<b>VEKTOR KVANTLANGAN VAQTINCHALIK ASSOTSIATIV XOTIRA ASOSIDA</b><b> </b><b>DINAMIK OBYEKTLARNI IDENTIFIKATSIYALASH</b><b> ALGORITMI</b>

In tasks of identifying relationships between technological parameters of automated control objects, the capabilities of neural networks are used. One of the important problems considered in such identification tasks is the choice of the neural network architecture for the modeled dynamic object. The article discusses the use of a neural network based on vector-quantized temporal associative memory for dynamic object identification. The input and output parameters of neural networks, the identification scheme, and the method for solving the problem of approximating a given function in a self-learning network are presented. A scheme for using a self-learning network for identification and the architecture of the Kohonen network used are presented. An algorithm for training network parameters is written.

Hali tarjima qilinmagan