论文部分内容阅读
首次提出用人工神经网络理论处理磁测资料，进行磁异常场分区的定量计算方法。并将该方法应用于南海研究。该方法将研究区域划分为若干统计单元，确定各统计单元中的特征变量、选择标准模型，并对BP神经网络进行训练。用学习成功的网络识别、分析整个研究区的磁异常场，从而将南海的磁异常场划分为12个特征区。
For the first time, artificial neural network theory is used to deal with the magnetic data to carry out quantitative calculation of magnetic anomalous field partition. The method is applied to the South China Sea research. The method divides the study area into several statistical units, determines the characteristic variables in each statistical unit, selects the standard model, and trains the BP neural network. With the successful learning of network identification, the magnetic anomaly fields in the whole study area are analyzed, and the magnetic anomalies in the South China Sea are divided into 12 characteristic regions.