Model description:
$$\begin{align*} y(t) &= y(t-1) + u(t-1) + 1.3u(t-2) + 0.3u(t-1)y(t-1) \\ &+0.5u(t-2)y(t-2)+e(t)/\Delta, \end{align*}$$
where $e(t)$ is normal school white noise signal with covariance 0.1.
Type:
Form:
Model order:
2
Time domain:
Publication details:
| Title | Generalized Predictive Control for a Class Of Bilinear Systems |
| Publication Type | Conference Paper |
| Year of Publication | 1970 |
| Authors | Liu, Guizhi, and Li and Ping |
| Conference Name | Control, Automation, Robotics and Vision |
| Date Published | 2006 |
| Abstract | A new generalized predictive control algorithm for a kind of input-output bilinear system is proposed in the paper (BGPC). The algorithm combines bilinear and linear terms of I/O bilinear system, and constitutes an ARIMA model analogous to linear systems. Using optimization predictive information fully, the algorithm carries out multi-step predictions by recursive approximation. The heavy computation of generic nonlinear optimization is avoided with control law of analytical form being used to the non-minimum phase bilinear systems. Simulation results show the effectiveness of the algorithm and the performance of the algorithm is better than linear generalized predictive control (LGPC). Key words: bilinear systems; bilinear generalized predictive control (BGPC); recursive approaches; non-minimum phase systems; analytical control laws |
| DOI | 10.1109/ICARCV.2006.345181 |