Fractures interaction and propagation mechanism of multi-cluster fracturing on laminated shale oil reservoir

The continental shale reservoirs of Jurassic Lianggaoshan Formation in Sichuan Basin contain thin lamina,which is characterized by strong plasticity and developed longitudinal shell limestone interlayer.To improve the production efficiency of reservoirs by multi-cluster fracturing,it is necessary to consider the unbalanced propagation of hydraulic fractures and the penetration effect of fractures.This paper constructed a numerical model of multi-fracture propagation and penetration based on the finite element coupling cohesive zone method;considering the construction cluster spacing,pump rate,lamina strength and other parameters studied the influencing factors of multi-cluster fracture interaction propagation;combined with AE energy data and fracture mode reconstruction method,quantitatively characterized the comprehensive impact of the strength of thin interlayer rock interfaces on the initiation and propagation of fractures that penetrate layers,and accurately predicted the propagation pattern of hydraulic fractures through laminated shale oil reservoirs.Simulation results revealed that in the process of multi-cluster fracturing,the proportion of shear damage is low,and mainly occurs in bedding fractures activated by outer fractures.Reducing the cluster spacing enhances the fracture system's penetration ability,though it lowers the activation efficiency of lamina.The high plasticity of the limestone interlayer may impact the vertical propagation distance of the main fracture.Improving the interface strength is beneficial to the reconstruction of the fracture height,but the interface communication effect is limited.Reasonable selection of layers with moderate lamina strength for fracturing stimulation,increasing the pump rate during fracturing and setting the cluster spacing reasonably are beneficial to improve the effect of reservoir stimulation.

Fracture propagation law of temporary plugging and diversion fracturing in shale reservoirs under completion experiments of horizontal well with multi-cluster sand jetting perforation

This study conducted temporary plugging and diversion fracturing(TPDF)experiments using a true triaxial fracturing simulation system within a laboratory setting that replicated a lab-based horizontal well completion with multi-cluster sand jetting perforation.The effects of temporary plugging agent(TPA)particle size,TPA concentration,single-cluster perforation number and cluster number on plugging pressure,multi-fracture diversion pattern and distribution of TPAs were investigated.A combination of TPAs with small particle sizes within the fracture and large particle sizes within the segment is conducive to increasing the plugging pressure and promoting the diversion of multi-fractures.The addition of fibers can quickly achieve ultra-high pressure,but it may lead to longitudinal fractures extending along the wellbore.The temporary plugging peak pressure increases with an increase in the concentration of the TPA,reaching a peak at a certain concentration,and further increases do not significantly improve the temporary plugging peak pressure.The breaking pressure and temporary plugging peak pressure show a decreasing trend with an increase in single-cluster perforation number.A lower number of single-cluster perforations is beneficial for increasing the breaking pressure and temporary plugging peak pressure,and it has a more significant control on the propagation of multi-cluster fractures.A lower number of clusters is not conducive to increasing the total number and complexity of artificial fractures,while a higher number of clusters makes it difficult to achieve effective plugging.The TPAs within the fracture is mainly concentrated in the complex fracture areas,especially at the intersections of fractures.Meanwhile,the TPAs within the segment are primarily distributed near the perforation cluster apertures which initiated complex fractures.

Simulation Method and Feature Analysis of Shutdown Pressure Evolution During Multi-Cluster Fracturing Stimulation

Multistage multi-cluster hydraulic fracturing has enabled the economic exploitation of shale reservoirs,but the interpretation of hydraulic fracture parameters is challenging.The pressure signals after pump shutdown are influenced by hydraulic fractures,which can reflect the geometric features of hydraulic fracture.The shutdown pressure can be used to interpret the hydraulic fracture parameters in a real-time and cost-effective manner.In this paper,a mathematical model for shutdown pressure evolution is developed considering the effects of wellbore friction,perforation friction and fluid loss in fractures.An efficient numerical simulation method is established by using the method of characteristics.Based on this method,the impacts of fracture half-length,fracture height,opened cluster and perforation number,and filtration coefficient on the evolution of shutdown pressure are analyzed.The results indicate that a larger fracture half-length may hasten the decay of shutdown pressure,while a larger fracture height can slow down the decay of shutdown pressure.A smaller number of opened clusters and perforations can significantly increase the perforation friction and decrease the overall level of shutdown pressure.A larger filtration coefficient may accelerate the fluid filtration in the fracture and hasten the drop of the shutdown pressure.The simulation method of shutdown pressure,as well as the analysis results,has important implications for the interpretation of hydraulic fracture parameters.

Polynomial Approach to Optimal One-wafer Cyclic Scheduling of Treelike Hybrid Multi-Cluster Tools via Petri Nets

A treelike hybrid multi-cluster tool is composed of both single-arm and dual-arm cluster tools with a treelike topology. Scheduling such a tool is challenging. For a hybrid treelike multi-cluster tool whose bottleneck individual tool is process-bound, this work aims at finding its optimal one-wafer cyclic schedule. It is modeled with Petri nets such that a onewafer cyclic schedule is parameterized as its robots' waiting time.Based on the model, this work proves the existence of its onewafer cyclic schedule that features with the ease of industrial implementation. Then, computationally efficient algorithms are proposed to find the minimal cycle time and optimal onewafer cyclic schedule. Multi-cluster tool examples are given to illustrate the proposed approach. The use of the found schedules enables industrial multi-cluster tools to operate with their highest productivity.

Optimal Integrated Schedule of Entire Process of Dual-blade Multi-cluster Tools From Start-up to Close-down

Multi-cluster tools are widely used in majority of wafer fabrication processes in semiconductor industry. Smaller lot production, thinner circuit width in wafers, larger wafer size, and maintenance have resulted in a large quantity of their start-up and close-down transient periods. Yet, most of existing efforts have been concentrated on scheduling their steady states.Different from such efforts, this work schedules their transient and steady-state periods subject to wafer residency constraints. It gives the schedulability conditions for the steady-state scheduling of dual-blade robotic multi-cluster tools and a corresponding algorithm for finding an optimal schedule. Based on the robot synchronization conditions, a linear program is proposed to figure out an optimal schedule for a start-up period, which ensures a tool to enter the desired optimal steady state. Another linear program is proposed to find an optimal schedule for a closedown period that evolves from the steady state period. Finally,industrial cases are presented to illustrate how the provided method outperforms the existing approach in terms of system throughput improvement.

Scheduling a Single-Arm Multi-Cluster Tool With a Condition-Based Cleaning Operation

As wafer circuit widths shrink less than 10 nm,stringent quality control is imposed on the wafer fabrication processes. Therefore, wafer residency time constraints and chamber cleaning operations are widely required in chemical vapor deposition, coating processes, etc. They increase scheduling complexity in cluster tools. In this paper, we focus on scheduling single-arm multi-cluster tools with chamber cleaning operations subject to wafer residency time constraints. When a chamber is being cleaned, it can be viewed as processing a virtual wafer. In this way, chamber cleaning operations can be performed while wafer residency time constraints for real wafers are not violated. Based on such a method, we present the necessary and sufficient conditions to analytically check whether a single-arm multi-cluster tool can be scheduled with a chamber cleaning operation and wafer residency time constraints. An algorithm is proposed to adjust the cycle time for a cleaning operation that lasts a long cleaning time.Meanwhile, algorithms for a feasible schedule are also derived.And an algorithm is presented for operating a multi-cluster tool back to a steady state after the cleaning. Illustrative examples are given to show the application and effectiveness of the proposed method.

基于Multi-Agent的无人机集群体系自主作战系统设计

针对无人集群自主作战体系设计中的关键问题,提出基于Multi-Agent的无人集群自主作战系统设计方法。建立无人集群各节点的Agent模型及其推演规则;对于仿真系统模块化和通用化的需求,设计系统互操作式接口和无人集群自主作战的交互关系;开展无人集群系统仿真推演验证。仿真结果表明,所提设计方案不仅能够有效开展并完成自主作战网络生成-集群演化-效能评估的全过程动态演示验证,而且能够通过重复随机试验进一步评估无人集群的协同作战效能,最后总结了集群协同作战的策略和经验。

脉冲控制下基于合作机制的多智能体系统分组一致性研究

针对在脉冲控制下非线性多智能体系统的分组一致性协同控制问题,本文基于复杂系统中智能体间的合作机制,提出了一种新颖的具有领导-跟随网络的分组一致性控制协议.该协议控制智能体仅在脉冲时刻接收来自邻居节点的信息并更新自己的状态,减少智能体因持续接收邻居信息更新自己状态而造成的网络开销.针对不同组之间智能体的信息交互,该协议具有防止组间干扰的机制,降低了不同组之间的通信干扰,提高系统鲁棒性.此外,本文还介绍了一种适用于非线性系统的方案,增强协议的适应性.基于实现分组一致性条件,构建了多智能体系统的动力学模型,并进行了协议的详细设计和理论推导.最后,通过仿真验证了协议的正确性和有效性.

Effect of height-diameter ratio on the mechanical characteristics of shale with different bedding orientations

Geological exploration cores obtained from shale gas wells several kilometers deep often show different height-diameter ratios(H/D)because of complex geological conditions(core disking or developed fractures),which makes further standard specimen preparation for mechanical evaluation of reservoirs difficult.In multi-cluster hydraulic fracturing,shale reservoirs between planes of hydraulic fractures with different lengths could be simplified to have different H/D ratios.Discovering the effect of H/D on the mechanical characteristics of shale specimens with different bedding orientations will support mechanical evaluation tests of reservoirs based on disked geological cores and help to optimize multicluster fracturing programs.In this study,we performed uniaxial compression tests and acoustic emission(AE)monitoring on cylindrical Longmaxi shale specimens under five bedding orientations and four H/D ratios.The experimental results showed that both the H/D-dependent mechanical properties and AE parameters demonstrated significant anisotropy.Increasing H/D did not change the uniaxial compressive strength(UCS)evolution versus bedding orientation,demonstrating a V-shaped relationship,but enhanced the curve shape.The stress level of crack damage for the specimens significantly increased with increasing H/D,excluding the specimens with a bedding orientation of 0°.With increasing H/D,the cumulative AE counts of the specimens with each bedding orientation tended to exhibit a stepped jump against the loading time.The proportion of low-average-frequency AE signals(below 100 kHz)in specimens with bedding orientations of 45°and 60°increased to over 70%by increasing H/D,but it only increased to 40%in specimens with bedding orientations of 0°,30°,and 90°.Finally,an empirical model that can reveal the effect of H/D on anisotropic UCS of shale reservoir was proposed,the anisotropic proportion of tensile and shear failure cracks in specimens under four H/D ratios was classified based on the AE data,and the effect of H/D on the anisotropic crack growth of specimens was discussed.

基于多智能体一致性的多车道车辆集群编队控制

为安全、高效地控制智能网联汽车(CAV)编队,研究了基于多智能体系统(MAS)一致性的多车道、零散随机分布的车辆集群编队控制策略,并进行数值仿真验证。建立了基于交错式结构的车群期望几何拓扑,提出了包括间距调整阶段、变道控制阶段和队形收敛阶段的3阶段式编队控制流程。基于一致性理论,设计了车群编队控制器,并进行稳定性论证。选取2种典型场景,对三车道场景中初始分布较为极端的车群开展编队研究。结果表明:在该策略的控制过程中,纵向车间距与期望间距误差在0.5 m以内,横向位置与期望位置误差在5 cm以内,且速度在变化后能快速收敛至期望值。因而,该策略能有效控制车群车辆安全、高效地达到车群期望队形。

多智能体系统在自适应控制下的聚类-分量一致性

为减少不必要的浪费,研究了多智能体系统在自适应控制下的聚类-分量一致性问题。通过引入相应的自适应控制协议,讨论了相应误差系统零解的渐近稳定性以及零解关于部分变元的稳定性,得出系统实现聚类-分量一致性的充分条件。最后通过实例验证了理论结果的可行性。

基于多智能体系统的分布式仪表集群协同控制机制分析

阐述基于多智能体系统的分布式仪表集群协同控制机制。通过分析智能体概念、多智能体系统特征,设计了系统架构和协同规划算法,并探讨协同反馈控制和故障容错机制。分析结果表明,该机制能够提高系统的鲁棒性和可靠性,实现更灵活、更高效的控制。

中国式现代化背景下安徽经济高质量发展测评及策略研究

文章从创新、协调、绿色、开放、共享发展以及提质增效六个方面选取了32个二级指标,并运用多层次因子分析法对在中国式现代化战略目标下安徽省16个地级市的经济高质量发展水平进行测度,最后使用系统聚类分析方法将测度结果分为5个层次进行深入分析。研究表明:安徽省各地经济高质量发展水平差距较大,合肥市作为安徽省的省会,发展基础较好。在创新发展、开放发展以及提质增效三个方面均排名第一,遥遥领先于其他城市。亳州市、黄山市以及芜湖市在协调发展、绿色发展以及共享发展方面分别排名第一。安徽省经济高质量发展整体上呈现“皖南>皖中>皖北”的不均衡分布格局,安徽省16个地级市各层次之间存在明显的区域内相似性和区域间差异性。

园区供电系统广义负荷需求响应建模与分析

产业园区供能系统采用的是多能耦合、源网荷储协同优化的方式,其电力负荷定义也转变为新型综合性的广义负荷。考虑电价信号对园区主动负荷响应特性的影响,建立了计及响应电价的园区供电系统广义负荷模型,并采用层次聚类方法提取典型负荷模式,研究园区供电系统负荷需求响应特征。为验证模型与方法的有效性,采用中国南方城市某园区供电系统负荷进行测算,模拟了实时电价和分时电价2种市场机制下主动负荷的响应特性。分析了不同主动负荷占比对园区广义负荷特征的影响,并定量讨论了不可控负荷、可转移负荷、可中断负荷对广义负荷需求响应效果的贡献程度。结果表明,适当的电价机制将有序引导园区供电系统广义负荷主动削峰填谷,有助于虚拟电厂建设,实现源网荷灵活互动。

多核集群系统下的混合并行遗传算法研究

为应对传统遗传算法在处理大规模组合优化问题面临的进化速度缓慢,难以达到实时要求的严峻挑战,提出了一种在多核PC集群系统上实现"粗粒度-主从式"混合并行遗传算法的模型:通过把"粗粒度-主从式"并行遗传算法映射到多核PC集群上,结合消息传递和共享存储两种并行编程模型,在节点间使用消息传递模型(MPI),对应的遗传算法为粗粒度并行遗传算法,在节点内使用共享存储模型(OpenMP),对应的遗传算法为主从式并行遗传算法,用MPI和OpenMP混合编程的方式以进程和线程两级并行在多核集群上实现具体的混合并行遗传算法。理论分析和实验结果表明,提出的实现模型有较好的性能,可大大改进传统遗传算法的缺陷。为利用并行遗传算法在普通多核PC集群上处理大规模组合优化问题提出了一种有效、可行的解决方案。

区域分布式电源集群监控系统

为了更好地管理分布式电源的运行,增强与配电网的友好互动能力,提高分布式电源并网运行后配电网的安全性和经济性,研发了一套区域分布式电源集群自动监控系统。提出了区域分布式电源集群监控系统的结构设计和功能体系,设计开发了系统的关键功能(集群划分、多目标优化、集群控制等),建立了分布式电源集群控制的多目标优化模型,并给出基于二阶锥的求解策略。基于实时数字仿真器(RTDS)构建了浙江杭湾工业园区某实际10kV馈线的网架,并以此为信源测试了系统在两种时间尺度下的运行情况,验证了系统的正确性和可靠性。

长三角区域多机场复合系统耦合协调发展度评价研究

大多数机场亏损严重,设施资源闲置浪费现象突出等问题引发了长三角机场建设密度是否过高的争论.基于这种背景,本文通过解析系统协调发展机理,在耦合度、协调度和综合发展指数模型基础上,构建了复合系统耦合协调发展度评价模型,并对长三角多机场复合系统进行了实证分析.结果表明:长三角多机场复合系统中,耦合协调发展度较高的机场占75%,耦合协调发展度与机场等级存在较强关联性;机场规模等级越高,其与经济的耦合协调发展度越好,大型综合枢纽机场普遍高于中小型机场;地区经济发展水平的高低是影响机场区域耦合协调发展程度的重要因素,经济发展水平越好的地区,机场与经济的耦合协调发展度越高.

基于多Agent的能量管理系统支持平台

将多Agent技术应用于构建高效、可靠、灵活的新一代能量管理系统(N-EMS)支持平台。提出了适用于N-EMS的二维联盟多智能体架构,从增强并发和负载均衡这2个方面讨论了提高系统实时性的方法,利用黑板模型实现了Agent之间的通信、协调,将不同功能的Agent组织成完整的电力系统应用。根据该方法构建的N-EMS已经在2个省级电网投入使用,现场运行情况证明了该方法的有效性。

基于多分簇级联融合的差异数据库安全访问

提出一种基于多分簇系统级联融合的安全访问方法,首先对所有的访问数据进行多级分簇,依照数据的访问特性,将不同的数据分配到不同的簇结构,然后对各个簇的特征进行融合处理,提取簇数据的深度特征,从而区分正常访问数据与异常攻击数据。最后采用一组正常数据加异常数据联合访问进行测试,结果显示,采用基于多分簇系统级联融合的安全访问控制,即使是在攻击数据率很高时,系统也能很好的分辨出正常数据与异常攻击,具有很好的应用价值。

集群系统下多弹道实时测量数据处理的UKF算法

随着外弹道测量数据采样频率的提高和多发导弹齐射技术的成熟,普通计算机受限于运算能力和内存已无法满足计算要求。文中采用无迹卡尔曼滤波(UKF)算法,结合质点弹道模型,基于曙光TC4000A集群系统设计了针对多弹道实时测量数据处理的并行计算软件,完成了对高频率采样数据的多弹道并行计算。该算法提高了弹道计算精度和计算效率,对集群系统下的并行计算在现代军事中的应用有较强的借鉴意义。