保精度-稀疏特性核回归模型的非线性动态系统辨识

针对一组有限测量数据的非线性动态系统建模方法存在模型结构复杂且易出现过拟合等问题,从建模精度及模型稀疏特性出发,提出了保精度-稀疏特性的核回归模型用于辨识非线性动态系统。该方法将逼近误差的L_∞范数思想与结构风险最小化理论相结合,建立求解非线性动态系统所对应的核回归模型优化问题,再应用较简单的线性规划对其求解。提出的方法具有如下三个显著特性:①应用逼近误差的L_∞范数最小化可保证非线性动态系统的辨识精度;②引入支持向量回归架构下的结构风险L_1范数对模型结构复杂性进行有效控制可保证模型稀疏特性;③模型的泛化性能可通过提出的方法从建模精度与模型稀疏特性之间取其平衡。最后,通过实验分析论证了提出方法在辨识非线性动态系统上的保精度-稀疏特性的合理性与优越性。     

一维守恒的Lagrangian ADER格式

基于欧拉框架下ADER格式,构造一维守恒只有一个时间步的、高精度中心型拉格朗日ADER(LADER)格式.构造r阶LADER格式包括:从欧拉方程出发推导拉格朗日框架下积分形式的方程、采用WENO方法高精度重构节点处守恒量和从1阶到r-1阶的空间导数、求拉氏框架下这些变量的Godunov值,并计算1阶到r-1阶的时间全导数,最后高精度离散积分形式的流通量函数.对光滑流场的模拟表明,LADER格式达到设计的精度;对含强间断的流场模拟表明,数值解在间断附近基本无振荡.     

Measurement accuracy analysis of the free carrier absorption determination of the electronic transport properties of silicon wafers

By introducing the random and systematic errors in simulated data computed from conventional frequency-scan and laterally resolved modulated free carrier absorption theory models, we investigate the relative determination sensitivities of three electronic transport properties, namely, carrier lifetime, carrier diffusivity and front surface recombination velocity of silicon wafers determined by frequency-scan and laterally resolved techniques. The phase and amplitude data with random errors as functions of the modulation frequency at zero pump-probe-beam separation or of the two-beam separation at four different modulation frequencies are simultaneously fitted to an appreciated carrier diffusion model to extract three transport parameters. The statistical results and fitted accuracies of the transport parameter determined by both techniques are theoretically analysed. Corresponding experimental results are carried out to compare to the simulated results. The simulated and experimental results show that the determination of the transport properties of silicon wafers by the laterally resolved technique are more accurate, as compared with that by the frequency-scan technique.     

修正荷载的Wilson-θ法的稳定性与精度分析

Wilson-θ法求得的位移、速度与加速度不满足t时刻的动力平衡方程。提出修正荷载的Wilson-θ法:增加一个荷载δF(t),使得t时刻的动力平衡方程得以满足;将-θ′δF(t)作为荷载加入t+Δt时刻的计算中,当θ′=0时,修正荷载的Wilson-θ法退化为Wilson-θ法。对应于不同的θ′值,在无条件稳定的前提下,θ的取值范围也不同。定义了逐步积分法中的计算误差。计算结果表明:计算误差与θ值成正相关,当θ′=0.6,无条件稳定的θ为最小值1.24,因而θ′=0.6,θ=1.24时,计算误差最小,建议在计算中采用。保持Wilson-θ法无条件稳定、几乎不增加计算量的条件下,修正荷载的Wilson-θ法可以提高计算精度。     

Effects of Quantum Noise on Quantum Clock Synchronization

In laboratory environment, the channel apparatus will generate particular dominant quantum noise. The noise then will give rise to some errors during synchronization. In this work, the accuracies of one qubit transport protocol and entangled states transport protocol in the presence of noise have been studied. With the help of three important and familiar noise models, the quantum noise will degrade the accuracy has been proved. Due to the influence of quantum noise, the accuracy of entangled qubits decrease faster than that of one qubit. The entangled states will improve the accuracy in noise-free channel, and will degrade the accuracy in noise channel.     

光谱分辨率和信噪比对光谱线指数测量精度的影响

分析光谱分辨率和信噪比对19个Lick原子吸收线指数测量精度的影响。采用高斯卷积方法,将光谱变换到不同分辨率下,通过测量并比较光谱在不同分辨率下的线指数值探究分辨率改变对线指数测量精度的影响;通过向光谱叠加符合不同高斯分布的随机噪声方法,将无噪理论光谱改变到不同信噪下,通过测量并比较光谱在不同信噪比时的线指数值分析信噪比对线指数测量精度的影响。结果表明：改变光谱分辨率能够引起线指数测量值的变化,此变化程度依赖于分辨率的改变程度,且此依赖关系对不同指数呈不同变化趋势;信噪比对线指数测量精度的影响随信噪比的降低而增大,当信噪比高于25时,此影响可忽略。     

烟雾箱模拟丙烯-NOx的大气光化学反应

利用自制烟雾箱对丙烯-NO_x的光化学反应进行了一系列模拟实验, 研究了相对湿度及丙烯与NO_x的初始浓度比值对丙烯-NO_x光化学反应产物如O₃和CO的影响. 实验得到了臭氧最大值、CO最大值及反应掉的C₃H₆均随相对湿度的增加而减少, 丙烯的臭氧生成活性最大值(IR_max)也随着相对湿度的增加而减小, 且两者间存在显著的线性关系. 丙烯与NO_x的比值对反应生成臭氧最大值的影响出现两种情况: 当初始浓度比值在1.0～3.4之间时, 臭氧最大值随着该比值的增加而有大幅度提高|而当该比值分别在3.7～7.4之间时, 臭氧最大值随比值的增加而有小幅度的下降. 此外, 本研究还利用丙烯MCMv3.1机理对反应进行了数值模拟.     

Practical Reference-Frame-Independent Measurement-Device-Independent Quantum Key Distribution Systems Against the Worst Relative Rotation of Reference Frames

Reference-frame-independent measurement-device-independent QKD (RFI-MDI-QKD), immune to both the slow unknown drift of reference frames and detector side channel attacks, can generate information-theoretic secure keys. Despite its natural resistance to the slow drift of reference frames, the performance of practical RFI-MDI-QKD systems descends with the increasing drift of reference frames. In this paper, we demonstrate the worst relative rotation of reference frames for practical RFI-MDI-QKD systems, and investigate their performance against the worst-case scenario, both of which are unclear to date. Simulation results show that, practical RFI-MDI-QKD systems can achieve quite good performance even against the worst-case scenario, which clearly demonstrates that it is possible to implement practical MDI-QKD systems with freely drifting reference frames.     

Sufficient Conditions of the Same State Order Induced by Coherence

In this paper, we study coherence-induced state ordering with Tsallis relative entropy of coherence, relative entropy of coherence and l₁ norm of coherence, and give the sufficient conditions of the same state order induced by above coherence measures. First, we show that the above measures give the same ordering for single-qubit states in some special cases. Second, we consider some special states in a d-dimensional quantum system. We show that the above measures generate the same ordering for these special states. Finally, we discuss dynamics of coherence-induced state ordering under Markovian channels. We find amplitude damping channel changes the coherence-induced ordering even though for single-qubit states with fixed mixedness.     

一种用于时差提取的卡尔曼-最优阶互相关算法

在定位系统的时差提取方法中,传统的互相关算法抗噪能力较弱、时差提取精度较低,针对以上缺点,提出卡尔曼-最优阶互相关算法,并且通过仿真对比分析相同条件下该算法和基本互相关算法、广义加权互相关算法的时差提取精度,验证了其更优的性能.最后,通过FPGA设计,将所提算法运用到实际工程中.仿真和试验结果表明,相对于传统算法,卡尔曼-最优阶互相关算法在较低信噪比的情况下具有更强的抗噪能力和更高的时差提取精度.