Performance Evaluation of Multistage Fractured Horizontal Wells in Tight Gas Reservoirs at Block M, Ordos Basin

Block M of the Ordos Basin is a typical low-permeability tight sandstone gas accumulation. To develop these reservoirs, various horizontal well fracturing technologies, such as hydra-jet fracturing, open-hole packer multistage fracturing, and perf-and-plug multistage fracturing, have been implemented in practice, showing greatly varying performance. In this paper, six fracturing technologies adopted in Block M are reviewed in terms of principle, applicability, advantages, and disadvantages, and their field application effects are compared from the technical and economic perspectives. Furthermore, the main factors affecting the productivity of fractured horizontal wells are determined using the entropy method, the causes for the difference in application effects of the fracturing technologies are analyzed, and a comprehensive productivity impact index (CPII) in good correlation with the single-well production of fractured horizontal wells is constructed. This article provides a simple and applicable method for predicting the performance of multi-frac horizontal wells that takes multiple factors into account. The results can be used to select completion methods and optimize fracturing parameters in similar reservoirs.

Controlled Frac Height Growth Technique to Avoid Contacting Water-Bearing Layer

Many techniques have been used to model, diagnose and detect fracture dimension and propagation during hydraulic fracturing. Diagnosing fracture dimension growth vs time is of paramount importance to reach the desired geometry to maximize hydrocarbon production potential and prevent contacting undesired fluid zones. The study presented here describes a technique implemented to control vertical fracture growth in a tight sandstone formation being stimulated near a water zone. This gas well was completed vertically as openhole with Multi- Stage Fracturing (MSF). Pre-Fracturing diagnostic tests in combination with high-resolution temperature logs provided evidence of vertical fracture height growth downward toward water zone. Pre-fracturing flowback indicated water presence that was confirmed by lab test. Several actions were taken to mitigate fracture vertical growth during the placement of main treatment. An artificial barrier with proppant was placed in the lower zone of the reservoir before main fracturing execution. The rate and viscosity of fracturing fluids were also adjusted to control the net pressure aiming to enhance fracture length into the reservoir. The redesigned proppant fracturing job was placed into the formation as planned. Production results showed the effectiveness of the artificial lower barrier placed to prevent fracture vertical growth down into the water zone. Noise log consists of Sonic Noise Log (SNL) and High Precision Temperature (HPT) was performed. The log analysis indicated that two major fractures were initiated away from water-bearing zone with minimum water production. Additionally, in- situ minimum stress profile indicated no enough contrast between layers to help confine fracture into the targeted reservoir. Commercial gas production was achieved after applying this stimulation technique while keeping water production rate controlled within the desired range. The approach described in this paper to optimize gas production in tight formation with nearby water contact during hydraulic fracturing treatments has been applied with a significant improvement in well production. This will serve as reference for future intervention under same challenging completion conditions.

Diagenesis Evolution and Pore Types in Tight Sandstone of Shanxi Formation Reservoir in Hangjinqi Area, Ordos Basin, Northern China

Diagenesis and pore evolution of tight sandstone reservoir is one of the most important issues surrounding clastic reservoirs. The tight sandstone of the Shanxi Formation is an important oil and gas producing layer of the Upper Paleozoic in Ordos Basin, and its densification process has an important impact on reservoir quality. This study determined the physical properties and diagenetic evolution of Shanxi Formation sandstones and quantitatively calculated the pore loss in the diagenetic process. Microscopic identification, cathodoluminescence, and a scanning electron microscope were used identify diagenesis, and the diagenesis evolution process was clarified along with inclusion analysis. In addition, reservoir quality was determined based on the identification of pore types and physical porosity. Results show that rock types are mainly sublitharenite and litharenite. The reservoir has numerous secondary pores after experiencing compaction, cementation, and dissolution. We obtained insight into the relationship between homogenous temperature and two hydrocarbon charges. The results indicated that there were two hydrocarbon charges in the Late Triassic–Early Jurassic (70–90 °C) and Middle Jurassic–Early Cretaceous (110–130 °C) before reservoir densification. The quantitative calculation of pore loss shows that the average apparent compaction, cementation, and dissolution rates are 67.36%, 22.24%, and 80.76%, respectively. Compaction directly affected the reservoir tightness, and intense dissolution was beneficial to improve the physical properties of the reservoir.

Permeability Predictions for Tight Sandstone Reservoir Using Explainable Machine Learning and Particle Swarm Optimization

High-precision permeability prediction is of great significance to tight sandstone reservoirs. However, while considerable progress has recently been made in the machine learning based prediction of reservoir permeability, the generalization of this approach is limited by weak interpretability. Hence, an interpretable XGBoost model is proposed herein based on particle swarm optimization to predict the permeability of tight sandstone reservoirs with higher accuracy and robust interpretability. The porosity and permeability of 202 core plugs and 6 logging curves (namely, the gamma-ray (GR) curve, the acoustic curve (AC), the spontaneous potential (SP) curve, the caliper (CAL) curve, the deep lateral resistivity (RILD) curve, and eight lateral resistivity (RFOC) curve) are extracted along with three derived variables (i.e., the shale content, the AC slope, and the GR slope) as data sets. Based on the data preprocessing, global and local interpretations are performed according to the Shapley additive explanations (SHAP) analysis, and the redundant features in the data set are screened to identify the porosity, AC, CAL, and GR slope as the four most important features. The particle swarm optimization algorithm is then used to optimize the hyperparameters of the XGBoost model. The prediction results of the PSO-XGBoost model indicate a superior performance compared with that of the benchmark XGBoost model. In addition, the reliable application of the interpretable PSO-XGBoost model in the prediction of tight sandstone reservoir permeability is examined by comparing the results with those of two traditional mathematical regression models, five machine learning models, and three deep learning models. Thus, the interpretable PSO-XGBoost model is shown to have more advantages in permeability prediction along with the lowest root mean square error, thereby confirming the effectiveness and practicability of this method.