Parametric identification of nonlinear model: a study of high-pressure sodium lamp

The problem was posed and an algorithm developed for neural network parametric identification of nonlinear dynamic models of objects with a computational experiment on the model by varying the identified parameters. The training samples were formed on its basis, followed by sequential training of two neural networks that bijectively map the parameters of the original model into the output variables of the second neural network. The efficiency of the proposed algorithm is estimated using the example of parametric identification of a nonlinear model of a high-pressure sodium lamp.