 |
| Model Based Fault Detection |
|
|
Model Based Fault Detection Technology
Model based fault detection systems are based on the notion of comparing a calculated output of a system (derived through the mathematical model of the system) with the actual output, based on a certain set of inputs. The magnitude of the difference between the actual output and the derived (expected) output, points to a fault or a developing fault in the system. Although the mathematical model is generally a black box without many physically meaningful parameters, the end result, which is the deviation from the expected or the norm, is highly reliable, trendable and thus valuable information. Combining this information with other diagnostic methods such as power spectrum density analysis provides for a robust fault detection and diagnostic system for electric motors and the driven equipment. ArtesisMCM is the only such system in the market for electric motors. The scope of the faults that are detected through MCM is wide and includes both electrical and mechanical faults. This makes MCM a great tool for condition monitoring and predictive maintenance of electric motors and any driven equipment such as pumps, compressors and fans. Although, it is a relatively new technology it already has a great track record in various industries and critical processes. Its installed base and adoption rate is growing rapidly.
Application on Condtion Monitoring and Preventing Unscheduled Downtime
The basic need of preventing unscheduled downtime involves detecting faults that are likely to develop into equipment failures. The information on the severity of the fault, thus the timeline of the prognosis is also important for timing the corrective action to prevent unscheduled shutdown. Model based fault detection is an holistic approach to this aim, as it involves assessing system-wide deviation from the norm and the magnitude of that deviation. It is best utilized when coupled with a diagnostic tool that could provide insight to the root cause of the fault and thus, an indication for a corrective action. Artesis MCM combines model based fault detection technology with diagnostic features derived from the model parameters it generates, power spectrum density analysis and other system performance measurements. Thus it provides a complete package for fault detection, failure prediction and corrective action indication. It has been utilized as the only such tool in various installations. It can also be utilized as a supplemental technology to other monitoring and diagnostic technologies, such as vibration analysis. Or such technologies can be used to supplement a program that has MCM at its core.
|
Simplifying Predictive Maintenance with MCM
MCM was developed to utilize model based fault detection technology in simplifying predictive maintanance. The following article describes how MCM was conceptualized, developed and now applied as a great tool in predictive maintainance. It is as article published in GE's Orbit magazine. The GE's recent label for MCM is Anomalert.
|
|