Available pure data analytics tools do not match market expectations. These tools are centered only in some drive train components neglecting ageing of the main structural components which size the financial life expectancy of the asset (blades, hub, mainframe, main shaft, yaw, tower, foundation. Also, they are base the learning process on high uncertainty wind data based on nacelle anemometers and/or sensorics which do not represent the entire turbine behavior, delivering a reactive learning data systems that waits to be taught based on the experience. Therefore, they are not able to predict failures that have not occurred before, thus not able to identify failure modes or track root causes either.
Real multitechnology fleets must be covered by advanced tools, able to provide independent and direct visibility and transparency to asset owners. In many cases classic data analytic tools do not deliver the understanding of the performance of each site, because normally the data owned by operators are very poor base for work (limited in frequency: 10 min samples and limited in channels and variables: especially in mature turbine models).
Turbines are 95% physics, and data analytics need to be linked with design criteria of the turbines. Thanks to physical models we can predict failure modes patterns and track root causes, but to go beyond the existing limited SCADA needs to be used in the most intelligent way, not only for machine learning.
Available data can be made credible measuring the wind adequately; through strategic placed sensorics we can create a very good set of inputs for turbine complete monitoring, answering the need of brining an independent, easy to understand, transparent and multitechnology tool for active asset management, which monitors asset performance, structural risks, life consumption and cash flows in detail.
This is key for condition based predictive maintenance strategies on all turbine components, in a flexible approach to be embedded within own clients tools.
Creates baselines and KPIs for health surveillance for each individual based on physical models.
Based on individual specific conditions.
Based on genetics and congenital factors.
Carries out periodic measurements, tests and analysis comparison with baselines.
Comparing real performance with baselines for proper health surveillance and triggering of actions.
Creates baselines and KPIs for performance and structural health based on physical models.
Based on site specific conditions.
Based on turbine design characteristics.
Measures independently and more precisely wind and operation conditions.
Carries out wind, vibrations and SCADA measurements comparison with measurements and baselines.
Comparing real performance with baselines for proper health surveillance and triggering of actions.