【摘要】 多参数神经网络油气检测技术首先提取四大类２６ 个地震波特征参数，并定量描述这些参数，然后用神经网络技术对这些参数进行分析，预测油气分布，使油气识别工作定量化、计算机化。本次研究开发了一套神经网络油气识别软件，并利用其对东濮凹陷胡１９ 块进行油藏精细描述，预测沙三下砂组含油面积０－７ ｋｍ２ ，石油地质储量７０ ×１０４ 吨。更多还原

【Abstract】 Using nervenetwork technique with the full utilization of petroleum informations taken by seismic waves,hydrocarbon can be detected First, four types of 26 characteristic parameters were extracted from seismic waves and discribed quantitively Then these parameters were analyzed by nervenetwork to recognize and predict petroleum distribution with raising resolutions This method was quantified and realised by computer with the corresponding software As an application, the reserving in Hu 19 zone of Dongpu depression was fine described by this software The predicted oilbearing area of sand groups from 3 to 10 in lower Sha 3 member was 0 7 km 2, and the predicted petroleum geological reserve was 70×10 4 tons [更多还原