Research On Calculation Method Of Recovery Of Tight Water-bearing Gas Reservoir

The tight water-bearing gas reservoir has the characteristics of low porosity,low permeability and high water saturation.Due to the special pore structure of this type of gas reservoir,the seepage law of fluid is different from that of conventional reservoir,resulting in a large deviation between the calculation results of various conventional recovery calibration methods of this type of gas reservoir and the actual situation.Therefore,it is urgent to establish a new recovery calculation method of tight waterbearing gas reservoir considering the influence factors of complex seepage.In view of geological and seepage characteristics of tight water-bearing gas reservoirs,this paper first uses the microscopic seepage physical experiment to study the influence of stress sensitivity and threshold pressure gradient on the recovery of gas reservoirs.The experimental results show that the greater the stress sensitivity coefficient and the threshold pressure gradient,the smaller the gas reservoirs recovery.According to the experimental data,the value of the average stress sensitivity coefficient of gas reservoirs and the empirical formula of the starting pressure gradient are obtained.Using random forest and XGBoost algorithm,the importance of macroscopic influencing factors affecting gas reservoir recovery is calculated,and five main controlling factors affecting gas reservoir recovery are clarified.Then,based on the conventional material balance recovery calculation method,a new material balance recovery calculation method considering stress sensitivity and threshold pressure gradient is established.At the same time,the productivity equation of fractured vertical wells considering micro influencing factors is used to form the recovery calculation method of productivity equation.Secondly,the particle swarm optimization support vector machine method is used to construct the artificial intelligence prediction method for recovery of tight water-bearing gas reservoirs by combining the main control factors affecting the recovery of gas reservoirs with the actual final recoverable reserves of gas reservoirs.Finally,taking the tight water-bearing gas reservoirs in the west of Sulige Gas Field as an example,the case verification and applicability analysis of the three gas reservoir recovery calculation methods are carried out.The results show that the particle swarm optimization support vector machine recovery calculation method has the highest accuracy,and the productivity equation recovery calculation method has the lowest accuracy among the three methods.The particle swarm optimization support vector machine method can be used in the initial stage of gas reservoir development.The material balance method is suitable for the use of gas reservoir recovery degree more than 10% and later,and the productivity equation method should be used in the stable stage of gas reservoir development and later.The recovery calculation method of tight water-bearing gas reservoir established in this paper fully considers the characteristics of tight water-bearing gas reservoir,and can accurately predict the recovery of this type of gas reservoir,which provides method guidance for rational and efficient development of tight water-bearing gas reservoir.