The imbibition has an important influence on the water injection development of ultra-low-permeability reservoirs. In this paper, the nuclear magnetic resonance (NMR) high-temperature and high-pressure displacement system was used to simulate the formation temperature and pressure and the spontaneous imbibition of single porosity medium and dual porosity medium (containing fractures), and imbibition displacement experiments at different injection rates were carried out. The NMR T2 spectrum curves of simulated oil signals in the pores, throats, and fractures were obtained. The characteristics of oil content change and oil displacement efficiency in dual porosity medium under different experimental conditions were quantitatively evaluated, and the contribution of spontaneous imbibition to oil displacement efficiency was clarified. The experimental results show that the oil displacement efficiency of single porosity medium is lower than that of dual porosity medium. The smaller the pore is, the greater the displacement speed is, and the greater the contribution rate of imbibition is. The porosity, permeability, and maximum pore throat radius are positively correlated with the oil displacement efficiency of spontaneous imbibition but are poorly correlated with the oil displacement efficiency of under the dual action of spontaneous imbibition and displacement. The poor correlation also includes the median pressure, maximum mercury saturation, median radius, and displacement rate. Fractures play a positive role in improving oil displacement efficiency.

There are many research methods and experimental means for quantitative and semiquantitative evaluation of low permeability reservoirs. Generally, people do not use a single means to study them but use a variety of experimental means to verify and complement each other. Conventional mercury penetration and constant velocity mercury penetration are two important experimental methods for the quantitative evaluation of tight reservoirs. The micro characteristic parameters of reservoirs obtained by them are quite different, which bring some difficulties to people’s research. This paper first analyzes the reasons for the differences between the two from the aspects of experimental theory and model, experimental conditions, and experimental process. Taking C 6 and C 7 reservoirs in Ordos Basin as an example, a total of 13 representative pairs of samples were selected to analyze the difference in capillary pressure curve shape and pore throat distribution characteristics between the two experiments and to clarify the reasons for the difference in microscopic pore characteristic parameters measured by the two experiments. Finally, the correlation between the microscopic pore characteristic parameters and the movable fluid saturation parameters is analyzed. The results show that the theoretical model of conventional mercury penetration experiment is a capillary tube bundle model with different radii. The maximum injection pressure of experimental mercury is high and the experimental speed is fast. The theoretical model of constant velocity mercury penetration experiment is the pore and throat capillary model with different radii. The maximum injection pressure is low, the experimental speed is very slow, and the process is quasistatic. The parameters such as displacement pressure, total mercury saturation, and separation coefficient obtained by the latter are smaller than those measured by the former; however, the maximum throat radius, average throat radius, and other parameters obtained by the latter are larger than those measured by the former. According to the correlation chart drawn, it can be concluded that the correlation between the microscopic pore throat characteristic parameters and the movable fluid saturation in the constant velocity mercury penetration experiment is better than that in the conventional mercury penetration experiment. The influencing factors mainly include permeability, porosity, displacement pressure, maximum pore throat radius, and sorting coefficient. The maximum mercury injection saturation has little correlation.

After the oilfield development enters the medium and high water cut stage, the fine three-dimensional description of the single sand body, the distribution rules of the remaining oil, and its main controlling factors have become the focus of research. The premise is to understand the original oil and water distribution characteristics and the important influencing factors of the reservoir. Taking the Yan 932 reservoir of Y oil area in Dingbian as an example, combined with the meandering river sedimentary model, this paper uses dense well pattern logging data to dissect single sand body on the foundation of the core and logging results. The conclusion shows that, first, Yan 932 layer is divided into two stages: Yan 932-1 and Yan 932-2, which constitute four single sand body superposition configuration modes, namely, type I, type II, type I & II, and type I/II. Second, type I & II is mainly developed in the sand body of the main channel. In the west of the main channel sand body, type I, type II, and type I/II are developed. Only type II is developed in the east of the main channel sand body. Thirdly, sedimentary microfacies, sand body thickness, sand body configuration, and structural characteristics are the key factors influencing the oil-water distribution law of the Yan 932 reservoir.