Abstract
This paper considers the parameter identification for a class of nonlinear stochastic systems with colored noise. We filter the input-output data by using an estimated noise transfer function and obtain two identification models, one containing the parameters of the noise model, and the other containing the parameters of the system model. A data filtering based recursive generalized extended least squares algorithm is proposed by using the data filtering technique, and a recursive generalized extended least squares algorithm is derived for comparison. Finally, an example is given to support the proposed algorithms. Compared with the recursive generalized extended least squares algorithm, the data filtering based recursive generalized extended least squares algorithm can not only reduce the computational burden, but also enhance the parameter estimation accuracy.
Highlights
Mathematical models are the basis of controller design [1]–[4] and system analysis [5]–[7]
Zhang et al proposed several state-space recursive identification algorithms for the bilinear systems including a state filtering-based least squares algorithm with the hierarchical identification principle [42], a hierarchical approach for joint parameter and state estimation algorithm [43] and a combined state and parameter estimation algorithm [44], which can directly provide the state-space model, but the computational complexity increases as the dimensions of the parameter vectors increase
As the practical processes are usually disturbed by stochastic noises, we introduce a noise term ω(t) ∈ R to (3), and we obtain a bilinear system with colored noise
Summary
Mathematical models are the basis of controller design [1]–[4] and system analysis [5]–[7]. Zhang et al proposed several state-space recursive identification algorithms for the bilinear systems including a state filtering-based least squares algorithm with the hierarchical identification principle [42], a hierarchical approach for joint parameter and state estimation algorithm [43] and a combined state and parameter estimation algorithm [44], which can directly provide the state-space model, but the computational complexity increases as the dimensions of the parameter vectors increase. Different from the iterative algorithms and the recursive state-space identification algorithms, this paper derives an recursive identification algorithm using the data filtering technique to reduce the computational burden and enhance the parameter estimation accuracy.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have