Lai, Xianghua (Second Institute of Oceanography) | Zhang, Heng (Shanghai Water Engineering Design & Research Institute Co., Ltd) | Li, Dong (Second Institute of Oceanography) | Hu, Taojun (Second Institute of Oceanography) | Wang, Yayao (State Grid Zhejiang Electric Power Corporation) | Zhang, Lijun (State Grid Zhejiang Electric Power Corporation) | Yan, Xunping (Zhoushan Power Supply Company)
Based on a series of physical property tests and static triaxial tests of marine soft soil near shore in Zhejiang Province, China, the hyperbolic relationship is gained between the mean stress and the generalized shear stress value. Then the modified Rhodes stress angle is considered in the model. The Cambridge model is modified and make it more responsive to the static characteristics of marine soft soil. The UMAT material constitutive subroutine of finite element software Abaqus is compiled, and a modified Cambridge model of marine soft soil is developed. The static triaxial simulation test is carried out by Abaqus, and the rationality of the improved model is verified. The results show that the modified Cambridge model of marine soft soil effectively simulates the stress-strain characteristics of marine soft soil, the nonlinear characteristics of the soft soil and the characteristics of the plastic flow.
As one of the stress-strain models describing incremental theory of plasticity (Roscoe, Schofield and Thurairajah, 1963), the Cambridge model (Roscoe, Schofield and Wroth, 1958; Roscoe, Barland, 1968; Zienkiewicz, Taylor and Zhu, 2013) is widely used in soil mechanics. Due to the differential soil properties, this model is modified in various specific applications, based on which, the cambridge model of marine soft soil should be modified on its shortcomings and the particular characteristics of marine soft soil. Y Chen (2002) considered the rheology of soil, based on the modified Cambridge model present a new modified Cambridge visco elastic plasticity model. YH Guo (2015) improved Cambridge model and got the accuracy of the elastic-plastic matrix improvement model, by using the Cambridge model to calculate the volumetric strain and shear strain of soil theoretically. There are also some researchers have improved the Modified Cam-Clay Model to make it compatible for overconsolidated soils (Amerasinghe, Kraft, 1983; Xu, Qi and Gao, 2008).
In this paper, based on the physical mechanics experiments, by fitting the mean stress and the generalized shear stress value, the state line of marine soft soil is hyperbolic, and the modified Rhodes stress angle are considered in the model, the further improvement of the Cambridge model and make it more responsive to the static characteristics of marine soft soil, which is validated by the secondary development of UMAT material constitutive subroutine in Abaqus.
Hou, Tengfei (China University of Petroleum, CUPB) | Zhang, Shicheng (China University of Petroleum, CUPB) | Li, Dong (China University of Petroleum, CUPB) | Ma, Xinfang (China University of Petroleum, CUPB)
Uniform proppant distribution in multiple perforation clusters plays a crucial role on sufficiently propping fractures conductivity in hydraulic fracturing. These propped fractures and their effectiveness is critically influenced by the in situ stress in the formation. As great uncertainty exists in uneven propped fracture, this paper examines the impact of proppant distribution and fracture conductivity variation on the gas productivity for shale gas reservoirs, by developing a reservoir simulation model.
In this paper, numerical reservoir simulation, which involves application of a constantly decreasing permeability to the propped fracture, are used to model the uneven proppant distribution and geomechanics effect. The decrease of permeability, along from the wellbore toward the tip, is simulated using an exponential approach, as well as a linear approach. Moreover, Effects of gas desorption and stress-dependent fracture conductivity are taken into account in this model. Sensitivity analysis is carried out on critical parameters to quantify the key parameters affecting gas productivity between uniform and nonuniform proppant distribution. The degree of non-uniform proppant distribution is also investigated and divided into four types of proppant distribution scenarios.
The following conclusions can be obtained based on the simulation results. A big difference on well performance between the case of linear and exponential permeability degradation is observed. The pressure distribution comparison shows higher pressure drops in the exponentially decreasing permeability case, which results in a lower gas production. Reservoir permeability plays a critical role in cumulative gas production, no matter in case of permeability exponentially degrading or linear degrading, followed by fracture half-length, primary fracture conductivity, Fracture complexity, permeability anisotropy. Furthermore, the effect of uneven proppant distribution between different clusters can significantly reduce the gas recovery, especially in low proppant concentration and small fracture conductivity.
The model presented in this paper takes the uneven proppant distribution and geomechanics effect into consideration and shows good agreement with real field production. This paper can demonstrate its own merits on the optimization of hydraulic fracturing treatments, and provide a better understanding of the effect of proppant distribution on well performance.
Hou, Tengfei (China University of Petroleum, CUPB) | Zhang, Shicheng (China University of Petroleum, CUPB) | Yu, Baihui (China University of Petroleum, CUPB) | Lv, Xinrun (China University of Petroleum, CUPB) | Zhang, Jingchen (Heriot-Watt University) | Han, Jingyu (China University of Petroleum, CUPB) | Li, Dong (China University of Petroleum, CUPB)
Channel fracturing, which greatly increase fracture conductivity by the creation of open channels inside fracture, has proved to be a novel stimulation technology that widely used in unconventional reservoir. The objective of this paper is to study the stimulation mechanism of channel fracturing by the combination of theoretical analysis and experimental research. However, for channel fracturing scenario, the currently available models are not accurate and appropriate in terms of prediction of proppant embedment and fracture conductivity in channel fracturing.
In this paper, new analytical models are derived to compute the proppant embedment, proppant deformation and fracture conductivity in channel fracturing. The mass deformation model and creeping deformation model are adopted to predict the change of proppant embedment and fracture conductivity over time. Many factors affecting the results of proppant embedment and conductivity, including closure pressure, elastic-plastic properties, properties of viscoelastic proppant and rock are investigated. Experimental researches are also conducted to evaluate conductivity at different closure pressures for the fractures of steel plate, shale and sandstone. Besides, the proppant embedment and proppant deformation are measured through the proppant embedment testing instrument, and the proppant distribution before and after experiments are comparatively analyzed.
The results show that the new analytical model proposed fits well with the experimental data, which verifies the accuracy and the feasibility of this model, though the decline rate of experimental data is a little bit faster than that of the model. The fracture conductivity is directly proportional to proppant viscosity, elastic modulus of proppant and inversely proportional to closure pressure, while elastic modulus of rock and large value of formation rock viscosity have slight impact on fracture conductivity. Moreover, the steady state of conductivity has been studied, and Comparisons between channel fracturing and conventional fracturing are analyzed in several aspects. The experimental results also reveal that the overall dimensions of created open channel may decrease or disappear due to the forced of formation stress.
Technical innovations in this paper are (a) new analytical models, including the mass deformation model and creeping deformation model, are adopted to predict the change of proppant embedment and fracture conductivity (b) Experimental tests are also performed to measure conductivity and proppant embedment at different closure pressures. This paper can demonstrate its own merits to show the advantage of channel fracturing technology.
Petrophysical properties are important characteristic parameters of reservoir quality. Rock physics models provide the link between elastic parameters and reservoir properties of interest. Pre-stack seismic inversion is an efficient approach to obtain these petrophysical parameters. This study proposes a direct estimation method of petrophysical properties based on AVO inversion. Firstly, a linear regression analysis of the well data is conducted and the rock physics model is obtained. Secondly, we derived the reflection coefficient approximate equation in terms of porosity, shale content and density through the incorporation of the rock physics model into the linear expression of Aki’s reflection coefficient. The equation establishes the theory foundation for the inversion of reservoir properties. Finally, we introduce a robust AVO inversion procedure to invert for reservoir parameters directly. The AVO inversion of the model and real data shows that the proposed inversion method can obtain reliable reservoir parameters from the seismic data directly.
The final goal of using geophysical methods for reservoir characterization is to predict the reservoir properties that include the lithology, fluid, and porosity. Reservoir properties are important characteristic parameters of reservoir quality. Prediction of reservoir patameters is crucial for the exploration and development of hydrocarbon reservoir because these parameters provide information for field appraisal and selection of optimal well location. Seismic data contain invaluable sources of information for reservoir characterization and pre-stack seismic inversion is an efficient approach to obtain these reservoir parameters.
Over the years, there have been many studies on how to obtain petro-physical properties from seismic data. Doyen (1988) estimates porosity from seismic data by using cokring. Tian (2002) combines the genetic algorithm and the BP algorithm to give GA-BP mixed algorithm which has higher accuracy and faster convergence speed. The new algorithm also provides improved predict accuracy of thin interbeded reservoir parameters (e.g., porosity, permeability, and saturation). Based on the BISQ model, Nie (2004) performs the inversion of reservoir parameters by using the Niche genetic algorithms. Mukerji et al. (2001) and Avseth et al. (2001) combine statistical rock physics, well log data analysis, and prestack seismic inversion to estimate reservoir parameters from prestack seismic data and to evaluate the associated uncertainty.
Li, Yi-Bo (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China) | Cheng-Yuan, Dong (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China) | Pu, Wan-Fen (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China) | Jin, Fa-Yang (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China) | Chen, Ya-Fei (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China) | Li, Dong (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China) | Zhao, Jiang-yu (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China) | Zhao, Jin-Zhou (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, People's Republic of China)
With the decline of conventional oil production, developing and producing heavy oil resources efficiently are becoming more important. High pressure air injection (HPAI) is generally considered as an efficient method to drive the crude oil in light oil reservoir. But there is a debate about whether the released heat from oxidation reaction is able to activate the heavy oil effectively. Comparing to the in-situ combustion (ISC) technique, air injection without ignition will reduce the operation difficulty and eliminate the security risk. Thus studying the changing of the oil property in presence of injected air is the prerequisite to apply the HPAI technique in heavy oil reservoir. For this purpose, the constant temperature oxidation experiments have been carried to study the influence of pressure on heavy oil in the air injection process at reservoir condition of Tahe oilfield.
The results showed that the pressure has obvious influence on the crude oil property. By the increment of pressure, the oxygen in the air presented a decreasing trend while the amount of carbon dioxide did not exhibit an increasing trend. The oxidation addition reaction dominated the reaction type. For the oil phase, the viscosity of the oxidized oil presented a slight decrease under relative low pressure condition. When the pressure reached 50MPa, the crude oil has transferred into coke completely. During the oxidation process, the property and amount of the coke directly related to the released heat amount in the high temperature oxidation (HTO) stage. According to the thermogravimetry (TG) and differential scanning calorimetry (DSC) results, the oxidized oil achieved lower HTO trigger temperature. So it is believed that the higher pressure has a positive influence on the coke deposit process. The formation of sufficient coke will bring the possibility for the application of HPAI technique in heavy oil reservoirs.
Li, Dong (SINOPEC Geophysical Research Institute) | Wang, Lixin (SINOPEC Geophysical Research Institute) | Xu, Zhaotao (SINOPEC Geophysical Research Institute) | Zheng, Xiaopeng (SINOPEC Geophysical Research Institute) | Mu, Jie (SINOPEC Geophysical Research Institute)
The seismic data acquired in the mountains area are generally irregular or sparely because of the complex surface, which may not fulfill the processing requirements and degrades processing quality, so these data should often be interpolated. A projection onto convex sets (POCS) algorithm using Fourier transforms is a well-known technique to reconstruct the irregular seismic data, helping on the processing of data with different acquisition problems. We proposed the interpolation procedure using POCS method based on OVT domain and applied it to the field data. Numerical examples indicates that the proposed scheme is effective and applicable, as it can reconstruct missing traces of complex data acquisition.
There are abundant oil and gas resources in the mountainous area of South China, which has broad prospects for exploration and development. However, seismic data is usually irregularly or sparsely distributed along the spatial direction in the complex surface area, which may not meet the processing requirements and then degrades processing quality, so it is important and necessary to regulate and interpolate seismic data at the missing spatial locations where measurements are not acquired in the seismic data processing stage.
Due to their simplicity, empirical production forecasting methods have been used by the petroleum industry for decades. Since 2008, a number of empirical methods have been introduced to the petroleum industry, specifically for wells located in tight/shale reservoirs. However, most of these new methods are not reliable for forecasting remaining reserves, although they may appear to be very good for forecasting EUR in wells in which a high percentage of the EUR has already been produced.
The Stretched Exponential Production Decline (SEPD) Method was introduced in 2010. Our results from analysis of both synthetic and actual field data by using SEPD have indicated that this method will most likely underestimate EUR in reservoirs with permeability ranging from 0.1mD to 0.0001mD. A modified SEPD (YM-SEPD) Method has therefore been developed to eliminate the SEPD Method's shortcoming by employing a new specialized plot to find all related parameters. This newly developed method is very easy to use and, most importantly, it will yield a much more reliable production and remaining reserve prediction for tight horizontal wells. With longer production histories, remaining reserves can be forecasted even more accurately and with a high confidence level.
Hundreds of horizontal wells including oil wells from various formations (Cadomin, Montney, Notikewin, Cardium, Barnett Shale, Muskwa, etc.), hydraulically fractured in various ways, have been analyzed using the modified SEPD (YM-SEPD) method. Results indicate that reliable EURs and production profiles can be predicted readily for wells having only two to three years of production history. For wells having less than two years of production history, the modified SEPD (YM-SEPD) Method can also yield reasonable production forecasts when coupled with Duong's empirical method.
This paper presents the application of the modified SEPD (YM-SEPD) Method to a number of actual and synthetic oil and gas wells to estimate their proved reserves, including horizontal wells producing dry, wet and retrograde gas as well as tight oil. These examples have had production histories with either observed or non-observed boundary-dominated flow (BDF). The examples also illustrate how the modified SEPD (YM-SEPD) method is capable of estimating proven reserves with high confidence.