Abstract
The co-estimation of the ultracapacitor states and parameters using the Unscented Kalman Filter (UKF) is fulfilled in this paper. In the proposed method, the state-space model of the ultracapacitor based on a three-branch equivalent circuit model is used for joint state and parameter estimation. To demonstrate the observability of this model, the dynamic interdependence of the system states is analyzed using a graphical approach (GA). Unlike the KF and Extended KF (EKF) methods that use linearized system models, the UKF uses a nonlinear transformation (unscented transform), which improves the estimation accuracy. In addition, the correlations between the states and parameters are taken into account during the estimation process, which further improves the state estimation accuracy. Some experiments are conducted in the laboratory for testing the usefulness of the suggested approach. The results demonstrate that the suggested approach can provide accurate state and parameter estimation results while it is also able to run in real-time on low-cost digital signal processors (DSPs).
Original language | English |
---|---|
Title of host publication | 2020 11th Power Electronics, Drive Systems, and Technologies Conference, PEDSTC 2020 |
Publisher | IEEE |
Publication date | Feb 2020 |
Article number | 9088356 |
ISBN (Electronic) | 978-1-7281-5849-5 |
DOIs | |
Publication status | Published - Feb 2020 |
Event | 11th Power Electronics, Drive Systems, and Technologies Conference, PEDSTC 2020 - Tehran, Iran, Islamic Republic of Duration: 4 Feb 2020 → 6 Feb 2020 |
Conference
Conference | 11th Power Electronics, Drive Systems, and Technologies Conference, PEDSTC 2020 |
---|---|
Country/Territory | Iran, Islamic Republic of |
City | Tehran |
Period | 04/02/2020 → 06/02/2020 |
Keywords
- Condition Monitoring
- Electric Vehicles (EVs)
- State Estimation
- Ultracapacitor