Blades' damage detection through AI

Artificial intelligence (AI) is becoming increasingly present in our daily lives, as evidenced by the development of tools like ChatGPT in recent times. SOLUTE, in its commitment to R&D, understands the importance of the new paradigm provided by artificial intelligence to offer increasingly comprehensive technical solutions to our clients.

In the last months, a team with experience in AI from our office in Badajoz has been assigned the task of developing an artificial intelligence model with the aim of automating blade damage inspection using TSR Inspector, the end-to-end platform for this purpose provided by TSR Wind, a company in which SOLUTE is the majority shareholder. The implementation of this algorithm within the platform will accelerate the generation of blade damage reports for our clients, thereby reducing the delivery time and allowing our clients to take necessary preventive measures in advance.

As with any artificial intelligence-focused project, challenges have arisen in terms of dataset preparation, as it must be properly labeled and contextualized, as well as in the processing of the actual images, which need to be adapted to feed an artificial intelligence model.

The development and improvement of this algorithm is iterative. In an initial version, the algorithm was prepared only to potentially identify the damage, so that our inspectors only had to classify the damage. Subsequently, the team has been improving the model's ability to not only detect but also automatically classify the type of damage, being able to automatically detect damage such as erosions or cracks, further facilitating the inspection process.