Research Of Dynamic Prediction Model And Analysis Method Of Fire Flooding Reservoir Development

Fire flooding is one of the important methods to improve the heavy oil recovery.It has the characteristics of wide application range and high oil displacement efficiency.However,due to the low success rate of fire flooding projects at home and abroad,the greater economic risk of fire flooding development,and the less research results on the prediction and analysis of fire flooding dynamic prediction,especially intelligent analysis method is almost blank in the research field of fire flooding.Therefore,in order to fully tap the thermal recovery potential of heavy oil and realize the scientific and optimization of heavy oil development technology,this paper is deeply carried out application research on dynamic prediction and intelligent analysis method of fire flooding development,and obtained new understanding and main results are as follows:Based on the principle of mass conservation and energy conservation,combined with the distribution law of oil,water,gas,temperature and pressure in the fire flooding reservoir,this paper innovatively considered the gravity overlap factors influence of gas phase and steam phase on fire flooding effect,and established a complete theoretical model of development dynamic prediction for fire flooding reservoir(FFDPM).The FFDPM model can be used for the selection of fire flooding reservoir,the design of development scheme and the real-time tracking analysis of fire flooding production dynamic,which provides an important theoretical basis for the timely adjustment of fire flooding development scheme and the optimization and determination of injection and production technical parameters.Studied and solved the effective running problem of FFDPM model numerical simulation.Based on the FFDPM theoretical model,the analytical expressions of the relevant physical parameters required by the FFDPM model are given by theoretical derivation and numerical analysis.The acquisition method of the prediction model important parameters is proposed,such as the mass fraction of the gas phase component and the three-phase saturation in the condensation zone.The processing method for the unexpected interruption of numerical simulation caused by parameter assignment error is proposed,which improved the effectiveness and accuracy of fire flooding dynamic numerical simulation.Studied and developed the numerical simulation system of fire flooding dynamic analysis and development effect prediction.On the base of the fire flooding mathematical model and relevant parameters determination method,a perfect numerical simulation system of fire flooding dynamic analysis and development effect prediction is researched and developed.The system can track and analyze the production dynamic of fire flooding in real time,and simulate the changing rules and theoretical optimal values of various physical parameters in the process of fire flooding.The theoretical simulation results in this paper are compared with the results of previous one-dimensional combustion tube experiment and fireflooding field test.The results show that the simulation system of this paper has fewer input parameters,faster calculation speed,and the simulation results are in good agreement with laboratory experiment and field fire flooding actual production results.By introducing modern intelligent computing technology,a prediction method of fire flooding development effect based on neural network is proposed.This method combines the adaptive chaotic clonal particle swarm optimization(ACCPSO) algorithm with RBF neural network,and establishes fire flooding development effect prediction theoretical model of the RBF neural network based on ACCPSO algorithm.The actual data of foreign 42 fire flooding field projects is taken as an example to trial calculation.The results show that the new method proposed in this paper has the advantages of fast convergence speed and high prediction accuracy.It has high prediction ability and accuracy in the development effect prediction of fire flooding.In this paper,the author also combined the improved artificial fish swarm intelligence algorithm with the fire flooding mathematical model established in this paper,and proposed injection and production parameters optimization method of fire flooding based on adaptive jump fish swarm intelligence algorithm.This paper taked Du66 block of Liaohe oilfield as an example to trial calculation,the results show that it is feasible to optimize the injection and production parameters of fire flooding by using intelligent calculation method.It can dynamically optimize and adjust injection and production scheme of fire flooding in real time,so as to ensure that the fire flooding reservoir exploitation is always in an efficient and stable operation state.