HZHaijie Zhang

Haijie Zhang

*This is an algorithmically generated author record based on publicly available information.

Haijie Zhang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Interpretable Machine Learning for Shale Gas Productivity Prediction: Western Chongqing Block Case Study

    by Haijie Zhang, Ye Zhao, Yaqi Li, Chaoya Sun, Weiming Chen, Dongxu Zhang

    The strong heterogeneity in and complex engineering conditions of deep shale gas reservoirs make productivity prediction challenging, especially in nascent blocks where data is scarce. This scarcity constitutes a critical research gap for the application of data-driven methods. To bridge this gap, we develop an interpretable framework by combining grey relational analysis (GRA) with three machine learning algorithms: Random Forest (RF), Support Vector Machine (SVR), and eXtreme Gradient Boosting (XGBoost). Utilizing small-sample data from 87 shale gas wells in the study area, eight key controlling factors were identified, namely, total fracturing fluid volume, proppant intensity, average tubing head pressure, pipeline transfer pressure, casing head pressure, ceramic proppant fraction, fluid placement intensity, and flowback recovery ratio. These factors were used to train, optimize, and validate a productivity prediction model tailored for deep shale gas horizontal wells. The results demonstrate that XGBoost delivers the highest predictive accuracy and generalization capability, achieving an R2 of 0.907 for productivity prediction—surpassing RF and SVR by 12.11% and 131.38%, respectively. Integrating SHapley Additive exPlanations (SHAP) interpretability analysis further enabled immediate post-fracturing productivity assessment and engineering parameter optimization. This research provides a reliable, data-driven strategy for predicting productivity and optimizing operations within the studied block, offering a valuable template for development in geologically similar areas.


  • Open Access

    Enrichment Mechanism and Development Technology of Deep Marine Shale Gas near Denudation Area, SW CHINA: Insights from Petrology, Mineralogy and Seismic Interpretation

    by Haijie Zhang, Ziyi Shi, Lin Jiang, Weiming Chen, Tongtong Luo, Lin Qi

    As an important target for deep marine shale gas exploration, shale reservoirs near denudation areas have enormous resource potential. Based on the impression method, the sedimentary paleogeomorphology near the denudation area is identified as three units: the first terrace, the second terrace, and the third terrace. At the second terrace where Well Z212 is located, the thickness of the Longmaxi Formation first section is only 0.8 m, and the continuous thickness of the target interval is only 4.3 m, making it a typical thin shale reservoir. By integrating petrology, mineralogy and the seismic method, the thin shale reservoir is characterized. Compared to shale reservoirs far away from the denudation area, the Well Z212 (near denudation area) production interval (Wufeng Formation first section) has high porosity (6%–10%), moderate TOC (3%–4%), a high carbonate mineral content (10%–35%), and a high gas content (>7 m3/t). The correlation between the total porosity of shale and the density of high-frequency laminations is the strongest, indicating that the silt laminations have a positive effect on pore preservation. There is a significant positive correlation between carbonate content and the volume of mesopores and macropores, as well as the porosity of inorganic pores. It is suggested that carbonate minerals are the main carrier of inorganic pores in Well Z212, and the pores are mainly composed of mesopores and macropores. Under the condition of being far away from the fault zone, even near the denudation area, it has good shale gas preservation characteristics. The key development technologies consist of integrated geo-steering technology, acidification, and volume fracking technology. Based on geological characteristics, the fracturing process optimization of Well Z212 has achieved shale reservoir stimulation.


  • Open Access

    Gas Production Prediction Model of Volcanic Reservoir Based on Data-Driven Method

    by Haijie Zhang, Junwei Pu, Li Zhang, Hengjian Deng, Jihao Yu, Yingming Xie, Xiaochang Tong, Xiangjie Man, Zhonghua Liu

    Based on on-site construction experience, considering the time-varying characteristics of gas well quantity, production time, effective reservoir thickness, controlled reserves, reserve abundance, formation pressure, and the energy storage coefficient, a data-driven method was used to establish a natural gas production prediction model based on differential simulation theory. The calculation results showed that the average error between the actual production and predicted production was 12.49%, and the model determination coefficient was 0.99, indicating that the model can effectively predict natural gas production. Additionally, we observed that the influence of factors such as reserve abundance, the number of wells in operation, controlled reserves, the previous year’s gas production, formation pressure, the energy storage coefficient, effective matrix thickness, and annual production time on the annual gas production increases progressively as the F-values decrease. These insights are pivotal to a more profound understanding of gas production dynamics in volcanic reservoirs and are instrumental in optimizing stimulation treatments and enhancing resource recovery in such reservoirs and other unconventional hydrocarbon formations.


  • Determination of Total Organic Carbon Content in Shale Formations With Regression Analysis

    by Jianguo Wang, Daihong Gu, Wei Guo, Haijie Zhang, Daoyong Yang

    By correcting both the positive and negative ΔlogR separation resulting from the resistivity in organic-deficient shales, the traditional ΔlogR correlation is modified, validated, and applied to determine the total organic carbon (TOC) content in shale formations. The TOC content is determined once the Fisher distribution, which represents the significance of each model, and Student's t-distribution, which denotes the significance of every variable in the models, have achieved values equal to or higher than their respective threshold values at a confidence level of 95%. Using a total of 45 sets of logging measurements, the newly proposed correlation is found to be able to reproduce the measured TOC values with a root mean-squared absolute difference (RMSAD) of 0.30 wt % and root mean-squared relative difference (RMSRD) of 23.8%, respectively. Uranium concentration, apart from interval transit time and resistivity, is found to be key in determining the TOC content in organic-rich shale without other radioactive minerals. By combining the reading of DGR (i.e., the difference between the spectral gamma ray with the radioactivity and the computed gamma ray without uranium), the traditional ΔlogR technique has now been improved and extended to the negative ΔlogR separation resulting from the resistivity in organic-deficient shale higher than that in organic-rich shale.