Font Size: a A A

Nonlinear Predictive Control Algorithm Based On Least Squares Support Vector Machine

Posted on:2014-05-12Degree:MasterType:Thesis
Country:ChinaCandidate:H X WangFull Text:PDF
GTID:2298330452462653Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
Statistical Learning Theory (SLT) is a theory that specialized in machinelearning with finite samples, and it is useful for practical system. Support vectormachine theory is a new learning method basing on this theory. During the machinelearning process of small samples, support vector machine can still maintain highfitting accuracy, global optimization, good generalization and a series of veryimportant and outstanding performance. As the traditional neural network innonlinear system modeling, Support vector machine have the ability to model andpredict models. So these characteristics of the support vector machine used innonlinear system modeling and predictive control is to be studied in the modeling andcontrol process. The article describes the basic principles of statistical learning theoryand support vector machines and their mutual relations in detail and the supportvector machine regression for data classification and implementation process. Thesupport vector machine regression modeling and model predictions is studied andvector machine model is taken advantage to the predictive control.The main content of the article are summarized as follows:First, the system of learning, support vector machine theory is learned. We studythe support vector machine in data classification and prediction, and make simulationof examples to indicate the feasibility and effectiveness of support vector machinelearning algorithm.Secondly, we will do study and research of the support vector machineregression modeling, using least squares support vector machine as the main tool, anonlinear system support vector machine model is made, and we will verify theaccuracy of the established model. Finally, the generalized predictive control of nonlinear systems based on supportvector machine model is studied. The ability of predicting of support vector machinemodel is used to to predict system output and to strike a control. Support vectormachine model of the nonlinear system-CSTR is built and we do predicting based onthis model. On this basis, generalized predictive control is applied and a good controleffect is obtained.
Keywords/Search Tags:LS-SVM, nonlinear systems, model prediction, predictive control
PDF Full Text Request
Related items