Energy assets such as power plants, photovoltaic panels and wind turbines must be inspected regularly. These inspections use a large amount of data or images. All these images must be analyzed to detect, categorize and classify defects.
Most of the time, this task is done manually: people look at the images one by one. This takes time and some defects can be missed or misclassified by a human operator (fatigue, lack of attention, etc.).
Computer vision facilitates quality inspections. It enables thorough, fast and high-quality checks: defective parts can be identified quickly thanks to the rapid processing capability of artificial vision. And at the same time, the possibility of error is reduced.
Over the past three years, the Digital Innovation team at the UK R&D Centre has developed expertise in the field of computer vision. It now works with various business units, such as EDF Renewables and EDF UK Customers to detect defects on wind turbine blades or install communicating meters in a secure way.
For EDF Renewable, for example, the team is developing models to detect defects on wind turbine blades using AI, with an accuracy of more than 85% from two different data sets.
When setting up communicating meters, an artificial intelligence mobile application checks the quality and automatically checks if the cables of the communicating meters are connected in the right order: