Search

History matching and quantifying uncertainty in hydrocarbon production forecasts are considered important steps in reservoir simulations and are key processes for providing crucial information for decision making in the development of oil and gas fields. However, history matching of unconventional reservoirs with complicated fracture system is very computationally expensive and time consuming, and most of the data used are subject to high uncertainty. In addition, the objectives of history matching are usually conflicting, such as improving match of oil rate will deteriorate match of water rate. Previous studies on assisted history matching have mainly focused on optimizing a single-objective function, where the production and pressure data are combined as a single misfit value. In this study, the proposed algorithm extends the application of the proxy model to deal with multiobjective optimization in the context of reservoir history matching. Then, the resulted proxy model were conducted to analyze history-matching quality and obtain uncertainty prediction. Result suggests that the multiobjective surrogate reservoir model scheme not only could obtain reliable history matches and provide accurate uncertainty estimation, but the complexity of gridded fractures is mitigated through the workflow. The benefit of using the multiobjective surrogate reservoir model is to obtain a various set of history matched models while reducing the number of expensive reservoir simulation runs. The proposed methodology might also be useful for reservoirs in which simulation models could not reflect the production history and for those with high uncertainties in reservoir characterization.

The CO2 water-alternating-gas flooding (CO2-WAG) is a key technology to improve the oil recovery of low permeability reservoirs. The effect of CO2 flooding to enhance the oil recovery is affected by geological conditions and production systems. The effect of CO2 flooding parameters on the enhanced recovery factor should be clarified to optimize the production system. In this paper, the machine learning algorithms are used to carry out the study and establish a set of procedures for optimizing CO2 flooding parameters based on the artificial neural network (ANN) and the particle swarm optimization (PSO) algorithm. Firstly, large amounts of basic data are generated by the Monte Carlo sampling method. Then, the recovery factor by the water flooding and the CO2-WAG and the enhanced recovery factor by CO2-WAG in different models are calculated in the reservoir numerical simulator. Moreover, the machine learning method is used to establish a neural network model, and analysis of the sensitivity of parameters of the enhanced oil recovery (EOR) is carried out by combining with the Sobol method. Finally, the neural network model and the particle swarm algorithm are combined to optimize the parameters of CO2-WAG flooding. The results show that the established model has a good prediction accuracy (97.6%), thus it could be used to predict the enhanced recovery factor by CO2-WAG, and it is applicable in potential evaluation of enhancing the oil recovery and optimization for parameters in the CO2-WAG well group.

The parallel reservoir perforating technology enables each shot hole in the gun to be fired in a direction parallel to the reservoir at a set angle, realizing a new technique in which the perforating projectile and the reservoir are parallel. The technology can be combined with the azimuth perforating technology. For the oblique angle of the reservoir and the wellbore, the oil pipe conveying type azimuth perforating technology, the inclined well self-orientation perforating technology and the modular cable perforating technology can be used for construction. The technology can effectively reduce the fracture pressure of reservoir construction, improve the productivity of medium and low permeability reservoir oil wells, and meet the needs of low permeability, thin poor layer and special well development and transformation.