Novel Rotating Machinery Structural Faults Signal Adaptive Multiband Filtering and Automatic Diagnosis

To realize an automatic diagnosis of rotating machinery structure faults, this paper presents a novel fault diagnosis model based on adaptive multiband filter and stacked autoencoders (SAEs). First, to solve the problem where the actual rotating frequency and its harmonics cannot be accurately extracted in engineering applications, an improved adaptive multiband filtering method is designed. This method takes the theoretical rotating frequency as the search center, extracts the maximum within the positive and negative deviation as the actual rotating frequency, and sets a threshold according to the actual value to realize multiband filtering. This method can effectively remove background noise and accurately extract the actual rotating frequency and its harmonics. Second, an unsupervised SAE multiclassification model is established to realize an automatic diagnosis of fault types. This model can automatically extract the in-depth features of the filtered signal and improve the fault classification accuracy. Third, engineering and comparative experiments were carried out to verify the effectiveness and superiority of this model. Results show that the proposed automatic diagnosis model can extract the characteristic components abundantly and accurately recognize rotating machinery structural faults.

A Generalizable Sample Resolution Augmentation Method for Mechanical Fault Diagnosis Based on ESPCN

Data augmentation has become a hot topic in the field of mechanical intelligent fault diagnosis. It can expand the limited training dataset by generating simulated samples, but there is still no effective method augmenting the resolution of low resolution sample. In this paper, a simple algorithm, namely, efficient subpixel convolutional neural network (ESPCN), is proposed to solve this deficiency. The ESPCN model performs the arrange operation on the raw low resolution data through the subpixel layer and outputs the result of four-channel multifeature maps. Then, the sample resolution is increased to four times compared with the raw low resolution sample. Finally, the generated high resolution dataset is employed to train the stacked autoencoders (SAE) for fault classification, and the raw high resolution dataset is used for testing. Two fault diagnosis cases with different sample dimensions and rotating speeds are set up to simulate the low resolution situation, and the experimental results verify the feasibility of the proposed algorithm.

Broad Autoencoder Features Learning for Classification Problem

Activation functions such as Tanh and Sigmoid functions are widely used in Deep Neural Networks (DNNs) and pattern classification problems. To take advantages of different activation functions, the Broad Autoencoder Features (BAF) is proposed in this work. The BAF consists of four parallel-connected Stacked Autoencoders (SAEs) and each of them uses a different activation function, including Sigmoid, Tanh, ReLU, and Softplus. The final learned features can merge such features by various nonlinear mappings from original input features with such a broad setting. This helps to excavate more information from the original input features. Experimental results show that the BAF yields better-learned features and classification performances.

Broad Autoencoder Features Learning for Classification Problem

Activation functions such as Tanh and Sigmoid functions are widely used in Deep Neural Networks (DNNs) and pattern classification problems. To take advantages of different activation functions, the Broad Autoencoder Features (BAF) is proposed in this work. The BAF consists of four parallel-connected Stacked Autoencoders (SAEs) and each of them uses a different activation function, including Sigmoid, Tanh, ReLU, and Softplus. The final learned features can merge such features by various nonlinear mappings from original input features with such a broad setting. This helps to excavate more information from the original input features. Experimental results show that the BAF yields better-learned features and classification performances.

DFTSA-Net: Deep Feature Transfer-Based Stacked Autoencoder Network for DME Diagnosis

Diabetic macular edema (DME) is the most common cause of irreversible vision loss in diabetes patients. Early diagnosis of DME is necessary for effective treatment of the disease. Visual detection of DME in retinal screening images by ophthalmologists is a time-consuming process. Recently, many computer-aided diagnosis systems have been developed to assist doctors by detecting DME automatically. In this paper, a new deep feature transfer-based stacked autoencoder neural network system is proposed for the automatic diagnosis of DME in fundus images. The proposed system integrates the power of pretrained convolutional neural networks as automatic feature extractors with the power of stacked autoencoders in feature selection and classification. Moreover, the system enables extracting a large set of features from a small input dataset using four standard pretrained deep networks: ResNet-50, SqueezeNet, Inception-v3, and GoogLeNet. The most informative features are then selected by a stacked autoencoder neural network. The stacked network is trained in a semi-supervised manner and is used for the classification of DME. It is found that the introduced system achieves a maximum classification accuracy of 96.8%, sensitivity of 97.5%, and specificity of 95.5%. The proposed system shows a superior performance over the original pretrained network classifiers and state-of-the-art findings.

DensePILAE: a feature reuse pseudoinverse learning algorithm for deep stacked autoencoder

Autoencoder has been widely used as a feature learning technique. In many works of autoencoder, the features of the original input are usually extracted layer by layer using multi-layer nonlinear mapping, and only the features of the last layer are used for classification or regression. Therefore, the features of the previous layer aren’t used explicitly. The loss of information and waste of computation is obvious. In addition, faster training and reasoning speed is generally required in the Internet of Things applications. But the stacked autoencoders model is usually trained by the BP algorithm, which has the problem of slow convergence. To solve the above two problems, the paper proposes a dense connection pseudoinverse learning autoencoder (DensePILAE) from reuse perspective. Pseudoinverse learning autoencoder (PILAE) can extract features in the form of analytic solution, without multiple iterations. Therefore, the time cost can be greatly reduced. At the same time, the features of all the previous layers in stacked PILAE are combined as the input of next layer. In this way, the information of all the previous layers not only has no loss, but also can be strengthened and refined, so that better features could be learned. The experimental results in 8 data sets of different domains show that the proposed DensePILAE is effective.

A Novel Method for Food Market Regulation by Emotional Tendencies Predictions from Food Reviews Based on Blockchain and SAEs

As a part of food safety research, researches on food transactions safety has attracted increasing attention recently. Food choice is an important factor affecting food transactions safety: It can reflect consumer preferences and provide a basis for market regulation. Therefore, this paper proposes a food market regulation method based on blockchain and a deep learning model: Stacked autoencoders (SAEs). Blockchain is used to ensure the fairness of transactions and achieve transparency within the transaction process, thereby reducing the complexity of the trading environment. In order to enhance the usability, relevant Web pages have been developed to make it more friendly and conduct a security analysis for using blockchain. Consumers’ reviews after the transactions are finished can be used to train SAEs in order to perform emotional tendencies predictions. Compared with different advanced models for predictions, the test results show that SAEs have a better performance. Furthermore, in order to provide a basis for the formulation of regulation strategies and its related policies, case studies of different traders and commodities have also been conducted, proving the effectiveness of the proposed method.