جستجوی مقالات مرتبط با کلیدواژه « fault diagnosis » در نشریات گروه « مکانیک »

This research pioneers the application of thermographic principles to diagnose faults, specifically cavitation and air entrainment, in centrifugal pumps. The study represents the inaugural investigation into the feasibility of leveraging infrared thermography for this purpose, underpinned by rigorous experimental methodologies to validate its efficacy. By capturing thermal images of pumps operating under varying conditions, a pseudo-coloring technique for precise temperature range segmentation was employed. This technique facilitated the assessment of fault severity, quantified through the computation of the . This index emerged as a quantifiable metric of fault severity, with elevated values correlating to more pronounced degrees of fault occurrence. Notably, in the case of air entrainment faults, a maximum temperature escalation of 3.9°C was recorded after 60 min run time, and the corresponding thermal index was found to be 5.12. The investigation employs the HSV model, extracting regions of thermal variation through hue differences for fault detection. This process is combined with edge detection methods like Sobel, Prewitt, Roberts, Canny, and Otsu. The Otsu technique consistently outperformed alternative approaches. Specifically, for high cavitation and air entrainment faults, the Otsu method had the highest mean of 0.1730 and 0.1253, respectively. Key findings include the effectiveness of image processing techniques, statistical measures, and edge detection methods for fault diagnosis, as well as insights into temperature differentials and motor load reductions with increasing fault severity. The research improves maintenance, enhancing efficiency and reducing downtime. It emphasizes infrared thermography's potential for fault diagnosis while identifying constraints and advocating further research.

This paper is about the application of instantaneous angular speed (IAS) signal in a 3-liter six-cylinder gasoline engine. The study is in continuation of former work in which a measurement system was developed for this signal on a rotating machine. The future trend of the research is to measure IAS in an I.C. engine. Therefore, the objective of the current work is to provide a verified software tool which can run simulated experiments with IAS signal output under healthy/faulty conditions. An engine model with detailed crankshaft elements is established in the GT-SUITE[1]</sup> software. Under the GT-SUITE environment, IAS signal output is obtained through simulated experiments. In order to validate the tool, the first torsional natural frequency of the crankshaft obtained from frequency analysis on the IAS signal is compared with the result of modal analysis on the crankshaft structure using the F.E. method. Also, the value is compared with the prediction from the GT CrankAnalysis module. A good match is found, which shows the validity of the developed software tool. Faulty condition of misfiring in one cylinder is simulated using this tool, and expected observations on the IAS output signal are verified to address the future trend of the research using the developed tool in this study.

In this article, an intelligent system is introduced in order to detection and classification of some common mechanical faults of an engine alternator based on the frequency analysis of vibration signals. For this purpose, firstly the vibration signals of an alternator under four conditions, including healthy, bearing corrosion, cracked rotor and unbalanced excited shaft, were captured by an accelerometer. Time-domain signals were then transformed into frequency-domain with the aid of FFT. At the next step, power spectral density (PSD) method was used for the secondary frequency signal processing level. Afterward, in data mining step, twelve statistical features were extracted from the PSD values of the signals, which were fed as the input data into the ANN classifier to detect and classify the alternator faults. The results indicate that the proposed method has the capable of detecting the different alternator faults with an accuracy higher than 92%.

In this paper, ‎the multi-sensor fault diagnosis in the exhaust temperature sensors of a V94.2 heavy duty gas turbine is presented‎. ‎A Laguerre network-based fuzzy modeling approach is presented to predict the output temperature of the gas turbine for sensor fault diagnosis‎. Due to the nonlinear dynamics of the gas turbine, in these models the Laguerre filter parts are related to the linear dynamic part of the models and ‎the nonlinear parts of models are considered as neuro-fuzzy models. In order to deal with the dimensionality problems associated with fuzzy models‎, ‎the nonlinear parts of models are considered as hierarchical fuzzy systems. In the residual evaluation phase, model error modeling adaptive threshold approach is used to increase fault detection robustness against the noise and disturbance. A new expert fuzzy system by multi-sensor information fusion is presented for the fault diagnosis system‎, ‎which can examine the performance of all the sensors simultaneously‎. The result shows that the proposed fault diagnosis system could considerably increase reliability and safety‎.

This paper present new method based on adaptive diagnosis, which can self-adaptively diagnose faults of bearings under varying operation conditions without any prior information. For this purpose, a new method using information-geometric support vector machine (IG-SVM) based on Empirical Mode Decomposition and Singular Value Decomposition (EMD–SVD) is presented. Firstly, the vibration signal is decomposed to singular features by the EMD-SVD. Then, the IG-SVM, which uses information geometry to modify SVM in a data-dependent way, is employed for fault clustering. The results show that the proposed method has an efficient approach for fault diagnosis of bearings.

Identifying fault categories, especially for compound faults, is a challenging task in mechanical fault diagnosis. For this task, this paper proposes a novel intelligent method based on wavelet packet transform (WPT) and multiple classifier fusion. An unexpected damage on the gearbox may break the whole transmission line down. It is therefore crucial for engineers and researchers to monitor the health condition of the gearbox in a timely manner to eliminate the impending faults. However, useful fault detection information is often submerged in heavy background noise. The non-stationary vibration signals were analyzed to reveal the operation state of the gearbox. The proposed method is applied to the fault diagnosis of gears and bearings in the gearbox. The diagnosis results show that the proposed method is able to reliably identify the different fault categories which include both single fault and compound faults, which has a better classification performance compared to any one of the individual classifiers. The vibration dataset is used from a test rig in Shahrekord University and a gearbox from Sepahan Cement. Eventually, the gearbox faults are classified using these statistical features as input to WSVM.

A classification technique using Support Vector Machine (SVM) classifier for detection of rolling element bearing fault is presented here. The SVM was fed from features that were extracted from of vibration signals obtained from experimental setup consisting of rotating driveline that was mounted on rolling element bearings which were run in normal and with artificially faults induced conditions. The time-domain vibration signals were divided into 40 segments and simple features such as peaks in time domain and spectrum along with statistical features such as standard deviation, skewness, kurtosis etc. were extracted. Effectiveness of SVM classifier was compared with the performance of Artificial Neural Network (ANN) classifier and it was found that the performance of SVM classifier is superior to that of ANN. The effect of pre-processing of the vibration signal by Discreet Wavelet Transform (DWT) prior to feature extraction is also studied and it is shown that pre-processing of vibration signal with DWT enhances the effectiveness of both ANN and SVM classifiers. It has been demonstrated from experiment results that performance of SVM classifier is better than ANN in detection of bearing condition and pre-processing the vibration signal with DWT improves the performance of SVM classifier.

This paper concentrates on a new procedure which experimentally recognises gears and bearings faults of a typical gearbox system using a least square support vector machine (LSSVM). Two wavelet selection criteria Maximum Energy to Shannon Entropy ratio and Maximum Relative Wavelet Energy are used and compared to select an appropriate wavelet for feature extraction. The fault diagnosis method consists of three steps, firstly the six different base wavelets are considered. Out of these six wavelets, the base wavelet is selected based on wavelet selection criterion to extract statistical features from wavelet coefficients of raw vibration signals. Based on wavelet selection criterion, Daubechies wavelet and Meyer are selected as the best base wavelet among the other wavelets considered from the Maximum Relative Energy and Maximum Energy to Shannon Entropy criteria respectively. Finally, the gearbox faults are classified using these statistical features as input to LSSVM technique. The optimal decomposition level of wavelet is selected based on the Maximum Energy to Shannon Entropy ratio criteria. In addition to this, Energy and Shannon Entropy of the wavelet coefficients are used as two new features along with other statistical parameters as input of the classifier. Some kernel functions and multi kernel function as a new method are used with three strategies for multi classification of gearboxes. The results of fault classification demonstrate that the LSSVM identified the fault categories of gearbox more accurately with multi kernel and OAOT strategy.

Rotating machinery is the most common machinery in industry. The root of the faults in rotating machinery is often faulty rolling element bearings. This paper presents a technique using optimized artificial neural network by the Bees Algorithm for automated diagnosis of localized faults in rolling element bearings. The inputs of this technique are a number of features (maximum likelihood estimation values), which are derived from the vibration signals of test data. The results show that the performance of the proposed optimized system is better than most previous studies, even though it uses only two features. Effectiveness of the above method is illustrated using obtained bearing vibration data.