石油勘探开发中的数据场可视化技术的研究
|
摘要
石油勘探开发是石油工业的重要部分，直接决定了我国的石油产量。在石油勘探、开发过程中，会采集到大量的数据，形成三维数据场，如某一区块的密度场、孔隙度场、高压射流的流场，目前使用这些数据时，通常采用的方法是画出数据场的等值线，然后由专业人员进行分析、解释。但是由于等值线是二维的，不能形成一个直观的、清晰的概念。利用可视化技术，尤其是三维数据场可视化技术，可以从大量数据中构造出三维图像。所以数据场可视化技术在石油勘探开发中的应用前景十分广泛，同时三维数据场的可视化技术也广泛地应用于医学影像学、气象学等领域。
光线投射法假定三维空间的数据f(x_i，y_j，z_k)分布在均匀的网格或规则网格的网格点上。算法过程中的预处理包括原始数据的格式转换、剔除冗余的数据及导出所需数据等功能。然后进行数据分类，其目的是根据数据值的不同，正确地将其分为若干类，并给每一类赋予不同的颜色值和不透明度值，以求正确地显示多种物质的不同分布或单一物质的不同属性。重采样，是从屏幕上的每一个象素根据设定的观察方向发出一条射线，这条射线穿过三维数据场，沿着这条射线选择K个等距的采样点，并由距离某一采样点最近的8个数据点的颜色值和不透明度值做三次线性插值，求出该采样点的不透明度值及颜色值。最后进行图像合成，将每一条射线上的各采样点的颜色值及不透明度值由前向后或由后向前的加以合成，即可得到发出该射线的象素点处的颜色值。
本文是将三维数据场可视化技术应用到石油勘探开发过程中的地质建模和数值模拟之中。先研究了石油勘探开发过程中形成的三维数据场以及三维数据场可视化技术中的直接体绘制技术，把直接体绘制技术应用到该领域中达到了较好的效果，直观、准确地再现了油藏分布以及石油勘探开发过程中的参数的动态变化，辅助相关领域工作人员的分析、设计，具有广泛应用前景。将体绘制技术中光线投射法应用到石油勘探开发中产生的地质数据中，直观地再现石油在油气藏中的状态，指导设计人员进行井位的确定，以及在石油开采中确定采油工艺的过程，提高石油生产的效益。
Petroleum exploration and exploitation is an important part of the petroleum industry. It directly decides the oil output of our country. In the process of oil exploration and exploitation, a great deal of data can be collected. These data form 3D data sets, for instance, density sets, hole degree, and flow sets of high-pressure shoot in one block. At present, the usual adopting method is to draw isoclines of data sets, and ask the career men to analyze and explain them. But because isoclines are 2D, they can't form an intuitive, clear concept. Using visualization technology, especially 3D data sets visualization technology has quite an abroad application foreground in Petroleum exploration and exploitation. Meanwhile 3D data sets visualization technology can also widely apply to medicine image field, meteorology field, etc.
Ray-casting Algorithm assumes that 3D space data f(xi,yj,zk) distributes on the
dots of even or formula reseau. He pretreatment in course of algorithm includes the function of transforming of original data, eliminating unwanted data and educing needed data, etc. Then to classify the data, whose aim is to classify them according to data value differences, and to put different color values and opacity degree values to every class in order to show different distributions of multi-matter or different properties of a single matte correctly. Resampling is that every image element in the screen gives off one ray according to the planned observing orientation. This ray penetrates 3D data sets, and to chooses K sampling dots of equidistance along with it, and does three linear insert values according to color and opacity values of eight data dots that mostly close to certain sampling dot, then evaluate the opacity value and color value of it. Finally compose the image that is composing color value and opacity value of every sampling dot on every ray according to front-back or back-front, so the color value of the image dot that gives off the ray can be gained.
The article applies 3D data sets visualization technique to the geological modeling and numerical value simulation in course of petroleum exploration and exploitation. First,
we research 3D data sets that formed during petroleum exploration and exploitation & Direct Volume Rendering technique in 3D data sets visualization technique, and apply Direct Volume Rendering technique to this field to achieve a better effect. It reappears oil distributions and parameter dynamic change intuitively and truly in course of Petroleum exploration and exploitation, and assistant the working personnel of the certain field to analyze and design, having wide application foreground. Applying Ray-casting Algorithm of Volume Rendering to the geological data produced by petroleum exploration and exploitation can directly reappear the petroleum's status in oil-gas storage, and direct the designers to make sure of the wells' locations and to processes of oil extraction techniques in oil exploitation, in order to raise the profit of the oil production.
光线投射法假定三维空间的数据f(x_i，y_j，z_k)分布在均匀的网格或规则网格的网格点上。算法过程中的预处理包括原始数据的格式转换、剔除冗余的数据及导出所需数据等功能。然后进行数据分类，其目的是根据数据值的不同，正确地将其分为若干类，并给每一类赋予不同的颜色值和不透明度值，以求正确地显示多种物质的不同分布或单一物质的不同属性。重采样，是从屏幕上的每一个象素根据设定的观察方向发出一条射线，这条射线穿过三维数据场，沿着这条射线选择K个等距的采样点，并由距离某一采样点最近的8个数据点的颜色值和不透明度值做三次线性插值，求出该采样点的不透明度值及颜色值。最后进行图像合成，将每一条射线上的各采样点的颜色值及不透明度值由前向后或由后向前的加以合成，即可得到发出该射线的象素点处的颜色值。
本文是将三维数据场可视化技术应用到石油勘探开发过程中的地质建模和数值模拟之中。先研究了石油勘探开发过程中形成的三维数据场以及三维数据场可视化技术中的直接体绘制技术，把直接体绘制技术应用到该领域中达到了较好的效果，直观、准确地再现了油藏分布以及石油勘探开发过程中的参数的动态变化，辅助相关领域工作人员的分析、设计，具有广泛应用前景。将体绘制技术中光线投射法应用到石油勘探开发中产生的地质数据中，直观地再现石油在油气藏中的状态，指导设计人员进行井位的确定，以及在石油开采中确定采油工艺的过程，提高石油生产的效益。
Petroleum exploration and exploitation is an important part of the petroleum industry. It directly decides the oil output of our country. In the process of oil exploration and exploitation, a great deal of data can be collected. These data form 3D data sets, for instance, density sets, hole degree, and flow sets of high-pressure shoot in one block. At present, the usual adopting method is to draw isoclines of data sets, and ask the career men to analyze and explain them. But because isoclines are 2D, they can't form an intuitive, clear concept. Using visualization technology, especially 3D data sets visualization technology has quite an abroad application foreground in Petroleum exploration and exploitation. Meanwhile 3D data sets visualization technology can also widely apply to medicine image field, meteorology field, etc.
Ray-casting Algorithm assumes that 3D space data f(xi,yj,zk) distributes on the
dots of even or formula reseau. He pretreatment in course of algorithm includes the function of transforming of original data, eliminating unwanted data and educing needed data, etc. Then to classify the data, whose aim is to classify them according to data value differences, and to put different color values and opacity degree values to every class in order to show different distributions of multi-matter or different properties of a single matte correctly. Resampling is that every image element in the screen gives off one ray according to the planned observing orientation. This ray penetrates 3D data sets, and to chooses K sampling dots of equidistance along with it, and does three linear insert values according to color and opacity values of eight data dots that mostly close to certain sampling dot, then evaluate the opacity value and color value of it. Finally compose the image that is composing color value and opacity value of every sampling dot on every ray according to front-back or back-front, so the color value of the image dot that gives off the ray can be gained.
The article applies 3D data sets visualization technique to the geological modeling and numerical value simulation in course of petroleum exploration and exploitation. First,
we research 3D data sets that formed during petroleum exploration and exploitation & Direct Volume Rendering technique in 3D data sets visualization technique, and apply Direct Volume Rendering technique to this field to achieve a better effect. It reappears oil distributions and parameter dynamic change intuitively and truly in course of Petroleum exploration and exploitation, and assistant the working personnel of the certain field to analyze and design, having wide application foreground. Applying Ray-casting Algorithm of Volume Rendering to the geological data produced by petroleum exploration and exploitation can directly reappear the petroleum's status in oil-gas storage, and direct the designers to make sure of the wells' locations and to processes of oil extraction techniques in oil exploitation, in order to raise the profit of the oil production.
引文
1) [唐泽圣94] 《用图像空间为序的体绘制技术显示三维数据场》 唐泽圣 袁骏 清华大学计算机科学与技术系 《计算机学报》
2) [黄文静98] 《体绘制及三维交互技术在地质数据可视化中的应用》 黄文静 唐龙 唐泽圣 清华大学计算机系CAD中心
3) [邓俊辉00] 《基于虚拟机的并行体绘制》 邓俊辉 唐泽圣 清华大学计算机科学与技术系
4) [任继成98] 《非规则数据场并行体绘制技术》 任继成 刘慎权 中科院《计算机学报》
5) [董峰97] 《基于空间分割的三维不规则数据场的体绘制算法》 董峰 蔡文立 石教英 浙江大学CAD&CG国家重点实验室《计算机辅助设计与图形学学报》
6) [王文成02] 《加速体绘制技术》 王文成 孙汉秋 吴恩华 中国科学院软件所计算机重点实验室 北京 100080计算机辅助设计与图形学学报 第14卷 第9期 2002年9月
7) [李允98] 《油藏模拟》 李允 石油大学出版社 98年1月
8) [潘景为92] 《石油开采》 潘景为 石油工业出版 92年8月
9) [罗平 03] 《油气田开发工程》 罗平亚 杜志敏 中国石化出版社 03年1月
10) [童晓光01] 《油气勘探原理和方法》 童晓光 何登发 石油工业出版
11) Meagher, Donald J. Efficient synthetic image generation of arbitrary 3-D objects. In proceeding of IEEE conference on Pattern Recognition and Image Processing: 473-478,1982.
12) Danskin, John and Pat Hanrahan. Fast algorithms for volume ray tracing. In 1992 workshop on volume visualizationg:91-98, Boston, MA, October 1992.
13) Subramanian, K. R. and Donald S. Fusell. Applying space subdivision techniques to volume rendering. In Proceedings of Visualization' 90: 150-159, October 1990.
14) Zuiderveld, Karel J. Acceleration of ray-casting using 3D distance transforms. In Proceedings of Visualization in Biomedical Computing 1992, 324-335, October 1992
15) Schroder, Peter and Gordon Stoll. Data parallel volume rendering as line drawing. In 1992 workshop on volume visualizationg, 25-32, Boston, MA, October 1992.
16) Vezina, Guy. Volume rendering on the MasPar. In 1992 workshop on volume visualizationg, 3-8, Boston, MA, October 1992.
17) Levoy, Marc. Volume rendering by adaptive refinement. The Visual Computer, 6(1) :2-7, February 1990.
18) A. V. Aho, J. E. Hopcroft, J. D. Ullman. The Design and AnaTysfs of Computer Algorithms, Addision-Wesley, Reading, MA, 1974
19) Philip Lacroute. Fast volume rendering using shear-warp factorization of the viewing transformation. Technical report: CSL-TR-95-678, September 1995
20) Marc Levoy. Display of surfaces from volume data. IEEE Computer Graphics & Applications, 8(3): 29-37, May 1988
21) Marc Levoy. Efficient ray tracing of volume data. ACM Transactions on Graphics, 9(3): 245-261, July 1990
22) Philip Lacroute, Marc Levoy. Fast volume rendering using shear-warp factorization of the viewing transformation. Computer Graphics (ACM Siggraph Proceedings), pp.451-458, 1994
23) Garrity M P. Rayracing irregular volume data. Computer Graphics, 24(5), 35-40, 1990
24) [刁麓弘03] 《Shear-warp算法的改进》 刁麓弘 山东大学硕士论文
2) [黄文静98] 《体绘制及三维交互技术在地质数据可视化中的应用》 黄文静 唐龙 唐泽圣 清华大学计算机系CAD中心
3) [邓俊辉00] 《基于虚拟机的并行体绘制》 邓俊辉 唐泽圣 清华大学计算机科学与技术系
4) [任继成98] 《非规则数据场并行体绘制技术》 任继成 刘慎权 中科院《计算机学报》
5) [董峰97] 《基于空间分割的三维不规则数据场的体绘制算法》 董峰 蔡文立 石教英 浙江大学CAD&CG国家重点实验室《计算机辅助设计与图形学学报》
6) [王文成02] 《加速体绘制技术》 王文成 孙汉秋 吴恩华 中国科学院软件所计算机重点实验室 北京 100080计算机辅助设计与图形学学报 第14卷 第9期 2002年9月
7) [李允98] 《油藏模拟》 李允 石油大学出版社 98年1月
8) [潘景为92] 《石油开采》 潘景为 石油工业出版 92年8月
9) [罗平 03] 《油气田开发工程》 罗平亚 杜志敏 中国石化出版社 03年1月
10) [童晓光01] 《油气勘探原理和方法》 童晓光 何登发 石油工业出版
11) Meagher, Donald J. Efficient synthetic image generation of arbitrary 3-D objects. In proceeding of IEEE conference on Pattern Recognition and Image Processing: 473-478,1982.
12) Danskin, John and Pat Hanrahan. Fast algorithms for volume ray tracing. In 1992 workshop on volume visualizationg:91-98, Boston, MA, October 1992.
13) Subramanian, K. R. and Donald S. Fusell. Applying space subdivision techniques to volume rendering. In Proceedings of Visualization' 90: 150-159, October 1990.
14) Zuiderveld, Karel J. Acceleration of ray-casting using 3D distance transforms. In Proceedings of Visualization in Biomedical Computing 1992, 324-335, October 1992
15) Schroder, Peter and Gordon Stoll. Data parallel volume rendering as line drawing. In 1992 workshop on volume visualizationg, 25-32, Boston, MA, October 1992.
16) Vezina, Guy. Volume rendering on the MasPar. In 1992 workshop on volume visualizationg, 3-8, Boston, MA, October 1992.
17) Levoy, Marc. Volume rendering by adaptive refinement. The Visual Computer, 6(1) :2-7, February 1990.
18) A. V. Aho, J. E. Hopcroft, J. D. Ullman. The Design and AnaTysfs of Computer Algorithms, Addision-Wesley, Reading, MA, 1974
19) Philip Lacroute. Fast volume rendering using shear-warp factorization of the viewing transformation. Technical report: CSL-TR-95-678, September 1995
20) Marc Levoy. Display of surfaces from volume data. IEEE Computer Graphics & Applications, 8(3): 29-37, May 1988
21) Marc Levoy. Efficient ray tracing of volume data. ACM Transactions on Graphics, 9(3): 245-261, July 1990
22) Philip Lacroute, Marc Levoy. Fast volume rendering using shear-warp factorization of the viewing transformation. Computer Graphics (ACM Siggraph Proceedings), pp.451-458, 1994
23) Garrity M P. Rayracing irregular volume data. Computer Graphics, 24(5), 35-40, 1990
24) [刁麓弘03] 《Shear-warp算法的改进》 刁麓弘 山东大学硕士论文