Fluorescence analysis of oil inclusions based on BP algorithm

Artificial-neural-network (ANN) technology has excellent ability of non-linear mapping, generalization, self-organization and self-learning, so ANN technology has been proved to be successful and widespread utility in engineering. Scientists of all fields have interested for its developments, and applied this technology to solve many petroleum-engineering problems. BP network is one of the most widely used neural network models. In this paper, we discuss the basic principle of BP neural network and its application in analysis inclusions fluorescence, for predicting and tasting stragraphic division. By forming neural network model and wavelet transforming, we can obtain variation of oil inclusion fluorescence at different depths of one well. Especially, in petroleum prospecting, this method can help establish reasonable stratigraphic framework, reveal the reservoir heterogeneity, and guide the rational development of oil and gas fields.