network model for robot vision detection caking. The following improvements visual algorithm based YOLOv5 are proposed article. Firstly, Convolutional Block Attention Module (CBAM) improved Total Cross...

The condition diagnosis of rotating machinery depends largely on the feature analysis vibration signals measured for diagnosis. However, from usually are nonstationary and nonlinear contain noise. useful fault features hidden in heavy background In this paper, a novel method based multiwavelet adaptive threshold denoising mutation particle swarm optimization (MPSO) is proposed. Geronimo, Hardin, Massopust (GHM) employed extracting weak under noise, adaptively selecting appropriate with...

Convolutional neural networks (CNN) are widely used on sequential data since it can capture local context dependencies and temporal order information inside sequences. Attention (ATT) mechanisms have also attracted enormous interests due to its capability of capturing the important parts a sequence. These two extract different features from In combine advantages CNN ATT, we propose convolutional attention network (CAN), which merges structure ATT into single serve as new basic module in...

A new denoising method was proposed in the paper according to characteristics of insulator infrared image with impulse noise. First, based on pulse coupled neural network (PCNN) detect location noise pixels, while maintaining same non-noise pixels. and then noise, window size filter adaptively determined by calculating intensity image. The pixels maximum minimum gray value filtering are excluded, using left similarity calculation out weights. weighted algorithm is used experiments show that...

When using meta-heuristic optimization approaches for optimization, a large number of samples are required. In particular, when generating subgeneration, the utilization existing is low and individuals high. Therefore, surrogate-based has been developed, which greatly reduces in subgeneration cost optimization. complex air supply scenarios, single-objective results may not be comprehensive; therefore, this paper developed double-objective method based on Kriging surrogate model Non-dominated...

Inflection graphs are highly complex networks representing relationships between inflectional forms of words in human languages. For so-called synthetic languages, such as Latin or Polish, they have particularly interesting structure due to abundance forms. We construct the simplest form inflection graphs, namely a bipartite graph which one group vertices corresponds dictionary headwords and other inflected encountered given text. then study projection this on set headwords. The decomposes...

Inflection graphs are highly complex networks representing relationships between inflectional forms of words in human languages. For so-called synthetic languages, such as Latin or Polish, they have particularly interesting structure due to the abundance forms. We construct simplest form inflection graphs, namely a bipartite graph which one group vertices corresponds dictionary headwords and other inflected encountered given text. We, then, study projection this on set headwords. The...

Recommender systems are aimed at generating a personalized ranked list of items that an end user might be interested in. With the unprecedented success deep learning in computer vision and speech recognition, recently it has been hot topic to bridge gap between recommender neural network. And methods have shown achieve state-of-the-art on many recommendation tasks. For example, recent model, NeuMF, first projects users into some shared low-dimensional latent feature space, then employs nets...