Predicting intensive care unit bed occupancy for integrated operating room scheduling via neural networks

In a master surgery scheduling (MSS) problem, a hospital's operating room (OR) capacity is assigned to different medical specialties. This task is critical since the risk of assigning too much or too little OR time to a specialty is associated with overtime or deficit hours of the staff, deferral or delay of surgeries, and unsatisfied—or even endangered—patients. Most MSS approaches in the literature focus only on the OR while neglecting the impact on downstream units or reflect a simplified version of the real‐world situation. We present the first prediction model for the integrated OR scheduling problem based on machine learning. Our three‐step approach focuses on the intensive care unit (ICU) and reflects elective and urgent patients, inpatients and outpatients, and all possible paths through the hospital. We provide an empirical evaluation of our method with surgery data for Universitätsklinikum Augsburg, a German tertiary care hospital with 1700 beds. We show that our model outperforms a state‐of‐the‐art model by 43% in number of predicted beds. Our model can be used as supporting tool for hospital managers or incorporated in an optimization model. Eventually, we provide guidance to support hospital managers in scheduling surgeries more efficiently.     

Adaptive control of track tension estimation using radial basis function neural network

Track tension is a major factor influencing the reliability of a track. In order to reduce the risk of track peel-off, it is necessary to keep track tension constant. However, it is difficult to measure the dynamic tension during off-road operation. Based on the analysis of the relation and external forces depending on free body diagrams of the idler, idler arm, road wheel and road arm, a theoretical estimation model of track tension is built. Comparing estimation results with multibody dynamics simulation results, the rationality of track tension monitor is validated. By the aid of this monitor, a track tension control system is designed, which includes a self-tuning proportional-integral-derivative (PID) controller based on radial basis function neural network, an electro-hydraulic servo system and an idler arm. The tightness of track can be adjusted by turning the idler arm. Simulation results of the vehicle starting process indicate that the controller can reach different expected tensions quickly and accurately. Compared with a traditional PID controller, the proposed controller has a stronger anti-disturbance ability by amending control pa-rameters online.     

无标度网络上的队列资源重分配策略

在实际的通信系统中,由于物理条件的限制,每个节点都具有有限长度的队列。研究了在有限队列资源的条件下,队列资源对无标度数据流动力学的影响。提出了一种队列资源重分配策略,策略中节点的队列长度与节点的介数成正比。仿真结果表明,在无标度网络中使用最短路径路由的条件下,所提出的重分配策略可以有效地改进网络的容量。同时对有限队列资源条件下的网络容量进行了理论分析。     

面向定位的分布式无线传感器网络细粒度调整方法

针对在初始的部署条件下不可定位的无线传感器网络,提出一种分布式的面向定位的网络调整方法。该方法通过节点所处的路径信息判别确定该节点的调整策略,从而使初始状况下不可定位的网络达到可定位条件,该方法仅需调整约11%节点就能够将稀疏的网络调整至可定位,比当前最好的网络调整方法减少了约40%。此外,该方法采用分布式执行策略,从而将定位所产生的通信负载和能耗均衡到网络中的多个节点,克服了先前集中式方法的可扩展性限制。大量的仿真实验结果表明,该方法较现有方法而言具有更高的执行效率。     

基于BP神经网络的雷达网抗干扰性能综合评估

在分析目前综合评估雷达网抗干扰性能时遇到困难后,探讨了利用BP人工神经网络良好的非线性函数逼近特性来解决这一问题的可行性,并分析了在用这种方法建模评估的过程中需要注意的各种问题;其中侧重对在具体构造评估模型时输入输出评估指标集的确定方法和模型的训练验证方法进行了研究,并进一步分析了建立的模型可能的用途。最后指出这种思路也可以推广到对类似的复杂信息系统的综合评估中。     

基于局域网的分布式战斗机空战仿真系统

简要介绍了一种实时分布交互式战斗机空战仿真系统的开发。整个系统按照分层模块思想设计,对空战中战斗机关键子系统和关键行为进行了精确的数学建模,减少了建模的复杂度、保持了平台的开放性和可维护性,据此实现了局域网环境下的分布交互式空战仿真研究。初步的仿真结果表明系统具有实时、精确、高逼真度、高效费比的特点,可作为飞机设计和空战研究的重要辅助手段。     

国防科技体系评价

如何科学、合理、有效的评价国防科技体系发展水平,识别国防关键技术,建立健全国防科技体系,是国防建设重中之重。基于未确知测度理论构建国防科技发展水平的评价原则,结合网络分析法确定指标体系权重,有效解决了国防科技体系评价的复杂性、系统性问题。研究表明,国防科技体系评价可以为国防科技发展战略的科学性和可操作性提供参考。     

基于神经网络的广义两点边值问题

在航天领域中 ,多数制导问题均可以抽象为广义两点边值问题。本文探讨了神经网络在广义两点边值问题求解中的应用。首先讨论了离线制导这一类静态两点边值问题的求解 ,在此基础上 ,对在线制导这一类动态两点边值问题进行了分析 ,并给出了有效的解决方法     

基于遗传算法的通信网络可靠性优化设计

在可靠性条件约束下 ,使网络成本最低是网络规划NP hard问题 .文章提出一种基于遗传算法的优化方法解决了这类问题 .仿真结果表明这种算法是有效的 .     

多重结构神经网络在平台罗经故障诊断中的应用

提出了基于层次分类诊断模型的多重结构神经网络 (MNN)在平台罗经故障诊断中的应用方法 ,并对其温控子系统的人工神经网络 (ANN)进行了Matlab实现