This paper describes the further development of the virtual flow meter concept based on the author’s simulator of an unsteady gas–liquid flow in wells. The results of comparison with commercial simulators based on real well data are given as practical applications. The results of the comparison of the simulators demonstrated high correspondence (<10% error) for a number of target parameters. The description of the architecture and results of testing the algorithm for automatic settings of the model parameters are given. Operating speed was the key criterion in the architecture development. According to the test results, it became possible to achieve the adaptation accuracy of 5% specified.

The paper investigated the problem of selecting/finding the optimal process conditions for gas condensate wells. The well process conditions imply a set of parameters that characterize its operation. The optimization of process conditions provides for the efficient operation of an oil and gas field while meeting the defined boundary and initial conditions, and allows for the process/production goal to be achieved. This paper proposed using the tree-structured Parzen estimator (TPE), which allows for the results from previous iterations to be considered, in order to identify the most promising region of conditions, thereby increasing the optimization efficiency. The movement of multiphase fluid inside the pipeline system (also in the borehole) must be calculated to solve the process optimization problem. The optimization module was integrated into the hydraulic and unsteady state multiphase flow calculations inside the well and the pipeline. The platform created allows for the process conditions at gas condensate fields to be identified via the use of numerical methods. The proposed optimization algorithm was tested in delivering the task of optimizing the process conditions in 13 producing wells in a part of a real gas condensate field in Western Siberia. The engineering problem of optimizing the production of gas and the gas condensate was solved as a consequence of the calculations performed.