|
|
Data Analysis for Predictive Maintenance of a Robotic Arm
Žitný, Roland ; Rozman, Jaroslav (oponent) ; Janoušek, Vladimír (vedoucí práce)
The Mitsubishi MELFA robotic arms used in modern factories work almost without interruption and produce sensory data about their operation. Various analysis techniques can be applied to such data for predictive maintenance, which provide information on the condition and maintenance needs of such robotic arms. The proposed predictive maintenance process consists of a sensory data acquisition system using the slmpclient and mitsubishi-monitor libraries, an analysis method system with anomaly detection using a convolutional autoencoder, anomaly classification using convolutional neural networks, and data segmentation into segments of individual robot actions using hidden Markov models. Such analysis techniques provide information on the severity, type, and location of emerging faults and abnormalities in behavior, which then determine the time required to perform the required maintenance. This work presents a created chain of predictive maintenance processes, where the obtained findings provide valuable insights into the application of predictive maintenance of Mitsubishi MELFA robotic arms in an industrial environment.
|
|
|
Data Analysis for Predictive Maintenance of a Robotic Arm
Žitný, Roland ; Rozman, Jaroslav (oponent) ; Janoušek, Vladimír (vedoucí práce)
The Mitsubishi MELFA robotic arms used in modern factories work almost without interruption and produce sensory data about their operation. Various analysis techniques can be applied to such data for predictive maintenance, which provide information on the condition and maintenance needs of such robotic arms. The proposed predictive maintenance process consists of a sensory data acquisition system using the slmpclient and mitsubishi-monitor libraries, an analysis method system with anomaly detection using a convolutional autoencoder, anomaly classification using convolutional neural networks, and data segmentation into segments of individual robot actions using hidden Markov models. Such analysis techniques provide information on the severity, type, and location of emerging faults and abnormalities in behavior, which then determine the time required to perform the required maintenance. This work presents a created chain of predictive maintenance processes, where the obtained findings provide valuable insights into the application of predictive maintenance of Mitsubishi MELFA robotic arms in an industrial environment.
|
|
|
Feature Space Reduction As Data Preprocessing For The Anomaly Detection
Bilik, Simon
In this paper, we present two pipelines in order to reduce the feature space for anomalydetection using the One Class SVM. As a first stage of both pipelines, we compare the performanceof three convolutional autoencoders. We use the PCA method together with t-SNE as the first pipelineand the reconstruction errors based method as the second. Both methods have potential for theanomaly detection, but the reconstruction error metrics prove to be more robust for this task. Weshow that the convolutional autoencoder architecture doesn’t have a significant effect for this task andwe prove the potential of our approach on the real world dataset.
|
host ::
RSS.