Open AccessINTEGRATED APPLICATION OF SYNERGETIC APPROACH AND NEURAL NETWORK STRUCTURES TO THE PROBLEM OF ELECTRIC DRIVE INTELLECTUAL CONTROL SYSTEM SYNTHESIS
Author(s) -
А. А. Колесников,
Daniil Marshakov,
Andrey Razmikovich Aydinyan
Publication year - 2014
Publication title -
vestnik donskogo gosudarstvennogo tehničeskogo universiteta
Language(s) - English
Resource type - Journals
eISSN - 1992-6006
pISSN - 1992-5980
DOI - 10.12737/7165
Subject(s) - artificial neural network , controller (irrigation) , control engineering , computer science , control theory (sociology) , parametric statistics , key (lock) , control (management) , control system , artificial intelligence , engineering , mathematics , statistics , computer security , electrical engineering , agronomy , biology
The applicability of the artificial neural networks to the synthesis of the intellectual systems with the synergetic control law is considered. The key points of the synergetic approach to the system synthesis problem are stated. An example of the synergetic synthesis of the robot electrodrive control law under the conditions of its uncontrolled parametric variation and external disturbances is considered. An algorithm for synthesis of the neural network controller based on a predetermined synergetic control law is presented. The algorithm is based on the synergetic controller behavior simulated by an artificial neural network as a result of its training through harmonic signals of different frequencies. The key element of the proposed integrated approach to the synthesis of the intelligent control system is that it involves a combination of the self-organization unity principle and the training of the neural network at the preliminary stage. This ensures a subsequent stable system operation.