Meaning
Mathematical flow of measurement errors through estimation algorithms, which corrupts the calculated output variables. Tracking sensor noise propagation ensures that battery state of charge calculations remain reliable even with cheap current sensors. The process helps in specifying the required accuracy of voltage and current sensors.
Error Amplification
Small measurement errors in cell voltage can lead to large errors in state of charge estimation when the open-circuit voltage curve is flat. Analyzing sensor noise propagation reveals how these voltage errors are amplified by the estimation filter. This analysis is necessary for designing filters that can handle noisy environments.
Filter Resilience
Extended Kalman filters and recursive least squares algorithms are designed to minimize the impact of measurement noise. By accounting for sensor noise propagation, engineers can tune the filter covariance matrices to reject high-frequency noise. This tuning ensures that the estimated states do not drift during long driving cycles, which is critical for preventing the battery management system from prematurely cutting off power due to false state-of-charge readings.
Design Constraint
High-precision sensors can reduce estimation errors but increase the overall cost of the battery pack. Understanding sensor noise propagation allows engineers to find the optimal trade-off between sensor cost and estimation accuracy. This optimization is essential for mass-market electric vehicles where every dollar counts.
