CAJ | 학술논문

基于BP神经网络的算法原理，本文通过对渤海油田某工区录井、测井、测试数据等解释结论的学习和训练，建立了录井油气水层神经网络解释模型。运用该模型可进行储层流体性质的识别和划分，通过验证该模型解释符合率达到了80%以上。
기우BP신경망락적산법원리，본문통과대발해유전모공구록정、측정、측시수거등해석결론적학습화훈련，건립료록정유기수층신경망락해석모형。운용해모형가진행저층류체성질적식별화화분，통과험증해모형해석부합솔체도료80%이상。
Based on the principle of BP neural network algorithm, interpretation results of mud logging, well logging and testing in Bohai oilfield were analyzed with the method of learning and training in this paper, and the model of BP neural network for the interpretation was built. By using this model, properties of reservoir fluids could be identified and divided. With the validation data, the interpretation coincidence rate of this model is proved to be above 80%.