Singapore’s tropical climate affects battery energy storage in terms of safety, performance, and lifespan. Project FXMS is one unique approach to solve these challenges for batteries.
FXMS uses a virtual battery-based digital twin technology to combine real-time data through the IoT, leveraging advanced machine learning and data mining algorithms. The project was funded by Energy Market Authority (EMA) Singapore.
To verify battery's performance and longevity, chamber testing is required. We have done chamber testing under various temperature and humidity ranges. The battery is cycled continuously and the battery states are monitored and stored to develop mathematical models.
A Deep learning based framework that can enhance the richness of obtained temperature information and prediction accuracy with limited sensor data.
This algorithm enables real-time temperature reconstruction from sparse sensor data, facilitating early detection of thermal fault issues in batteries.
An online self-adaptation mechanism based temperature estiamtion model to adjust to changing operational conditions without the need for retraining.
A mahematical model that looks at a battery’s degradation history in a T-shaped pattern. This work incorporated several Machine Learning techniques that help it learn better and work with different batteries.
A dynamic framework for offering the optimal feasible operation space. The improved lifespan is achieved if the battery is confined within such evolving feasible domains throughout its service time.
A Probablisitc model for estiamting battery's remaining useful life. This model is specifically developed for the cases suffering from limited data availability.
A model predictive control (MPC)-based framework for optimal disaptch of grid-scale batteries. This frameowrk considers stress due to both depth of discharge as well as temperature under partial charge-discharge cycles.
A battery control model that establishes the coupled relationship among the dendrite growth rate, battery power dispatch, and temperature. The resulting model can be used for regulation of dendrite growth.
We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.