Historical fitting of reservoir numerical simulation, simulation method of reservoir numerical simul

This is an extremely important task in reservoir simulation. After a reservoir model is established, it has not been tested whether it fully reflects the actual oil and gas reservoir. It is only by feeding historical data from production and injection into the model and running the simulator that the results of the calculations are compared to the actual dynamics of the reservoir that the reservoir description used in the model is valid.

If the calculated dynamic data is far from the actual reservoir dynamic data, we must constantly adjust the basic data entered into the model until the dynamics calculated by the simulator are in satisfactory agreement with the actual dynamics of reservoir production. Since the purpose of historical fitting adjustment parameters is to make the description of the real reservoir as accurate as possible, it is an important step in reservoir simulation.

Obviously, the model used for the simulation should be similar to the actual reservoir. Significant discrepancies between the data used in the numerical simulations describing the reservoir and the actual data that control the reservoir dynamics can lead to significant distortion of the simulation results. Unfortunately, we knew very little about the accuracy of the model and what parameters should be modified to ensure that it was similar to the actual reservoir until it was tested.

In this case, one of the most effective and most frequently used verification methods is to simulate the past dynamics of the reservoir and compare the results of the simulation calculations with the actual past dynamics of the reservoir, which is known as historical fitting. Historical fitting helps us find and correct errors in reservoir characterization data to improve the model and verify the reliability of the reservoir characterization. If the modified model simulation calculation dynamics are consistent with the past historical dynamics of the reservoir, and the reservoir description is reasonable, then it should be said that historical fitting itself is an effective reservoir description method.

The numerical reservoir simulation method is the only method to quantitatively describe the flow law of multiphase fluids in heterogeneous formations. For example, many conventional methods assume that the reservoir is a circular homogeneous medium, and if the reservoir geometry is slightly more complex and is a heterogeneous medium, it is very difficult or even impossible to solve. For the numerical simulation of oil and gas reservoirs, the workload of calculating the homogeneous reservoir with complex form is almost the same as that of calculating the homogeneous reservoir with simple form.

Therefore, reservoir numerical simulation can solve problems that cannot be solved by other methods. For problems that can be solved by other methods, numerical simulation methods can be solved faster, cheaper, more conveniently, and more reliably, and increase the credibility of other analytical methods.

An oil and gas reservoir can only be developed once in reality. However, by applying reservoir numerical simulation, it is easy to recalculate the development process of different development methods, so one can choose the best development method from them.

Therefore, for reservoir engineers, numerical simulation provides a fast, accurate and comprehensive method for dynamic analysis. For managers, numerical simulations provide a comparison of different mining plans; Numerical simulation is an effective training tool for inexperienced engineers.

The main working procedure of numerical simulation research is to carry out a comprehensive digital simulation study of an oil and gas reservoir, which often requires a large amount of effort and a long time (sometimes up to a year or even longer), and also has high requirements for computer hardware and technicians, but although the problems faced in different projects will be very different, the basic research process of most reservoir numerical simulation is the same. In order to give the reader a clear overall concept of digital simulation research from the beginning, the following is a brief introduction to the main working procedures of reservoir numerical simulation.

Definition of the problem: The first step in carrying out reservoir modeling work is to determine the objectives and scope of the study. That is, it is necessary to give a clear positioning of this digital model study, clarify what are the main problems to be solved in this simulation, which reservoir dynamic characteristics need to be studied, and what impact the completion of these projects will have on reservoir managers, etc.

In order to carry out numerical simulation research program design according to the requirements of the project, and collect relevant reservoir basic geology, fluid and production dynamic data.

This means that after the simulator is selected, we must design a suitable mesh model. The design of the mesh model is influenced by factors such as the type of simulation process, the complexity of the fluid motion in the heterogeneous reservoir, the selected research objectives, the accuracy of the reservoir description, and the allowable computational time and cost budget. The higher the number of meshes, the more detailed the simulated dynamics of a single well, but the longer the number of meshes will take to calculate, the higher the cost, and sometimes even unacceptable, so we often have to design our mesh models within the overall framework determined by the research objectives, according to the allowable computational time and cost constraints.

Reservoir numerical simulation is a new discipline that has grown with the emergence and development of electronic computers, which refers to the use of computers to solve the mathematical model of the reservoir, simulate the flow of water in the ground, and give the distribution of oil and water at a certain time. 1953 American G.After publishing "Calculation of Unstable Gas Seepage in Pore Media", he opened the way for calculating the seepage problem of oil and gas reservoirs by numerical methods.

For more than 30 years, due to the rapid development of large-scale fast electronic computers, Daxin die has promoted the wide application of numerical simulation methods. In the early 60s of the 20th century, the multi-dimensional and multi-phase black oil model was studied. In the early 70s of the 20th century, the component model, the miscible model and the thermal oil recovery model were studied. In the late 70s of the 20th century, various chemical flooding models were studied.

The final step in the study of digital modeling is to systematically organize the calculated results and draw clear conclusions into a clear, concise report. Depending on the purpose of the study, the format of the report can range from a simple thematic report to a set of multi-volume reports with a large amount of text, data, figures and figures, as well as multiple color drawings. However, regardless of the form and length of the reports, they should clearly state the model used in the study, the basis for the calculations, and the main findings and conclusions, with appropriate length and sufficient arguments.