Object-Oriented Decision Trees

In the context of optimized Operation & Maintenance of wind energy infrastructure, it is important to develop decision support tools, able to guide operators and engineers in the management of these assets. This task is particularly challenging given the multiplicity of uncertainties involved, from the point of view of the aggregated data, the available knowledge with respect to the wind turbine structures, and sub-components, as well as the constantly varying operational and environmental loads. We propose to use run wind turbine telemetry data through the decision tree learning algorithms to detect faults, errors, damage, patterns, anomalies and abnormal operation. The use of decision trees is motivated by the fact that they tend to be easier to implement and interpret than other quantitative data-driven methods in addition to being a natural fit for learning from big data on wind turbine fleets. The telemetry consists of data from condition monitoring systems (CM) and data from specialized structural health monitoring (SHM).

In addition we are working on extending the concept of decision tree learning to Object-Oriented decision tree learning:
- For systems with behaviour involving feedback (e.g., after a repair/ update in the system) or for evolving systems (e.g. when new sensors are integrated)
- WT is viewed as a multi-layered system of objects (e.g. structure, controller, actuator, etc.) that are defined on the basis of abstract super-classes, attributed with specific properties and methods