Deepli to Supply Listen AI for Predictive Maintenance to Multiaxis Robots at Global Advanced Parts Maker
TECHWORLD ·
✦ AI Summary
Deepli announced on the 1st that it will supply its predictive maintenance solution, "Listen AI," to a global advanced parts manufacturer's multiaxis transfer robots.
"Listen AI" is an acoustic analysis AI that analyzes sounds on manufacturing floors in real time and diagnoses abnormalities using microphones on robot drive units.
Deepli plans to advance the system beyond abnormal sound alerts to fault-axis diagnosis and remaining useful life (RUL) prediction.
Deepli announced on the 1st that it will supply its predictive maintenance solution, "Listen AI," to a global advanced parts manufacturer's multiaxis transfer robots. "Listen AI" is a solution based on acoustic analysis AI that analyzes sounds on manufacturing floors in real time and provides meaningful data.
Deepli has experience inspecting the quality of drive components such as motors, gears, and bearings, and also has experience inspecting fastening sounds from connectors, bolts, and other parts. It also has experience analyzing more than 70 sound types and holds more than 15TB of manufacturing acoustic data collected directly.
Deepli entered the predictive maintenance market based on this experience and data. According to Deepli, "Listen AI" uses sound to diagnose abnormalities in the motors and reducers of multiaxis robots for advanced parts inspection and transfer.
Deepli said it has identified the correlation between robot load and sound. Based on this, "Listen AI" tracks deterioration trends in parts and provides early warning indicators.
The system uses microphones installed on the robot drive unit to continuously monitor sound. It determines sounds other than normal ones, calculates a score for the judgment result, and then sends abnormality information to an automated notification system. Listen AI performs continuous sound-based monitoring, alerts users in advance to signs of abnormality before a defect occurs, and records related data 100%, contributing to data assetization.
The system extracts meaningful results even in wideband noise environments where multiple robots operate simultaneously. Listen AI is also being applied at the same time to detect abnormal bearing sounds in raw material transfer equipment for displays. As a result, it shifts away from the previous method of inspecting all robots when a problem occurs on one line, and toward selectively inspecting only the line where an abnormal signal is detected.
Deepli emphasized that this shift helps reduce maintenance manpower and costs. Deepli also announced plans to further advance this sound-based predictive maintenance case. It is pursuing an advanced plan that goes beyond generating alerts when abnormal sounds are detected, and it plans to realize fault-axis diagnosis and remaining useful life (RUL) prediction in the future.
Based on this, Deepli plans to discuss expanding the lines to which predictive maintenance is applied. It also plans to discuss expansion to other factories.
Lee Su-ji, CEO of Deepli, said that while automation processes using robots are rapidly increasing, robot predictive maintenance is still at a stage that requires advanced technology. Lee said Listen AI has demonstrated the potential and validity of sound-based predictive maintenance and has turned it into a real-world supply, and added that the company aims to establish itself as a leader in robot predictive maintenance by pursuing greater analytical precision in the future.
Source: TECHWORLD · Lee Gwang-jae
Original: https://www.epnc.co.kr/news/articleView.html?idxno=407599
References
This article was produced with the help of an automated content generation algorithm.
Source: TECHWORLD
View originalThis article was summarized and organized by BizCrush based on the original article from TECHWORLD. For exact quotations and full details, please refer to the original article.