FORECASTING THE EVAPORATION OF MOTOR FUEL IN STORAGE WITH THE USE OF NEURAL NETWORKS

Abstract

A neural network was obtained to predict the evaporation of fuel when stored from horizontal reservoirs with a volume of 5 and 10 m3 with a forecast accuracy of not less than 95%. Applying the obtained network to the data from the program "Automated control system for horizontal reservoirs at the oil depot", it was possible to obtain the results of the operational forecast for 3 days for storage of fuel in a 5 and 10 m3 tank. The average estimate of the accuracy of the forecast for 1 month was 90.5 and 93.4%, respectively.