In the process of gear meshing, the tooth surface will produce friction heat, which will in turn act on the friction coefficient. The temperature does not directly affect the friction coefficient of the gear tooth surface, but indirectly through affecting the gear material parameters. Based on the relationship between gear material parameters and temperature and the analysis of friction under different states of mixed lubrication, the relationship model between gear surface friction coefficient and temperature was established, and the rotating pin-on-disc tribotest of 18CrNiMo7-6 is carried out. The experimental results show that the friction coefficient does not always rise with the increase of temperature, but fluctuates in the condition of mixed lubrication between 25 and 150°C. There are two low data points, the friction coefficient corresponding to 45-65℃ and 120-140℃, respectively. At the same time, the experimental results can be used to get the model coefficient and verify the correctness of the model.
Induction motors are prime component in the industries. Hence, condition monitoring and fault diagnosis of induction motor are important to avoid shutdowns and unplanned maintenance. A technique based on time-domain grayscale current signal imaging (TDGCI) and convolutional neural network (CNN) is proposed for intelligent fault detection of broken rotor bar in an induction motor. The standard current signal dataset made available by the Aline Elly Treml Western Parana State University is used for analysis. This dataset is acquired by simulating the healthy and broken rotor bar (BRB) fault conditions with the four increasing severity levels (1BRB, 2BRB, 3BRB, and 4BRB) at eight loading conditions varying from no load to full load. Conventional machine learning techniques have the limitations of feature selection, while the proposed technique can automatically extract the features from the given input image. The TDGCIs obtained from the time-domain current signal is used as input to exploit the enormous capability of CNN to carry out the image classification, thereby classifying faults features embedded in the images. The efforts are presented to design CNN parameters to achieve the fault classification accuracy of 99.58% for all cases with optimized computational time. The significant reduction in the computational time for fault classification compared to the peer published work is an important contribution.
Work analysis is one of critical tools for improving order picking efficiency in logistics centers. Traditionally, video analysis has been used to analyze work time for various tasks. However, video analysis generally requires a substantial amount of time and labor, which makes it too inefficient for real-time work improvement. To address this, a novel automatic work analysis method is proposed for analyzing the order picking process in logistics centers in this study. The automatic work analysis method can be divided into two parts: dynamic positioning and motion analysis. With the proposed method, an ultrasonic network and a tracking camera are used for measuring a worker’s real-time position while he/she is moving, and an acceleration sensor is used for checking the acceleration of the dominant working hand. Using the position and acceleration data, the worker’s motion is then estimated by an estimation model pre-tested for accuracy. To test the effectiveness of the proposed method, an experiment was conducted in which the measurement error of positioning was found to be approximately 0.01±0.19 m for the x-coordinate values and 0.03±0.45 m for the y-coordinate values. The total measurement error was confirmed to be within a one-step stride. The results of the motion analysis were more than 90% consistent with those obtained by traditional video analysis.