In energy storage systems (batteries and supercapacitors), researchers use various methods to adjust the microstructure, surface area and porosity of active electrode materials. The macroscopic and microcosmic structure features of electrode materials affect electrochemical performance.
However, the widely-accepted structure-performance relationship is still unclear. Therefore, the biggest challenge is to obtain the relations between the microscopic structures and the macroscopic performances for predicting the capacity/capacitance, life cycles and many other material characteristics.
In this project, we'll develop machine learning technology to predict the performance of batteries and supercapacitors. Machine learning provides powerful tools that use existing data to better understand the relationships between the structure, properties and performance of origin materials. We expect to analyse the function between the structure and performance of materials from available databases with the help of machine learning technology.
During this project, we'll develop ML technology which can predict the performance of energy storage systems in terms of their energy, life, egradation, speed, etc.
Research activities can include:
- collecting experimental data, including the physical and chemical properties of electrode materials or the testing parameters
- developing algorithms
- applying different machine learning models, such as artificial neural network models
- evaluating different machine learning models ability to predict performance, using correlation coefficient and root mean square error.
The supervisory team has a strong track record and a wide range of expertise in energy storage.You'll work in motivated and highly-collaborative research group. In return, the project will provide you with an effective and rich learning experience.
The project aims to:
- develop proper theoretical models that predicts the electrochemical properties of the electrode materials as well as cell characteristic
- publish high-quality results in international journals
- gain an advanced understanding of machine learning for energy storage devices
- lead to a potential breakthrough in creating powerful tools for predicting materials characteristics and performance.
Skills and experience
As the ideal candidate, knowledge in computational technology, computer science and machine learning technology is highly desirable.
You may be able to apply for a research scholarship in our annual scholarship round.
Contact the supervisor for more information.