Purpose: This paper deals with a method for the parametric system identification of a nonlinear system to obtain its parametric representation using a linear transfer function. Such representation is applicable in off-line profile correction methods minimizing the error between a reference input signal and a signal performed by the test rig. In turn, a test signal can be perfectly tracked by a servo-hydraulic test rig. This is the requirement in massive production where short test sequences are repeated to validate the products. Design/methodology/approach: A numerical and experimental case studies are presented in the paper. The numerical study presents a system identification process of a nonlinear system consisting of a linear transfer function and a nonlinear output component, being a static function. The experimental study presents a system identification process of a nonlinear system which is a servo-hydraulic test rig. The simulation data has been used to illustrate the feasibility study of the proposed approach, while the experimental data have been used to validate advanced model structures under operational conditions. Findings: The advanced model structures confirmed their better performance by means of the model fit in the time domain. Research limitations/implications: The method applies to analysis of such mechanical and hydraulic systems for which measurements are corrupted by residual harmonic disturbances resulting from system nonlinearities. Practical implications: The advanced model structures are intended to be used as inverse models in off-line signal profile correction. Originality/value: The results state the foundation for the off-line parametric error cancellation method which aims in improving tracking of load signals on servo-hydraulic test rigs.
