Abstract

The development of modern automated systems for monitoring, control and management of technological processes is associated with the improvement of methods for identifying and studying the primary means of collecting and processing information. An algorithm for neural network parametric identification using reservoir calculations is proposed. An algorithm for parametric identification orient to conducting a computational experiment on a given nonlinear model, forming training samples based on the results of the experiment, training dynamic and static neural networks and calculating estimates of parameters of a nonlinear model using trained networks according to experimental data. Keywords eddy current sensor, identification, dynamic model, equivalent circuit, static neural network, dynamic neural network, reservoir computing, mathematical model

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