A Cost-Efficient MFCC-Based Fault Detection and Isolation Technology for Electromagnetic Pumps

Abstract

Fluid pumps serve critical purposes in hydraulic systems so their failure affects productivity, profitability, safety, etc. The need for proper condition monitoring and health assessment of these pumps cannot be overemphasized and this has resulted in extensive research studies on standard techniques for ensuring optimum fault detection and isolation (FDI) results for these pumps. Interestingly, mechanical vibrational signals reflect operating conditions and by exploring the robust time–frequency-domain feature extraction techniques, the underlying nonlinear characteristics can be captured for reliable fault diagnosis/condition assessment. This study is based on the use of vibrational signals for fault isolation of electromagnetic pumps. From the vibrational signals, Mel frequency cepstral coefficients (MFCCs), the first-order and the second-order differentials were extracted and the salient features selected by a rank-based recursive feature elimination (RFE) of uncorrelated features. The proposed framework was tested and validated on five VSC63A5 electromagnetic pumps at various fault conditions and isolated/classified using the Gaussian kernel SVM (SVM-RBF-RFE). Results show that the proposed feature selection approach is computationally cheaper and significantly improves diagnostics performance. In addition, the proposed framework yields a comparatively better diagnostics results on electromagnetic pumps in comparison with other diagnostics methods, hence a more reliable diagnostics tool for electromagnetic pumps.

D-dCNN: A Novel Hybrid Deep Learning-Based Tool for Vibration-Based Diagnostics

“…Vibration monitoring is one of the most popular (and reliable) FDI techniques for industrial purposes. Its success is highly attributed to the availability of time domain, frequency domain, and time-frequency domain (TFD) signal processing techniques for discriminative feature engineering-hand-crafted feature extraction, selection, and manipulation [3,4]. For instance, the invention of TFD signals processing techniques like the short-time Fourier transform (STFT), empirical mode decomposition (EMD), wavelet transform (WT), Mel frequency cepstral coefficients (MFCCs), etc.…”

“…For instance, the invention of TFD signals processing techniques like the short-time Fourier transform (STFT), empirical mode decomposition (EMD), wavelet transform (WT), Mel frequency cepstral coefficients (MFCCs), etc. provided strong comparative efficiencies over the more traditional statistical time domain and frequency domain techniques for vibration monitoring [4,5]. Against their efficiencies, the reliability of all these hand-crafted features is limited by expensive statistical assumptions and trade-offs.…”

Gearbox Failure Diagnosis Using a Multisensor Data-Fusion Machine-Learning-Based Approach

“…The achieved accuracy results given in Table 3 highlight the improvement brought by MFCC (about 7%) compared to that of the LPC-LSTM methodology. Confusion-matrix analysis in Figures 17-19 confirmed the better classification tendency of the failure majority because of MFCC spectrum representation providing more time and frequency details from nonlinear and nonstationary signals [29], in addition to CNNs, which are known for their strong ability to extract useful features. Despite the improvement in accuracy, this approach deals with computational-burden issues related to the convolutional layers' slow training [42] due to successive convolutional operations during training (convolution, pooling, etc).…”

“…Indeed, as shown in [23,24], it could be more efficient to add convolution layers to benefit from their ability to extract useful information for diagnosis [25,27] or filter the signals through an autoencoder network [26], but signal processing remain the step of choice for improving the accuracy of the failure-detection process. For fault detection and isolation, Ugochukwu et al [29] proposed to extract useful features using MFCC. Most discriminant features were then chosen as input for the classification using SVM.…”

Automatic Database Segmentation using Hybrid Spectrum -Visual Approach

“…The first order of MFCC (∆-MFCC) and the second-order derivatives of MFCC (∆∆-MFCC) are then added, and these are referred to as differential (13-coefficients) and acceleration, (13-coefficients) respectively. Equation ( 5) [21] is used to get the ∆-MFCC coefficients where (n=1), while its derivative yields the ∆∆-MFCC coefficients, (n=2). Where, dt is the ∆ coefficient at time t, from frame t computed in terms of the static coefficients − , + and N is the window size of the delta.…”