A Method for Accelerating Historical Fitting of Reservoir Numerical Simulation Based on Mass Equilibrium and Monte Carlo Algorithm

To address the challenges arising from substantial reservoir model uncertainties, historical fitting operation and multiple solutions of calculation results, an expedited approach for optimizing historical fitting in the context of reservoir numerical simulations was developed. This approach, grounded in material balance analysis, involves the creation of a three-phase material balance model encompassing oil, gas, and water. This model is constructed by utilizing pressure, production, and fluid PVT data measurements. The global sensitive parameters like geological reserves of oil and water parameters were determined by material balance calculation. The uncertainty analysis was carried out by Monte-Carlo simulation method, and the initial optimization parameter combination and the adjustable range of oil input parameters were obtained. To clarify the adjustment direction of historical fitting, reduce the number of simulation operations, and optimize and accelerate the process of historical fitting. Based on the analysis of theoretical examples, the geological reserves obtained by the new method are 17.159 × 106sm3, and the geological reserves obtained by the volumetric method are 17.159 × 106sm3. The error is within 2%, which is basically consistent with the water invasion amount measured by the water invasion model and numerical simulation. Based on the case calculation of Jiefang Qudong oilfield in Xinjiang, the parameters of water invasion were measured by the material balance method. Volumetric coefficients and dissolved gas-oil ratios were stochastically simulated through the Monte Carlo method for 3000 iterations after gas production. Subsequently, the average values, as well as the corresponding percentiles P10, P50, and P90, were determined through numerical simulation. The formation pressure was predicted to be between 37–56 MPa. The measured pressure was 36–54 Mpa, and the two were basically consistent. The methodology founded on material balance principles and employing the Monte Carlo algorithm to expedite the historical calibration process in reservoir numerical simulations leads to a reduction in the quantity of required simulation iterationsclarifies the adjustment direction of historical fitting, reduces the uncertainty, and optimizes the process of accelerated historical fitting. Based on the uncertainty of the model in the previous research, this paper carries out Monte-carlo simulation on the basis of the calculation of material balance, compared with the measured data, the adjustable range of the uncertain parameters in the numerical simulation is determined. By the research on analysis of uncertainty and global sensitivity factors. It facilitates the development of a comprehensive macroscopic understanding of the reservoir, thereby expediting the history matching process.