Non-Intrusive Operation Status Tracking for Legacy Machines via Sound Recognition

Ooi, Boon-Yaik; Lim, Jason Jing-Wei; Lee, Wai-Kong; Shirmohammadi, Shervin

doi:10.1109/i2mtc43012.2020.9129526

Cited by 6 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, Hesser et al suggested retrofitting a milling machine with a vibration sensor to collect data and predict tool wear [5]. Considering an obvious challenge of these vibration sensors for retrofitting (placement constraints), Ooi et al have successfully implemented a sound-based retrofitting solution to track the status of a legacy machine [6]. The so collected raw data or processed information can then be visualized using conventional visual devices or futuristic visual devices like AR glasses as proposed in Guerreiro et al [1] There are many such proposals for using an AR based visual aid for data monitoring as it makes user interaction effortless and efficient [7], [8] Though these State-Of-The-Art articles discuss about smart retrofitting and AR based data visualization, they tend to make two assumptions:…”

Section: State-of-the-artmentioning

confidence: 99%

Augmented Reality Based Machine Monitoring for Legacy Machines: a retrofitting use case

Kancharla¹,

Bekaert²,

Lannoo³

et al. 2021

2021 XXX International Scientific Conference Electronics (ET)

View full text Add to dashboard Cite

Monitoring a machine and the insight it provides for appropriate maintenance is of prime importance for the modern industry. While this is fully supported in Industry 4.0, many manufacturing units today are unable to use its technologies. One of the major reasons behind that is the lack of onboard sensors to capture and communicate appropriate data from a legacy machine. Retrofitting is considered an efficient solution to include legacy machines in an industry 4.0 ecosystem by adding external sensing and connectivity. As a proof of concept, in this paper we consider an induction motor for retrofitting, a major component of any manufacturing unit. We further show how retrofitting can help in making legacy machines compliant to the new technologies from Industry 4.0. Finally, visualization of the collected data is performed over a web based Augmented Reality application. The choice for a web-based Augmented Reality application was made to enable monitoring of the sensor data over a multitude of devices.

show abstract

Section: State-of-the-artmentioning

confidence: 99%

Augmented Reality Based Machine Monitoring for Legacy Machines: a retrofitting use case

Kancharla¹,

Bekaert²,

Lannoo³

et al. 2021

2021 XXX International Scientific Conference Electronics (ET)

View full text Add to dashboard Cite

show abstract

Detecting of Tree Cutting Sound in the Forest by Machine Learning Intelligence

Ramineni,

Shankar,

Shabana

et al. 2024

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Operation and Productivity Monitoring from Sound Signal of Legacy Pipe Bending Machine via Convolutional Neural Network (CNN)

Kim,

Mun,

Jun

et al. 2024

Int. J. Precis. Eng. Manuf.

View full text Add to dashboard Cite

This study introduces a non-invasive approach to monitor operation and productivity of a legacy pipe bending machine in real-time based on a lightweight convolutional neural network (CNN) model and internal sound as input data. Various sensors were deployed to determine the optimal sensor type and placement, and labels for training and testing the CNN model were generated through the meticulous collection of sound data in conjunction with webcam videos. The CNN model, which was optimized through hyperparameter tuning via grid search and utilized feature extraction using Log-Mel spectrogram, demonstrated notable prediction accuracies in the test. However, when applied in a real-world manufacturing scenario, the model encountered a significant number of errors in predicting productivity. To navigate through this challenge and enhance the predictive accuracy of the system, a buffer algorithm using the inferences of CNN models was proposed. This algorithm employs a queuing method for continuous sound monitoring securing robust predictions, refines the interpretation of the CNN model inferences, and enhances prediction outcomes in actual implementation where accuracy of monitoring productivity information is crucial. The proposed lightweight CNN model alongside the buffer algorithm was successfully deployed on an edge computer, enabling real-time remote monitoring.

show abstract

Non-Intrusive Operation Status Tracking for Legacy Machines via Sound Recognition

Cited by 6 publications

References 26 publications

Augmented Reality Based Machine Monitoring for Legacy Machines: a retrofitting use case

Augmented Reality Based Machine Monitoring for Legacy Machines: a retrofitting use case

Detecting of Tree Cutting Sound in the Forest by Machine Learning Intelligence

Operation and Productivity Monitoring from Sound Signal of Legacy Pipe Bending Machine via Convolutional Neural Network (CNN)

Contact Info

Product

Resources

About