Optimization Method for Water-Flooded Beach-Bar Sand Bodies: A Case Study of the Fourth Member Red Beds of the Paleogene Shahejie Formation in the Dongying Depression

Abstract

As water flooding continues to advance in mature oil fields, conventional well logging curve responses exhibit anomalies, particularly in deep-seated beach-bar thin-layer sand bodies. These sand bodies exhibit strong vertical and planar heterogeneity, which hinders a clear understanding of their distribution and connectivity. This paper conducts a sensitivity analysis of the lower part of the 4th member of the Shahejie Formation in fault block C26, based on core data analysis and the integration of drilling, logging, recording, and production dynamic data. Natural potential and resistivity curves are selected as the sensitive curves for identifying sand bodies. Preliminary processing of well logging curves is carried out using the “standard layer constraint and multi-standardization method comparison” principle. Employing sedimentological research methods and the principles of wavelet transformation, the well logging curves within the study area undergo extraction of both high and low frequencies. This procedure accentuates details related to thin sand bodies and responses indicative of sedimentary cyclicity. Through a thoughtful amalgamation of multiple curves, the investigation achieves a systematic fusion of natural potential curves via multi-curve frequency division fusion, employing reconstruction optimization. This method adeptly mitigates interference stemming from water-flooded layers, effectively addressing challenges such as excessive calibration and ambiguous identification of sand bodies. As a result, a comprehensive analytical approach is established for assessing the distribution and connectivity of deep-seated, beach-bar thin sand bodies influenced by water-flooded layers, providing clarity on the connectivity relationships among the sand bodies. Additionally, in combination with the results of mercury injection experiments in the water-flooded layer segments, favorable reservoir criteria for the study area are determined, providing a scientific basis for adjusting future development plans.

1. Introduction

With the continuous progress of water flooding in oil field development, the extent of water flooding in oil fields is increasing, resulting in various alterations in reservoir characteristics and the distribution status of oil, gas, and water. Specifically, within the Shasixia sub-member of the Dongying depression, there is a development of deep-layer shoals and thin-layer sand bodies. The water-flooded layer induces less conspicuous responses in sedimentary logging cycle characteristics, significantly impacting the development of such reservoir sand bodies [1,2,3,4,5]. Water-flooded high-permeability zones, influenced by favorable lithofacies belts, display weaker compaction and cementation effects, as well as being influenced by faults and fractures, leading to the emergence of localized advantageous permeable channels. Additionally, water flooding may elevate the likelihood of fractures occurring in the sand bodies. Utilizing modeling techniques such as EPM, DFN, and CN has proven effective in studying the internal heterogeneity of the sand bodies. Through methods encompassing core observation, thin-section identification, and physical property testing, it has been observed that high-permeability channels are closely linked to favorable lithofacies belts in the study area, with a relatively minor impact from faults and fractures. The sand bodies in the study area display rapid lateral and vertical variations, internally demonstrating relatively weak diagenetic processes, development of primary pores, visible secondary pores, and an absence of commonly observed advantageous permeable channels formed by fractures. The primary channels for permeability development are predominantly associated with the development of primary pores in shoal-bank sand bodies [6,7,8,9].

Various methods are currently employed for identifying and processing water-flooded layers in well logging curves. These methods include the natural potential method, logging curve feature analysis [10], crossplot method, movable water analysis method, hierarchical analysis method [11], grey correlation analysis, curve reconstruction method [12,13,14], and artificial intelligence algorithms, among others [15,16,17,18,19,20,21]. In this study, using C26 in a well-defined oil field as a case study, we conducted observations and analyses using core data, with a particular focus on the sand body and sedimentary cyclical logging response characteristics within the study area. We selected sensitive logging curves specific to the study area. Employing sedimentological research methods and in conjunction with the fundamental principles of wavelet transformation [22,23,24,25,26,27,28,29], this research highlights sedimentary cyclical response characteristics and clarifies sand body connectivity relationships. We established a method for multi-curve frequency fusion reconstruction of logging curves within the study area and established relationships between response characteristics of different reservoir types. These findings provide a solid foundation for future development adjustments.

• Conduct a sensitivity analysis of sand bodies and sedimentary logging curves in the study area based on systematic observation and core data analysis. Optimize the selection of sensitive logging curves in the study area.
• In response to the challenge of identifying thin-layer sand bodies and sedimentary cycles in the presence of water-flooded layers, apply sedimentological research methods and wavelet transformation principles. Explore the utilization of multi-curve frequency fusion and reconstruction methods for thin-layer sand bodies, enabling a rational prediction of logging curves in water-flooded layers. Establish the groundwork for subsequent thin-layer sand body identification and sedimentary cycle analysis.
• Evaluate sand bodies and sedimentary cycles by integrating the results of sedimentary microfacies studies in the study area. Investigate favorable sedimentary microfacies influenced by water-flooded layers and elucidate the relationship between response characteristics and different reservoir types.

2. Geological Overview

The Bohai Bay Basin, situated in the eastern part of China, stands as a significant oil and gas basin. Within this basin, the Dongying Depression, located in the southern region, emerges as a prolific oil and gas depression. Originating as a Mesozoic-Cenozoic sagging basin against the backdrop of Precambrian basement rocks, the Dongying Depression is geographically positioned adjacent to the Qingtuozhi uplift in the east, overlapping with the Guangrao uplift and the Luxi uplift to the south. Bounded by the Qingcheng uplift and Linfanjia uplift in the west, and delineated by the Chenjiazhuang uplift and the Bin County uplift in the north, it represents a complex half-graben extensional basin characterized by northward faulting and a gradual southward dip (Figure 1). During the Paleogene rifting period, influenced by the Xishan movement, the basin underwent an evolutionary process encompassing initial rifting, intense rifting, and rifting contraction. This process led to the formation of a complete sedimentary cycle featuring shallow lake facies, deep lake facies, river deposits, and shallow lake facies [30,31,32]. Block 26, situated in the southern gentle slope zone of the Dongying Depression [33,34], is positioned within the Purity-Caoqiao nose-shaped structural belt. Governed by boundary normal faults to the north and west, it exhibits extensional faults. The northern boundary normal fault has a downthrown block to the south, while the western boundary normal fault has a downthrown block to the east. Within the block, predominantly east-west trending normal faults are observed, resulting in a relatively simple structure with the development of structural lithological oil and gas reservoirs (Figure 2). The study area’s stratigraphic sequence, from bottom to top, includes the Paleogene Kudian Formation, Paleogene Shahejie Formation, Paleogene Dongying Formation, Neogene Guantao Formation, and Neogene Minghuazhen Formation.

The target layer is the 4th member of the Paleogene Shahejie Formation (Figure 2), specifically the Hongyi sand group (H1) of the lower sub-member of the 4th member of the Shahejie Formation (Es₄^L). Formed approximately 45.4 million years ago, this layer, with a thickness of about 60 m, represents a set of strata formed in a marginal-shallow lake sedimentary environment with intermittent saline water under arid to semi-arid climatic conditions. The target layer in the study area encompasses multiple sets of counter-cyclical beach-bar sand bodies, characterized by their thinness, often of around 2 m, and common occurrences of 1-m thick sand bodies. These sand bodies exhibit rapid vertical and horizontal variations with poor connectivity. The curves analyzed include SP (spontaneous potential curve), GR (natural gamma curve), AC (acoustic time difference curve), R4 (resistivity curve), and CON (induction conductivity curve).

The study area is situated in the southern gentle slope belt with a relatively weak sediment supply, and sediment deposition lacks a distinct directional pattern. Previous research had suggested the presence of fluvial facies, floodplain lake, and other sedimentary facies in the study area. However, based on comprehensive core analysis and well log facies analysis results, this study has identified the development of multiple sets of retrogradational sand bodies in the study area. No clear progradational responses, such as meandering or braided channel-type sand bodies, were observed, leading to the interpretation of the study area as having shore-shallow lacustrine beach-bar.

Among these subfacies, the barrier bodies consist of fine-grained sandstones, fine sandstones, and thick-bedded fine sandstones, typically exceeding 1 m in thickness. These can be further categorized into the beach-bar crest and the side edge of beach-bar, highlighting the high points and favorable distribution of sand bodies. The sheet-like sand sedimentary microfacies are characterized by fine sandstones and muddy fine sandstones, generally less than 1 m in thickness. Using well C26-25 as an example (Figure 3), it is evident that the lithology in the study area exhibits retrogradational responses, transitioning from mudstone to muddy fine sandstone to fine sandstone. The oil-bearing quality gradually improves from the base to the top of the retrogradation, with the top of the no. 3 sand body reaching a level of oil saturation.

The prioritized selection of sensitive curves forms the basis for subsequent sand body identification and the analysis of various sedimentary microfacies deposition cycles. In this study, the optimization of critical well logs, such as Spontaneous Potential (SP), Natural Gamma (GR), Acoustic Time Difference (AC), Resistivity (R4), and Induction Conductivity (CON), was conducted using intersection diagrams reflecting diverse lithologies and sand-mudrock characteristics (Figure 4a–d). Notably, the SP curve exhibits significant sensitivity in sand body identification, showcasing effective performance, particularly in the classification of fine sandstone. While the GR curve demonstrated some effectiveness in recognizing sand-mud rock, its performance in identifying fine-grained lithology was relatively poor, particularly in distinguishing between fine sandstones, muddy fine sandstones, and fine sandstones. A comprehensive analysis of well logging response characteristics in the study area revealed that the natural potential curve exhibited the most pronounced response characteristics to sand bodies and sedimentary cycles (Figure 2 and Figure 3). Following closely in sensitivity were the resistivity curve, induction curve, natural gamma curve, and sonic travel time curve. Consequently, SP, R4, CON, GR, and AC were selected as the sensitive well logging curves in this study.

3. Materials and Methods

Block 26 is predominantly characterized by shore-shallow lacustrine beach-bar composed of mudstone, muddy fine sandstone, and fine sandstone, all of which exhibit favorable oil-bearing characteristics. Nevertheless, a significant challenge arises due to the thin and rapidly changing nature of these sand bodies, making their vertical identification and prediction a formidable task. Additionally, the field has reached the middle to late stages of development, leading to frequent water inundation with various types and degrees of severity. Well logging curve responses display variations, particularly the natural potential curve, which is highly sensitive to inundation and exhibits noticeable baseline shifts, thereby impairing the identification of sand bodies and sedimentary cycle responses. To effectively address the challenge of identifying thin-layer sand bodies under the influence of water inundation, this paper standardizes the sensitive curves within the target layer in the study area. Building upon this foundation, the natural potential curve (SP), natural gamma curve (GR), sonic travel time curve (AC), resistivity curve (R4), and induction conductivity curve (CON) undergo wavelet transformations. This process highlights different scale features and separates the high and low-frequency components of well logging curves, allowing for a clear representation of their detailed characteristics and trends. The high-frequency components significantly enhance the identification of thin-layer sand bodies and sedimentary cycle analysis in the study area, while the low-frequency components provide insights into the patterns of water ingress and retreat. The specific expressions are as follows:

$$C\left[a,b;f\left(t\right),\psi \left(t\right)\right]={\int }_{-\mathrm{\infty }}^{\mathrm{\infty }}f\left(t\right)\frac{1}{\sqrt{a}}{\psi }^{\mathrm{*}}\left(\frac{t-b}{a}\right)\mathrm{d}t$$

(1)

Drawing on our comprehension of sedimentary cycles and sand bodies, this paper, leveraging the outcomes of high and low-frequency analysis, devises a multi-curve frequency fusion reconstruction approach tailored to the study area. This method aptly tackles the difficulties associated with the limited response characteristics of well logging natural potential curves in the presence of water inundation.

3.1. Low Frequency Multiple Regression

Prior to conducting frequency decomposition, the sensitive well logging curves are initially normalized to derive the respective normalized result matrices BGY and YGY for B and Y, respectively.

$$AC={\left[a{c}_1 a{c}_2\dots {ac}_n\right]}^T$$

(2)

$$CON={\left[{con}_1 {con}_2\dots {con}_n\right]}^T$$

(3)

$$GR={\left[{gr}_1 {gr}_2\dots {gr}_n\right]}^T$$

(4)

$$R4={\left[r_1 r_2\dots r_n\right]}^T$$

(5)

$$Y={\left[{sp}_1 {sp}_2 \dots { sp}_n\right]}^T$$

(6)

$$B=\left[AC CON GR R4\right]$$

(7)

In this research, the db4 wavelet is selected for the high- and low-frequency decomposition of the aforementioned BGY and YGY corresponding curves (Figure 5), resulting in the corresponding high- and low-frequency result matrices. “S” represents the original curve, while “a1–a6” denotes the corresponding six low-frequency decomposition curves, and “d1–d6” represents the corresponding six high-frequency decomposition curves. Among these matrices, BGYA signifies the matrix of low-frequency approximation coefficients derived from the six-level decomposition of BGY corresponding curves, and BGYD denotes the matrix of detail coefficients from the six-level decomposition of BGY curves. Likewise, YGYA represents the matrix of low-frequency approximation coefficients obtained from the six-level decomposition of YGY corresponding curves, while YGYD represents the matrix of detail coefficients derived from the six-level decomposition of YGY curves. Performing multivariate regression analysis on the low-frequency component curves:

$$Z={\left[1 1 \dots 1\right]}^T$$

(8)

$$W={\left[w_1 w_2 w_3 w_4{ w}_5\right]}^T$$

(9)

$$YGYA=\left[Z BGYA\right] W=X_1 W$$

(10)

$$J(W)={\sum }_{i=1}^n{\left(f\left(x^{\left(i\right)}\right)-y^{\left(i\right)}\right)}^2=(X_1 W-YGYA{)}^T\left(X_1 W-YGYA\right)$$

(11)

$$\frac{\partial J\left(W\right)}{\partial W}=2{X_1}^T{X}_1\theta -2{X_1}^{T}YGYA=0$$

(12)

$$W={\left({X_1}^T{X}_1\right)}^{-1}{X_1}^{T}YGYA$$

(13)

A multivariate regression analysis was conducted for the low-frequency component curves, leading to the derivation of the corresponding weighted parameters W (

Table 1. Detailed Table of Low-Frequency Multivariate Regression Parameters.

Well	Label	W	AAC6	ACON6	AGR6	AR46	R2-A
C26-48	Weighting Factor	0.605	1.019	−0.714	0.106	−2.422	0.9338
	Confidence Interval	0.570 ~0.640	0.966 ~1.072	−0.753 ~−0.675	0.063 ~0.149	−2.499 ~−2.345
C26-42	Weighting Factor	0.820	−1.034	1.866	−0.595	−0.491	0.9375
	Confidence Interval	0.761 ~0.879	−1.154 ~−0.914	1.770 ~1.962	−0.639 ~−0.551	−0.704 ~−0.278
C26-GX23	Weighting Factor	0.402	−2.588	−0.785	2.935	−0.733	0.8848
	Confidence Interval	0.325 ~0.479	−2.717 ~−2.460	−0.931 ~−0.640	2.757 ~3.113	−0.847 ~−0.620

). It is noteworthy that the goodness of fit, as indicated by R2-A, consistently exceeded 85%, thereby establishing a solid foundation for the subsequent multivariate regression of high-frequency components from multiple curves. In this context, AAC6, ACON6, AGR6, AR406, and ASP6, respectively, signify the low-frequency approximation curves corresponding to the AC, CON, GR, R4, and SP curves.

3.2. Multiple Regression of High Frequency Component Fusion

Building upon the low-frequency multiple regression results mentioned earlier, a weighted coefficient matrix W is derived, corresponding to the approximate coefficients. Subsequent analysis of high-frequency components leads to the establishment of the detail coefficient matrix U (

Table 2. Detailed Table of High-Frequency Multi-Curve Multivariate Fusion Regression Parameters.

Well	Label	U	DAC6	DCON6	DGR6	DR46	SP-A	R2-D
C26-48	Weighting Factor	−0.024	−0.274	0.789	0.247	0.376	1.040	0.8246
	Confidence Interval	−0.060 ~0.012	−0.334 ~−0.214	0.701 ~0.878	0.174 ~0.321	0.291 ~ 0.461	−0.989 ~1.090
C26-42	Weighting Factor	0.155	−0.378	0.455	0.113	−0.223	0.775	0.8186
	Confidence Interval	0.123 ~0.188	−0.449 ~−0.307	0.389 ~0.520	0.055 ~0.171	−0.381 ~−0.065	−0.729 ~−0.822
C26-GX23	Weighting Factor	0.003	−0.121	0.354	0.066	−0.068	1.004	0.8048
	Confidence Interval	−0.019 ~0.026	−0.195 ~−0.47	0.254 ~−0.453	−0.028 ~0.161	−0.155 ~0.019	0.948 ~1.060

). Specifically, DAC6, DCON6, DGR6, DR406, and DSP6 represent the high-frequency detail curves for the AC, CON, GR, R4, and SP curves, respectively.

$$U={\left[u_1 u_2 u_3 u_4{ u}_5 { u}_6\right]}^T$$

(14)

$$YGY=\left[Z BGYD X1W\right] U=X_2 U$$

(15)

$$J(U)={\sum }_{i=1}^n{\left(f\left(x^{\left(i\right)}\right)-y^{\left(i\right)}\right)}^2=(X_2 W-YGY{)}^T\left(X_2 W-YGY\right)$$

(16)

$$\frac{\partial J\left(U\right)}{\partial U}=2{X_2}^T{X}_2\theta -2{X_2}^{T}YGY=0$$

(17)

$$U={\left({X_2}^T{X}_2\right)}^{-1}{X_2}^{T}YGY$$

(18)

3.3. Optimal Treatment of Watered-Out Layer Curve

Using the multi-curve frequency fusion reconstruction method described above, the well logging curves within the water-flooded layers in the study area were optimized. The target layer, which is the H1 sand group in the lower part of the 4th member of the Shahejie Formation, is influenced by water inundation, resulting in baseline drift and weakened response characteristics in the well logging curves. Nonetheless, the well logging facies still exhibit funnel-shaped retrogradational response characteristics, and the core data reveal a pattern of mudstone-muddy fine sandstone-fine sandstone retrogradational features. Additionally, the sedimentary facies types in adjacent layers are similar, and the well logging responses are largely consistent. In this paper, based on the multi-curve frequency fusion optimization method and utilizing AC, CON, GR, and R4 curves, the natural potential curve is effectively adjusted to accentuate the response characteristics of sand bodies and sedimentary cycles.

In this context, AAC6, ACON6, AGR6, AR406, and ASP6 represent the low-frequency approximation curves of their respective curves, while DAC6, DCON6, DGR6, DR406, and DSP6 represent the high-frequency detail curves of those same curves. SPA signifies the low-frequency approximation curve obtained through the multivariate regression method, and SP-AD designates the multi-curve frequency fusion reconstruction curve. Taking well C26-GX23 as an example (Figure 6), where ‘A’ denotes the standard curve segment of the multi-curve frequency fusion reconstruction. It is observable that segments ‘a’, ‘b’, and ‘c’ in the new curve SP-AD4 more distinctly delineate the presence of sand bodies compared to the original SP curve, resulting in higher recognition accuracy and more pronounced retrogradational response characteristics. ‘B’ signifies the predicted curve segment of the multi-curve frequency fusion reconstruction. Owing to the influence of water inundation, segment ‘B’ displays significant drift in the original SP curve, with minimal or weak response characteristics. Notably, segment ‘d’ in the new curve SP-AD4 unambiguously indicates the presence of sand bodies, segment ‘e’ exhibits more pronounced retrogradational response characteristics in the lower part, and segment ‘f’ also clearly illustrates two sets of retrogradational sand bodies, especially in the lower part, which are approximately 1 m thick. Segment ‘g’ similarly reveals distinct retrogradational sand bodies, whereas the original SP curve displays nearly no response.

Combining the effectiveness chart of the water-flooded layers with the multi-curve frequency fusion reconstruction analysis (Figure 7), it becomes evident that the lower part of well C26-GX23, affected by water inundation, provides a clearer and more distinct representation of sand bodies. This confirms the practicality of the multi-curve frequency fusion reconstruction analysis method in identifying and analyzing thin-layer sand bodies in the presence of water inundation.

4. Discussion

By employing the previously described multi-curve frequency fusion reconstruction analysis method, the utility of this approach in identifying thin-layer shoreface-barrier sand bodies amid water inundation has been validated. Utilizing well C26-X55 as an illustrative case for a single-well sand body study within water-flooded layers (Figure 8), an exploration of the vertical variations in sand body characteristics has contributed to the delineation of a favorable reservoir distribution within the inter-well and planar context.

The original natural potential curve is represented by SP, while SP-AD stands for the multi-curve frequency fusion reconstruction curve. In the upper figure, the fourth trace denotes the drilling-verified sand bodies. It is noticeable that segment ‘a’ provides a clearer depiction of sand bodies, with a prominent response feature approximately 2 m in thickness at the top of the retrogradational cycle (Figure 8).

In segments ‘c’ and ‘d’, the SP-AD curve clearly exhibits retrogradational sand body features with distinct funnel-shaped characteristics. The lower part of segment ‘d’ reveals a sand body with an approximate thickness of 1 m. In segments ‘e’, ‘f’, ‘g’, and ‘h’ of well C26-X55, as it enters the water-flooded layers, the original SP curve exhibits a noticeable baseline drift, and the well logging response characteristics for the sand bodies are extremely poor. However, the SP-AD curve, created through multi-curve frequency fusion reconstruction, maintains good well logging response characteristics within the water-flooded layers. In segment ‘e’, the sand body is more distinctly represented, while segment ‘f’ exhibits clear retrogradational features. Segments ‘g’ and ‘h’ correspond to the sand bodies with relatively good response characteristics, whereas the original SP curve displays almost no response features.

Taking into account the well logging processing and analysis results mentioned previously, we made use of data from water-flooded layers and well data with well logging response characteristics that are clearly evident in standard non-water-flooded wells, such as C26-X55 and C26-X52 wells. These data were normalized based on sensitive curves and multi-curve frequency fusion reconstruction curves. Through an integrated analysis of well logging and sedimentary cycles, a multi-factor cross-plot method was developed. SP-GYN is the normalized curve for spontaneous potential, COND-GYN is the normalized curve for induction conductivity, AC-GYN is the normalized curve for acoustic time difference, R-GYN is the normalized curve for resistivity, and GR-GYN is the normalized curve for natural gamma. Combining a fundamental understanding of sedimentology and reservoir porosity characteristics, the target layer was categorized into three distinct groups, designated as I, II, and III (refer to Figure 9a–d), each with specific characteristics as follows:

Category I reservoirs correspond to beach-bar crest, where the hydrodynamics are more pronounced, resulting in thorough lake wave scouring. These reservoirs are primarily composed of fine sandstone, offering the best permeability and are conducive to forming negative anomalies in the natural potential curve. Consequently, the natural potential curve proves effective in identifying Category I reservoirs (

Table 3. Results Table of Sedimentary Microfacies Reservoir Response Characteristics.

Category Type		Microfacies	Lithology	Porosity Characteristics	SP (MV)	GR (API)	AC (μs/ft)	R4 (OMM)	CON (MMO)
Reservoir	I	Beach-Bar Crest	Fine Sandstone	Strong hydraulic conditions, good grain sorting, good roundness, well-developed primary porosity, visible secondary dissolution porosity.	20–60	50–105	80–95	2–7	50–450
	II	Side edge of Beach-Bar	Fine Siltstone	Relatively weak hydraulic conditions, good grain sorting, well-developed primary porosity.	50–110	60–120	70–105	1.8–4	50–900
		Sheet sand	Siltstone
Non-Reservoir	III	Shallow shore Lake	Mudstone	Poor physical properties	60–120	90–150	80–120	0.01–2	600–2000

). Category I reservoirs possess favorable oil-bearing characteristics, marked by distinct high resistivity and low induction conductivity features, which are consistent with oil testing results. Category II reservoirs correspond to the side edge of beach-bar and sheet sand sedimentary microfacies. The hydrodynamics in this area are relatively weaker compared to the reservoir top. These reservoirs mainly comprise fine sandstone and siltstone, exhibiting moderate physical properties. The lithological transitions occur rapidly, and the acoustic log spans are relatively wide. While the permeability is slightly lower than that of Category I, it remains a relatively permeable layer. Some intervals within this category still exhibit good oil-bearing characteristics, characterized by slightly negative anomalies in the natural potential curve, low induction conductivity, and high resistivity values, which can reach up to 6 ohm-m. Category III represents non-reservoirs, distinguished by mudstone sediment within the shore-shallow lacustrine beach-bar. These non-reservoirs display low resistivity and high induction conductivity, along with high natural gamma.

Building upon the optimization results of individual wells and a comprehensive understanding of shoreface-barrier thin-layer sand bodies, we established a methodology for identifying sand bodies between wells within the study area. This enabled us to conduct an in-depth examination of the lateral variations in sand bodies and lay the foundation for an analytical framework for the sand bodies within the C26 block. Ultimately, this approach shed light on the spatial distribution characteristics of sand bodies in the study area. To illustrate, we can consider Sand Layer 4 of the lower sub-member of the Red One Sand Layer Group in the lower part of the 4th member of the Shahejie Formation. The sand bodies within this layer exhibit a point-like and patchy distribution pattern (Figure 10), which aligns with the geological understanding of shoreface-barrier sand bodies prevalent in the study area.

5. Conclusions

Addressing the challenges presented by variations in logging curves within water-flooded layers and the complexities associated with identifying thin-layer beach-bar sand bodies and sedimentary cycles, this study employs sedimentological research methods. It conducts comprehensive analyses of core and log facies, systematically investigating the matter at hand. The conclusions derived from this analysis are outlined below:

(1).

This passage delves into sedimentology research and its application within a specific geological area. It indicates that the study area features beach-bar subfacies and underscores the sensitivity of natural potential curves in identifying sand bodies, sedimentary cycles, and water-bearing layers, with natural gamma and resistivity curves exhibiting somewhat lower sensitivity.

(2).

The text introduces a multi-curve frequency fusion reconstruction approach, segmenting the study area into low-frequency components reflecting large-scale sedimentary cycles and high-frequency components reflecting variations in sand bodies. It discusses the use of regression analysis on these components and establishes a method for identifying beach-bar thin-layer sand bodies and sedimentary cycles.

(3).

To address challenges associated with the response characteristics of water-bearing layers in natural potential curves, it introduces a multi-curve frequency fusion reconstruction and prediction analysis method specifically tailored to water-bearing layers. This method effectively tackles the complexities of studying sand bodies affected by water, rendering sand body variations more distinct.

(4).

The text classifies the target layers into three categories (I, II, III) through mercury injection experiments and log curve analysis after water layer treatment. It establishes a multi-factor intersection identification chart and a quantitative identification result table for water-bearing layer treatment, offering theoretical guidance for future oil field development.

Following the extensive investigation outlined earlier, a specialized approach has been formulated to optimize the processing of logging curves, specifically designed for thin-layer sand bodies in water-flooded conditions. Utilizing a multi-curve frequency fusion reconstruction analysis method, this strategy highlights the response characteristics of sand bodies and sedimentary cycles. This sets the foundation for subsequent refinements in oilfield exploration and development.