LSTM-Based Condition Monitoring and Fault Prognostics of Rolling Element Bearings Using Raw Vibrational Data

Abstract

The 4.0 industry revolution and the prevailing technological advancements have made industrial units more intricate. These complex electro-mechanical units now aim to improve efficiency and increase reliability. Downtime of such essential units in the current competitive age is unaffordable. The paradigm of fault diagnostics is being shifted from conventional to proactive predictive approaches. As a result, Condition-based Monitoring and prognostics are now essential components of complex industrial systems. This research is focused on developing a fault prognostic system using Long Short-Term Memory for rolling element bearings because they are a critical component of industrial systems and have one of the highest fault frequencies. Compared to other research, feature engineering is minimized by using raw time series sensor data as an input to the model. Our model achieved the lowest root mean square error and outperformed similar research models where time domain, frequency domain, or time-frequency domain features were used as input to the model. Furthermore, using raw vibration data also enabled better generalization of the model. This has been confirmed by evaluating the performance of the developed model against vibration data generated by distinct sources, including hydro and wind power turbines.

AdaBoost Ensemble Approach with Weak Classifiers for Gear Fault Diagnosis and Prognosis in DC Motors

“…Long Short-Term Memory (LSTM) is a type of recurrent neural network (RNN) architecture designed to model sequential data [24]. It addresses the vanishing gradient problem of traditional RNNs by introducing gating mechanisms to control the flow of information.…”

“…These parameters are learned during training. This method utilizes Long Short-Term Memory (LSTM) networks to advance the field of fault prognosis for rolling element bearings, which are crucial components in industrial setups [24]. The model achieves improved performance by directly importing raw time series sensor data, minimizing the need for feature engineering.…”

“…The model achieves improved performance by directly importing raw time series sensor data, minimizing the need for feature engineering. LSTM, or Long Short-Term Memory, allows the model to surpass other models that rely on conventional characteristics, demonstrating its ability to accurately forecast faults [24]. Moreover, using unprocessed vibration data improves the model's capacity to make accurate predictions across different types of energy sources, including hydro and wind power turbines, demonstrating the robustness of LSTM in proactive, predictive maintenance applications [24].…”

An In-Depth Study of Vibration Sensors for Condition Monitoring

“…Predictive maintenance makes use of condition monitoring data, which provides significant insight into equipment performance [20], health, and behavior. This data-driven [21] strategy promotes informed decision making as well as ongoing development. Predictive maintenance minimizes the under or over-utilization of maintenance resources that are associated with traditional techniques, resulting in more effective resource allocation and cost savings.…”

“…Predictive maintenance helps prevent the occurrence of these breakdowns by predicting possible problems before they become catastrophic failures. These machines are equipped with sensors [25][26][27][28][29] that continuously collect data on parameters like temperature [9,20,25,30], pressure [7,20] and vibration [21]. These data are then evaluated by software, which uses cutting-edge methods such as machine learning [31], previous data, and contextual information to construct predictive models that identify expected issues or maintenance requirements based on usage patterns and environmental factors.…”