中文摘要：

利用多地震属性和BP神经网络可以得到胜利油田垦71地区的岩性预测．由井附近的地震道中可以提取井数据和多地震属性，并由此得到岩性信息，再用BP网络对岩性信息进行标定．岩性分布是基于训练好的网络和该地区的多地震属性进行计算的，结果与该区域未参加训练的井资料相比符合率为75％．

英文摘要：

The lithology prediction in the Ken-71 area of Shengli Oil Field is achieved by multiple seismic attributes and Backpropagation（BP） neural network. We use BP neural network to calibrate the lithology information extracted from the well logs and multiple seismic attributes extracted from the trace near the well. The lithology distributions are calculated based on well trained network and multiple seismic attributes of the area. The calculated results are about 75 % agreement with the data of the wells in the area which are not in the training network.