July 16, 2015 By Vijay Anand
Electrical maintenance engineers in process industries face a unique set of problems when it comes to supporting operation of motors. On one hand, every minute of motor downtime leads to thousands of dollars of lost revenue, and on the other hand, the knowledge required to provide predictive information on the condition of a motor is limited and not easily accessible.
On-line condition monitoring of motors is quiet complex as it involves mechanical failures, such as bearings, rotor issues, such as bar cracks or eccentricity, and the need for an experienced analyst which involves precious time and money. How can industry overcome this hurdle and supplement revenues by getting advanced intelligence on motor condition?
Automated diagnosis techniques focus on the early identification of defects in motors by using powerful algorithms. It carries out ‘on-site’ analysis replacing a traditional analyst to identify defects early and to quantify the severity of the defect. This enables plant maintenance managers to schedule timely interventions and remedy problems.
These defects include, but are not limited to, bearing problems, motor electromagnetic anomalies and any negative influences on reliability or performance arising from the connected environment. Analysis software can consist of libraries of algorithms used collectively to achieve the best possible discrimination of defects.
The deployment of these algorithms form the backbone of the monitoring system, which uses a physics-of-failure approach, combined with sophisticated signal processing of the data, to derive key condition parameters that reflect the development of defects.
Refer to the case below where (chart 1) traditional measurements do not pick up any fault whereas (chart 2) the algorithm-based automated diagnosis picks up bearing fault during the same period, helping the customer to get early warning of an impending failure.
While identifying faults in motors early is critical, it is equally critical to have this intelligence communicated for the maintenance team to plan maintenance actions.
Gone are the days when data is stored on a desk-based computer and monitored manually. Welcome to the new age era where the health status of a motor can be brought to palms of one’s hand through the power of the internet, making the monitoring truly remote.
In such an environment, an alarm is triggered via SMS/e-mail if a key condition parameter exceeds preset threshold limits, giving the plant operator an early warning that maintenance is needed. Customers can now access information and observe live trends of analyzed data (KCP), spectrum, and reports 24 hours a day via remote connections, providing them inputs on motor condition directly.
Remote condition monitoring using advanced automated diagnosis with the power of the internet can provide customers peaceful sleep at night by being ‘always on site, always monitoring’, even when they aren’t.