عنوان :
طراحي كنترل كننده شبكه موجك فازي تطبيقي براي سيستمهاي غير خطي
محل تحصيل :
اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر
صفحه شمار :
ده، 169، [II]ص.: مصور، جدول، نمودار
يادداشت :
ص.ع.به: فارسي و انگليسي
استاد راهنما :
فريد شيخ الاسلام، سعيد صدري
استاد مشاور :
فرح تركمني آذر، بهزاد مشيري
توصيفگر ها :
روشOLS , فيلتر كالمن گسسته , فيلتر كالمن توسعه يافته
تاريخ نمايه سازي :
14/12/87
استاد داور :
علي اكبر صفوي، محمدباقر شمس اللهي،سعيد حسين نيا،محمدرضا احمدزاده
دانشكده :
مهندسي برق و كامپيوتر
كد ايرانداك :
ID223 دكتري
چكيده فارسي :
به فارسي و انگليسي: قابل رؤيت در نسخه ديجيتال
چكيده انگليسي :
Abstract In recent years by utilizing soft computing and wavelet theory a numberof efficient techniques are represented among which are wavelet networks andfuzzy wavelet networks In wavelet networks the Universal Approximation property is guaranteed and an explicit link between the network coefficients andthe wavelet transform is fulfilled and accordingly an initial guess for networkparameters can be derived Also potential achievement of the same extent ofapproximation is provided with a network of reduced size On the other hand the wavelet networks are optimal approximators because they require thesmallest number of bits to obtain an arbitrary precision The localizationproperty of wavelet decomposition is reflected in the important properties ofwavelet networks The wavelet neural networks can approximate any function toan arbitrary precision with a finite sum of wavelets and can capture differentbehaviours of global or local approximated function Also the waveletnetwork provides an adaptive discretization of the wavelet transform bychoosing influential wavelets based on a given data set and it is possible tohandle problems of large dimension In this thesis we have presented a newadaptive fuzzy wavelet network controller A FWNC for control of nonlinearaffine systems inspired by the theory of multiresolution analysis MRA ofwavelet transforms and fuzzy concepts The proposed adaptive gain controller which results from the direct adaptive approach has the ability to tune theadaptation parameter in the THEN part of each fuzzy rule during real timeoperation Each fuzzy rule corresponds to a sub wavelet neural network sub WNN and one adaptation parameter Each sub WNN consists of wavelets witha specified dilation value The degree of contribution of each sub WNN can becontrolled flexibly Orthogonal Least Square OLS method is used to determinethe number of fuzzy rules and to purify the wavelets for each sub WNN Sincethe efficient procedure of selecting wavelets used in the OLS method is not verysensitive to the input dimension the dimension of the approximated functiondoes not cause the bottleneck for constructing FWN Fuzzy wavelet network isconstructed based on the training data set of the nominal system and theconstructed fuzzy rules can be adjusted by learning the translation parameters ofthe selected wavelets and also determining the shape of membership functions Then the constructed adaptive FWN controller is employed such that thefeedback linearization control input can be best approximated and the closed loop stability is guaranteed The performance of the proposed A FWNC isillustrated by applying a second order nonlinear inverted pendulum system andcompared with previously published methods Simulation results indicate theremarkable capabilities of the proposed control algorithm It is worth noting thatthe proposed controller significantly improves the transient responsecharacteristics and also the number of fuzzy rules and on line adjustableparameters are reduced
استاد راهنما :
فريد شيخ الاسلام، سعيد صدري
استاد مشاور :
فرح تركمني آذر، بهزاد مشيري
استاد داور :
علي اكبر صفوي، محمدباقر شمس اللهي،سعيد حسين نيا،محمدرضا احمدزاده
