Resumen
This article presents an approach based on artificial intelligence techniques to predict critical deposition in heat exchangers used in petroleum preprocessing. Deposition on the heat exchange surface during operation can reduce equipment efficiency and cause maintenance problems. The proposed method uses a recurrent neural network in a Multi Layer Perceptron (MLP) model. Data was collected from exchangers of a petroleum preheating battery with data from 2014 to 2021. The neural network was trained with the data to predict the occurrence of deposition. The results showed that the neural network is capable of accurately predicting the occurrence of deposition in heat exchangers. Predicting deposition in advance can help minimize maintenance costs and increase energy efficiency, making operations safer and more efficient. Thus, the proposed approach can bring significant benefits to the petroleum industry by allowing early prediction of critical deposition in heat exchangers.
DOI:https://doi.org/10.56238/sevened2023.006-022