Meaning
Mathematical estimation technique used to calculate the internal distributed variables of a system across its physical dimensions using limited external sensors. Implementing spatial state estimation helps identify local hot spots and lithium depletion zones in long cell designs. This approach relies on physical models to fill in the gaps between sensor locations.
State Reconstruction
Advanced algorithms use a small number of surface-mounted sensors to reconstruct the full 3D temperature and state-of-charge profiles inside the cell. When performing spatial state estimation, the algorithm must run in real time on the battery management processor. This reconstruction is important to preventing localized degradation in high-energy packs.
Sensor Placement
The locations of the sensors on the cell surface must be carefully selected to maximize the information they capture. By optimizing sensor placement, engineers can improve the accuracy of the spatial state estimation without increasing the number of sensors. This selection process is often guided by observability analysis.
Numerical Overhead
Running these multi-dimensional models in real time requires significant processing power. To reduce the computational load, spatial state estimation algorithms often use reduced-order models that capture the essential dynamics with fewer equations. This reduction is necessary to make the algorithm compatible with standard microcontrollers, which typically have limited memory and processing speed, thereby enabling cost-effective implementation in automotive-grade battery management systems.
