模糊逻辑系统在船舶风险评估中的应用

为了合理进行船舶风险评估,提出了一种基于模糊逻辑系统的船舶风险评估方法.根据模糊逻辑系统的基本原理,建立了船舶风险评估的模糊逻辑模型;按照国家标准、规范,并结合经验建立了风险指标的模糊集合、风险矩阵及模糊规则,然后进行了模糊推理,得到了系统的风险等级,并制定了预防风险的措施.以某艇消防系统的风险评估为例,证明了其能够合理地评估出系统的风险,可为决策提供参考.     

基于小波神经网络的蒸汽发生器水位辨识与控制仿真

提出一种基于小波神经网络的控制方法,对蒸汽发生器水位进行控制仿真.该方法利用小波神经网络作为控制系统的辨识器和控制器来构成控制系统.小波神经网络辨识器能更准确逼近非线性对象,小波神经网络控制器能自适应产生最佳的控制规律.仿真结果表明,该方案具有响应快、超调量小、较强抑制干扰能力等良好性能.     

DESIGNING A VIABLE PREDICTION MARKET TO FORECAST DEFENSE ACQUISITION COST AND SCHEDULE OUTCOMES

This paper is the first to demonstrate a viable prediction market that successfully forecasts defense acquisition cost and schedule outcomes, and to provide insights for defense executive decision‐making. Already used in private industry, prediction markets can also accurately forecast outcomes and their associated risks for government programs. Using virtual money, prediction markets allow traders to ‘bet’ on some future outcome. This market mechanism turns out to be a relatively simple and accurate way to discover, aggregate, and communicate to a defense executive the collective market’s beliefs about the likelihood of an eventual outcome of an acquisition program of interest.     

The management of security incidents by private security

This article is about the management of security incidents in organisations and companies which are under the protection of private security personnel, whether in-house or contracted. Incidents can be defined as accidental or anecdotal (bird flies into a camera) to a violation of law or company policy. Managing security incidents is one of the key functions of a private security service provider. The purpose of this study is to explore the management of security incidents and the information related thereto, identify shortcomings and find solutions for managing these shortcomings. Interviews were used to gain insight from personnel within the private and government sectors which are served by the private security industry. It was found that security incidents are handled in a routine way by organisations from the different sectors. One major shortcoming is that security incidents and the information related thereto are not managed by security service providers according to an operational framework or model with the required infrastructure and resources. The significance of this study is for security incidents to be managed correctly, so that the information related thereto may be captured accurately, analysed and used proactively and reactively to improve physical protection systems, develop preventative strategies and provide actionable information products in order to reduce crime, increase detection rates and prevent losses.     

A theory of war finance

This paper analyses the financial and war‐spending policies of a state that faces a conflict in which defeat would result in the loss of sovereign power and in which the material consequences, conditional on avoiding defeat, are stochastic. The analysis takes explicit account of the historical experiences of lenders, who face debt repudiation if the state to whom they have lent is defeated and who also face partial default if the material consequences of the war are unfavorable for the debtor state, even if it avoids defeat. In this analysis, the state uses war debt to smooth expected consumption intertemporally in response to temporary war spending, and the state also uses contingent debt servicing to insure realized consumption against the risk associated with the material consequences of the war. An important innovation in the analysis is to treat the equilibrium amount of war spending, the state's resulting probability of avoiding defeat, as well as the equilibrium amount of borrowing as a set of endogenous variables to be determined simultaneously.     

网络化装备维修保障概念与体系结构研究

摘要：针对传统的装备维修保障理论不能适应现代装备发展的问题,在网络技术高速发展的今天,提出了网络化装备维修保障的概念．主要探讨了网络化维修保障的概念和特点、网络化维修保障理念与传统维修保障理念的关系、网络化维修保障体系构想等问题．对网络化装备维修保障的概念和体系进行了初步的定义和分析,为进一步研究网络化维修保障理论奠定了基础．     

模糊推理在软件维护性评估和预测中的应用

摘要：在基于活动的质量模型维护性分析方法基础上,对贝叶斯方法进行了扩展,将模糊推理技术与现有的贝叶斯方法进行比较,开发了一个基于模糊推理的系统来对软件维护性进行量化评估和预测,描述了模糊技术在维护性评估和预测准确度方面的优势．     

基于改进半监督FCM聚类的复合材料超声检测脱粘缺陷识别

针对径向基（RBF）神经网络在进行超声检测脱粘缺陷识别时存在参数选择不确定、网络结构鲁棒性差等问题,提出一种改进的自适应半监督模糊C均值聚类（FCM）的RBF神经网络的方法,将kw近邻估计法和半监督模糊C均值聚类方法相结合,改进了隶属度函数,自适应确定聚类数目。将改进的RBF神经网络应用于超声检测脱粘缺陷识别,实验结果表明：与传统RBF神经网络相比,本方法减弱了孤立样本对网络结构的影响,增强了网络结构的鲁棒性,提高了脱粘缺陷识别的准确率,是一种较好的超声检测脱粘缺陷识别分类方法．     

舰艇编队协同反导网络抗毁性研究

在深入研究舰艇编队协同反导作战体系结构的基础上,构建了舰艇编队协同反导网络模型。针对构建网络,提出了网络抗毁性测度指标,分析网络遭受随机攻击和选择性攻击后网络测度的变化,仿真结果表明决策器之间的连接对网络的抗毁性更加敏感。     

失去润滑条件下弧齿锥齿轮传动系统瞬态温度场Simulink仿真分析

根据传热学理论和热网络法,以某型直升机尾减速器弧齿锥齿轮传动系统为研究对象,建立了其温度场Simulink仿真计算模型。基于该模型,计算获得该传动系统正常润滑条件下的稳态温度场,并以它作为初始条件计算分析该传动系统失去润滑条件下的瞬态温度场,为预测该弧齿锥齿轮传动系统在失去润滑条件下的干运转生存能力提供理论依据。