The early identification of a defect that is developing in a bearing is crucial for avoiding failures in rotating machinery. Frequency domain analysis of the vibration signals has been shown to contribute to a better understanding of the nature of a developing defect. Early signs of degradation might be more noticeable in certain frequency bands. The advantages in identifying and monitoring these bandwidths are several: prevention of serious machinery damages, reduction of the loss of investments, and improvement of the accuracy in failure predicting models. This paper presents and compares two approaches for the diagnosis of bearing faults. The first approach was knowledge-based. It relied on principles of mechanics to interpret the measured vibration signals and utilized prior knowledge of the bearing characteristics and testing parameters. The second approach was data-driven whereby data were acquired exclusively from the vibration signal. Both approaches were successfully applied for fault diagnosis by identifying the frequencies of the vibration spectra characteristic for the bearing under study. From this, bandwidths of interest for early fault detection could be determined. The diagnostic abilities of both approaches were studied to analyze and compare their individual strengths regarding the aspects of implementation time, domain knowledge, data processing associated knowledge, data requirements, diagnostic performance, and practical applicability. The advantages, apparent limitations as well as avenues for further improvement of both approaches are discussed.
22 January 2022 in Robotics and Computer-Integrated Manufacturing
This article presents the measurement and analysis of the influence of velocity on the quasi-static deflections of industrial manipulators of three different manufacturers. Quasi-static deflection refers to the deflection of the end effector position of articulated robots during movement at low velocity along a predefined trajectory. Based on earlier reported observations by the authors, there exists a difference in the static and quasi-static deflections considering the same points along a trajectory. This work investigates this difference to assess the applicability of robotic compliance calibration at low velocities. For this assessment, the deflections of three industrial articulated robots were measured at different speeds and loads. Considering the similarity among the robot models used in this investigation, this work also elaborates on the potential influence of the measurement procedure on the measured deflections and its implications for the compliance calibration of articulated robots. For all industrial articulated robots in this investigation, the quasi-static deflections are significantly larger than the static ones but similar in trend. Additionally, the magnitude of the quasi-static deflections presents a proportional relationship to the Cartesian velocity.
As prediction models for automated decision-making in manufacturing become more widespread, enhancing their reliability has become increasingly critical. One effective approach is improving the quality of training data through data curation or cleaning. However, most existing data curation methods rely on domain-specific expertise in data science, limiting their accessibility. To address this challenge, a general-purpose time-series data curation toolbox with a user-friendly interface was developed. Designed to minimize user input while maintaining traceability, the toolbox was tested on various publicly available time-series datasets. Its performance was evaluated by comparing the prediction accuracy of machine learning models trained on original versus cleaned data. The results demonstrate the significant potential of the general-purpose toolbox for enhancing data quality and model performance across a range of applications.