In the paper a fuzzy model based predictive control algorithm is presented. The proposed algorithm is developed in the state space and is given in analytical form, which is an advantage in comparison with optimisation based control schemes. Fuzzy model-based predictive control is potentially interesting in the case of batch reactors, heat-exchangers, furnaces and all the processes with strong nonlinear dynamics and high transport delays. In our case it is implemented to a continuous stirred-tank simulated reactor and compared to optimal PI control. Some stability and design issues of fuzzy model-based predictive control are also given.
The fundamental methods which are essentially based on the principal of predictive control are Generalized Predictive Control , Model Algorithmic Control  and Predictive Functional Control , Dynamic Matrix Control , Extended Prediction Self-Adaptive Control  and Extended Horizon Adaptive Control . All those methods are developed for linear process models. The principle is based on the process model output prediction and calculation of control signal which brings the output of the process to the reference trajectory in a way to minimise the difference between the reference and the output signal in a certain interval, between two prediction horizons, or to minimise the difference in a certain horizon, called coincidence horizon. The control signal can be found by means of optimisation or it can be calculated using the explicit control law formula [3, 11]. The nature of processes is inherently nonlinear and this implies the use of nonlinear approaches in predictive control schemes. Here, we can distinguish between two main group of approaches: the first group is based on the nonlinear mathematical models of the process in any form and convex optimisation , while the second group relies on approximation of nonlinear process dynamics with nonlinear approximators such as neural networks [28, 29], piecewise-linear models , Volterra and Wiener models , multi-models and multi-variables [16, 23], and fuzzy models [1, 25]. The advantage of the latter approaches is the possibility of stating the control law in the explicit analytical form.
Publisher : Springer
By : Sašo Blažic · Igor Škrjanc
File Information: English Language/ 14 Page / size: 358 KB
سال : 2007
ناشر : Springer
کاری از : Sašo Blažic · Igor Škrjanc
اطلاعات فایل : زبان انگلیسی / 14 صفحه / حجم : KB 358
لینک دانلود : روی همین لینک کلیک کنید