Research Of Dynamic Prediction Methods And Optimization Technique Of Steam Flooding

Because of limited ability to increase the final oil recovery for steam stimulation (generally by 15%-20%), therefore, now it is an important faced subject that how to improve steam flooding technology, and how to realize the suitable development of steam flooding in oil field and scientific management. Although numerical simulation method is available and can be used to predict the steam flooding productions,it requires more data informations and research cost,and the computing process is complex.At such conditions, this paper develops and researches the new steam flooding theory model,providing a method to predict in real time, analyze and optimize the parameters of steam flooding,and it has important significance for enriching thermal recovery theory,selecting the oil field for steam flooding, drawing up development program, optimizing steam flooding process and improving the effect of steam flooding.According to the laws of matter conservation and energy conservation, combining with distribution of oil, water and gas in the reservoir of steam flooding, mass balance equation and energy balance equation with strict integral form are built.these equations laid theory foundation for analyzing the dynamicity and complexity of steam flooding dynamic, and they also provide a way to deal with the problem.This paper proposes the conceptions of equivalent position and pseudo- equivalent position of mutual conversion under the condition of steam overlay and steam front pushing,and their calculation methods are also put forward.An geological physical model is set up,and this paper has solved the compatibility issures of dynamic instability problems happened in the application of Van Lookeren theory,providing the basis for quantitatively analyzing the phenomenon of steam overlay and steam front push in steam flooding.Steam flooding dynamic prediction theory model (SFPM) consisting of steam front pushing prediction model (FAPM) and steam overlay prediction (SOPM) has been set up.SFPM can be used for selecting the reservoir for steam flood, designing the implementation plan of steam flood, analyzing the sensitive parameters,predicting and adjusting production dynamic.SFPM is the objective function of optimizing the technical parameters in steam flood.It just needs 35 basic parameters to run the SFPM,and the useful information will be available within a few seconds.Then the information can be used for selecting the reservoir for steam flood, designing originally the process and analyzing the feasibility of the scheme.the process of analyzing is easy and convenient,with high precision of results.This paper has designed and developed the performance prediction, optimization and adjustment system of the steam flooding, and developed numerical calculation software. This software can be used for selecting steam flooding reservoir, feasibility study, reasonable development plan establishment, oil production history fitting, and analyse the key technical factor for recovery.Through theoretically analyzing steam flooding sensitive parameters and "three fields" distribution for Qi 40 block in liaohe oilfield, the theory agreeing with practical conclusions has been obtained. The results show that oil-gas ratios will be lower than 0.15 after steam flooding for 67 months in medium-deep heavy oil reservoirs of Qi 40 block in liaohe oilfield, and economic benefits become worse.And steam injection quantity, injection method and even development method of this block should be adjusted, so the better development effect can be expected.The dynamic optimization and adjustment technical analysis based on particle swarm algorithm for steam flood is proposed. The results show that:(1) particle swarm optimization scheme has significantly increased cumulative oil-gas ratio for each month, and improved oil recovery; (2) the using of particle swarm optimization steam injection scheme can keep cumulative oil-gas ratio more than 0.15 for 120 months, obviously improving steam flooding effect; (3) accumulative oil-gas ratio curve declines slowly in the particle swarm optimization scheme, and production can continue after 120 months.The development, optimizing and adjusting system of steam flooding theoretical model and the predictions of the software agree with oilfield productions, and it has important application value.