Driving Safe and Efficiently

The electric vehicle battery challenge is central to battery producers, OEMs, and users. Each electric vehicle is exposed to different environmental conditions and usage patterns, resulting in unique batteries that require specific operational performance calculations. Monitoring and predicting battery characteristics such as the state of health (SoH) and state of charge (SoC) allows addressing topics such as range anxiety, predictive maintenance, residual value, and second life for EVs. However, today’s physics-based models developed for calculating such parameters in batteries have two main downsides: low accuracy and high computation costs.

The solution deployed by MHP and AWS uses a hybrid physics and data-driven approach, allowing organizations to create operational models that can run in near-real time and include the effects of real-world factors by leveraging the operational vehicle data. The solution needs to be scalable to a great number of vehicles and modular to ensure additional components can be added as the digital twin becomes more complex. Additionally, it needs to be able to provide a scalable mechanism to automatically recalibrate the models of individual EVs to ensure they continue to make accurate predictions.

The screenshot in Figure 1 below is of the deployed solution for a fleet of four vehicles. We’ve also created a video walkthrough.

The architecture of the solution uses AWS IoT FleetWise to feed simulated vehicle data into an Amazon Timestream database. The data is consumed by the custom-built MHP digital twin service, which stores the prediction metadata in Amazon DynamoDB and the predictions results back into Amazon Timestream. Prediction results are fed into AWS IoT TwinMaker and are used in Amazon Managed Service for Grafana to allow a fleet operator to monitor the state of health and other vehicle-based information in an easy-to-use dashboard. This is shown in Figure 2.

The MHP digital twin service is key to the Level 4 digital twin solution and handles passing operational data to the digital twin model, running the model, calculating the model error, and recalibrating the digital twin model when needed. The service leverages model calibration techniques developed as part of the aws-do-pm open-source framework.

The MHP digital twin service has two key features to enable a modular digital twin that can grow as new components are added. First, the digital twin is divided into submodules, each of which are represented by their own digital twin. Second, each digital twin is divided into three modules (Figure 3) – a data service, update service, and prediction service – to enable a Level 4 living digital twin to use measurements to update the model in order to improve predictions as new data becomes available.

In summary, MHP and AWS have collaborated to build a modular, scalable digital twin solution to address the challenges of electric vehicles, specifically EV battery performance. This solution is a Level 4 living digital twin per the 4-level index proposed by AWS. The solution focuses on predicting the state of health of an EV battery based on driving patterns, using a driver digital twin and a battery digital twin.