Quantum-Inspired Self-Organization Network And Its Application In Oilfield Water Flooded Layer Identification

Oil is an indispensable resource in the course of human development.As a strategic reserve,it has always attracted the attention of countries around the world.At present,most of China's oil fields have entered the middle and late stages of extraction,and many reservoirs are flooded when natural crude oil is about to dry up,increasing production and recovery.China has become the country with the highest proportion of waterflood extraction in the world.Most oil fields have entered a high water cut stage.With the long-term displacement of underground crude oil by injection water,the degree of flooding is serious,which is accompanied by related problems such as increased difficulty in extraction.The identification factors that affect the flooded layers in oil fields are more complicated,and the distribution of groundwater and the degree of salinity are unpredictable.Therefore,it is an urgent task to correctly identify and evaluate the level of flooded layers and maximize recovery.This article cuts through this problem,and designs a quantum-inspired self-organizing network model for identification of water flooded layers in oil fields,and uses related experiments to verify.The main research contents are as follows:(1)Using the core logging data to predict the flooding level of the flooded oil field,the structure of the characteristic index set is very important.In this paper,the actual logging data obtained from the mine are sorted,and the forward transformation and inverse transformation of the Walsh filter are used to realize the preprocessing of the logging data,and the characteristic index set for identifying the flooded layer is constructed.(2)A quantum-inspired self-organizing network model is studied.The sample data and the weights of the competition layer are described by qubits.With the efficient computer system of quantum computing and the ability of nonlinear mapping,and based on relevant experiments,the network constructed The model has the advantage of high recognition rate.(3)Combined with the actual logging data obtained in the mine,the water-flooded layer is identified,and the comparison experimental results prove that the quantum-inspired self-organizing network model proposed in this paper has a higher recognition rate on the problem of water-flooded layer identification and has certain reference Value and meaning.