2. 199 1 won the third prize of scientific and technological progress of China Oil and Gas Corporation (ministerial level): the application research of kriging technology.
3. 1996 won the second prize of scientific and technological progress of China petroleum and natural gas group (ministerial level): technical research on three-dimensional quantitative reservoir geological model of fluvial facies of Guantao Formation in Gudong Oilfield, Shengli.
4. 1996 won the Double Top Ten Award promoted by China Oil and Gas Corporation: Development and Application of Rolling Exploration and Development Software System,
5. 1999 won the first prize of scientific and technological progress of Shaanxi provincial education commission: intelligent information integration system for oil and gas field development;
6. 1996 won the second prize of scientific and technological progress of Shaanxi Education Commission and the first prize of scientific and technological progress of Shengli Petroleum Administration Bureau; Geostatistical study on heterogeneity of fluvial reservoirs in Gudong Oilfield
7. 1999 won the first prize of the 6th Natural Science Award of Shaanxi Province (199965438+February). Simulation of River Structure in Mijingwang Oilfield in Eastern China, 1997 Annual Technical Conference and Exhibition of American Society of Petroleum Engineers, San Antonio, 38678.
8. 1999 won the third prize of Shaanxi province for scientific and technological progress: intelligent information integration system for oil and gas field development;
After more than 30 years of work accumulation, Professor Wang Jiahua and his colleagues have formed a stable research advantage in the methods and software research of reservoir stochastic modeling and the application of risk analysis and decision analysis in reservoir management.
Research direction:
2. 1 computer software system gas reservoir geological statistical analysis system
GASOR, a statistical analysis system of reservoir geology, is a project team composed of teachers and students from the Computer Department of Xi Petroleum Institute under the auspices of Professor Wang Jiahua. After fourteen years' research, it has developed into a large-scale computer software, which is used for stochastic modeling of reservoirs and establishing three-dimensional quantitative geological models of reservoirs. At present, it has 6.5438+0.5 million sentences. After the development of microcomputer version, 1.0, 2.0 and other versions, GASOR 3.0 has been introduced.
The system is used for SUN series workstations and other compatible series workstations, and its grade requirements are SUN 4/75 and above workstations. Other performance indicators are: the required memory capacity is more than 32M; Its disk space requirements are above 70M.
The system has eight functional modules: data loading, histogram analysis, data conversion, two-dimensional variogram, three-dimensional variogram, structural analysis, effective thickness analysis, model verification, random simulation, grid coarsening and three-dimensional graphic display.
Among them, the random simulation module is a main module. At present, GASOR 3.0 includes three models: sequential index model, truncated Gaussian model and random walk model. The first two models were put forward by Professor A. Journel of Stanford University and Professor G. Matheron of French Geostatistics Center respectively. The random walk model was put forward by Professor Wang Jiahua and Associate Professor Zhang Tuanfeng according to the actual situation of China's oil fields, and was recognized by Daqing, Liaohe and other oil fields.
Three-dimensional graphic display module is another main module of this system. Can display two-dimensional contour map and three-dimensional geological map. In 3D map display, the module has the functions of scaling, rotation, illumination, uncovering, cutting, Chinese characters, color labeling and so on, which can adapt to display the surface and internal properties of various geological bodies.
2.2 Application of Risk Analysis and Decision Analysis in Reservoir Management
The oil industry is full of risks and uncertainties, and it is also recognized as an area that needs to accurately assess various risks. Correctly evaluating risks gives oil companies a competitive advantage. The University of Edinburgh in Scotland studied the decision-making behavior of 20 companies operating in Beihai Oilfield. The results show that the accuracy of each company's decision analysis is directly proportional to the success rate of investment decision. At the same time, oil companies are using various software systems for risk analysis and decision analysis.
1) decision analysis
The application of decision analysis in petroleum exploration (Newendorp, 1996) is discussed in detail. It is generally believed that decision tree method is the most commonly used method in decision analysis.
The decision tree method is formed by abstracting this decision-making process. Decision tree is a model of exploratory decision-making process. Decision tree visualizes the decision-making process: what needs to be decided is the trunk, and every branch on the trunk represents every decision. So which branch should I choose? That is, what kind of decision should be made? For simple things, you can make a qualitative analysis to make a decision. But in the face of complex things, qualitative analysis is not enough, which will inevitably lead to decision-making mistakes, thus affecting the success or failure of investment. So it is necessary to quantify the decision-making process. Decision tree is specifically aimed at this problem. Decision tree analysis method is a powerful method to describe and solve complex decisions. Once the problem is clear, the decision tree method is helpful to find the way to optimize the scheme.
2) Risk analysis
At present, the risk analysis of petroleum industry mainly refers to the application of Monte Carlo simulation method. Generally speaking, Monte Carlo simulation combined with decision tree method constitutes the most commonly used risk analysis method.
Monte Carlo simulation method mainly focuses on the uncertainty of parameter values, and describes these parameters with various statistical distributions. For example, cash flow can be expressed according to several key parameters every year: usually oil production, oil price, production consumption, franchise rights and taxes. The standard probability distribution, such as normal, lognormal, triangular and uniform distribution, is used to describe the variation of each parameter. In general, it is assumed that the variables are independent of each other, because this will greatly simplify the calculation. Choose a value from the probability distribution of each parameter and substitute it into the equation to get a possible value. This work will be repeated hundreds of times until the possible NPV frequency distribution map is given. Therefore, Monte Carlo simulation method is a natural extension of standard net present value method.
3) Application in reservoir management
In recent years, the number of SPE papers reporting the application of risk analysis and decision analysis in reservoir management has increased dramatically. The application of decision analysis and risk analysis includes risk analysis of sidetracking development, optimization of oil and gas production scheme, decision analysis of oilfield produced water treatment, application of risk analysis and decision analysis in oilfield development scheme selection, and application of decision analysis in economic evaluation of measure wells.
We have developed a software system for risk analysis and decision analysis.
2.3 Technical Route, Key Technologies and Solutions
Stochastic reservoir modeling is a technology developed by many international oil companies, research institutions and universities. As a new technology to promote the development of reservoir description in quantitative direction, a large number of papers and research reports are published every year. The biggest advantages of using stochastic reservoir modeling method to describe reservoir heterogeneity are:
1, statistical method has its outstanding advantages in dealing with uncertainty. The lack of data information will inevitably lead to the uncertainty of modeling results. When stochastic modeling is used, the uncertainty of modeling results will be greater when there is less data. On the contrary, the more data, the less uncertainty.
2. It is conducive to the comprehensive utilization of various data. For example, seismic, well testing and logging data have different resolutions, but they can also be combined. Sedimentary facies data are discrete, and parameters such as porosity, permeability and layer thickness are continuous, which can also be used in combination with bundling.
3. Permeability is an important parameter in reservoir engineering. Experience tells people that it is impossible to accurately determine the spatial distribution of permeability by using the permeability data of well points and any simple interpolation method. The spatial distribution of sedimentary facies can be accurately determined by stochastic modeling, and then the spatial distribution of permeability can be accurately determined.
Stochastic reservoir modeling is an important part of geostatistics. It is an important part of the development of reservoir description. Its purpose is to establish the spatial distribution of sedimentary facies in the reservoir, and on this basis, establish the spatial distribution of physical parameters such as porosity and permeability in the reservoir. Using the results of stochastic modeling of oil and gas reservoirs, the description and understanding of reservoir heterogeneity are more reasonable, and a fine three-dimensional quantitative geological model can be provided, so as to predict the spatial distribution of remaining oil through reservoir numerical simulation.
The spatial distribution of sedimentary facies (or sedimentary subfacies and sedimentary microfacies) in reservoirs is an important property of reservoirs. Its characteristics control the distribution and flow of fluid in the reservoir and dominate a series of important factors affecting the production of oil and gas reservoirs. For example, the spatial distribution of permeability and porosity, the spatial distribution and geometric size of mudstone suspended in sand bodies, the spatial distribution of shielding zones in reservoirs, the continuity between different sand bodies and the geometric position and size of reservoirs are all controlled by sedimentary facies, especially sedimentary microfacies. The location, direction, length and spatial analysis of faults and fractures in reservoirs also have great influence on oil and gas production.
Reservoir heterogeneity includes rock heterogeneity and fluid heterogeneity, which is the expression of the inherent characteristics of geological-physical factors in the reservoir. Physical parameters such as sedimentary facies and permeability, faults and fractures have great influence on underground oil and gas flow and oil and gas production, which belongs to reservoir heterogeneity. Reservoir heterogeneity modeling is to predict the spatial distribution of reservoir heterogeneity, and the result is a three-dimensional quantitative geological model of reservoir. Modeling with geostatistics and statistics is the work content of reservoir stochastic modeling. Because the well point data used are generally less and the heterogeneity is serious, the obtained three-dimensional quantitative geological model should be obviously uncertain. Especially when dealing with the spatial distribution of sedimentary facies, it is more difficult because the object is a discrete spatial variable.
It should be pointed out that the prediction of spatial distribution of sedimentary facies is the most challenging in the stochastic modeling of the whole reservoir. The reasons are the different types of sedimentary facies, the existence of uncertainty and the complexity of sedimentary facies distribution in three-dimensional space.
Using the method and results of reservoir stochastic modeling, we can optimize the oil and gas development scheme in the whole development process of oil and gas fields, improve the method of reservoir numerical simulation, improve its accuracy, determine the reasonable well location and horizontal well trajectory, and predict the spatial distribution of remaining oil and oil and gas resources.
China's main oil producing area is in the east. Most of its oil fields have entered the middle and late stage of development, and the average water production of oil wells has reached 80%. The difficulty of oil and gas field development has greatly increased. Most of these oilfields are continental deposits with complex geological conditions and serious heterogeneity.
According to the estimation of domestic experts, 20% of the movable oil is not affected by the secondary oil displacement agent because of the heterogeneity of the reservoir. At this time, by deepening the understanding of reservoir heterogeneity and improving the secondary oil recovery technology, this part of movable oil can be completely produced. In order to describe the distribution of underground remaining oil in detail, the description of reservoir heterogeneity is required to develop quantitatively in a smaller scale. Professor Qiu Yinan, a well-known domestic oilfield development geologist, has published many works in recent years, expounding the importance of reservoir stochastic modeling and establishing three-dimensional quantitative geological model in oilfield development.
Reservoir stochastic modeling methods are usually divided into two categories. The first kind is to infer the variation function of spatial attribute parameters, and then establish a stochastic model based on the variation function to get the results of stochastic modeling. Sequential Gaussian simulation method, sequential index simulation method and truncated Gaussian simulation method all belong to this category. The other is to study the distribution of spatial shape, and then give the spatial distribution of the research object by modeling its geometric parameters. This method is called object-oriented method. Characteristic point process simulation method belongs to this category. Random walk simulation also belongs to this second method. The development and application of random walk method will greatly enrich peace and promote the development of the second kind of reservoir stochastic modeling method.
3. Monographs and papers:
3. 1 stochastic reservoir modeling: 15 papers (10 domestic, 6 international).
1.Wang J and MacDonald, A., 1997, "Modeling of River Structure in Intensive Drilling Oilfield in Eastern China", presented at the annual technical meeting of American Society of Petroleum Engineers 1997, American Society of Petroleum Engineers 386781October 5-8, San Antonio, Texas.
2. Wang, Zhang, t. 1995, Three-stage stochastic modeling method for fluvial reservoirs, American Society of Petroleum Engineers, 29965, 10.
3. Wang, Zhang, Huang, 1997, application of random walk model in two-dimensional braided river simulation. In 1930 ... George. Conger. , volume 25, 1 15- 124, Int. Scie。 Publisher, Dutch
4. Gao, Wang, Jia Hua, updated random variance and optimal sampling design, mathematics. Geology, April, 1996
5. Gao, Wang, Jia Hua, Drilling Location Optimization and Spatial Sampling Probability, 30th National Drilling Technology Conference. Geological congress, April-14, August, 1996
6. Gao, Wang, Jia Hua, Application of Identification Probability and Pseudoentropy Criterion in Drilling Positioning, 30th International Drilling Technology Conference. Geological congress, April-14, August, 1996
7. Zhang Tuanfeng, Wang Jiahua, Geological application of stochastic simulation of oil and gas reservoirs, Mathematical Geology of China (5), 1994.
8. Zhang Tuanfeng, Wang Jiahua, Jing Ping, Yan: Research on 3D reservoir stochastic modeling and stochastic simulation technology, Mathematical Geology of China (7), 1996.
9. Wang Jiahua et al.: Stochastic Simulation Algorithm in Reservoir Description, Journal of xidian university, Vol.22, 1995.
10. Wang Jiahua et al.: Simulation method based on variogram in reservoir evaluation, Mathematical Geology of China (6), 1995.
1 1. Wang Jiahua et al.: Quantitative evaluation of spatial distribution uncertainty of reservoir characteristics, Proceedings of the Science and Technology Conference of An Petroleum Institute, Shaanxi Science and Technology Press, 1996.
12. Wang Jiahua et al.: Improving the effect of reservoir numerical simulation with stochastic simulation, Journal of An Petroleum Institute, 1 1, No.3, 1996.
13. Wang Jiahua et al.: On the essential difference between Kriging estimation and stochastic simulation, Journal of An Petroleum Institute, 12, No.2, 1997.
14. Wang Jiahua et al.: Random Geometry and Its Application in Geological Sedimentary Facies, Proceedings of the Third China Society of Industrial and Applied Mathematics, Tsinghua University Press, 1994.
15. Wang Jiahua et al.: Basic principle of stochastic simulation of oil and gas reservoirs, logging technology, No.4, 1994.
Monograph 2
1. Wang Jiahua, Gao,, Kriging Geological Mapping Technology-Computer Model and Algorithm, Petroleum Industry Press, 1999.
2. Wang Jiahua, Zhang Tuanfeng, Stochastic Modeling of Oil and Gas Reservoir, Petroleum Industry Press, 200 1
3.2 Graduation thesis
Doctor:
1. Gao: Oil and gas exploration well location model and its application: (Professor Zhao, academician of China Academy of Sciences and president of China Geo University, is the tutor, and I am the deputy tutor)
Master:
1. Guo: research on reservoir data analysis system (oil and gas field development specialty);
2. vilen, research and application of permeability coarsening method and software (oil and gas field development major);
3. He Juhou, visualization of stochastic modeling results of oil and gas reservoirs based on polygonal regions; (computer application major),
4. Wang Hongxia, Research and application of distributed stochastic simulation of oil and gas reservoirs based on CORBA, (computer application major);
5. Dong Chen, application research of CORBA technology in: development, (computer application major);
6. Zhou Chongli, Research on CORBA-based network parallel computing: application in stochastic modeling of oil and gas reservoirs (computer application major),
7. Yang Hui, research and application of multi-database system model based on middleware, (computer application major);
8. Liu Chongtao, research on two-stage estimation algorithm of fault visualization in reservoir description, (computer application major);
9. Yang Huabin, Application of Computer Graphics Algorithm in Fine Reservoir Description, (Computer Application Major)
10. Liu Xingyu, research on visualization interior point judgment algorithm of two-dimensional geological map of oil reservoir (computer application major);
1 1. Chen fengxi: Research on fault display software and its application in reservoir description (oil and gas field development engineering major);
12. Yang Xiaofei: Research on 3D visualization of borehole trajectory based on OpenGL, (computer application major);