Research Article
Regularized Least Squares Recursive Algorithm with Forgetting Factor for Identifying Parameters in the Grinding Process
| Symbols | Definition |
| | | | Coefficient of delay link | | Transfer function with delay link and inertia link | | Transfer function with inertia link | | The actual value of ore feeding quantity | | The actual value of classifier adding water quantity | | The actual value of the return sand quantity | | The actual value of classifier overflow concentration (%) | | Observed times | | | , U1K | The actual input signal of recorded data | | The actual output signal of recorded data | a | Parameters to be identified | | Uncorrelated random variable that subjects to distribution | | Normal distribution | | | | | | | | | F | The sum of residuals squares | N | Number of data group | | Forgetting factor ) | | A factor | | Regularization parameter | | Gain matrix | | | | Criterion function of parameter estimation at | | True value of model parameters | | Eigenvalue of matrix ) | | Nonzero eigenvector | | System input of equation (43) | | System output of equation (43) | | Parameters to be identified of equation (43) | | Length of observation data |
|
|