Identification of wiener fractional model using Self-Adaptive Velocity Particle Swarm Optimization

Support vector machine and higher‐order cumulants based blind identification for non‐linear Wiener models

“…Similarly, when the parameters [ a 1 , a 2 ] and [ l 1 , l 2 , l 3 ] (i.e. the parameter vectors w and w 1 ) are further estimated by using the PSOA [26], the self‐adaptive velocity particle swarm optimisation (SVPSO) [27], the GA [28], the DEA [33], the recursive LSM (RLSM) [35] and the HGA [36], respectively, the estimates and estimated errors of the parameters are also shown in Tables 1 and 2, respectively. TOC is also used to identify the model's order n and parameters [ b 1 , b 2 ], respectively.…”

“…Among different non-linear system identification structures, Wiener models have been found very useful in modelling nonlinear systems such as pH neutralisation processes [20], distillation column [21], and electroencephalograph [22]. Recently, the existing literature related to the identification of Wiener models includes the correlation analysis method [23], maximum likelihood method [24], subspace method [25], particle swarm optimisation algorithm (PSOM) [26,27], genetic algorithm (GA) [28], frequently method [29], semi-parametric Bayesian method [30], recursive identification method [31], iterative method [32], differential evolution algorithm (DEA) [33], orthonormal basis functions method [34], recursive least square method (LSM) [35], hierarchical gradient approach (HGA) [36] etc. Most of the existing identification methods for Wiener models often assume that input and output measurements are available.…”