System of a Remotely Control of Technical Condition of Complex Electrical Objects

Abstract

Complex technical objects of diagnostics are multifunctional and are characterized by complex organization of interrelations of information paths in the structure from the entrance to the exit. Therefore, the aim of this work is to solve the problem of synthesis of multidimensional models of adaptation along input and output of distributed system of collection and processing of measuring information. This model will solve the problem of building an adaptive system for remote control of the technical condition of a complex electromechanical complex. The proposed model allows changing the structure and/or parameters of the elements of the control system to retrain the neural network model to provide the required value of the output signal at a given signal at the input, while providing a given reliability of the data obtained by the operation of the electrical complex, by changing the weight functions of the neural network.