With the rapid development of shale gas in northeastern Sichuan province, problems such as low drilling rate, long drilling cycle and serious bit wear are becoming more and more prominent in shale gas drilling process. This paper proposes new prediction theory with two steps based on logging data sensitive to lithology by introducing Bayesian classifier and Self Adaptive Double Chain Quantum Genetic Algorithm (S-DC-QGA) which could optimize BP neural network prediction model. In the first step, the Bayesian network classifier achieved the classification and recognition of lithology, improved the correlation of model sample data, and enhanced the accuracy and speed of lithology classification. In the second step, S-DC-QGA characterized as the global optimization is used to optimize the number of hidden layer nodes, weights and thresholds of the BP neural network. Results showed that Bayesian lithologies classification enhanced the correlation of samples, moreover optimized BP neural network solved overfitting, random initial weights of BP itself and oscillation of fitting and generalization ability with slight change of network parameters, improving the prediction accuracy of rock drillability parameters and the generalization ability of the model. The findings of this study can help for better drilling design and improvement of mechanical drilling speed.

In order better to evaluate the brittleness of tight reservoir, some studies were conducted to build the new prediction method of brittleness index. First, mechanical properties of tight formation are obtained by triaxial mechanical tests. Subsequently, the correlation analysis between brittleness index and logging parameters was conducted to determine the potential sensitive parameters to brittleness index. Then, based on grey correlation, the initial sequence of sensitive parameters was determined and normalized. On this basis, based on the analytic hierarchy process, the two-two comparison judgment matrix of sensitive parameters was built, the weight vector was determined. Finally, the relationship model between the brittle index and sensitive parameter was established, which meant that a new prediction model of brittleness index was established. Compared with experiment measurement and elastic parameters methods, new method could also have high precision, and reflect the brittleness of reservoir, moreover not depending on the shear wave logging. Therefore, new method provides a new idea for predicting brittleness index and guidance for reservoir fracturing.

Rock typing is an extremely critical step in the estimation of carbonate reservoir quality and reserves in the Middle East. In order to recognize the rock types of carbonate reservoirs in the Mishrif Formation better, classify the reservoirs accurately, and establish the permeability model in line with the study area precisely, it is necessary to study the recognition method conforming to the actual situation of the study area. The practice shows that the current recognition methods based on capillary pressure curve, flow unit and NMR logging data can effectively distinguish rock types, but a large number of accurate experimental data are required, which can only be applied in a few cored well, however, cannot be applied in the whole oil field. In this study, based on core, thin section, logging data, the sedimentary characteristics of carbonate reservoir, logging response of four rock types as well as porosity and permeability characteristics of Mishrif Formation in W are comprehensively studied. Based on Bayesian stepwise discriminant theory in multivariate statistics, the Bayesian discrimination model based on conventional logging data is established. The examining results showed that, compared with the description of logging and coring, the accuracy of Bayesian discriminant model and cross confirmation rate have achieved more than 80% for the original sample. Reliability verification showed that the matching degree of the rock type recognized in the non-cored well with the core and mud logging was as high as 90%, which matched the depositional environment of the entire region. The study results confirm the validity and generalizability of the Bayesian method to identify and predict rock types, which can be applied to the entire Middle East region to solve the problem of the lack of core data to accurately evaluate the quality of non-cored wells and accurately predict production, meeting the needs of actual reservoir evaluation and production development in the Middle East.

Mud shale, used in drilling engineering, is prone to hydration and expansion, resulting in creep deformation that leads to wellbore shrinkage and pipe sticking incidents. Studying the creep characteristics of mud shale is significant for designing a reasonable well structure and determining the lower limit of drilling fluid density. The influence of moisture content on rock strength and creep mechanical properties were studied using water absorption, uniaxial compression, and creep tests. Test results show that with an increase in the moisture content, the mud shale was damaged and softened; moreover, the elastic modulus decreased with increase in moisture content. Under the same load level, the instantaneous strain increased with increasing moisture content. Under different loading stresses, the creep of the rock had nonlinear characteristics, which could be divided into three different creep stages: attenuation, second, and accelerated creep. A new improved creep model based on the Nishihara model was established to describe the accelerated creep characteristics of mud shale under different moisture contents. The ageing degradation and water-bearing weakening effects were introduced. The Levenberg–Marquardt nonlinear least-squares method was applied to invert the creep parameters. The results show that the simulated creep curves, generated using the new creep model, conform to the experimental ones. The relationship between the drilling fluid density and wellbore shrinkage ratio can be defined using this model; it provides a reference for reasonably determining the drilling fluid density.

The exploration and development of the tight reservoirs in the Lucaogou Formation of the Junggar Basin have been poorly stable due to poorly understanding of rock mechanical properties and the lack of a reasonable fracability evaluation plan. Firstly, the impacts of confining pressure on the rock mechanical parameters such as compressive strength, Young’s modulus, Poisson’s ratio, and tensile strength were analyzed, through the trial mechanical test under multilevel confining pressure, the whole mineral analysis with X-ray diffraction (XRD), and Brazilian test on hybrid sedimentary rocks. Subsequently, the relation between fracability and brittleness, fracture toughness, difference coefficient of horizontal principal stress and compressive strength of the rock were systematically studied. Then, quantitative evaluation model of fracability was established via the analytic hierarchy process. The results indicated that the compressive strength of rock increased, the elastic modulus increased, Poisson’s ratio decreased, along with the increase of confining pressure. Additionally, the difference of horizontal principal stress determined the ductility of fractures, and the horizontal principal stress difference coefficient determined the complexity of fractures. Finally, test results illustrated that productivity index was positively correlated with fracability index. The micro-seismic monitoring results further confirmed that the fracability evaluation model established had good applicability to the tight hybrid sedimentary reservoir of Lucaogou Formation. This study could provide important reference and theoretical basis for and completion practice of deep tight oil and gas evaluation and drilling \({\sigma _c}\) \({\sigma _V}\) \({\sigma _h}\) \({\sigma _H}\).