Gearbox fault diagnosis

Dişli çarklar güç ve hareket iletimi için endüstrinin her alanında, değişik ortam ve koşullarda kullanılan makine elemanlarıdır. Makinelerin bu önemli elemanlarının zorlayıcı şartlar altında çalışması sonucu, dişilerde hasarlar meydana gelmekte ve daha büyük olumsuzlukları önleyebilmek için bu hasarların bilinmesi ve oluşum nedenlerinin tespit edilmesi gerekmektedir. Bu çalışmada amaç, dişli arızalarının sebep olduğu titreşimlerin istatistiksel proses kontrol grafiklerinde analiz edilerek düz dişli hata tespitini gerçekleştirmektir. Bu amaçla bir deney düzeneği kurulmuş, önce sağlam dişlilerin kullanılması ile farklı yükleme ve devir sayıları durumları için rulmanlı yataklar üzerinden radyal yöndeki titreşimler kaydedilerek referans spektrumları elde edilmiştir. Daha sonra sağlam dişliler üzerinde yapılan işlemlerle suni olarak aşınma ve kırık hasarları oluşturulmuştur. Düzenek çeşitli ön yük ve devir sayılarında çalıştırılarak rulman yataklarındaki titreşimler kaydedilmiştir. Bu titreşimlerin frekans spektrumları ve istatistiksel proses kontrol grafikleri çizdirilmiştir. Sağlam ve hasarlı dişli durumları için oluşturulan grafikler karşılaştırıldığında aşınmış dişli, tek dişi kırık pinyon dişli ve tek dişi kırık pinyon ve tekerlek dişli hasarlarının varlığı tespit edilebilmiştir.

The object of this paper is to propose a new phase-cycle analysis for gear and bearing diagnostics of rotating machinery which synthesises the theory of angle-time cyclostationary (AT-CS). The motivation for this research came from bearing diagnostics which is confronted by the inevitable smearing phenomenon due to bearing slip. The bearing signal under non-stationary conditions is neither jointly second-order cyclostationary (CS2) nor pseudo-cyclostationary (pseudo-CS) and, hence, most CS2 tools are subjected to loss of diagnostics information. The proposed phase-cycle analysis is believed to propound the study of the kinematics of gears and bearings, and it will significantly modify the prevailing historical interpretation of the bearing slip. In this paper, the phase-cycle analysis which is based on the theory of AT-CS is named the “cyclomap” because it can provide much more information about the kinematics of cyclostationary and cyclo-non-stationary signals. The cyclomap shows some interesting features in delineating the energy distribution of the kinematical properties of machines at an angular period of interest. In order to further study the properties of the cyclomap, it is systematically compared to the well-known order-frequency cyclic modulation coherence in diagnosing the CS2, pseudo-CS, and second-order cyclo-non-stationary signals of the experimental datasets.

An analysis of gearbox vibration signals is almost always the default choice when diagnosing the condition of a gearbox because of the rich information contained in the vibration signals and their ease of measurement. Gearbox vibration signals are mostly non-stationary owing to uncertainties associated with the drive and load mechanisms. The non-stationary nature of gearbox vibration signals is evident from an unequal number of samples between successive tachometer pulses. In the present work, independent angular re-sampling (IAR) technique is employed to convert non-stationary vibration signals in the time domain into quasi-stationary signals in the angular domain. The resulting angular domain signals for each gear health condition are merged and then partitioned into a number of data samples which are decomposed using wavelet packet transform. Standard deviation of wavelet packet coefficients is then computed and the distance evaluation technique employed to determine the frequency bands which have a higher distance evaluation factor. Normalized standard deviation values from the optimal frequency bands are then used to train and test an adaptive neurofuzzy inference system (ANFIS). Promising results are obtained when the proposed method is employed to diagnose common faults such as a cracked tooth, chipped tooth and missing tooth in a single stage spur gearbox under fluctuating speed conditions. Gearbox fault diagnosis under fluctuating speed conditions is a more realistic scenario where very limited work has been accomplished.

Signal processing for electrocardiogram (ECG) records cardiac activity to unveil any abnormality in the heart through electrocardiograph. The pictorial representation comes in graph to indicate electric potential changes occurring between electrodes when patients’ cardiovascular state is being examined. The electrical functioning of the heart is translated into a waveform, being utilized to find the heart condition. An ECG signal tracks heart diseases, such as poor blood flow to the heart and structural abnormalities. Analysis of ECG signal and removal of interference for clinical diagnosis in presented in this paper.

Most of the mechanical systems use gears as means to transfer power from one shaft to another. Due to continuous operation and heavy load, they tend to develop some faults, which if not treated properly create a permanent damage to the structure. This eventually affects the efficiency of the system, which creates an alarming situation and thus need to be looked into. Various research tasks have already been taken up to study these faulty conditions and their effects. In this study, the vibrations created during the meshing of teeth of gears are used for acquiring the data set. Machine learning algorithms are then applied to this data set to achieve the information which then can be interpreted into a form which people can easily understand. Here, logistic function classifier and REP tree are employed to obtain the output. The whole process was executed in a series of steps namely feature/data extraction, classification of data and then the results obtained are further discussed.

The vibration signal of a gearbox carries abundant information about its condition and is widely utilized to diagnose the condition of the gearbox. Most of the research efforts in diagnosing gearbox faults, however, assume the acquired vibration signal to be stationary. This paper attempts to highlight the non-stationary nature of gearbox vibration signals owing to uncertainties of the drive and load mechanisms and to synchronize such signals from the revolution point of view. This synchronization is achieved by a simple process referred to as the independent angular re-sampling (IAR) technique. The basis of the IAR technique is the assumption of constant angular acceleration over each independent revolution of the gearbox drive shaft. Through the IAR process, non-stationary signals sampled at constant time increments are converted into quasi-stationary signals sampled at constant angular increments. The angular domain signals obtained from the IAR technique are merged to generate the synchronized angular domain vibration signal for the complete rotational period. The angular domain signal for each gear health condition is partitioned into a number of data samples which are then analyzed using wavelet packet decomposition. Standard deviation values of wavelet packet coefficients are then computed and the resulting vectors utilized as inputs to a multilayer perceptron neural network for identifying the condition of the gearbox.

—Automatic fault diagnosis is an inseparable part of today's electromechanical systems. Advanced signal processing and machine learning techniques are required to address variabilities and uncertainties associated with the monitoring signals. In this paper, deep neural networks are employed to diagnose five classes of gearbox faults applied to three common monitoring signals, i.e. vibration, acoustic and torque. Discrete wavelet transform is used to provide the initial features as the inputs of the network. A test-rig based on a 250W three-phase squirrel cage induction machine shaft connected to a single stage helical gear is built for validation of the proposed method. The experimental results indicate accurate fault diagnosis in various conditions such as different modalities, signal variabilities, and load conditions.

This paper proposes a novel technique for the condition monitoring of gearboxes based on a self-organizing feature maps (SOF M ) network. In order to visualize the learned SOF M results more clearly, an improved method based on the uni ed distance matrix (U -matrix) method is presented, in which the overall topological information condensed into the map units is considered so as to project the high-dimensional input vectors into a two-dimensional space and give a better picture of their intrinsic structure than the original U -matrix method. The feature data extracted from industrial gearbox vibration signals measured under different operating conditions are analysed using the proposed technique. The results show that trained with the SOF M network and visualized with the improved method, the feature data are mapped into a two-dimensional space and formed clustering regions, each indicative of a speci c gearbox condition. Therefore, the gearbox operating condition with a fatigue crack or broken tooth compared with the normal condition is identi ed clearly. F urthermore, with the trajectory of the image points for the feature data in two-dimensional space, the variation of gearbox conditions is observed visually, and the development of gearbox early-stage failures is monitored in time.

Gears are the most commonly used components in mechanical transmission systems. Their failures may cause transmission system breakdown and result in economic loss. Identification of different gear crack levels is important to prevent any unexpected gear failure because gear cracks lead to gear tooth breakage. Signal processing based methods mainly require expertize to explain gear fault signatures which is usually not easy to be achieved by ordinary users. In order to automatically identify different gear crack levels, intelligent gear crack identification methods should be developed. The previous case studies experimentally proved that K-nearest neighbors based methods exhibit high prediction accuracies for identification of 3 different gear crack levels under different motor speeds and loads. In this short communication, to further enhance prediction accuracies of existing K-nearest neighbors based methods and extend identification of 3 different gear crack levels to identification of 5 different gear crack levels, redundant statistical features are constructed by using Daubechies 44 (db44) binary wavelet packet transform at different wavelet decomposition levels, prior to the use of a K-nearest neighbors method. The dimensionality of redundant statistical features is 620, which provides richer gear fault signatures. Since many of these statistical features are redundant and highly correlated with each other, dimensionality reduction of redundant statistical features is conducted to obtain new significant statistical features. At last, the K-nearest neighbors method is used to identify 5 different gear crack levels under different motor speeds and loads. A case study including 3 experiments is investigated to demonstrate that the developed method provides higher prediction accuracies than the existing K-nearest neighbors based methods for recognizing different gear crack levels under different motor speeds and loads. Based on the new significant statistical features, some other popular statistical models including linear discriminant analysis, quadratic discriminant analysis, classification and regression tree and naive Bayes classifier, are compared with the developed method. The results show that the developed method has the highest prediction accuracies among these statistical models. Additionally, selection of the number of new significant features and parameter selection of K-nearest neighbors are thoroughly investigated.