Condition monitoring of electric motors with motor current analysis and smart diagnostics

Electric motors and machines are mechanically connected to each other via belts, couplings, gears or directly. Via this coupling, vibrations of the driven machine or mechanical process disturbances are transmitted back to the motor. There, the effects are reflected in the motor current and voltage. With suitable measurement technology, the feedback effects can still be measured in the electrical supply lines of the motor in the control cabinet, evaluated and assigned to the causative symptoms and monitored. This is exactly what the e-MCM device does fully automatically, using the electric motor as a sensor.

In addition, the most important electrical health data of the motor and its performance values are determined at the same time. e-MCM for stationary monitoring and AMTpro for mobile applications and route receivers.

The engine as a sophisticated sensor for condition monitoring

e-MCM was developed by Artesis for predictive maintenance of critical rotating AC machines. The e-MCM's patented machine-learning algorithm enables comprehensive fault detection in advance of incipient damage. With continuous monitoring and real-time model-based voltage and current analysis, e-MCM can detect electrical, mechanical and process faults in fixed and variable speed motors and generators. An e-MCM device uses the motor itself as a sophisticated sensor. This enables continuous online fault monitoring and simultaneous power measurement in one.

e-MCM is installed using current transformers connected to the three mains phases of the motor and the monitoring monitor in any control panel. e-MCM is typically located in or near the motor control panel and is particularly useful in environments where motors are not easily accessible, either because they are remote or the environment is dangerous or inaccessible.

Automated motor current analysis with expert know-how

The technology, developed by NASA, offers machine learning capabilities in a compact built-in instrument. More than 100,000 motor signals have been recorded and evaluated. This experiential knowledge is incorporated in the software and automatically makes it an experienced expert. Due to the machine learning algorithm, it can recognise normal operation under a variety of conditions, such as different speeds or loads - and thus enables close monitoring without false alarms.

When first switched on, e-MCM begins an automatic self-learning process in which it learns the normal operating condition of the plant. If e-MCM identifies a new operating condition that it did not learn during the self-learning phase, it offers the possibility to include this condition in future monitoring. e-MCM permanently monitors the rotating machinery, continuously takes measurements and compares them with the digital twin created during the self-learning process.