Development of HT- BP Neural Network System for the Identification of Well Test Interpretation Model

摘要: The neural network technique that is a field of artificial intelligence (AI) has proved to be good model classifier in all areas engineering and especially, it gained considerable acceptance well test interpretation (WTIM) identification petroleum engineering. Conventionally, the WTIM been approached by graphical analysis method requires an experienced expert. Recently, equipped with back propagation (BP) learning algorithm was presented differs from AI such as symbolic approach must accompanied data preparation procedures smoothing, segmenting, transformation. In this paper, we developed BP Hough transform (HT) overcome selection problem use single rather sequential nets. powerful tool for shape detection image processing computer vision technologies. Along these lines, number exercises were conducted actual two steps. First, newly model, namely, ANNIS (Artificial Neural Network Identification System) utilized identify WTIM. Secondly, obtained reservoir characteristics modified Levenberg-Marquartmore » method. results show quite reliable having noisy, missing, extraneous points. They also demonstrate parameters successfully estimated.« less