The National Institute of Technical Teachers Training & Research (NITTTR) Chandigarh, is an autonomous educational institution under the government of India’s Ministry of Education (MoE). As part of their digital twin program, NITTTR needed to include a practical model with sufficient complexity in both its design and operation that was representative of real-life scenario. NITTTR chose to model wind turbines because they are complex machines that face unique challenges.
The development and functionality of the digital twin of a vertical wind turbine were performed using the Altair® RapidMiner® data analytics platform along with Altair Embed®, Altair® HyperGraph®, and Altair® OptiStruct®. These tools span data acquisition, modeling, data visualization and processing, and finite element (FE) modeling and analysis. With Altair’s digital twin solutions, NITTTR was able to continuously observe the virtual wind turbine’s parameters – thus improving the physical asset’s performance and illustrating the basic concepts of digital twin technology to students and users alike. They were able to leverage a machine learning model developed using two data sets consisting of 18,000 tests collected using an accelerometer and a strain gauge. Overall, the model was able to predict the “good” condition of the wind turbine structure with an accuracy of 97%, and the “faulty” condition of the wind turbine structure with an accuracy of 95%.