A method for controlling a technological process, comprising forming information nodes of lattice structures of signals of technological information, forming three-dimensional time series of signals of controlled parameters of the technological process, connecting information nodes of these series with an artificial neural network, training the artificial neural network and selecting a level of the controlled variable of the technological process with its help, and forming a two-dimensional time series of the controlled parameter, as well as two-dimensional orthogonal vectors to the three-dimensional time series of the controlled parameter,` find the correlation coefficients between the time series of the controlled parameter and the two-dimensional time series of the orthogonal vectors of the controlled parameters, use the output information node of the orthogonal vector with the highest value of the correlation coefficient of the corresponding controlled parameter for communication with the artificial neural network, in the process of training which finds the most informative nodes of lattice structures (lines, fields, volumes) that dominate the control process.
Comments are closed