| 摘要 |
PROBLEM TO BE SOLVED: To provide recursive algorithm of adjustable multiplex parameter regression with a parameter determined in regressed analysis and reinforced by a proper forgetting factor. SOLUTION: The algorithm application includes a lead acid storage battery, a nickel hydrogen battery, and a lithium ion battery. The controlling algorithm has an arbitrary number of model parameters having each exclusive time weighting factor. Recently obtained data are more largely influenced in order to determine a state of system by a method for determining an optimal value of time weighting factor. A weighted recursive least squares method is employed, and time weighting corresponds to an exponential forgetting formalism. A drived mathematical result does not involve inverse matrix operation, and the method is iterative. Each parameter is regressed individually at every time step. COPYRIGHT: (C)2007,JPO&INPIT |
