الفهرس 
SUPPORT VECTOR MACHINESOnly 14 pages are availabe for public view
 |  | from | 166 |  |  |
| | | from | 166 | | |
Abstract
Support Vector Machines (SVMs) is a new paradigm that uses a group of supervised learning methods for both classification (pattern recognition) and regression (function estimation) problems. However Neural Networks are used extensively in modeling of dynamic systems, they suffer from many problems such as many local minima, determining the structure of the network and the number of parameters to be determined is increased with growing the network structure. Least Squares Support Vector Machines (LS-SVMs) is a modified version of SVM. In this case, training means solving a set of linear equations, instead of the quadratic programming problem involved by the standard SVM. Traditional linear control techniques are mainly based on the quality of the model obtained, if the model is of high quality, the controller will track the desired output.Electro-hydraulic servo-mechanisms are well known in control systems for their fast dynamic response, high power to inertia ratio and control accuracy. If the system dynamics can be precisely described and the system dynamics vary around the designed operating point, a fixed parameter controller may be designed based on conventional control theory to precisely track the desired output. For most industrial systems, it is very difficult to model the system precisely with traditional methods. In addition due to disturbances, variations of loads, and changing process dynamics, the system parameters may vary, so a good modeling technique is required. LS-SVM is a nonlinear modeling paradigm used extensively to model nonlinear systems is used here to precisely model this hydraulic system.The major objective of the research undertaken in this thesis was to apply a predictive Controller to a nonlinear electro-hydraulic servo-system to improve its speed output performance. This objective was achieved through the use of an on-line adaptive Generalized Predictive Control (GPC) based on LS-SVM using linear Kernel Function (KF). The on-line adaptive GPC controller based on LS-SVM was implemented to control the real-time electro-hydraulic servo-system to track the desired outputs under different operating conditions and load variations. Experiments were conducted to illustrate the feasibility and benefits of the online adaptive GPC controller based on LS-SVM in comparison with the traditional PI, its velocity-form and self-tuning PID strategies. The speed output of the hydraulic motor followed the desired output successfully. The proposed control scheme forced the hydraulic motor speed output to track the reference outputs simultaneously under changes of the load disturbances.