Abstract

Sejak kebelakangan ini, ramai penyelidik telah memberi perhatian kepada aplikasi rangkaian neural untuk pengenalpastian sistem. Dalam penulisan ini, model Hammerstein untuk suspensi pasif kereta suku telah dikenal pasti menggunakan rangkaian perceptron neural anggapan berbilang lapis. Data masukan dan keluaran telah diperolehi dengan memandu sebuah kereta pada satu paras jalan yang khas. Struktur daripada rangkaian adalah berdasarkan kepada model daripada sistem. Rangkaian algoritma pembelajaran adalah berdasarkan kaedah pemarkahan Fisher. Maklumat Fisher ini diberikan sebagai pemberat matrix kovarian masukan dan keluaran lapisan rangkaian yang tersembunyi. Unitwise daripada kaedah pemarkahan Fisher berkurangan kepada algoritma di mana setiap unit menganggarkan sendiri pemberat dengan menggunakan kaedah pemberat kuasa dua terkecil. Hasilnya telah menunjukkan bahawa nilai ralat minimum punca knasa dua proses pembelajaran diperoleh dengan satu lelaran yang singkat. Kata kunci: Pengenalpastian sistem, model Hammerstein, rangkaian perceptron neural berbilang lapis, pemberat kuasa dua terkecil, maklumat Fisher, pemarkahan Fisher Recently, some researchers have focused on the applications of neural networks for system identification. In this paper, a Hammerstein model of a quarter car passive suspension system is identified using multilayer perceptron neural networks. Input and output data are acquired by driving a car on a special road event. The networks structure is based on system model. The network learning algorithm is based on Fisher’s scoring method. Fisher information is given as a weighted covariance matrix of inputs and outputs of the network hidden layer. Unitwise, Fisher’s scoring method reduces to the algorithm in which each unit estimates its own weights by a weighted least square method. The results show that the minimum mean square error (MSE) value of the training process was found with a short record. Key words: System identification, Hammerstein model, multilayer perceptron, weighted least square, Fisher information, Fisher’s scoring

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.