Higher‐order Markov chain models for categorical data sequences*

In this paper we study higher‐order Markov chain models for analyzing categorical data sequences. We propose an efficient estimation method for the model parameters. Data sequences such as DNA and sales demand are used to illustrate the predicting power of our proposed models. In particular, we apply the developed higher‐order Markov chain model to the server logs data. The objective here is to model the users' behavior in accessing information and to predict their behavior in the future. Our tests are based on a realistic web log and our model shows an improvement in prediction. © 2004 Wiley Periodicals, Inc. Naval Research Logistics, 2004     

逆边界无网格法声源识别的理论与实验

针对逆向求解声源识别中的声辐射传输建模问题,采用无网格法将Kirchhoff- Helmholtz边界积分方程离散为受边界条件约束的有限维线性方程组,通过分块矩阵法对该约束方程组进行求解,得到了离散后声辐射传输模型的数值表达式.在此基础上,进一步研究了逆向求解声源识别问题的基本原理及其不适定性.为克服其不适定性,采用Tikhonov正则化和L曲线正则化参数选取方法,从而确立了有效的逆向求解方法.此外,还进行了扬声器阵列声源识别实验,实验结果验证了逆边界无网格声源识别理论和方法的可行性及可靠性.     

基于串联谐振变换器的感应式电能传输系统研究

摘要：为提高装甲车辆蓄电池充电装置的环境适应性和安全性,解决传导式电能传输不适合旋转部件的难题,提出一种基于串联谐振式变换器的感应式电能传输系统．该系统是采用旋转式耦合器,以耦合器的漏感为谐振电感,应用全桥式串联谐振功率变换器的电能传输系统．利用所设计的感应式电能传输试验装置,进行了不同气隙条件下的功率传输和初、次级电流对比试验,试验结果表明：随着耦合器气隙增大,电能传输效率呈现斜率渐缓式减小趋势,励磁电感减小,励磁电流变大,耦合器的铁损增加,系统的功率因数降低．     

基于DAC5686的线性调频信号产生方法研究

研究了基于DAC5686的线性调频信号产生方法。分析了线性调频信号及其产生原理,研究了DAC5686工作原理,设计了信号产生电路,给出了线性调频信号的实现实例．结果表明,该方法可以灵活地产生所需线性调频信号,从而满足某型地空导弹制导雷达中频接收机性能测试的需要．与传统的线性调频信号产生方法相比,该方法降低了系统设计难度,缩短了开发周期,提高了设计的可靠性,具有较高的实用价值,对于其它场合频线性调频信号的产生也具有一定的借鉴意义．     

Into the black box? Technology sharing in major arms transfers and beyond

From the recipients’ perspective, arms transfers have, through the use of offsets, technology transfers, and industrial participation, become an opportunity to receive not only advanced weapons, but also technologies not otherwise available. How important are friendly relations for securing a military export order? To what extent do buyers demand advanced military or commercial technology and how are these demands accepted by the supplier? How does this influence smaller producers in relation to major producers? These questions are addressed by studying (a) the Joint Strike Fighter/F-35 by Lockheed Martin, USA, and the JAS-39 Gripen aircraft by Saab, Sweden, (b) the Medium Multi-Role Combat Aircraft competition in India for 126 combat aircraft, and (c) the Indian offset policy. In the final section, some long-term consequences are discussed.     

高硬再制造表面强化层的加热切削研究

应用高性能硬质合金、陶瓷刀具和CBN刀具对常温条件下难以加工的高硬堆焊材料进行加热切削试验。试验结果表明：尽管加热条件下冲击效应降低,但CBN刀具仍然破损严重;陶瓷刀具虽然也无法实现理想的切削,但切削过程中刀具破损模式从崩碎式失效转变为磨损失效;3种刀具中高性能硬质合金应用效果最佳,其最佳参数条件下,一个切削里程后刀面VB值仅为0．3mm,试验过程中,刀具磨损随表面加热温度、切削速度、切削深度的提高而增大,但随进给量取值的增大而减小,其中切削速度对刀具VB值的影响最大,而进给量和削切深度对刀具VB值的影响较小。综合分析还表明：试验中刀具磨损速率仍然受刀具红硬性极限的限制。     

基于循环移位和异或运算的对合线性变换研究

在对称密码算法的设计中,为达到良好的扩散作用,设计者一般均选择分支数较大的线性变换.基于循环移位和异或运算的线性变换由于其实现效率较高,已经在很多密码算法中被采用,比如分组密码SMS4、HIGHT,Hash函数SHA -2、MD6等.此外,如果线性变换是对合的,还为解密带来了方便.研究了基于循环移位和异或运算设计的对合线性变换,给出了这类线性变换的计数公式,指出它们的分支数上界为4,并讨论了循环移位的参数与分支数之间的关系,从而为基于这类运算设计的线性变换提供了理论依据.     

小波神经网络UCAV飞行控制系统

针对采用H∞控制设计方法需求解难度很大的偏微分方程,而导致工程实现难度大的问题,提出了采用小波神经网络进行无人攻击机(UCAV)飞行控制系统设计的方法,该方法集小波变换和神经网络的优点于一身,简化了网络训练,避免了系统达到局部最优解,使得系统具有更好的逼近精度。通过仿真实验表明,利用小波神经网络进行UCAV控制律设计,使系统具有较好的跟踪性和鲁棒性,避免了精确模型的建立,便于分析系统的稳定性。     

多目标攻击导弹捷联惯导的传递对准

在分析中距复合制导空空导弹捷联惯导传递对准的基础上,提出由机载火控系统根据导弹发射包线计算发出开始捷联惯导传递对准精对准的时刻,导弹惯导提前进行传递精对准的方法。利用此方法可以解决机载武器系统多目标攻击空空导弹传递对准满足连续发射多枚导弹实时性要求的问题。最后,在单目标情况下就开始传递对准时刻包线给出了算例。     

A branch and bound algorithm for the minimum storage‐time sequencing problem

The minimum storage‐time sequencing problem generalizes many well‐known problems in combinatorial optimization, such as the directed linear arrangement and the problem of minimizing the weighted sum of completion times, subject to precedence constraints on a single processor. In this paper we propose a new lower bound, based on a Lagrangian relaxation, which can be computed very efficiently. To improve upon this lower bound, we employ a bundle optimization algorithm. We also show that the best bound obtainable by this approach equals the one obtainable from the linear relaxation computed on a formulation whose first Chvàtal closure equals the convex hull of all the integer solutions of the problem. © 2001 John Wiley & Sons, Inc. Naval Research Logistics 48: 313–331, 2001