BOGDAN DOICIN, DORIN STANICA EZEANU, DIANA CURSARU RELATII DE ESTIMARE A CONVERSIEI ETANOLULUI SI RANDAMENTULUI IN HIDROGEN UTILIZAND RETEAUA NEURONALA PENTRU REFORMAREA CU ABUR A ETANOLULUI PE CATALIZATOR DE CO/AL2O3 An artificial neural network has been developed for estimation of ethanol conversion and hydrogen yield function of reaction conditions in ethanol steam reforming process based on Co/Al2O3. Ethanol (8% and 10% vol) was considered as raw material for hydrogen production by steam reforming. For this investigation, we have created an artificial neural network to describe the complex relationship between ethanol steam reforming and the Co/Al2O3 catalyst used in the process. To test the estimation accuracy of the created neural network, a set of input data, whose outputs are already known was used. For these input data, the output data are calculated and then compared with the known outputs.