Classification and Evaluation of Shale Oil Reservoirs of the Chang 7_1-2 Sub-Member in the Longdong Area

Abstract

Establishing a suitable classification and evaluation scheme is crucial for sweet spot prediction and efficient development of shale oil in the Chang 7_1-2 sub-member of the Longdong area. In this paper, a series of experiments, such as casting thin sections (CTS), scanning electron microscopy (SEM), low-temperature nitrogen adsorption (LTNA), high-pressure mercury intrusion porosimetry (HMIP), and nuclear magnetic resonance (NMR), were integrated to classify the pore throats and shale oil reservoirs in the study area. Moreover, the pore structure characteristics of different types of reservoirs and their contributions to productivity were revealed. The results show that the pore-throat system can be divided into four parts: large pore throats (>0.2 μm), medium pore throats (0.08~0.2 μm), small pore throats (0.03~0.08 μm), and micropore throats (<0.03 μm). Based on the development degree of various pore throats, the reservoir is divided into four types: type I (Φ ≥ 10%, K > 0.1 mD), type II (Φ ≥ 8%, 0.05 mD < K < 0.1 mD), type III (Φ ≥ 5%, 0.02 mD < K < 0.05 mD) and type IV (Φ < 5% or K < 0.02 mD). From type I to IV reservoirs, the proportion of dissolved pores and intergranular pores gradually decreases, and the proportion of intercrystalline pores increases. The proportion of large pore throats gradually decreases, and the proportions of medium pore throats and small pore throats increase initially and then decrease, while the proportion of micropore throats increases successively. The NMR pore size distribution changes from the right peak to the left peak. The developed section of the type I reservoir corresponds to the oil layer, and the developed section of the type I and II reservoirs corresponds to the poor oil layer. In contrast, the developed section of the type III and IV reservoirs corresponds to the dry layer. The daily production from single wells is primarily attributable to type I and II reservoirs.

1. Introduction

The increasing economic and social demand for energy and decreasing conventional oil and gas resources have led to the use of unconventional oil and gas resources [1,2,3,4,5]. Global energy trends have been vastly altered by technological and commercial breakthroughs in shale oil and gas development in the U.S. [6,7]. According to EIA data [8], by the end of 2019, the cumulative amount of U.S. shale oil production was 3.17 × 10⁸ t, with remaining proven reserves of 32.5 × 10⁸ t. In the U.S., shale oil accounted for more than 66% of the country’s crude oil production, and this proportion will continue to increase. China is rich in continental shale oil resources and has the third-largest amount of technically recoverable shale oil resources, after Russia and the United States [9]. The primary point of contention over the definition of shale oil at the moment is whether or not it should cover oils found in other kinds of tight inter-layers or neighboring layers of source rocks (such as siltstone, sandstone, limestone, and dolomite). Regarding the source-reservoir relationship, seepage mechanisms, as well as the primary factors regulating accumulation and development conditions, the oil resources accumulated in such interlayers or adjacent layers differ from those accumulated in pure mudstone and shale [10]. Continental shale oil in China can be divided into three primary types: interlayer type, hybrid sediment type, and pure shale type [11]. In 2019, the Qingcheng oilfield was discovered in the Chang 7 of the Mesozoic Yanchang Formation in the Ordos Basin, with reserves exceeding 10 × 10⁸ t of shale oil resources [12]. The distribution of favorable Chang 7 shale oil is primarily controlled by the vertical overlapping distribution of source rocks and gravity flow sand bodies [13]. The shale oil of the Chang 7_1-2 sub-member mainly is of the interlayer type [14]. This inter-layer reservoir is distinguished from pure mudstone, shale, and tight oil reservoirs by high permeability, huge reserve space, good reservoir connectivity, and great brittleness, which makes it easier to stimulate. Shale oil reservoirs have the characteristics of extremely low porosity (<12%) and permeability (<0.3 mD), complex pore structure, and strong heterogeneity [15], which result in difficult identification of high-quality reservoirs and significant differences in single-well productivity [16]. Establishing a classification evaluation scheme suitable for Chang 7_1-2 sub-member shale oil reservoirs in the Longdong area is urgent.

Li et al. [17] established a classification scheme for glutenite reservoirs based on sedimentary facies, diagenesis, and formation overpressure. Wang et al. [18] classified shale reservoirs based on X-ray diffraction (XRD) and low-temperature nitrogen adsorption (LTNA). Lu et al. [19] and Zhou et al. [20] characterized the microscopic pore-throat structure of shale oil reservoirs using high-pressure mercury intrusion porosimetry (HMIP) and determined classification schemes and lower limits of shale oil reservoirs. Yang et al. [21] quantitatively characterized shale microscopic pore structure at different scales by scanning electron microscopy (SEM), HMIP, LTNA, and NMR to classify shale reservoirs. Zhao et al. [22] established a classification scheme suitable for tight oil reservoirs based on sorting coefficient, pore-throat volume ratio, and mercury withdrawal efficiency. Zhang et al. [23] explored the upper and lower limits of the microscopic pore structure and classification scheme of tight gas reservoirs by HMIP, rate-controlled MIP, NMR, and wettability experiments. Huang et al. [24] established a reservoir classification system by a series of experiments (i.e., rate-controlled MIP, conventional MIP, contact angle measurement) and the mechanical equilibrium principle. The effective development of shale oil is dependent upon the factors such as porosity, oil saturation, permeability and the fluid flow properties in the shale matrix [25,26,27]. In addition, the size, distribution, and connectivity of pore throats control the quality of unconventional reservoirs [28]. Theoretically, the characterization and classification of the microscopic pore-throat can be carried out by comprehensively utilizing various high-end analysis techniques including high-resolution imaging technology, fluid injection technology, and the ray method technique. However, the primary ray methods, such small-angle X-ray scattering (SAXS), ultra-small-angle neutron scattering (USANS), etc., are uncommon and expensive. High-resolution imaging is qualitative, and there is a prominent contradiction between resolution and representation. Relatively speaking, the distribution of pore throats can be quantitatively shown by the fluid injection technology, which includes N₂/CO₂ adsorption and mercury intrusion technology. Moreover, the sample size used is large, which is more representative, and the characterization results are better suited for microscopic pore−throat classification [29]. NMR is a rapid and harmless method, full-scale integrated characterization of pore size distribution (PSD). A single testing technology cannot adequately explain the PSD of rock [30]. Therefore, the full PSD of rocks can be revealed more effectively by the comprehensive utilization of various techniques [31]. The above classification schemes are all established for conglomerate reservoirs, mud shale reservoirs, and tight oil and gas reservoirs. However, there have been few studies on the classification scheme of interlayer shale oil reservoirs.

This paper attempts to classify the interlayer shale oil reservoirs in the Chang 7_1-2 sub-member in the Longdong area. First, pore throats were classified based on the fractal characteristics of the pore throats. Thereby, a classification scheme of reservoirs was established based on the development degree of different types of pore throats. On this basis, the physical properties and microscopic pore structure characteristics of different reservoirs were revealed by a series of experiments, such as casting thin sections (CTS), SEM, LTNA, HMIP, and NMR. Moreover, the contributions of different types of reservoirs to productivity were revealed.

2. Geologic Setting

The Ordos Basin is a superimposed basin located in the midwestern regions of China. It can be divided into six secondary tectonic units: the western thrust belt, the Tianhuan depression, the Yishan slope, the Jinxi fault-fold belt, the Yimeng uplift, and the Weibei uplift [32]. The internal tectonics of the basin are relatively stable and simple, in which faults and folds are not developed, and the stratigraphy is gentle. The entire basin exhibits a regional tilted landform with high west and low east. The Longdong area is located southwest of the YiShan slope [33], with a gentle overall tectonic and a small-scale nose-shaped uplift locally (Figure 1) [34].

The Chang 7 member of the Yanchang Formation represents the maximum expansion period of the lake basin during most active basin deposition. More than 80% of the basin area is semi-deep lake to deep lake deposition, depositing a set of organic-rich oil-generating rock series with a thickness of more than 100 m. The Chang 7 sedimentary period mainly developed a deep lake-turbidite facies sedimentary system. The Chang 7 member can be subdivided into three sub-members: Chang 7₁, Chang 7₂, and Chang 7₃. The Chang 7₃ sub-member mainly developed a thick set of high-quality source rocks, including black shale and dark mudstone. The thickness of single-layer sand body is less than 2 m, and the sand ratio is less than 5%. The Chang 7_1-2 sub-members are mainly rich in organic mud shale intercalates with multiphase thin layers of silt-fine sandstone, with a single sand body thickness of 3.5 m on average; the average sand ratio is 17.8% [35], which is rich in shale oil. Thus, the Chang 7_1-2 sub-member is the target layer of our study.

3. Materials and Methods

3.1. Sample Characteristics

The dominant lithology of the Chang 7_1-2 sub-member in the study area consists of lithic arkose and feldspathic litharenites. The rock mineral fractions are mainly quartz (51.4%) and feldspar (24.2%). Followed by clay minerals (10.9%), clay minerals are dominated by illite (>8%), followed by chlorite (0.2%), and kaolinite as a whole is not developed. The rock fragments are metamorphic rocks (phyllite) and sedimentary rocks (dolomite). The cement is dominated by iron dolomite (1.9%), iron calcite (0.9%), and silica (0.6%), while the matrix is dominated by hydromica (11.5%). Grain shapes are mainly subangular, and grains are mainly moderately sorted. Therefore, the Chang 7_1-2 shale oil reservoirs are characterized by low compositional and structural maturity. According to data from 550 samples, the porosity of the Chang 7_1-2 shale oil reservoirs mainly ranges from 4% to 12%, and the permeability varies from 0.001 mD to 0.3 mD (Figure 2 and Figure 3).

3.2. Experimental Methods

We collected 550 physical property data, 56 CTS data, 44 SEM data, and 377 mercury intrusion data from Chang 7_1-2 shale oil samples in the Longdong area from the Changqing Oilfield Corporation. Additionally, 14 samples collected from the study area were extracted with a 9:1 solvent mixture of dichloromethane and methanol for 72 h and dried at 110 °C to remove residual oil and water. Subsequently, CTS, SEM, HMIP, LTNA, and NMR experiments were performed.

The CTS experiments were completed on a CIAS-2007 rock casting image analyzer. The CTS was resin-impregnated with blue epoxy; therefore, the blue color in the micrographs of the casts represents pores.

The SEM experiments were first performed by polishing the observation surface of the samples with an Ilion+ II 697C argon-ion polisher from the Gatan Company in the United States, then the samples were observed with a Quanta 450 field emission SEM produced by the American FEI Company.

The HMIP experiments were performed on corelab CMS300 and AutoPore IV 9500 high-pressure mercury intrusion instruments. In this study, the maximum mercury intrusion pressure was 200 MPa.

The LTNA tests were completed on a Micrometrics ASAP 2020 Analyzer, which used N₂ (77 K) as the adsorption gas. Adsorption branch data were interpreted using the Barrett–Joyner–Halenda (BJH) model to obtain the pore size distributions.

The NMR experiments were carried out on a MesoMR23-060H-1 NMR analyzer. First, dried shale oil samples were thoroughly saturated with simulated formation water. Then, the T₂ spectrum under saturated conditions was measured. The NMR parameters were set as follows: the echo time (T_E) was set at 0.08 ms; the waiting time (Tw) was 4 s; the number of echoes (NECH) was 10,000; and the number of scans (NS) was 32.

4. Results and Discussion

4.1. Fractal Characteristics of Pore Throats

Fractal theory is used to study the self-similarity and complexity of irregular shapes [36]. Currently, fractal theory is extensively used to study the microscopic pore structure of rocks [37]. The fractal dimension is a comprehensive parameter that characterizes the roughness of the pore surface morphology and the complexity of size distribution. Moreover, the pore structure inside the rock has self-similarity within a certain size range. Each type of pore throat should have the same fractal dimension, and different types of pore throats should have different fractal dimensions [38,39,40]. The fractal dimension can be determined based on SEM images or data from LTNA or HMIP experiments. Based on the LTNA data, there are three methods to calculate the fractal dimension, including the fractal Brunauer–Emmett–Teller (BET) model, the fractal Frenkel-Halsey Hill (FHH) model, and the thermodynamic method [41,42,43]. Among them, the fractal FHH model has been proven to be an effective method and is widely used to calculate the fractal dimension of coal and shale [44,45]. Due to its wide pore size coverage (from a few nm to hundreds of μm) [46,47], the HMIP method was used to characterize the fractal characteristics of shale oil samples in the study area.

The fractal dimension can be determined by the following mathematical equation (Equation (1)) [48,49,50]:

$$\log (1-S_{H\mathrm{g}}^{r+})=\left(3-\mathrm{D}\right)\log r-\left(3-\mathrm{D}\right)\log r_{max},$$

(1)

where S_Hg^r+ is the accumulative mercury saturation, r is the pore-throat radius, μm, r_max is the maximum pore-throat radius (μm), and D is the fractal dimension, which ranges from 2 to 3.

Plots of log (1 − S_Hg^r+) versus Log (r) based on HMIP data are shown in Figure 4, indicating that pores of different sizes have different fractal characteristics. According to Equation (1), four fractal dimensions were measured as r values of <0.03 μm, 0.03~0.08 μm, 0.08~0.2 μm, and >0.2 μm. Moreover, all of these linear fits have high correlation coefficients (R² > 0.88), indicating that the pore throats of the Chang 7_1-2 shale oil samples exhibit good fractal characteristics. In addition, due to the lack of large pore throats, some samples only exhibit three-stage fractal characteristics. (Figure 4a). Based on fractal characteristics, we divided the pore-throat system of the Chang 7 shale oil reservoir into four parts: micropore throats (radius < 0.03 μm), small pore throats (radius 0.03~0.08 μm), medium pore throats (radius 0.08~0.2 μm), and large pore throats (radius > 0.2 μm).

The relationships between the permeability and volume ratio of different pore throats are displayed in Figure 5. Permeability has a strong positive correlation with large pore throats and medium pore throats (R² > 0.7). The correlation between permeability and small pore throats is not strong (R² = 0.0174). However, permeability is strongly negatively correlated with micropore throats (R² = 0.7929) (Figure 5). Therefore, the pore-throat classification scheme obtained by the HMIP fractal method is reasonable.

4.2. Reservoir Classification

Cluster analysis is a method of classifying data based on some similarity in practice [51]. The cluster analysis process consists of the following four steps: data preprocessing, construction of the relationship matrix, selection of the clustering method, and determination of an optimal number of clusters. Data preprocessing refers to standardizing data so that different variables have the same scale, The standardization process is performed to make the data comparable. Data standardization methods are as follows [52]:

Standard deviation standardization:

$$x_{ij}^{\prime }=\frac{x_{ij}-{\overline{x}}_j}{s_j}\left(i=1,2,\dots ,n,j=1,2,\dots ,m\right),$$

(2)

$${\overline{x}}_j=\frac{1}{n}{{\displaystyle \sum }}_{i=1}^n{x}_{ij},s_j={\left[\frac{1}{n}{{\displaystyle \sum }}_{i=1}^n{\left(x_{ij}-{\overline{x}}_j\right)}^2\right]}^{\frac{1}{2}},\left(j=1,2,\dots ,\mathrm{m}\right),$$

(3)

where x_ij is expressed as the raw data of the jth index of the ith sample, and x′_ij is the standardized data of the jth index of the ith sample.

Min-Max standardization:

$$x_{ij}^{″}=\frac{x_{ij}^{\prime }-\underset{1\le i\le n}{\min }\left\{x_{ij}^{\prime }\right\}}{\underset{1\le i\le n}{\max }\left\{x_{ij}^{\prime }\right\}-\underset{1\le i\le n}{\min }\left\{x_{ij}^{\prime }\right\}},\left(j=1,2,\dots ,\mathrm{m}\right),$$

(4)

where $x_{ij}^{″}$ is the normalized data of the jth index of the ith sample. After range transformation, all index data $x_{ij}^{″}$ ∈ [0, 1].

Phenomenal advances in technological developments in information technology have led to the potential destruction and misuse of personal data information [53]. Therefore, to prevent loss of data information it is necessary to protect personal data using privacy preserving techniques. The systematic clustering approach minimizes information loss and ensures data quality [54]. This algorithm attempts to build all clusters simultaneously by first randomly selecting b n k c records as seeds. Then the algorithm allocates all records in the data set to their respective closest cluster and consequently updates feature weights to minimize information loss. This process is continued until the assignment of records to the cluster stops changing. If some clusters contain fewer than k records, those clusters are merged with another large cluster [55].

We used the systematic clustering method to classify shale oil reservoirs according to the proportion of large pore throats, medium pore throats, small pore throats, and micropore throats. First, Equations (2)–(4) were applied to standardize the proportions of different pore throats. Then, we selected the relationship matrix of squared Euclidean distance and the intergroup connection method to perform the systematic clustering analysis. Ultimately, the corresponding clustering result dendrogram and the various clustering numbers could be obtained (Figure 6).

Determining the optimal number of clusters is critical to obtaining more convincing reservoir classification results. To determine the optimal cluster number, three cluster numbers (i.e., 3, 4, and 5) were selected for the systematic clustering analysis. Figure 7 shows the relationship between the porosity and permeability of different types of reservoirs obtained by different cluster numbers. When the number of clusters is 3, the porosity and permeability data of different types of reservoirs overlap considerably, and the coincidence rate is only 69.4% (Figure 7a). When the number of clusters is 4, the conformity rate is 85.6% (Figure 7b). When the number of clusters is 5, the conformity rate is only 66.2% (Figure 7c). These results indicate that the optimal number of clusters is 4. Correspondingly, the Chang 7_1-2 shale oil reservoirs are classified into four types.

The type I reservoir is dominated by large pore throats (average 31.37%) and medium pore throats (average 25.74%). The type II reservoir is rich in medium pore throats (average 37.31%), followed by small pore throats (average 21.78%) and micropore throats (average 36.74%). The type III reservoir is dominated by small pore throats (average 36%) and micropore throats (average 52%). The type IV reservoir is characterized by a relatively high proportion of micropore throats, which reaches up to 90.57%. From type I to IV reservoirs, the proportion of large pore throats gradually decreased, the medium and small pore throats first decreased and then decreased, and the proportion of micropore throats increased (Figure 8).

4.3. Physical Property and Pore Structure Characteristics of Different Reservoirs

4.3.1. Physical Property Characteristics of Different Reservoirs

Physical parameters are the most basic and effective indicators for reservoir evaluation. The porosity, permeability, and displacement pressure boundaries of different types of reservoirs were determined (Figure 9). The porosity boundaries of type I, type II, type III, and type IV reservoirs were determined to be 10, 8, and 5%, respectively. The permeability boundaries of type I, type II, type III, and type IV reservoirs were determined to be 0.1, 0.05, and 0.02 mD, respectively. The displacement pressure boundaries of the type I, type II, type III, and type IV reservoirs were determined to be 2, 4.2, and 8 MPa, respectively.

4.3.2. Pore Types of Different Types of Reservoirs

Three pore types of the Chang 7_1-2 shale oil reservoir were identified from CTS and SEM images: dissolved pores, intergranular pores, and intercrystalline pores. Dissolved pores are formed by leaching, dissolution, metasomatism, and with irregular distribution and large diameter. Intergranular pores are mainly developed in the sediments with shallow burial, concentrated in the contact between mineral particles and crystals, between particles and between crystals. Intergranular pores are mostly elongated and irregular polygons. Intergranular pores are mainly formed between crystals of some authigenic clay minerals (mainly illite), with some connectivity [56]. A large number of dissolved pores and a small number of intergranular pores were observed in the type I reservoir under CTS (Figure 10a). The surface porosity of the dissolved pores in the type I reservoir varied from 0.3% to 4%, with an average of 1.58%. The surface porosity of intergranular pores ranged from 0.1% to 3.5%, averaging 0.52%. The total surface porosity was distributed between 0.3% and 6%, averaging 2.11% (Figure 11). Additionally, many dissolved pores and intercrystalline pores were observed under SEM (Figure 10e). Type I reservoirs experienced stronger to strong feldspar and rock chip dissolution during the diagenesis process, and developed better quality reservoirs.

Some dissolved pores and intergranular pores were observed in the type II reservoir under CTS (Figure 10b). The surface porosity of the dissolved pores was distributed between 0.1% and 4%, with an average of 0.97%. The surface porosity of intergranular pores was distributed between 0.1% and 3%, with an average of 0.28%. The total surface porosity was distributed between 0.1% and 5%, with an average of 1.27% (Figure 11). The proportion of dissolved pores and intergranular pores in the type II reservoir was lower than that in the type I reservoir. Many dissolved pores and intercrystalline pores were observed under SEM (Figure 10f). Type II reservoirs experienced strong compaction and weak cementation and weak dissolution during the diagenesis process, forming reservoirs of good quality.

A small number of dissolved pores and intergranular pores were observed in the type III reservoir under CTS (Figure 10c). The surface porosity of the dissolved pores varied from 0.1% to 2%, averaging 0.28%. The surface porosity of intergranular pores varied from 0 and 1.2%, averaging 0.064%. The total surface porosity varied from 0 to 3%, averaging 0.45% (Figure 11). A large number of intercrystalline pores and a small number of dissolved pores were also observed under SEM (Figure 10g). Type III reservoirs experienced strong compaction in the early stage of diagenesis, but strong cementation in the later stage, which dominated the whole diagenetic evolution, developing the medium quality reservoirs.

The dissolved pores and intergranular pores were virtually undetected in type IV reservoirs under CTS (Figure 10d). The surface porosity of the dissolved pores was distributed between 0 and 2.5%, averaging 0.175%. The surface porosity of intergranular pores was distributed between 0 and 1.3%, averaging 0.056% (Figure 11). Many intercrystalline pores were observed under SEM (Figure 10h). Type IV reservoirs experienced strong compaction in early diagenesis, and weak dissolution and weak cementation in the later stage of diagenesis, forming reservoirs of poor quality. Therefore, from type I to IV reservoirs, the surface porosity of dissolved pores and intergranular pores gradually decreased, as did the surface porosity of shale oil reservoirs. Meanwhile, complex diagenesis influenced pore development and evolution of the reservoir, which led to a gradual deterioration of both quality and pore structure from type I to IV reservoirs.

4.3.3. Pore Size Distribution of Different Types of Reservoirs

Pore information in the range of 1 nm to 200 nm can be acquired from LTNA experiments [57]. The pore size distribution (PSD) of shale oil reservoirs was calculated by the BJH model (Figure 12). The BJH PSDs of the different samples were similar, mainly ranging from 0.01 μm to 0.1 μm and peaking at approximately 0.03 μm.

The NMR technique is a rapid and nondestructive technique that was intensively adopted to characterize the full PSD of rocks. The NMR T2 spectra of shale oil samples under formation water saturated conditions exhibited a bimodal distribution pattern, with left peaks at approximately 0.1 ms and right peak at 20 ms. The left peaks of the NMR T2 spectrum and BJH PSD had good similarity. Therefore, the BJH PSD can be used for calibrating the NMR T2 spectra. The PSD of shale oil samples mainly ranged from 0.01 μm to 20 μm (Figure 13). The bimodal NMR PSD could be separated into small-sized pores (<0.2 μm) and large-sized pores (>0.2 μm). The type I reservoir was enriched in large-size pores, with the proportion reaching 73.8%. The type II reservoir was dominated by large-size pores (61.1%), followed by small-size pores (38.9%). In comparison, the type III reservoir was dominated by small-sized pores (60.3%), followed by large-sized pores (39.7%). However, small-sized pores dominated the type IV reservoir with a high proportion of 99.7%. Thus, from type I to IV reservoirs, the content of large-scale pores gradually decreased, while the content of small-scale pores increased sequentially (Figure 13).

The classification scheme for the Chang 7_1-2 sub-member shale oil reservoirs in the Longdong area is as follows (

Table 1. Classification and evaluation scheme of shale oil reservoirs in the 7_1-2 sub-member of the Longdong area.

Parameter	Type I	Type II	Type III	Type IV
Porosity/%	>10	8~10	5~8	<5
Permeability/mD	>0.1	0.05~0.1	0.02~0.05	<0.02
Displacement pressure/MPa	<2	2~4.2	4.2~8	>8
Pore size	large-sized pores	large-sized pores(main) + small-sized pores	small-sized pores(main) + large-sized pores	small-sized pores
Pore-throats Type	large pore-throats, medium pore-throats	mesoporous throats	small pore-throats, micropore throats	micropore throats
Pore Type	dissolved pores, intergranular pores	dissolved pores (main) + intercrystalline pores	intercrystalline pores (main) + dissolved pores	intergranular pores
Reservoir evaluation	better	good	medium	poor

). From type I reservoirs to type IV reservoirs, the physical properties and pore structure gradually deteriorate.

4.4. Relationship of Different Types of Reservoirs with Oil Layers and Productivity

In this section, we discuss the relationship of different types of shale oil reservoirs with oil layers and productivity based on the results of oil testing and log interpretation. As shown in Figure 14, the Chang 7₁ and Chang 7₂ sub-members mainly develop an oil layer (20.47% and 15.28%), a poor oil layer (21.16% and 21.84%), and a dry layer (37.69% and 37.43%). Furthermore, the Chang 7₁ sub-member is better than the Chang 7₂ sub-member in the degree of oil layer development. Moreover, statistical results show that the oil layers are dominated by type I reservoirs, the poor oil layers are dominated by type I and type II reservoirs, followed by type III reservoirs, and the dry layers are dominated by type III and IV reservoirs (Figure 15 and Figure 16).

Daily production of oil testing directly reflects the level of reservoir productivity. The correlation between the thickness of different types of reservoirs and daily oil production is significantly different (Figure 17), and type I and II reservoir thicknesses have a strong positive correlation with daily oil production. There is no apparent relationship between the type III reservoir thickness and daily oil production. However, the type IV reservoir thickness exhibits a specific negative correlation with the daily oil production. Further statistics on the single-well daily production for different types of reservoirs, indicating that the oil daily production is higher for reservoirs dominated by type I and type II. Therefore, the daily oil production of a single well is mainly attributable to the type I and II reservoirs (Figure 17,

Table 2. Different types of reservoirs with single well daily production.

Well	Main Reservoir Types	Oil Testing Layers	Daily Production (t/d)
Zhuang143	I	Chang71	30.60
Li348	I, II	Chang72	22.44
Li239	I, II	Chang71-2	21.42
Li99	I	Chang72	21.34
Li17	I, II, III	Chang72	21.08
Cheng98	I, II, III, IV	Chang71	12.24
Yue36	I, II	Chang71	11.82
Bai128	II	Chang72	11.10
Zhuang38	I, II	Chang71	11.05
Li190	I, II	Chang71	10.97
Xi268	I, II	Chang71	10.03
Cheng75	I, II	Chang71	7.74
Ban7	I, II	Chang71	7.60
Cheng95	II, III	Chang71	6.90
Zhuang89	I, II, III, IV	Chang71-2	6.69
Shan160	II, III	Chang71	5.52
Bai456	I, II	Chang72	5.30
Le23	II, III	Chang71	4.70
Ning156	II, III	Chang72	4.48
Zhen120	III, IV	Chang72	2.64
Ta37	III, IV	Chang71	1.10

).

The effective development of shale oil depends on the factor of oil saturation. The relationship between oil saturation and daily production of oil testing in the Chang 7_1-2 sub-member of the study area has a positive trend, but the relationship is slightly lower. The oil saturation is lower than 40%, and its daily oil production is less than 4 t, which is dominated by low-producing wells. When the oil content saturation is 40–50%, most of the daily oil production is more than 4 t, predominantly from industrial oil flow wells, including some low production wells. When the oil content saturation is greater than 50%, most of the daily oil production is over 4 t, primarily from industrial oil wells but containing some high production wells, and the number of low production wells is obviously reduced (Figure 18).

5. Conclusions

1.

According to fractal theory, the pore-throat system is divided into four parts: large pore throats (>0.2 μm), medium pore throats (0.08~0.2 μm), small pore throats (0.03~0.08 μm), and micropore throats (<0.03 μm).

2.

Based on the development degree of different pore throats, the shale samples were statistically clustered and analyzed. The Chang 7_1-2 shale oil reservoirs are divided into four types: type I reservoirs, type II reservoirs, type III reservoirs, and type IV reservoirs. From type I to IV reservoirs, the physical properties and pore structure gradually deteriorate. Type I, II, III, and IV reservoirs can be named better, good, medium, and poor reservoirs. A classification scheme of reservoirs was established.

3.

The Chang 7₁ and Chang 7₂ sub-members mainly develop oil layers, poor oil layers, and dry layers. The intervals where type I and II reservoirs are developed correspond to oil layers, the type III and IV reservoirs correspond to the dry layer. The higher daily oil production from a single well is primarily attributable to type I and II reservoirs.