Multi-Sensor Data Fusion Using a Relevance Vector Machine Based on an Ant Colony for Gearbox Fault Detection

Abstract

Sensors play an important role in the modern manufacturing and industrial processes. Their reliability is vital to ensure reliable and accurate information for condition based maintenance. For the gearbox, the critical machine component in the rotating machinery, the vibration signals collected by sensors are usually noisy. At the same time, the fault detection results based on the vibration signals from a single sensor may be unreliable and unstable. To solve this problem, this paper proposes an intelligent multi-sensor data fusion method using the relevance vector machine (RVM) based on an ant colony optimization algorithm (ACO-RVM) for gearboxes’ fault detection. RVM is a sparse probability model based on support vector machine (SVM). RVM not only has higher detection accuracy, but also better real-time accuracy compared with SVM. The ACO algorithm is used to determine kernel parameters of RVM. Moreover, the ensemble empirical mode decomposition (EEMD) is applied to preprocess the raw vibration signals to eliminate the influence caused by noise and other unrelated signals. The distance evaluation technique (DET) is employed to select dominant features as input of the ACO-RVM, so that the redundancy and inference in a large amount of features can be removed. Two gearboxes are used to demonstrate the performance of the proposed method. The experimental results show that the ACO-RVM has higher fault detection accuracy than the RVM with normal the cross-validation (CV).

Faults Diagnosis of Self-validating Air Data Sensing System Based on Ensemble EMD and Multiclass RVM

“…Firstly, the SVADS system concludes multiple pressure measuring channels, and each channel has complex faults modes such as the blockage of pressure ports, pipe pressure leak, faults of pressure sensors and faults of circuits; secondly, the faults feature of each mode is nonlinear, which the difficulty in faults feature extraction is bigger; thirdly, the mount of actual faults data is very little, and the small sample problem needs to be solved; lastly, the obtained fault information includes some uncertain extent and the classification results should be uncertain. Presently, some faults diagnosis algorithms have been proposed based on empirical mode decomposition (EMD) [10], neural networks [11], and relevance vector machine (RVM) [12]. Based on the adaptive decomposition of signals in frequency domain, EMD has applied to some feature exaction; however, its mode mixing problem is not suitable for nonlinear faults feature exaction of this paper.…”

Diagnosis methodology for identifying gearbox wear based on statistical time feature reduction

“…As aforementioned, in the proposed work, the acquired vibrations signals belong to those vibrations in the perpendicular plane of the gearbox rotating axis since some studies has reported that the occurrence of perpendicular vibrations on the rotating axis is related to the inappropriate working conditions of rotational machines [5], [13], [24]. Regarding the proposed methodology, the data acquisition is performed by carrying out different experiments at different operating frequencies for driving the induction motor: 5 Hz, 15 Hz and 50 Hz.…”

“…Despite their reliability, the appearance of unexpected faults in gearboxes may occur at any time, causing unscheduled breakdowns in the elements of the associated kinematic chain. It has been reported that the appearance of gear faults account for 80% of the breakdowns in transmission machinery systems and 10% of the faults in rotating machinery [4]- [5]. Therefore, strategies of condition monitoring and fault detection related to gearbox transmission systems play a key role to ensure the effectivity and safety of multiple industrial processes [6]- [8].…”

Research on Mechanical Fault Prediction Method Based on Multifeature Fusion of Vibration Sensing Data

“…Presently, widely-used mechanical fault prediction methods employ artificial neural networks (ANNs) [4][5][6], support vector machines (SVMs) [7,8], deep learning [9][10][11], and other artificial intelligence (AI) technologies. For example, Ben et al [12] proposed the use of empirical mode decomposition and energy entropy for feature extraction, which was combined with an ANN for multifeature fusion to make bearing fault predictions.…”