Light-Weighted Deep Learning Model to Detect Fault in IoT-Based Industrial Equipment.

Industry 4.0, with the widespread use of IoT, is a significant opportunity to improve the reliability of industrial equipment through problem detection. It is difficult to utilize a unified model to depict the working condition of devices in real-world industrial scenarios because of the complex and dynamic relationship between devices. The scope of this research is that it can detect equipment defects and deploys them in a natural production environment. The proposed research is describing an online detection method for system failures based on long short-term memory neural networks. In recent years, deep learning technology has taken over as the primary method for detecting faults. A neural network with a long short-term memory is used to develop an online defect detection model. Feature extraction from sensor data is done using the curve alignment method. Based on long-term memory neural networks, the fault detection model is built (LSTM). In the end, sliding window technology is used to identify and fix the problem: the model's online detection and update. The method's efficacy is demonstrated by experiments based on real data from power plant sensors.

Improved Capsule Network Optimization Hierarchical Convolution Algorithm for Mental Health Recognition.

To address the shortcomings of standard convolutional neural networks (CNNs), the model structure is complex, the training period is lengthy, and the data processing technique is single. A modified capsule network is presented to optimize hierarchical convolution-the algorithm for identifying mental health conditions. To begin, two types of data processing are performed on the original vibration data: wavelet noise reduction and wavelet packet noise reduction; this retains more valuable information for mental health identification in the original signal; secondly, the CNN employs the concept of hierarchical convolution, and three distinct scaled convolution kernels are utilized to extract features from numerous angles; ultimately, the convolution kernel's extracted features are fed into the pruning strategy's capsule network for mental health diagnosis. The enhanced capsule network has the potential to significantly speed up mental health identification while maintaining accuracy. It is time to address the issue of the CNN structure being too complex and the recognition impact being inadequate. The experimental findings indicate that the suggested algorithm achieves a high level of recognition accuracy while consuming a small amount of time.