摘要
提出一种基于渗透率约束的微观驱替模型图像生成算法,利用Kozeny-Carman方程将渗透率与统计特性相结合用于生成目标图像的两点概率函数,以模拟退火算法为框架利用生成的两点概率函数最终生成目标图像。同时在生成图像的过程中控制目标的孔隙平均径长,确保最终生成不同尺寸的图像时孔径不会差别过大。最后为了解决生成速率较慢的问题,引入纹理生成的方法加速小尺寸图像到大尺寸图像的生成。通过对生成的图像进行性质分析及数据测定,来验证算法的有效性。
In this paper,an image generation algorithm of microscopic displacement model based on permeability constraints is proposed.Kozeny-Carman equation is used to combine permeability and statistical characteristics to generate a two-point probability function of the target image,and the simulated annealing algorithm is used as the framework to generate the target image.At the same time,in the process of image generation,the average pore diameter length of the target is controlled to ensure that the aperture will not be too different when the image of different sizes is generated.Finally,in order to solve the problem of slow generation rate,the texture generation method is introduced to accelerate the generation of small-size images to large-size images.Through the property analysis and data measurement of the generated images,the effectiveness of the proposed algorithm is verified.
作者
文天昊
吴晓红
吕朝阳
滕奇志
Wen Tianhao;Wu Xiaohong;LV Chaoyang;Teng Qizhi(College of Electronics and Information Engineering,Sichuan university,Chengdu 610065;Chengdu Xitu Technology Co.,Ltd.,Chengdu 610000)
出处
《石化技术》
2025年第8期91-93,44,共4页
Petrochemical Industry Technology
关键词
图像生成
模拟退火
渗透率
孔隙平均径长
Image generation
Simulated annealing
Permeability
Average pore diameter
