Recognition of safety helmets wearing by construction workers is a common target detection topic in applications of deep learning-based image processing. This paper provides a study of an enhanced YOLOv5-based method, in which the challenges caused by complicated construction environment backgrounds, dense targets, and the irregular shape of safety helmets are addressed. In a trunk network, feature extraction is more based on the target shape by using the Deformable Convolution Net instead of the conventional convolution; in the Neck, a Convolutional Block Attention Module is introduced to weaken feature extraction of complex backgrounds by giving weights to enhance the characterization ability of target features; and the original network’s Generalized Intersection over Union Loss is replaced by Distance Intersection over Union Loss to overcome the problem of erroneous location when the population is dense. The dataset for the training network is created by mixing open-source datasets with autonomous collecting to evaluate the effectiveness of the algorithm. We observed that the improved model has a detection accuracy of 91.6%, up 2.3% over the original network model, and a detection speed of 29 frames per second, which is compliant with most security cameras’ capture frame rate.

With environmental issues arising from the excessive use of fossil fuels, clean energy has gained widespread attention, particularly the application of lithium-ion batteries. Lithium-ion batteries are integrated into various industrial products, which necessitates higher safety requirements. Narrowband Internet of Things (NB-IoT) is an LPWA (Low Power Wide Area Network) technology that provides IoT devices with low-power, low-cost, long-endurance, and wide-coverage wireless connectivity. This study addresses the shortcomings of existing lithium-ion battery pack detection systems and proposes a lithium-ion battery monitoring system based on NB-IoT-ZigBee technology. The system operates in a master-slave mode, with the subordinate module collecting and fusing multi-source sensor data, while the master control module uploads the data to local monitoring centers and cloud platforms via TCP and NB-IoT. Experimental validation demonstrates that the design functions effectively, accomplishing the monitoring and protection of lithium-ion battery packs in energy storage power stations.

A bearing is a critical component in the transmission of rotating machinery. However, due to prolonged exposure to heavy loads and high-speed environments, rolling bearings are highly susceptible to faults, Hence, it is crucial to enhance bearing fault diagnosis to ensure safe and reliable operation of rotating machinery. In order to achieve this, a rotating machinery fault diagnosis method based on a deep convolutional neural network (DCNN) and Whale Optimization Algorithm (WOA) optimized Deep Extreme Learning Machine (DELM) is proposed in this paper. DCNN is a combination of the Efficient Channel Attention Net (ECA-Net) and Bi-directional Long Short-Term Memory (BiLSTM). In this method, firstly, a DCNN classification network is constructed. The ECA-Net and BiLSTM are brought into the deep convolutional neural network to extract critical features. Next, the WOA is used to optimize the weight of the initial input layer of DELM to build the WOA-DELM classifier model. Finally, the features extracted by the Improved DCNN (IDCNN) are sent to the WOA-DELM model for bearing fault diagnosis. The diagnostic capability of the proposed IDCNN-WOA-DELM method was evaluated through multiple-condition fault diagnosis experiments using the CWRU-bearing dataset with various settings, and comparative tests against other methods were conducted as well. The results indicate that the proposed method demonstrates good diagnostic performance.

Accurate measurement of flue gas velocity is crucial for safe production, energy saving and emission reduction in power plants. In this paper, a real-time flue gas audio-range velocimeter was developed using a novel sonic driving mode, quadratic-convex frequency sweeping. Both numerical analysis and in situ experiments were conducted which had validated the developed velocimeter. The accuracy and robustness of the developed velocimeter was quantitatively assessed by root mean square error (RMSE) and standard deviation (SD) from the tests data. The test data was intentionally contaminated with varied levels of noises, forming sets of measurement data with varied signal-to-noise ratios (SNRs). The simulation results show that within the variation of SNR from 0 dB to −18 dB, the averaged relative RMSE obtained using this proposed quadratic-convex driving mode is 1.2 dB and 2.3 dB lower than those obtained using linear and quadratic-concave driving modes, respectively. The experiment results indicate that the RMSE and SD values based on the quadratic-convex driving mode are less than 0.1 for each given SNR. Compared with the traditionally adopted method of linear frequency sweeping, the proposed sonic driving scheme can significantly improve the accuracy and robustness of flue gas velocity measurement.