Classifier fusion of vibration and acoustic signals for fault diagnosis and classification of planetary gears based on Dempster–Shafer evidence theory

Abstract

Nowadays, the ever increasing need for higher accuracy, reliability and security in modern industries has given rise intensively to the use of multi-sensor data fusion method in fault diagnosis of industrial equipment. In this article, an effective and powerful method for precise fault diagnosis of planetary gearbox based on fusion of vibration and acoustic data using the Dempster–Shafer theory is presented. For this purpose, the vibration and acoustic signals in different modes of the gears were first received simultaneously by two separate sensors and then were transmitted from time domain to time–frequency domain using wavelet analysis. After signal processing, each sensor's data were transferred to a local classifier for primary fault diagnosis. Local classification was performed by artificial neural network classifier. The outputs of the local classification were used as the inputs into Dempster–Shafer rules for fusion of classifiers and achieving the final accuracy of the classification. In primary fault diagnosis, the accuracy of fault classification based on vibration and acoustic signals was obtained as 86% and 88%, respectively. After incorporating the outcomes of two sensors, the final accuracy of the classification was calculated as 98% which indicates a 10% jump compared to single-sensor mode. These results indicate the effectiveness of the data fusion method in condition monitoring and fault diagnosis of the equipment. Moreover, in this article, the capability of Dempster–Shafer theory in the fusion of uncertain data and the increase of accuracy in the classification was demonstrated to a quiet acceptable level.

Spark plug fault recognition based on sensor fusion and classifier combination using Dempster–Shafer evidence theory

“…It should be noted that after testing different signal processing methods such as discrete wavelet transform (DWT), it was found that the best recognition accuracy was achieved by applying these features to the de-noised signals in time-domain, as also reported in [26]. The 17-feature parameters are shown in Table 3 [27]. It is noted that these features can be used for fault diagnosis of all mechanical components, and do not belong a specific system.…”

“…The most important layer in ANN design is hidden layer that should be defined by trial and error [27]. The number of hidden layer neurons can be determined by iterative or random selection.…”

Integration of Artificial Intelligence into Dempster Shafer Theory: A Review on Decision Making in Condition Monitoring

“…Yibo et al in [3] have demonstrated the effectiveness of data fusion between vibration and oil analysis using D-S with BPA from BPNN output and for fault diagnosis of gear box. While Khazaee et al in [4] have reported an increase of accuracy in machinery fault diagnosis via ANN-…”

Two Stage Data Fusion of Acoustic, Electric and Vibration Signals for Diagnosing Faults in Induction Motors

“…Methods of fusing multiple signals for condition monitoring purposes have been reported by Safizadeh and Latifi [2], who fused vibration and load signals to diagnose bearing faults in induction motors, Khazee et al [3], who combined vibration and acoustic signals to diagnose faults in gears, Yang and Kim [5], who fused vibration and current signals and Loutasa et al [6], who combined vibration, acoustic and oil-debris signals to diagnose faults in gears. In each case, data fusion was shown to increase the accuracy of the system.…”

“…Recently several new condition monitoring methods have been developed, which fuse different types of signals in order to achieve better accuracy in identifying faults. It has been observed in multiple works (see for example [1]- [6]) that different signals are most informative for different faults: vibration signals are well suited for monitoring bearing faults [1], [2], current signals are good for detecting problems such as broken rotor bars or eccentricity relatedfaults [1], [3] and capacitor sensors may be used to measure the partial discharges that are indicative of stator winding insulation problems [1], [4].…”