Do the benefits of AI for solar and wind energy exist?

Solar and wind power are booming, but the world's transition to renewable electricity is still too slow to quickly meet climate goals. Harnessing wind and solar energy on a global scale is easier said than done for many reasons. One is that wind turbines and solar panels are complex, finicky engineered systems that are prone to failure. Frequent breakdowns reduce power output and make wind and solar farms expensive to operate and maintain.

Joyjit Chatterjee, a data scientist at the University of Hull in England, said the ability to use artificial intelligence to predict power production and component failures could make renewable electricity more economical and reliable to Accelerate widespread adoption. However, it is not used in this area as it is in many other areas such as e-commerce, manufacturing, and healthcare. "Artificial intelligence could have a real impact on climate change and sustainability," he said, "but there is very little work related to the renewable energy field."

So Chatterjee and his colleagues, He Nina Dethlefs, Director of Computer Science at MU, brought together experts in artificial intelligence and renewable energy at the recent International Conference on Learning Representations (ICLR). In a perspective paper published June 10 in the Data Science Journal Patterns, the pair present key takeaways from the conference, outlining the barriers limiting the impact of AI on renewable energy and how established and emerging technologies can be used to artificial intelligence methods to overcome these obstacles.

Wind turbines and solar panels on utility-scale farms are equipped with sensors that allow operators to remotely monitor their power generation and health. These sensors include vibration sensors, temperature sensors, accelerometers, and speed sensors. The data they generate offers an opportunity. AI models trained on historical power generation and failure data can predict unexpected failures in wind turbine gearboxes or solar panel inverters, helping operators prepare for outages and plan routine maintenance.

Chatterjee said reinforcement learning is an exciting new machine learning technique that can help improve these models. In reinforcement learning, an algorithm interacts with the world during training, receiving continuous feedback on reward or punishment decisions to learn how to achieve certain goals. This type of real interaction could come from humans.

“One of the dangers of AI is that it’s not perfect,” Chatterjee said. “We can have people involved to constantly help optimize the AI ​​model. People often worry that AI will replace the human part and make decisions. . But humans need to work with AI models to jointly optimize the models for decision support."

He added that a focus on natural language generation (the process of converting data into human-readable text) will enhance the Trust in artificial intelligence and increase its use. Due to a lack of transparency, industry engineers are reluctant to use the few failure prediction models created by researchers. Providing operators with brief natural language messages will facilitate interaction.

For the artificial intelligence community, one of the barriers to creating better models is the limited amount of publicly available data, given the commercial sensitivities of the wind and solar industries. Chatterjee said that in addition to industry reluctance to share data openly, a lack of standards also affects the development of AI models. "Wind farm operators in different parts of the world manage data differently, so it's really challenging for researchers to use resources together."

To solve this problem, the artificial intelligence community can use a method called Machine learning techniques for transfer learning. By identifying hidden patterns in various features in the data, this method allows data scientists to transfer the knowledge gained from solving one machine learning task to another related task, making it easier to train neural networks and develop deep learning models when data is limited. . "This will help you develop a model for turbine Y based on a model just for turbine X, even without historical data," Chatterjee said. However, neural networks are not necessarily Always the answer. These deep learning models have become popular because they are traditionally suitable for learning from images and text. The problem is, neural networks often fail. Furthermore, training these large-scale, computationally complex models requires energy-intensive high-performance computing infrastructure, which is difficult to achieve in developing countries.

At least for the renewable energy sector, sometimes it can be okay to be simple. The AI ​​community should first focus on using simpler machine learning models, such as decision trees, to see if they work. "Generally not every problem requires a neural network," Chatterjee said. "Why increase carbon emissions by training and developing more computationally complex neural networks? Future research needs to be conducted on less resource-intensive and carbon-intensive models." ”

The above is the detailed content of Do the benefits of AI for solar and wind energy exist?. For more information, please follow other related articles on the PHP Chinese website!