Dibeh G. Automation of intelligent control blocks linear electron accelerator

The thesis is devoted to the design of digital regulators for the blocks of stabilizing of power supply of magnetic elements and energy of bunch of electrons in the microprocessor control system of electron accelerators. The analysis of existent methods of description of nonlinear processes of the linear accelerator is conducted on expedience of fuzzy and neural network models application for digital control accelerating systems. The methods of the control energy of electrons for the linear accelerator, based on application of the neural network perceptron type model, is offered. It allows to draw conclusion about possibility of his application in the digital control the magnetic elements of accelerating system. The method of control, based on introduction of additive constituents, allowing to promote robustnesse of fuzzy regulator is offered. The variant of moderniza-tion of structure of control system the linear electron accelerator LA-40 is offered. Authenticity of the got dissertation job performances is confirmed experimental researches and results of introduction.