Research on the application of intelligent sensors based on the Internet of Things in fault diagnosis of mechanical and electrical equipment

Abstract

The purpose of this work to do is to solve the fault diagnosis of agricultural mechanical and electrical equipment and guarantee the smooth operation of production line and industrial process. The research begins by collecting operational data from electromechanical equipment based on Internet of Things (IoT) technology and utilizes Narrowband Internet of Things (NB-IoT) modules to achieve communication for terminal electromechanical devices. Subsequently, the Kernel Extreme Learning Machine (KELM) is introduced and combined with the Whale Optimization algorithm to construct a fault diagnosis model based on the Whale Optimization Kernel Extreme Learning Machine (WOKELM). Finally, the performance of the model is experimentally evaluated. The results indicate that, compared to other baseline algorithms, the proposed model algorithm achieves Accuracy values exceeding 90 %, with at least a 3.85 % improvement over the KELM baseline algorithm. Additionally, in the training a.nd testing sets, the F1 values of the proposed model algorithm reach 91.24 % and 85.85 %, respectively, which is at least 2.98 % higher than other model algorithms. Furthermore, an analysis of fault diagnosis error rates reveals that the Root Mean Squared Error (RMSE) for fault diagnosis is below 4.13 %. Therefore, the proposed fault diagnosis model demonstrates excellent performance in terms of accuracy and precision, providing robust support for improving the intelligence and accuracy of fault diagnosis in electromechanical equipment.