DOI: 10.14489/td.2016.08.pp.054-057

Yusupbekov N.R., Gulyamov Sh.M., Rasuleva M.A., Ergashev F.A., Temerbekov B.M.
SHORT-TERM FORECASTING IN THE OPERATIONAL MANAGEMENT TASKS OF CHEMICAL-TECHNOLOGICAL PROCESSES AND PRODUCTIONS
(pp. 54-57)

Abstract. The paper outlines the development of operational conduct short-term forecasting algorithms for dynamic objects on the basis of procedures consistent dynamic filtering primary measurement information. It is shown that the optimal kalmanovsky filter type is very effective in terms of improved accuracy primary measurement information in the case of a priori specified linear models of technological processes. It deals with the problem of increasing the efficiency of systems of automated operational dispatch control of complex chemical-technological processes that operate in real time, based on algorithms short forecasting parameters of objects management procedures and dynamic multi-step filtration. The ratios are analyzed to provide the optimal solution to the problem of  multi-step dynamic filtering for cases where there is a sufficiently representative statistical data about the behavior of the object. The allegation is proved that in the case of the normal distribution law with optimal information processing efficiency criterion (in this case, the criterion of A-optimality) cannot be improved by switching to a non-linear operators, i.e. the optimum linear operator is generally the best one. The general informative statement of the characteristics of the problem of forecasting the stochastic process in terms of the static theory of optimal estimation is performed. It is shown that the optimal adaptive kalmanovsky filter type is efficient in terms of the accuracy of the forecasts in the case of a priori specified linear mathematical models of chemical-technological processes and systems. Analytical design of an optimal filter is carried out within the framework of the correlation theory. Under correlation theory carried out analytical design of optimal filter. It is a filter of kalamanovsky type. The filter provides the best estimates of the parameters of the process being forecast for the generalized minimum variance estimates state (minimum trace of the covariance matrix) vector. It enables us to construct an algorithm predicting unobserved values of the studied object parameters.

Keywords: short-term forecasting, dynamic filtering, parameter estimation models of complex dynamic objects.

 

This article  is available in electronic format (PDF).

The cost of a single article is 350 rubles. (including VAT 18%). After you place an order within a few days, you will receive following documents to your specified e-mail: account on payment and receipt to pay in the bank.

After depositing your payment on our bank account we send you file of the article by e-mail.

To order articles please fill out the form below:

Purchase digital version of a single article Your name * Your e-mail * Your post address with index * Your phone number: * Comments to your order Type the characters you see in the picture below