ABB develops hybrid predictive maintenance technology

ABB develops hybrid predictive maintenance technology

Technology News |
ABB has combined machine learning and fault analysis algorithms to provide automated predictive maintenance for all kinds of applications across Industry 4.0
By Nick Flaherty


ABB has developed a hybrid approach for automated preventive maintenance in many different Industry 4.0 applications

The advances in digital technology, machine learning (ML) and cloud and edge computing mean a new approach for asset health maintenance is emerging. Process industries rely on a multitude of crucial equipment such as motors, pumps, fans, compressors and turbines running around-the-clock to ensure smooth production. Keeping these machines at peak health is critical as wear-and-tear is inevitable.

However forecasting and predicting maintenance is not straight-forward and varies from one indutry to another. To effectively schedule maintenance, it is necessary to predict how a detected abnormal condition is likely to develop in the future. Only then can valuable insight into the probable future consequences be gained

The hybrid approach being used by ABB combines machine learning (ML) models and Failure Modes and Effect Analysis (FMEA) to provide accurate information about actual asset health.

Successful predictive maintenance requires a three-pronged process:
• Condition monitoring that can provide early detection of faults
• Identification of specific failure mode(s) related to the fault detection
• Quantification of the extent of fault development to support maintenance planning

Despite the availability of various popular ML approaches to develop models for asset condition, eg, Principal Component Analysis (PCA), K-Nearest Neighbor (KNN), Local Outlier Factor (LOF), One Class Support Vector Machine (OCSVM), ML approaches are black box approaches and fully dependant on asset data; they make no assumptions about the asset or its failure modes.

Practical industrial experiences indicate that such approaches are not always successful and often lead to several types of false positives and false negatives. Raising an alarm when the asset is completely healthy or vice versa increases unplanned costs.

Related articles

ABB’s hybrid approach relies on historical data and engineering models of an asset to enable the implementation of predictive maintenance. The online condition monitoring data is combined with data science uses one of two techniques, either detection of anomalies where online measurements deviate from normal operating behaviour or the identification of known failure characteristics where online measurements closely match a “fault signature”

Both techniques use a data model: the former represents “good health” and the latter captures the “data signature” present under fault conditions.

The former is often the best approach to use because there is almost always sufficient historical data available that represents good health. This health data enables the model to be trained. Conversely, there is usually no- or insufficient data available to represent all possible failure conditions. The second model also relies on equipment characteristics, and these depend on the installation and operating conditions. This means the data that relates to a specific machine is usually insufficient to train an accurate fault model.

With the hybrid approach, an engineering model is used to quantify the extent of the deviation of the online measurements from the health model. A Failure Mode Analysis model is used.

FMEA is a core component of Reliability Centered Maintenance (RCM) programs that already exist for the most commonly encountered equipment and systems. It defines the potential detection and identification of a fault.

Next: Pilot project for predictive maintenance

ABB’s researchers have used a new three-step process for an asset health configuration that relies  on both data and fault analysis models. This hybrid approach generates indicators for predictive maintenance: a Key Diagnostic Indicator (KDI) and fault indicators. A KDI is calculated for each measurement in each model by comparing the deviation of the measured value from the “expected” reference value. The reference value is derived by searching the entire data for the closest fit to the current conditions. An ML algorithm such as KNN is used to efficiently compute the nearest neighbour.

All KDI’s are expressed as a percentage, allowing the user to easily interpret the value, irrespective of model, measured quantity or range.

Fault Probability Indicator (FPI) and Fault Severity Indicator (FSI) are calculated for every fault. This represents an aggregation of the deviations of all the weighted measurements to provide a distinction between severity and probability.

The fault model provides additional information related to cause analysis and corrective actions, which are easily viewed. This allows a fully automated workflow by integrating the fault indicator values and the fault information with a Computerized Maintenance Management System (CMMS). Such systems typically orchestrate maintenance activities based on priority and resource availability.

ABB has developed software using this technique that has been deployed in 33 hydropower plants by Enel Green. Real-time condition monitoring of a wide range of assets from hydro-turbines, pumps and motors to generators is ongoing with an anticipated project completion in 2022.

Based on the success of the initial pilot project, ABB plans to expand the use of this hybrid approach to predictive maintenance to different industrial verticals such as conventional power plants, refineries, cement mills and the oil and gas industry. This is possible as the ML- and FMA modelling s completely generic by nature and therefore not industry specific.

More details of the system are at

Other related articles 

Other articles on eeNews Europe

Linked Articles
eeNews Europe