Shear wave velocity is one of the important parameters reflecting the lithological and physical properties of reservoirs, and it is widely used in the fields of lithology and fluid property identification, reservoir evaluation, seismic data processing, and interpretation. However, due to the high cost and challenge of obtaining shear wave velocity, only a few key wells are measured. Considering the intricate nonlinear mapping relationship between shear wave velocity and conventional logging data, an integrated network incorporating an attention mechanism, a convolutional neural network, and a bidirectional gated recurrent unit (STACBiN) is proposed for predicting shear wave velocity. The impact of conventional logging data on shear wave velocity is analyzed, thus employing the attention mechanism to focus on data correlated with shear wave velocity, which can enable the prediction results of the method proposed superior to those of conventional methods. Additionally, the prediction results of this method are compared with the prediction results of the two-dimensional convolutional neural network (2DCNN) and bidirectional gated recurrent unit (BiGRU). It is verified that the network proposed can effectively predict the shear wave velocity, with minimal error between predicted and true values.

With the deepening of exploration and development, many low-resistivity reservoirs have been found in the XP area. We have found that the genesis of these low-resistivity reservoirs is that they contain a lot of very fine sand, which leads to the high content of bound water in the reservoirs and reduces resistivity. Compared with normal oil reservoirs, these reservoirs show low resistivity and high natural gamma, which makes it difficult to identify reservoirs qualitatively and calculate parameters quantitatively. Using conventional logging interpretation methods, the wrong shale content and porosity will be obtained, and such reservoirs may even be judged as mudstones. To solve this problem, we extract a characteristic parameter from the T2 spectrum of nuclear magnetic resonance (NMR) logging, which can effectively identify such reservoirs. In addition, we have innovatively adopted an optimized logging interpretation method for integrated NMR logging and conventional logging, which has effectively improved the parameter calculation accuracy of this type of reservoir and achieved a good application effect in the XP area.

The accurate identification of [ ] nonhydrocarbon gas layers plays an important role in the exploration and development of natural gas reservoirs. The logging response characteristics of layers containing nonhydrocarbon gas and hydrocarbon gas are very similar because there are gas fluids in both kinds of layers, which leads to the uncertainty in the interpretation results of gas types. We have developed a comprehensive method for the identification of [ ] nonhydrocarbon gas layers by using [ ]-relative content and apparent porosity. First, based on the formation component bulk volume model of the coexisting reservoir of the [ ] nonhydrocarbon gas layers and the hydrocarbon gas layers, we develop an optimization algorithm to quantitatively calculate the [ ] relative content in the reservoir. Second, by analyzing and determining the logging response value of the [ ] nonhydrocarbon gas under the formation temperature and pressure conditions, we establish a qualitative identification chart using the difference between the calculated value of the apparent porosity of [ ] and the hydrocarbon gas. Finally, we were able to accurately identify the [ ] nonhydrocarbon gas layers by comprehensively analyzing the calculation results of the [ ] relative content in the reservoir and the identification results of the [ ] apparent porosity identification chart. We apply the above workflow in the Baiyun deep-water area in the eastern South China Sea, and it shows high efficiency in the identification of [ ] nonhydrocarbon gas layers.

Recently, theoretical modeling based on rock physics has emerged as a pivotal approach to studying the resistivity of complex fractured–cavernous microstructures. In this work, to study the effects of fractured–cavernous structures on carbonate reservoir resistivity, electrical conductivity models were developed based on the effective medium theory and Ohm’s Law, and theoretical simulations were performed to examine how the porosity and resistivity of the rock matrix, the formation water resistivity, and the parameters of the fractured–cavernous microstructure affect the resistivity of rocks saturated with petroleum or water. Furthermore, the modeling results revealed the specific relationships between these factors in petroleum-saturated and water-saturated rocks. For vuggy reservoirs, a significant negative correlation between throat diameter and resistivity was revealed when variations in the rock matrix and formation water resistivity were negligible. Furthermore, the pore shape—especially the extension of pores in the direction of the current—severely reduced the resistivity of petroleum-saturated rocks. For fractured reservoirs, the porosity and resistivity of the rock matrix were the primary factors affecting resistivity, with the fracture inclination angle and width also exhibiting pronounced effects on the resistivity of water-saturated rocks. The rock cementation exponent was much smaller when the matrix pores were interconnected through fractures than when they were interconnected through throats. The findings reveal that the effects of the structural parameters of fractured–cavernous carbonate reservoirs on reservoir resistivity differ between petroleum-saturated and water-saturated rocks. The conventional Archie’s equation is insufficient for evaluating fluid saturation in carbonate reservoirs. A saturation evaluation model with a variable rock cementation exponent tailored to the specific reservoir type should thus be developed.

Deep and ultra-deep reservoirs are characterized by low porosity and permeability, pronounced heterogeneity, and complex pore structures, complicating permeability evaluations. Permeability, directly influencing the fluid production capacity of reservoirs, is a key parameter in comprehensive reservoir assessments. In the X Depression, low-porosity and low-permeability formations present highly discrete and variable core data points for porosity and permeability, rendering single-variable regression models ineffective. Consequently, accurately representing permeability in heterogeneous reservoirs proves challenging. In the following study, lithological and physical property data are integrated with mercury injection data to analyze pore structure types. The formation flow zone index (FZI) is utilized to differentiate reservoir types, and permeability is calculated based on core porosity–permeability relationships from logging data for each flow unit. Subsequently, the average permeability for each flow unit is computed according to reservoir classification, followed by a weighted average according to effective thickness. This approach transforms logging permeability into drill stem test permeability. Unlike traditional point-by-point averaging methods, this approach incorporates reservoir thickness and heterogeneity, making it more suitable for complex reservoir environments and resulting in more reasonable conversion outcomes.

Building a 3D porosity model for carbonate reservoirs is crucial for the planning of hydrocarbon development. However, properly estimating the porosity of carbonate reservoirs using wireline logs is still a challenge task due to the complexity of vugs. We have treated the porosity estimate of carbonate as an optimization problem and propose to estimate the porosity of carbonate reservoir by integrating several kinds of petrophysics data. We begin with establishing a series of linear equations between the porosity and wireline logs. The objective is computing a porosity value that simultaneously minimizes the defined linear equations. The multifunction minimization can successfully improve the robustness of the proposed method to noise and the "small" wireline log values due to the complex dissolution vug reservoirs. We build a 3D porosity model by applying the cooperative kriging stochastic simulation algorithm to the porosity estimated at well bore locations and inverted P impedance from seismic data. We examine the proposed workflow by applying it to the Ordovician carbonate dissolution vug reservoir in the L area of the Tahe oilfield. The statistical analysis shows that the built 3D model has a high correlation with the dynamic production data of the reservoir.