Search

In response to the unclear understanding of fracture propagation and intersection interference in zipper fracturing under the factory development model of deep shale gas wells, a coupled hydro-mechanical model for zipper fracturing considering the influence of natural fracture zones was established based on the finite element – discrete element method. The reliability of the model was verified using experimental data and field monitoring pressure increase data. Taking the deep shale gas reservoir in southern Sichuan as an example, the propagation and interference laws of fracturing fractures under the influence of natural fracture zones with different characteristics were studied. The results show that the large approaching angle fracture zone has a blocking effect on the forward propagation of fracturing fractures and the intersection of inter well fractures. During pump shutdown, hydraulic fractures exhibit continued expansion behavior under net pressure driving. Under high stress difference, as the approaching angle of the fracture zone increases, the response well pressure increase and the total length of the fractured fracture show a trend of first decreasing and then increasing, and first increasing and then decreasing, respectively. Compared to small approach angle fracture zones, natural fracture zones with large approach angles require longer time and have greater difficulty to intersect. The width of fractures and the length of natural fractures are negatively and positively correlated with the response well pressure increase, respectively, and positively and negatively correlated with the time required for intersection, the total length of hydraulic fractures, and fracturing efficiency, respectively. As the displacement distance of the well increases, the probability of fracture intersection decreases, but the regularity between displacement distance and the response well pressure increase and the total length of fractures is not obvious.

Based on the elastic theory of porous media, embedded discrete fracture model and finite volume method, and considering the micro-seepage mechanism of shale gas, a fully coupled seepage-geomechanical model suitable for fractured shale gas reservoirs is established, the optimization method of refracturing timing is proposed, and the influencing factors of refracturing timing are analyzed based on the data from shale gas well in Fuling of Sichuan Basin. The results show that due to the depletion of formation pressure, the percentage of the maximum horizontal principal stress reversal area in the total area increases and then decreases with time. The closer the area is to the hydraulic fracture, the shorter the time for the peak of the stress reversal area percentage curve to appear, and the shorter the time for the final zero return (to the initial state). The optimum time of refracturing is affected by matrix permeability, initial stress difference and natural fracture approach angle. The larger the matrix permeability and initial stress difference is, the shorter the time for stress reversal area percentage curve to reach peak and return to the initial state, and the earlier the time to take refracturing measures. The larger the natural fracture approach angle is, the more difficult it is for stress reversal to occur near the fracture, and the earlier the optimum refracturing time is. The more likely the stress reversal occurs at the far end of the artificial fracture, the later the optimal time of refracturing is. Reservoirs with low matrix permeability have a rapid decrease in single well productivity. To ensure economic efficiency, measures such as shut-in or gas injection can be taken to restore the stress, and refracturing can be implemented in advance.

A multi-process (fracturing, shut-in and production) multi-phase flow model was derived considering the osmotic pressure, membrane effect, elastic energy and capillary force, to determine the optimal shut-in time after multi-cluster staged hydraulic fracturing in shale reservoirs for the maximum production. The accuracy of the model was verified by using production data and commercial software. Based on this model and method, a physical model was made based on the inversion of fracture parameters from fracturing pressure data, to simulate the dynamic changes of pore pressure and oil saturation during fracturing, soaking and production, examine effects of 7 factors on the optimal shut-in time, and find out the main factors affecting the optimal shut-in time through orthogonal experiments. With the increase of shut-in time, the increment of cumulative production increases rapidly first and then tended to a stable value, and the shut-in time corresponding to the inflection point of the change was the optimal shut-in time. The optimal shut-in time has a nonlinear negative correlation with matrix permeability, porosity, capillary pressure multiple and fracture length, a nonlinear positive correlation with the membrane efficiency and total volume of injected fluid, and a nearly linear positive correlation with displacement. The seven factors in descending order of influence degree on optimal shut-in time are total volume of injected fluid, capillary force multiple, matrix permeability, porosity, membrane efficiency, salinity of fracturing fluid, fracturing fluid displacement.

【Objective】The recognition of tea buds is one of the core technologies to realize automatic tea picking.The growth posture of tea buds and the shooting angle during collecting images will bring difficulties to the recognition of tea buds, resulting in the problem of low recognition accuracy. The study is conducted to improve the difficulty in tea bud recognition and improve the recognition accuracy of the model.【Method】The study presents an improved YOLOX tea bud detection algorithm SS-YOLOX, which can accurately identify and classify tea buds including one bud with one leaf and one bud with two leaves. In this method, the feature extraction ability of the model is improved by adding the attention module(SE), the problem of missing detection of small targets is improved, the soft NMS algorithm is introduced to improve the scoring mechanism when the detection frame overlap is high, and the ability of the model to recognize the buds in different scenes is improved.【Result】The ablation test of YOLOX model shows that the detection accuracy of the model can be improved by introducing soft NMS algorithm and SE module, but the improvement result is more obvious by introducing SE module. The feasibility and accuracy of the algorithm are verified by comparing different bud images. The experimental results show that the average accuracy mAP of SS-YOLOX model is 2.2% higher than that of the original YOLOX model, reaching 86.3%, indicating that the recognition ability of the model is improved after the improvement. When the number of target buds is large, SSYOLOX model can effectively reduce the missed detection rate and false detection rate.【Conclusion】Therefore, SS-YOLOX model can accurately identify tea buds, and the recognition effect is better, which can provide a technical basis for intelligent tea picking.