Portuguese 
English 
Spanish 
3 min of reading 
0 comments 
New Technology Developed by Scientists Uses AI to Identify Failures in Wind Turbines, Optimizing Maintenance and Increasing the Efficiency of Wind Power Generation.
Researchers have proposed a new technology capable of automatically identifying, locating, and measuring damage in wind turbine blades. The solution was presented in a paper published in the journal Scientific Reports, part of the Nature group, and has the potential to improve the efficiency of preventive maintenance in wind power generation parks.
The project is the result of the work of scientists who developed a hierarchical machine learning model, dubbed HHMLM (Hybrid Hierarchical Machine Learning Model), which utilizes acoustic data for structural monitoring. The system is aimed at low-power devices, being compatible with tinyML applications, a branch of artificial intelligence applied to small devices with limited processing capabilities.
Intelligent and Continuous Monitoring of Wind Turbine Blades
The system proposed by the scientists works by capturing sounds emitted by internal or external damage to the turbine blades. These sounds, referred to as acoustic emissions, are processed by algorithms trained to identify patterns indicating structural problems.
-
Built in 1773 by two rival British inventors, the mechanical Silver Swan still works perfectly 253 years later inside the Bowes Museum, in the English county of Durham, without any electric motor, powered only by clock springs wound once a day, and Mark Twain described it in Paris in 1867 as a creature with lively eyes.
-
57 years and one day ago, the Soviet probe Venera 6 traversed the clouds of Venus for 51 minutes under parachutes and stopped transmitting 10 kilometers from the surface because the pressure of 60 bar and the heat of 320 degrees Celsius crushed its 405-kilogram hull, and no space agency has managed to replicate the feat to this day.
-
Harvard researchers discover that the more elderly people walk 3,000 to 7,500 steps per day, the more years they delay the progression of Alzheimer’s in the brain: a study by Mass General Brigham followed at-risk patients and showed a gain of up to seven years compared to sedentary individuals, while Brazil races against the clock with 5.7 million cases expected by 2050.
-
Between Alaska and Siberia, there is an abyss on the sea floor, a submarine trench 2,600 meters deep that surpasses the Grand Canyon in vertical scale and hosts one of the rarest ecosystems on the planet.
During testing, the new technology was applied to composite material wind turbine blades. The researchers simulated various types of damage, such as cracks, impacts, and wear caused by the environment, with the goal of training and evaluating the model.
Based on this data, the system was able to identify damage with an accuracy rate exceeding 96%, outperforming other traditional models that tend to have lower performance. Additionally, the solution can process the information in real-time, facilitating a quick response for repairs or replacements.
Scientists Seek Greater Efficiency for Wind Energy
Wind energy is one of the main renewable sources currently used in Brazil and worldwide. Efficient maintenance of turbines is crucial to ensure the continuous and safe production of this energy. The blades, in particular, are subject to high levels of stress, environmental impacts, and natural wear.
According to the study authors, the new technology can reduce operational costs, decrease downtime of wind turbines, and extend the lifespan of equipment. By automating the inspection process, which currently relies on specialized labor and expensive equipment, the system allows for a more efficient and economical approach for the sector.
Future Applications and Large-Scale Viability
In addition to detecting damage, the HHMLM model is also capable of estimating the type of failure and its exact location using only a single acoustic emission sensor. This represents an advance over other solutions that require multiple sensors or manual analysis steps.
The scientists highlight that the technology can be implemented in wireless sensors, allowing its use in remote or hard-to-access environments, such as offshore wind farms. This flexibility increases the feasibility of using the solution on a large scale.
The next step for researchers is to work on miniaturizing the system and integrating it with cloud monitoring platforms. This will allow the collected data to be accessed remotely and used for more precise and rapid decision-making.
Clean Energy Supported by Artificial Intelligence
As artificial intelligence advances and solutions like this are developed, scientists are broadening the possibilities for optimization in renewable energy sources. The application of cutting-edge technology in wind turbines is an example of how innovation can contribute to making the global energy matrix more sustainable.
The study reinforces the role of scientific research in creating tools that assist in the transition to a low-carbon economy. The new technology represents an important step in this direction by offering an intelligent and automated alternative for monitoring and maintaining the turbines used in wind energy generation.
Source: Scientific Reports
Be the first to react!