利用神经网络拾取叠加速度

鉴于三维地震和高密度二维地震的数据量很大，仍然采用人工方法拾取速度谱，不仅效率低，而且精度低。为此，人们提出许多新的方法，本文采用人工神经网络与模糊数学相结合的方法。首先对输入数学方法作出边界搜索和模糊聚类预测处理，然后通过人工神经网络误差反向传递算法（ＢＰ）算法，学习给定的样本值，训练网络模型，输入经过预处理的待识别数据，完成识别工作，自动提取叠加速度，该方法具有抗噪能力强，拾取速度精度高的特点

Stack velocity pickup using neural network

Manual vclocity spcctrum pickup results in low cfficiency and poor accuracy because both 3D seismic survey and high-density 2D seismic survey have very great data volume. So, people advance continuously many new mcthods to solve the problem. I put forward a new method which involves neural network and fuzzy mathematics. The method works in the following steps:Make boundary search and clustering processing of the lnput data with the aid of fuzzv mathematics.Learn thc given samples to train nctwork model by using BP algorithm for neural network error.Input the preproccssed data for recognition.Automatlcally pick up stack velocity after the data recognition.The rnethod has good noise-resistance and offers accurate velocity, so that ve-locity spectrum can be lnterpreted automatlcally.