calculating production performance and detecting various situations
such as unexpected system down and checking product quality through
AI deep learning analysis using sound data.
In the past, Specialized personnel or embedded applications
such as complex machines were required for monitoring machines.
Now, With various AI voice recognition technologies,
It’s possible to detect unusual signals in sound data
made by factories and machines.
System diagram
Monitoring equipment operation status, Determining its lifespan,
Checking defective products through deep learning analysis using sound sensors.
· Sound Analysis - Analysis of phenomena, predictions, and causes → Detection of equipment abnormalities
·Regression Analysis - phenomenon, optimization, cause analysis → facility lifecycle management and optimization operation
▲ Algorithm conceptual diagram – After featuring sound data with SFTP, the filtered data will be processed by CNN for deep learning.
The classified result will be used for determining valid production performance and equipment status.
Main Features
Real-time status monitoring for production
(operating time, maximum/actual production quantity),
and replacement related information.
Provision of facility management, mold management,
mold parts management, mold removal/installation history,
number of strokes by mold, production status
System features
-
01
Real-time monitoring
of machine
equipment status -
02
Early detection of
system anomalies -
03
Strategic utilization of
robot/facility data -
04
Reduce
unnecessary
maintenance costs -
05
Prevent accidents by
increasing system
safety and reliability
Benefits
-
01 Notification for maintenance schedule according to facility operation rate.
-
02 Automatic facility status detection and notification.
-
03 Flexible, Fast, and scalable data processing
-
04 Automatic data-based crash prediction.
-
05 Saving cost for pre-detection
