Stochastic contraction based online estimation of second order wiener system

Area of Science:

• Control Systems Engineering
• Signal Processing
• Nonlinear System Identification

Background:

• Wiener systems, characterized by a linear time-invariant dynamic system followed by a memoryless nonlinearity, are prevalent in various engineering applications.
• Accurate online estimation of parameters and states in these systems is crucial for effective control and monitoring.
• Existing methods often struggle with noisy data and unknown nonlinear structures.

Purpose of the Study:

• To design a novel stochastic estimator for the online estimation of second-order Wiener systems.
• To simultaneously estimate system parameters and states from noisy output data.
• To address systems with unknown nonlinearity structures.

Main Methods:

• Utilizes the differential mean value theorem.
• Applies results from stochastic contraction theory, specifically for semi-contracting systems.
• Employs analytical derivations to demonstrate convergence and boundedness of estimates.

Main Results:

• The proposed stochastic estimator demonstrates asymptotic convergence.
• Analytical proofs confirm the boundedness and convergence of both parameter and state estimates.
• Numerical simulations validate the estimator's accuracy using real-world system examples with additive noise.

Conclusions:

• The developed stochastic estimator accurately estimates parameters and states of Wiener systems online.
• The method is robust to measurement noise and unknown nonlinearities.
• This approach offers significant potential for real-time applications in system identification and control.