Search results

This study aims to perform the experimental work on a laboratory-constructed steel truss bridge model on which hammer blows are applied for excitation. The vibration response signals of the bridge structure are collected using sensors placed at different nodes. The different damaged states such as no damage, single damage, double damage and triple damage are introduced by cutting members of the bridge. The masked noise with recorded vibration responses generates challenge to properly analyze the health of bridge structure.

The analytical modal properties are obtained from finite element model (FEM) developed using SAP2000 software. The response signals are analyzed in frequency domain by power spectrum and in time-frequency domain using spectrogram and Stockwell transform. Various low pass signal-filtering techniques such as variational filter, lowpass sparse banded (AB) filter and Savitzky–Golay (SG) differentiator filter are also applied to refine vibration signals. The proposed methodology further comprises application of Hilbert transform in combination with MUSIC and ESPRIT techniques.

The outcomes of SG filter provided the denoised signals using appropriate polynomial degree with proper selected window length. However, certain unwanted frequency peaks still appeared in the outcomes of SG filter. The SG-filtered signals are further analyzed using fused methodology of Hilbert transform-ESPRIT, which shows high accuracy in identifying modal frequencies at different states of the steel truss bridge.

The sequence of proposed methodology for denoising vibration response signals using SG filter with Hilbert transform-ESPRIT is a novel approach. The outcomes of proposed methodology are much refined and take less computational time.

This study aims to include the diagnosis of an old concrete deck steel truss rural road bridge in the damaged and retrofitted state through vibration response signals.

The analysis of the vibration response signals is performed in time and time-frequency domains using statistical features-root mean square, impulse factor, crest factor, kurtosis, peak2peak and Stockwell transform. The proposed methodology uses the Hilbert transform in combination with spectral kurtosis and bandpass filtering technique for obtaining robust outcomes of modal frequencies.

The absence or low amplitude of considered mode shape frequencies is observed both before and after retrofitting of bridge indicates the deficient nodes. The kurtosis feature among all statistical approaches is able to reflect significant variation in the amplitude of different nodes of the bridge. The Stockwell transform showed better resolution of present modal frequencies but due to the yield of additional frequency peaks in the vicinity of the first three analytical modal frequencies no decisive conclusions are achieved. The methodology shows promising outcomes in eliminating noise and visualizing distinct modal frequencies of a steel truss bridge.

The findings of the present study help in analyzing noisy vibration signals obtained from various structures (civil or mechanical) and determine vulnerable locations of the structure using mode shape frequencies.

The literature review gave an insight into few experimental investigations related to the combined application of Hilbert transform with spectral kurtosis and bandpass filtering technique in determining mode frequencies of a steel truss bridge.

Taking into consideration the current human need for agricultural produce such as rice that requires water for growth, the optimal consumption of this valuable liquid is important. Unfortunately, the traditional use of water by humans for agricultural purposes contradicts the concept of optimal consumption. Therefore, designing and implementing a mechanized irrigation system is of the highest importance. This system includes hardware equipment such as liquid altimeter sensors, valves and pumps which have a failure phenomenon as an integral part, causing faults in the system. Naturally, these faults occur at probable time intervals, and the probability function with exponential distribution is used to simulate this interval. Thus, before the implementation of such high-cost systems, its evaluation is essential during the design phase.

The proposed approach included two main steps: offline and online. The offline phase included the simulation of the studied system (i.e. the irrigation system of paddy fields) and the acquisition of a data set for training machine learning algorithms such as decision trees to detect, locate (classification) and evaluate faults. In the online phase, C5.0 decision trees trained in the offline phase were used on a stream of data generated by the system.

The proposed approach is a comprehensive online component-oriented method, which is a combination of supervised machine learning methods to investigate system faults. Each of these methods is considered a component determined by the dimensions and complexity of the case study (to discover, classify and evaluate fault tolerance). These components are placed together in the form of a process framework so that the appropriate method for each component is obtained based on comparison with other machine learning methods. As a result, depending on the conditions under study, the most efficient method is selected in the components. Before the system implementation phase, its reliability is checked by evaluating the predicted faults (in the system design phase). Therefore, this approach avoids the construction of a high-risk system. Compared to existing methods, the proposed approach is more comprehensive and has greater flexibility.

By expanding the dimensions of the problem, the model verification space grows exponentially using automata.

Unlike the existing methods that only examine one or two aspects of fault analysis such as fault detection, classification and fault-tolerance evaluation, this paper proposes a comprehensive process-oriented approach that investigates all three aspects of fault analysis concurrently.

The purpose of this paper is to propose a new fault feature extraction scheme for the rolling element bearing.

The generalized Stockwell transform (GST) and the singular value ratio spectrum (SVRS) methods are combined. A time-frequency distribution measurement criterion named the energy concentration measurement (ECM) is initially used to determine the parameter of the optimal GST method. Then, the optimal GST is applied to conduct a time-frequency transformation for a raw signal. Subsequently, the two-dimensional time-frequency matrix is obtained. Finally, the improved singular value decomposition (SVD) analysis is used to conduct a noise reduction of the time-frequency matrix. The SVRS is proposed to select the effective singular values. Furthermore, the time-domain feature of the impact signal is obtained by taking the inverse GST transform.

The simulated and experimental signals are used to verify the superiority of the proposed method over conventional methods. The obtained results show that the proposed method can effectively extract fault features of the rolling element bearing.

This paper mainly discusses the application of GST and SVRS methods to analyze the weak fault feature extraction problem. The next research direction is to explore the application of the Hilbert Huang transform (HHT) and variational modal decomposition (VMD) in the impact feature extraction of rolling bearing.

In the present study, a new SVRS method is proposed to select the number of effective singular values. This paper proposed an effective way to obtain the fault feature in monitoring of rotating machinery.

The purpose of this paper is to compare the accuracy of tracking the amplitude and frequency changes of non‐stationary electric signals.

Short‐time fourier transform (STFT) and S‐transform algorithms were applied to analyze non‐stationary signals originating from switching of capacitor banks in a power system.

The S‐transform showed possibilities of sharp localization of the basic component, and allowed improvement of tracking dynamism the transient components in comparison to STFT.

S‐transform is a better tool for the analysis of non‐stationary waveforms in power systems and its properties can be used for diagnostic and power quality applications.

The dynamic tracking of the changes in time and frequency of real‐like signals originating from a power system are investigated in this paper.

Noise abatement is one of the key techniques for Partial Discharge (PD) on-line measurement and monitoring. However, how to enhance the efficiency of PD signal noise suppression is a challenging work. Hence, this study aims to improve the efficiency of PD signal noise abatement.

In this approach, the time–frequency characteristics of PD signal had been obtained based on fast kurtogram and S-transform time–frequency spectrum, and these characteristics were used to optimize the parameters for the signal matching over-complete dictionary. Subsequently, a self-adaptive selection of matching atoms was realized when using Matching Pursuit (MP) to analyze PD signals, which leading to seldom noise signal element was represented in sparse decomposition.

The de-noising of PD signals was achieved efficiently. Simulation and experimental results show that the proposed method has good adaptability and significant noise abatement effect compared with Empirical Mode Decomposition, Wavelet Threshold and global signal sparse decomposition of MP.

A self-adaptive noise abatement method was proposed to improve the efficiency of PD signal noise suppression based on the signal sparse representation and its MP algorithm, which is significant to on-line PD measurement.

This study aims to reveal the essential characteristics of nonstationary signals and explore the high-concentration representation in the joint time–frequency (TF) plane.

In this paper, the authors consider the effective TF analysis for nonstationary signals consisting of multiple components.

To make it, the authors propose the combined multi-window Gabor transform (CMGT) under the scheme of multi-window Gabor transform by introducing the combination operator. The authors establish the completeness utilizing the discrete piecewise Zak transform and provide the perfect-reconstruction conditions with respect to combined TF coefficients. The high-concentration is achieved by optimization. The authors establish the optimization function with considerations of TF concentration and computational complexity. Based on Bergman formulation, the iteration process is further analyzed to obtain the optimal solution.

With numerical experiments, it is verified that the proposed CMGT performs better in TF analysis for multi-component nonstationary signals.

An electrical power system is expected to deliver undistorted sinusoidal, rated voltage and current continuously to the end-users. The problem of power quality (PQ) occurs when there is (are) deviation(s) in voltage and/or current which cause(s) failure or mal-operation of the customer's equipments. Various methods are suggested to detect and classify single PQ event in a power system, the performance of such methods to classify composite PQ events is limited. The purpose of this paper is the classification of composite PQ events in emerging power systems.

This paper proposes an effective method to classify composite PQ events using Hilbert Huang transform (HHT). The performance of probabilistic neural network (PNN) classifier and support vector machine (SVM) classifier to efficiently classify composite PQ events is compared.

The features extracted from HHT are simple yet effective. SVMs and PNN classifiers are used for PQ classification. It is found that PNN classifier outperforms SVM with the classification accuracy of 100 percent.

Different PQ signals used for analysis are generated by simulating a practical distribution system of an Indian academic institution.

This work proposes a tertiary wavelet model based automatic epilepsy classification system using electroencephalogram (EEG) signals.

In this paper, a three-stage system has been proposed for automated classification of epilepsy signals. In the first stage, a tertiary wavelet model uses the orthonormal M-band wavelet transform. This model decomposes EEG signals into three bands of different frequencies. In the second stage, the decomposed EEG signals are analyzed to find novel statistical features. The statistical values of the features are demonstrated using multi-parameters graph comparing normal and epileptic signals. In the last stage, the features are inputted to different conventional classifiers that classify pre-ictal, inter-ictal (epileptic with seizure-free interval) and ictal (seizure) EEG segments.

For the proposed system the performance of five different classifiers, namely, KNN, DT, XGBoost, SVM and RF is evaluated for the University of BONN data set using different performance parameters. It is observed that RF classifier gives the best performance among the above said classifiers, with an average accuracy of 99.47%.

Epilepsy is a neurological condition in which two or more spontaneous seizures occur repeatedly. EEG signals are widely used and it is an important method for detecting epilepsy. EEG signals contain information about the brain's electrical activity. Clinicians manually examine the EEG waveforms to detect epileptic anomalies, which is a time-consuming and error-prone process. An automated epilepsy classification system is proposed in this paper based on combination of signal processing (tertiary wavelet model) and novel features-based classification using the EEG signals.

The individual machine learning methods used for fault detection and classification have accuracy performance at a certain level. A combined learning model composed of different base classifiers rather than an individual machine learning model is introduced to ensure diversity. In this way, this study aims to improve the generalization capability of fault detection and classification scheme.

This study presents a probabilistic weighted voting model (PWVM) with multiple learning models for fault detection and classification. The working principle of this study’s proposed model relies on weight selection and per-class possibilities corresponding to predictions of base classifiers. Moreover, it can improve the power of the prediction model and cope with imbalanced class distribution through validation metrics and F-score.

The performance of the proposed PWVM was better than the performance of the individual machine learning methods. Besides, the proposed voting model’s performance was compared with different voting mechanisms involving weighted and unweighted voting models. It can be seen from the results that the presented model is superior to voting mechanisms. The performance results revealed PWVM has a powerful predictive model even in noisy conditions. This study determines the optimal model from among voting models with the prioritization method on data sets partitioned different ratios. The obtained results with statistical analysis verified the validity of the proposed model. Besides, the comparative results from different benchmark data sets verified the effectiveness and robustness of this study’s proposed model.

The contribution of this study is that PWVM is an ensemble model with outstanding generalization capability. To the best of the authors’ knowledge, no study has been performed using a PWVM composed of multiple classifiers to detect no-faulted/faulted cases and classify faulted phases.