考虑地质分层约束的长短期记忆循环神经网络测井曲线重构

延长油田东部裸眼井区早期测井资料普遍只有自然电位（SP）、自然伽马（GR）及梯度电阻率(R2.5)三条曲线,因缺失声波（AC）、地层电阻率（RT）等测井曲线,难以满足精细油藏地质研究需求。东部裸眼井区开发时间长、单井产量低,重新测井缺乏可行性及经济性。采用长短期记忆循环神经网络(Long Short-Term Memory,LSTM)进行缺失测井曲线重构是一种经济有效方法,适用于地层测井序列数据。然而延长油田东部浅层油藏上覆黄土层段测井数据信号干扰大,直接应用模型精度较差。针对此问题,本文采用考虑地质分层约束的LSTM模型进行缺失测井曲线的重构,通过分层数据截取每口井长6层段测井数据作为样本数据,既保留了LSTM模型处理序列数据的优势,同时又避免了上覆黄土层测井数据对模型的干扰。利用裸眼井区完整测井数据进行模型训练优化和验证,讨论了考虑地质分层约束的LSTM测井曲线重构精度,结果表明通过引入地质分层约束,模型重构测井曲线精度更高。应用优化后模型实现裸眼井区50口仅有GR、SP、R2.5三条曲线数据井的AC、RT曲线重构,对50口井的142个射孔段进行二次解释,对比试油解释结论符合率达到89.4%,验证了该方法对测井曲线重构的实用性和有效性。     

静动态压缩下千枚岩长径比效应影响研究

为了研究长径比效应对层状千枚岩力学特性、能量耗散及破坏模式的影响,文中选用4种倾角（α=0°、30°、60°、90°）下不同长径比（L/D=0.5、0.6、0.8、1.0、1.2、1.6、2.0）的千枚岩分别进行了静载单轴压缩和分离式霍普金森杆（SHPB）试验。结果发现,静载压缩试验条件下,不同倾角下千枚岩随长径比的增大,峰值强度和峰值应变均减小。通过单轴动态压缩试验,发现4种层理倾角千枚岩动态抗压强度与试样长径比呈二次函数关系,随着长径比的增加,动态抗压强度出现一个峰值后逐渐降低;千枚岩峰值应变与试样长径比呈指数函数关系下降;对动态冲击压缩试验进行能量分析,发现不同工况的千枚岩在同一冲击气压下,入射能、反射能、透射能均呈现出先缓慢上升再快速上升最后趋于稳定的三段式变化;随着试样长径比增大,千枚岩反射能比先增大后减小,透射能比先减小后增大;采用能量比值法进行对比分析,发现在长径比L/D=1.2时,千枚岩的反射能比达到最大,透射能比达到最小;对千枚岩的宏观破坏模式进行分析,发现动态冲击压缩下千枚岩的宏观破坏模式受长径比影响较大,长径比越小破坏越完全;长径比越大,破坏越不充分。     

Yolov7-pose识别运动员姿态时关键点异常的消除方法

本研究首先构建了深度聚合网络（deep layer aggregation,DLA）,用于替换Yolov7-pose算法中原有的特征融合网络,增强了特征的表达;其次,将高效聚合网络（efficient layer aggregation Net, ELAN）和尺度感知注意力网络（scale-aware attention Net, SAAN）融合在一起,引入到Yolov7-pose算法中;然后,引入尺度因子到Yolov7-pose算法的损失函数中,并通过实验确定尺度因子最佳值。实验结果表明,经过本法改进的方法识别篮球比赛视频流中的多名运动员姿态时,不再出现关键点异常现象,姿态识别的精度达到95%以上,运行效率达到24 FPS以上,综合性能显著优于原始Yolov7-pose算法等算法。     

COMPARISON OF TWO METHODS OF OPERATING THEATRE PLANNING： APPLICATION IN BELGIAN HOSPITAL

Operating Theatre is the centre of the hospital management＇s efforts. It constitutes the most expensive sector with more than 10% of the intended operating budget of the hospital. To reduce the costs while maintaining a good quality of care, one of the solutions is to improve the existent planning and scheduling methods by improving the services and surgical specialty coordination or finding the best estimation of surgical case durations. The other solution is to construct an effective surgical case plan and schedule. The operating theatre planning and scheduling is the two important steps, which aim to make a surgical case programming with an objective of obtaining a realizable and efficient surgical case schedule. This paper focuses on the first step, the operating theatre planning problem. Two planning methods are introduced and compared. Real data of a Belgian university hospital ＂Tivoli＂ are used for the experiments.     

Cluster Synchronization in a Complex Dynamical Network with Two Nonidentical Clusters

This paper further investigates cluster synchronization in a complex dynamical network with two-cluster. Each cluster contains a number of identical dynamical systems, however, the subsystems composing the two clusters can be different, i.e., the individual dynamical system in one cluster can differ from that in the other cluster. Complete synchronization within each cluster is possible only if each node from one cluster receives the same input from nodes in other cluster. In this case, the stability condition of one-cluster synchronization is known to contain two terms： the first accounts for the contribution of the inner-cluster coupling structure while the second is simply an extra linear term, which can be deduced by the ＂same-input＂ condition. Applying the connection graph stability method, the authors obtain an upper bound of input strength for one cluster if the first account is known, by which the synchronizability of cluster can be scaled. For different clusters, there are different upper bound of input strength by virtue of different dynamics and the corresponding cluster structure. Moreover, two illustrative examples are presented and the numerical simulations coincide with the theoretical analysis.     

多智能体并行仿真框架

多智能体仿真具有广泛的应用前景。但是,现有的多智能体仿真和并行仿真无法解决数量庞大且具有复杂动力学特性的智能体仿真,因此研究这类多智能体的有效仿真具有重要的意义。为了能够有效仿真这类系统,设计了一个两层的多智能体并行仿真框架,指出了该框架中面临的挑战。除此之外,针对仿真中提高仿真效率的挑战,作者提出了一种加快仿真速度的方法。实验结果表明该方法可以明显提高仿真效率。     

基于接收方信用量调整的TCP新算法

传输控制协议(transport control protocol, TCP)发送窗口的大小由拥塞窗口和接收方确认窗口中较小的值来确定。在高速网络中,通过合理地增加拥塞窗口,可以提高TCP的性能。然而,一个相对较小的接收方确认窗口将限制发送窗口的增加,如果根据TCP接收方的数据变化率恰当地增加接收方确认窗口赋予发送方的信用量,就能减少这种限制。提出一种增强的TCP算法,命名为接收方信用量调整传输控制协议（receiver credit adjustment transport control protocol, RCA-TCP）,该算法根据TCP接收方缓冲区的动态数据变化率和TCP连接的最小往返时间之间的乘积的估计值来获得信用量增量。在ns-2(network simulator, version 2)下仿真表明,该算法能获得更好的性能。     

Coordinating production and recycling decisions with stochastic demand and return

In this paper, the joint production and recycling problem is investigated for a hybrid manufacturing and remanufacturing system where brand-new products are produced in the manufacturing plant and recycled products are remanufactured into as-new products in the remanufacturing facility. Both the brand-new products and remanufactured products are used to satisfy customer demands. Returns of used products that are recycled from customers are assumed to be stochastic and nonlinearly price-dependent. A mathematical model is proposed to maximize the overall profit of the system through simultaneously optimizing the production and recycling decisions, subject to two capacity constraints — the manufacturing capacity and the remanufacturing capacity. Based on Lagrangian relaxation method, subgradient algorithm and heuristic algorithm, a solution approach is developed to solve the problem. A representative example is presented to illustrate the system, and managerial analysis indicates that the uncertainties in demand and return have much influence on the production and recycling policy. In addition, twenty randomly produced examples are solved, and computational results show that the solution approach can obtain very good solutions for all examples in reasonable time.     

Coordinating pricing and inventory decisions in a bounded production system with uncertain yield

Most inventory researches assume that production level can fluctuate arbitrarily. However, large fluctuation of the production level in many firms may be much costly. This paper addresses the coordinating pricing and inventory control problem in a bounded production system with uncertain yield, in which the production level is constrained between a maximum and minimum level in each period and the price can be adjusted dynamically. We show that the optimal policy is the interval base-stock-list-price policy. Also, we study the impact of the production bounds and uncertainty of the yield on the production system. Numerical experiments are also performed to study the impact of parameters on the system.     

The M/PH/1 queue with working vacations and vacation interruption

We study an M/PH/1 queue with phase type working vacation and vacation interruption where the vacation time follows a phase type distribution. The server serves the customers at a lower rate in a vacation period. The server comes back to the regular busy period at a service completion without completing the vacation. Such policy is called vacation interruption. In terms of quasi birth and death process and matrix-geometric solution method, we obtain the stationary queue length distribution. Moreover we obtain the conditional stochastic decomposition structures of queue length and waiting time when the service time distribution in the regular busy period is exponential.