神经网络在固相颗粒损害仿真建模中的应用
|
摘要
地层损害是石油工业棘手的一个问题，而固相颗粒损害是地层损害的主要形式，因此，如何研究、量化固相颗粒损害问题对于保护储集层防止地层损害是很重要的一项工作。本文用神经网络的方法建立了固相颗粒损害的动态模型，然后对固相颗粒损害进行了仿真研究，取得了满意的效果，为量化固相颗粒损害问题和量化影响固相颗粒损害的诸多因素（颗粒尺寸、浓度、注入速度等）的相对作用大小提供了一个快速方便的方法。应用结果表明，该方法简单，通用性强，不仅适合于固相颗粒损害问题的量化研究，而且还适合于其它领域的类似问题的研究。
In petroleum industry,formation damage is a difficult problem and its main form is solid particles damage. So how to study and quantify the damage caused by solid particles is an important task for preventing reservoir from damage. In this paper,dynamically modeling of solid particles damage is set up by neural network. Then the damage result from solids invasion is simulated,and satisfactory result is obtained,which provides an efficient and fast method for the quantification of damage due to solids or particles factors such as solid size,concentration etc. The results of application demonstrate this method is simple and general for other study.
In petroleum industry,formation damage is a difficult problem and its main form is solid particles damage. So how to study and quantify the damage caused by solid particles is an important task for preventing reservoir from damage. In this paper,dynamically modeling of solid particles damage is set up by neural network. Then the damage result from solids invasion is simulated,and satisfactory result is obtained,which provides an efficient and fast method for the quantification of damage due to solids or particles factors such as solid size,concentration etc. The results of application demonstrate this method is simple and general for other study.
引文
1张绍槐，罗平亚．保护储集层技术，北京：石油工业出版社，1991 2 梅文荣，刘文忠译．地层损害评估进展与控制战略。国外钻井技术，1990（1、2） 3 梅文荣．用数值法研究固相颗粒损害．石油钻采工艺，1993（1） 4 刘瑞林．用神经网络学习识别矿层．江汉石油学院学报，1991（13） 5 朱广生，刘瑞林，用神经网络建立孔隙度预测模型．江汉石油学院学报，1993（1） 6 于建华，应用人工神经网络自动识别岩性，石油地球物理勘探，1993（1） 7周成当，金振武．神经网络在地震资料处理中的应用．江汉石油学院学报，1992 8 焦李成．神经网络系统理论．西安：西安电子科技大学出版社，1990
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心