فهرست مطالب hesam addin arghand

In this paper, the vibration analysis methods and shock pulse method (SPM) are compared in order to detect the unhealthy condition as well as fault type in the early stages of rolling element bearing (REB) degradation. To analyze vibration signals, three weak signature detection methods based on continuous wavelet transform (CWT), empirical mode decomposition (EMD) and envelope technique are employed. A set of accelerated life tests on REBs was designed and performed in CM lab of Sharif university of technology. Seven tests were conducted and vibration signals, as well as shock pulse signals, were recorded regularly. The trend of vibration level and shock pulse level are compared for early detection of the unhealthy condition in REBs. In addition, the extracted spectrums from SPM, CWT, EMD, and envelope techniques are studied to detect bearing characteristics frequencies (BCFs) to diagnostics. Results show that SPM has better performance on early fault detection of REBs rather than vibration analysis techniques.

Fault detection of rolling element bearing (REB), has a very effective role in increasing the reliability of machinery and improving future decisions for rotating machinery operation. In this study, a new method based on a convolutional neural network (CNN) is developed for fault detection of REB. Its performance will be compared with other artificial intelligence (AI) techniques, 2-layer, and deep feedforward neural network (FFNN). In this regard, a set of accelerated-life tests has been implemented on an experimental platform. The models are aimed to recognize the impact pattern in the raw signals generated by faulty REBs. The innovation of the present study is to convert the high-dimensional input as a raw temporal signal to low-dimensional output. The developed method does not need preprocessing of data. Using several types of accelerated tests prevents overfitting. The result shows that the accuracy of the developed CNN-based method is 98.6% for all data sets and 94.6% for the validation dataset. The accuracy of the 2-layer FFNN is 85% for all datasets and 74.2% for the validation dataset and the accuracy of the deep FFNN is 82% for all datasets and 67% for the validation dataset. Therefore, the developed CNN-based method has better performance than the FFNN-based models.

Early fault detection of the rolling element bearings has a very important role in increasing the reliability of rotating machines.It leads to better decision-making for maintenance activities.  In recent decades, the shock pulse method has been developed to detect faults in the early stage of rolling element bearings degradation. In this paper, the accuracy of the remaining useful life estimation using extracted features from vibration signals and that from the shock pulse method are compared. In this regard, a set of accelerated life tests on rolling element bearings were designed and performed. Both shock pulse signals and vibration signals of the under-test rolling element bearings were recorded. Then two models based on feed-forward neural-network are developed to predict the remaining useful life of rolling element bearings. In the first model, only extracted features from vibration signals are fed for remaining useful life prediction. In the second model, the extracted features from shock pulse method are fed too. The results show that using shock pulse method-based features improves the accuracy of remaining useful life estimation. Also, using the health indicators extracted from vibration analysis and shock pulse method leads to a better estimating of the degradation behavior.

Estimation of remaining useful life (RUL) of rolling element bearings (REBs) has a major effect on improving the reliability in the industrial plants. However, due to the complex nature of the fault propagation in these components, their prognosis is affected by various uncertainties. This effect is intensified when the recorded data is offline, which is very common for many industrial machines due to the lower cost rather than the online monitoring strategy. In the present paper, in order to overcome the shortcoming of the feed-forward neural network (FFNN) in REBs prognostics, a new method for considering two main uncertainties (caused by the measurement and process noises) is proposed, in the presence of offline data acquisition. Inthe proposed method, the primary RUL probability distribution corresponded to each offline measured data is predicted, utilizing the outputs of trained FFNNs. Then, the predicted RUL distribution will become more robust in confronting the temporal changes, by taking into account the approval of pervious stage predictions to the present prediction. As a result, the overall probability distribution of REBs RUL and also its confidence levels (CLs) areobtained. Finally, the evaluation of the proposed method is performed byutilizing bearing experimental datasets. The results show that the proposed method has the capability to express the estimated RUL CLs in the offline data acquisition method, effectively. By providing a probabilistic perspective, the proposed method can improve the reliability of the asset and also the decision-making about the future of the industrial plants.