×
The submission system is temporarily under maintenance. Please send your manuscripts to
Go to Editorial ManagerVibration analysis is indispensable for different mechanical applications for early fault diagnosis, and many methods are used to analyze signals. Order tracking is one of these methods that is necessary for vibration analysis, especially for rotating machines. One critical and widely used method of order tracking is Vold-Kalman order tracking (VK-OT), which is used to diagnose faults in non-stationary machines. However, it has complicated and intensive calculations, requires special analysis tools, needs large memory, and takes a very long time to extract the results. The proposed method aims to analyze signals by using Vold-Kalman filter order tracking with shorter time and less calculation memory with high accuracy by using partitioning method that separates the signal into many blocks with overlaps. The proposed method achieved less processing time and need much smaller memory than the original Vold-Kalman filter-based methods.
Monitoring the health of rotating machinery is essential to ensure system safety, achieve cost savings, and enhance overall reliability. The requirement for a reliable and clear method of identifying defects has prompted the development of several monitoring techniques. They utilize vibration, measurement of the motor's current signature, and acoustic emission data in the process of condition monitoring. The MFS (machinery fault simulator) equipment was used to determine bearing faults using vibration signal analysis. MFS conducts simulations and investigations of many bearing issues, including those occurring in the inner race, outer race, and balls. An accelerometer (type B & K 4366) was connected to a data acquisition device (IDAC-6C) to record vibration signals under different operating conditions. Furthermore, a tachometer equipped with an LCD display is employed to measure the rotational speed. Four types of defects in ball bearing (Koyo 1205C3 type) were studied, the slot in outer race with size 0.196 mm, the slot in inner race with size 0.191 mm, in ball with size 0.196 mm in additional to compound defect. In this paper, spectral correlation technique was employed to detect defects in ball bearings running at varying speed, along with spectral coherence and the corresponding Enhanced Envelope Spectrum (EES) in frequency-order domain and order-order domain. The results show that the adopted methods, that are used to analyze the real vibration signals for diagnosis the defected ball bearings, are suitable, accurate and less processing time for varying speed. The processing time of the FastACP method used to analyze the signals in order- order domain is less than that of the adopted method in the frequency-order domain for any defect type. Overall, using the FastACP method in the order-order domain significantly reduced processing time by approximately 27% compared to the adopted method in the frequency- order domain under varying speed conditions.
The accurate prediction of machinery faults is considered an effective strategy to increase the operation life of machines, ensure smooth operation, and provide a safe environment. Accordingly, the demands on predictive tools such as machine learning to detect machinery faults before catastrophic failure occurs has increased rapidly. In this research, a diagnosis algorithm based on using a 2D color-coded map as the input to a deep artificial neural network is proposed. These maps are called RDEgram after the processing of vibrational signals based on reverse dispersion entropy (RDE) method. The effectiveness of the proposed algorithm is investigated by testing its capability to detect different faults located at different locations on ball bearings under constant speed conditions. First, the squared envelope signal is extracted by applying the short time Fourier transform to vibration signal. Then, the RDE is used to process the squared envelope to detect the range of frequencies at which the transients occur. The RDEgram color-coded map is used to represent the RDE values as a function of frequency and frequency resolution. The maps from different fault features are collected to form the diagnostic patterns. Finally, a pretrained convolutional neural network (CNN) is applied to learn the feature pattern and diagnose the bearing faults. The CNN is trained using fixed- speed data and then it is applied to diagnose faults in the test data recorded at the same speed. The prediction method adopted in the current research shows a 100% level of accuracy for predicting two types of faults (pit and slot) located at various positions a ball bearing (KOYO 1205 C3 type) running at two constant speeds (25 and 30 Hz).