Euler-Lagrange Modeling and Passivity Based Control of Buck Converters with Input Filter

In this paper, buck converters with input filter are modeled using the Euler Lagrange formalism and then build a PBC （passivity based controller）. The model is validated, by comparing its response with those of two switched circuits： sylnmetric and asymmetric. In the former, both switches are realized by MOSFETS while in the second one of them is realized by a diode. It is then showed by simulation and, explained with energy-based arguments why the obtained model thoroughly represents only the symmetric circuit. The model is then used to build a passivity-based control law. As this control law assumes the stability of the zero dynamic, conditions under which this hypothesis is satisfied, are first given. It is shown by simulation with switched circuits the robustness of the proposed controller against load variations. Then, a prediction of the source variations is included in the controller in order to render it robust against source variations.

Krashen's Input Hypothesis and Affective Filter Hypothesis and Their Enlightenment to English Teaching in China

"The Input Hypothesis"and"The Affective Filter Hypothesis"are the most important branches of the"Monitor The ory"declared by the famous American linguist Krashen in 1980s.This thesis is based on Krashen's Input Hypothesis and Affec tive Filter Hypothesis,aiming to provide an applicable guide for foreign language teachers and learners in China.

Voltage-Mode Universal Biquad Filter Employing Single Voltage Differencing Differential Input Buffered Amplifier

A new multi function voltage-mode universal biquadratic filter using single Voltage Differencing Differential Input Buffered Amplifier (VD-DIBA), two capacitors and one resistor is proposed. The proposed configuration has four inputs and one output and can realize all the five standard filters from the same circuit configuration. The presented biquad filter offers low active and passive sensitivities. The validity of proposed universal biquadratic filter has been verified by SPICE simulation using 0.35 μm MIETEC technology.

A new layered space-time detection algorithm for frequency selective fading multiple-input multiple-output channels based on particle filter

In this paper, a new observation equation of non-Gaussian frequency selective fading Bell Labs layered space time （BLAST） architecture system is proposed, which is used for frequency selective fading channels and non-Gaussian noise in an application environment of BLAST system. With othogonal matrix triangularization （QR decomposition） of the channel matrix, the static observation equation of frequency selective fading BLAST system is transformed into a dynamic state space model, and then the particle filter is used for space-time layered detection. Making the full use of the finite alphabet of the digital modulation communication signal, the optimal proposal distribution can be chosen to produce particle and update the weight. Incorporated with current method of reducing error propagation, a new space-time layered detection algorithm is proposed. Simulation result shows the validity of the proposed algorithm.

Implications of Krashen's Input Hypothesis and Affective Filter Hypothesis in College English Teaching

American linguist Krashen proposed the input hypothesis and affective filter hypothesis which has influenced linguistic world deeply. These two assumptions play an important role in improving learners' language ability in the process of second language acquisition. Applying two hypothesis theories to college English teaching, teachers' aim is to establish a new mode of language teaching and improve the efficiency of college English teaching.

An Identical Inputs-Adaptive Filter for the Detection of Signal’s Breakdown Points

Detection of the signal’s breakdown points is important for many science and engineering applications. Numerous signal processing methods have been used for this purpose. Of these, the adaptive prediction is simple and easy to implement, however;its simplicity and robustness are hindered by the required delay in the input signal. This paper introduces an efficient alternative to the adaptive prediction in the application of breakdown and inflection points’ detection. Unlike the adaptive predictor, the proposed filter doesn’t require a delay in the primary input to produce the filter’s reference input, which significantly improves the computation speed and overcome the problem of performance sensitivity to the delay value. The Normalized Least-Mean Squares algorithm was used to realize both the adaptive predictor and the proposed filter. The filters were implemented in LabVIEW system design software. The performances of the filters were studied using simulated signals and the simulation results were verified using an experimental signal. The simulation and experimental results showed that the proposed filter efficiently detects the signal breakdowns. Furthermore, the simplicity of the filter offered a significant improvement in the computation speed.

Krashen's Input Hypothesis and Affective Filter Hypothesis'Enlightenment to the Vocational English Teaching

The Krashen's second language acquisition theory have two important assumptions: input hypothesis and the affective filter hypothesis. They can guide and inspire vocational teaching, because they start from the student's learning situation, guide teachers to adjust their teaching methods, and provide better service for teaching and learning. This paper mainly analyzes the Krashen's input hypothesis and the affective filter hypothesis'enlightenment to the vocational English teaching.

基于指令滤波器的无人艇抗饱和动态面跟踪控制器设计

本文对带有外界干扰、输入受限情形下的无人水面艇轨迹控制问题开展了研究。首先,给出了考虑前进、横漂和艏摇的无人艇三自由度跟踪控制模型。其次,基于反步法、跟踪微分器和抗饱和辅助系统,提出了有限时间饱和动态面跟踪控制器,并通过引入补偿信号消除了滤波器带来的微分误差,提高了控制系统的性能。最后,借助李雅普诺夫稳定性理论证明了所设计系统状态为有限时间稳定的,并通过数字仿真验证了所设计控制策略的有效性。

考虑输入时滞的舵机伺服系统自适应指令滤波输出反馈控制

针对舵机伺服系统运行过程中存在复杂非线性以及信号输入时滞、使系统的精度降低,严重时甚至造成系统失稳的问题,提出了一种考虑输入时滞的自适应指令滤波输出反馈控制策略。首先,将系统中存在的输入时滞考虑为控制器信号输入时滞,给出舵机伺服系统状态方程,选用梯度下降法设计时滞估计律对未知时滞进行估计;其次,将系统中存在的复杂非线性统一为扰动,选用扩展状态观测器(ESO)加以估计,将观测的系统状态值用于设计控制量实现输出反馈控制;然后,选用指令滤波将高阶求导过程转化为求积分,实现信号滤波以抑制微分噪声;最后,基于Lyapunov-Krasovskii泛函稳定性定理证明控制器实现系统有界稳定。仿真及实验表明,所提的控制方法与传统反步控制方法、自抗扰控制和比例微分积分(PID)控制方法相比,响应时间分别提升了92%、88%和51%,跟踪精度分别提升了90%、84%和26%。

5G基站射频滤波器发展和分析

大规模多输入多输出技术是5G的关键技术之一,包含了多个射频通道,系统中的滤波器数量大幅增加,使得5G滤波器在保持高性能的同时对小型化和高集成度提出了更高要求。分析了5G射频基站滤波器的基本要求,介绍了射频基站滤波器的形式和特点,包括微带、波导腔、同轴、介质加载、陶瓷介质滤波器。通过分析不同形式基站滤波器的特点及研究现状,并进行综合对比,从性能、尺寸、成本方面综合考虑,陶瓷介质滤波器在5G基站中具有更大的发展空间。

面向智能物联网的资源高效模型推理综述

从智慧城市到工业自动化,智能物联网在越来越多的场景中得到了广泛应用.模型推理作为实现智能决策和响应的核心技术,在智能物联网系统中扮演着举足轻重的角色.然而,智能物联网设备通常在计算能力、通信带宽、内存容量和电池寿命等资源上高度受限.这使得智能物联网中的模型推理资源开销成为一个关键技术挑战.本综述总结了在智能物联网场景中优化模型推理资源开销的相关技术,对当前在智能物联网应用中使用的主流模型推理优化技术进行概述,并深入分析它们在资源效率方面的优势和不足.本文从推理涉及的三大模块(传感器数据、智能模型、物联网硬件)和五类关键资源的角度出发设计新的技术分类,并首次提出了一套针对智能物联网模型推理的通用的优化流程,能够帮助相关研发人员定位和优化推理效率瓶颈.最后,本文讨论了智能物联网推理效率相关的四个未来研究方向.

模块化教学对英语文化语用石化的改善研究

高职高专学生虽具备了一定的英语基础知识和基本技能,但在实际运用中,英语文化语用中的石化现象普遍存在,制约了其英语运用能力的提高。对高职高专学生英语文化语用石化现象情况、石化原因进行分析,根据输入假说理论、情感过滤假说理论,初步从理论角度解释了模块化教学对于改善英语文化语用石化现象的可行性,并以学前教育专业大一学生为例,阐述了模块化教学的具体步骤,旨在提升高职高专学生跨文化交际的能力和水平。

噪声环境下基于等价输入干扰和卡尔曼滤波器的控制系统扰动抑制

控制系统中存在的未知扰动和测量噪声,限制了系统性能的提升,给控制设计带来了巨大挑战。提出了一种基于等价输入干扰(Equivalent Input Disturbance,EID)和卡尔曼滤波器(Kalman Filter,KF)的复合扰动抑制方法,以减小测量噪声对扰动抑制和系统控制性能的影响。首先,设计一种稳态卡尔曼滤波器,滤除测量信号中的高频噪声;其次,基于卡尔曼滤波器的输出,设计状态观测器和干扰估计器,计算等价输入干扰的估计值,并在控制通道中反向补偿,得到一种复合控制规律,以提高噪声环境下控制系统的扰动抑制性能;最后,设计电机转速控制系统仿真实验,验证该方法的有效性。与传统控制方案进行比较表明:新的结构使得系统在同时存在测量噪声和扰动的情况下,仍然具有良好的控制性能。

Particle filter based stochastic interference cancellation for frequency-selective MIMO channel equalization

To mitigate the effects of the previous symbol decision errors of a decision-feedback （DF） equalizer on the current decision, a particle filter （PF） based DF equalizer for frequency selective multiple-input-multiple-output （MIMO） channel is proposed. On the basis of the analyses of DF equalization for the MIMO wireless system, it is found that a stochastic interference cancellation （IC） scheme can be employed to prevent the error propagation in a severe space-time interference scenario. This is because the random rather than the deterministic scheme can reduce the probability of an error decision even if an error decision occurs. Besides, the signal-to-interference-plus-noise ratio （SINR） based IC order, which is obtained via pilot, can guarantee the optimality of the cancellation. The bit error rate （BER） performance of the proposed scheme is verified through simulation experiments under different multipath interference environment.

发射波束域MIMO-STAP雷达发射接收联合设计方法

研究了信号相关杂波背景下机载多输入多输出(multiple-input multiple-output,MIMO)雷达发射波束形成和接收滤波器的联合设计问题,建立了机载MIMO雷达发射波束域空时自适应处理(space-time adaptive processing,STAP)信号模型。为了提升杂波环境下目标的检测性能,通过最大化输出信干噪比构建了发射波束形成和接收滤波器的联合设计问题;然后,设计了一种新的基于优化最小化(majorization-minimization,MM)框架的迭代优化算法来解决联合设计问题。该算法通过合理寻找目标函数的下界可以有效提升算法的收敛速度并降低算法的运行时间。此外,与传统的相控阵雷达和MIMO雷达相比,优化后的发射波束形成和接收滤波器可以显著提升输出信干噪比,仿真实验验证了所提方法的有效性。

基于结构随机场的桥梁移动荷载统计矩识别

由于结构、材料不确定等因素,桥梁结构往往具有随机性,因此,基于桥梁结构的响应识别得到的桥梁所受移动荷载也是不确定的,但相关研究还较少。为此,本文研究桥梁结构为空间随机场的情况下,有效识别移动荷载的统计矩。提出的识别方法基于桥梁结构随机场的Karhunen⁃Loeve(KL)展开、未知输入的卡尔曼滤波与改进的两点估计法的结合。桥梁结构参数为空间相关的随机场,通过KL展开将随机场转化为多随机变量的组合。在多随机变量下的不确定传播中,利用改进的两点估计法,将识别移动荷载统计矩问题转化为若干个确定性识别移动荷载识别的逆问题。采用未知输入的卡尔曼滤波进行确定性识别移动荷载,可有效估计识别移动荷载的统计矩。通过数值模拟算例验证了提出的估计方法的有效性。

部分约束下空中目标拦截制导控制一体化方法

针对拦截空中目标的场景,提出一种考虑攻击角度的制导控制一体化(integrated guidance and control,IGC)方法,同时考虑输入饱和以及攻角的约束问题。首先,在俯仰平面下对包含不确定性的系统进行了建模。基于反演法和指令滤波器,处理了攻击角度约束问题和执行器机械限制的输入饱和问题,设计了含误差积分反馈的补偿项以处理跟踪误差,引入了障碍Lyapunov函数将攻角约束在预设区间。基于Lyapunov理论证明了闭环系统的稳定性和变量的有界性,以及各约束条件的成立性。仿真实验表明,方法能够以预设角度有效拦截目标,过程满足对输入以及攻角的约束条件,同时具备较强的鲁棒性。

未知非线性多智能体系统的分布式有限时间自适应神经网络控制

针对一类具有未知磁滞输入的非线性多智能体系统,提出了一种基于改进命令滤波器的分布式预设性能自适应神经网络渐近一致控制方案。首先,为保证系统的跟踪性能,提出了基于新型性能函数的有限时间Funnel控制方案,使得跟随者与领导者在有限的时间内输出一致。其次,利用径向基神经网络和放缩方法消除了未知非线性的影响。然后,使用命令滤波器解决了传统递归方法的“微分爆炸”难题。理论分析表明:该控制方案消除了未知磁滞输入的影响并且实现了一致性误差渐近收敛至0。最后通过实例仿真验证了文中控制方案的有效性。

输入受限下机器人关节神经网络自适应控制

针对机器人关节控制输入受限以及动力学模型中存在非线性摩擦、柔性变形和未知外部干扰力矩等问题,提出了一种基于径向基函数(radial basis function,RBF)神经网络的输入饱和指令滤波自适应控制方法。基于指令滤波反步法,采用饱和函数约束控制输入的幅值,使用RBF神经网络在线逼近未知干扰,并利用Lyapunov稳定性理论证明了闭环系统的所有误差最终一致有界。仿真结果表明,控制算法不仅使系统的控制输入幅值被严格约束在规定的范围之内,完成了对目标轨迹的高精度跟踪(跟踪误差约为±0.003 rad),而且还可抵抗外部阶跃干扰力矩和建模误差对控制系统的不良影响,保证系统的高精度与强鲁棒性,性能优于PID(propotional integral derivative)控制和普通指令滤波反步控制(command filter backstepping control,CFBC),对机器人关节在高精度领域应用与智能控制具有重要价值。

Ka波段回旋行波管高效率输入耦合器设计

为了实现Ka波段回旋行波管高纯度TE01模式输入,通过在输出口添加滤波结构,对二级功分TE01输入耦合器进行改进,改进后杂散模式传输效率由平均7%下降至2%。该结构主体为一个级联的两级Y型功分网络,能高效地实现矩形波导TE10模式到圆波导TE01模式的转换。在基于传输线理论对功分网络的传输性能理论分析的基础上,借助三维电磁仿真软件进行多次优化迭代,最终得到了一个在31 GHz附近模式转换效率大于99%的宽带TE01输入耦合器,该输入耦合器具有5 GHz的−0.1 dB插入损耗带宽,同时有效频带内的平均模式转换效率高达98.6%,模式纯度在99%以上并且回波损耗小于−15 dB。采用矢量网络分析仪对该器件进行了背靠背冷测实验,结果表明,其带内衰减约为0.5 dB,与仿真计算结果偏差较小,符合工程实际需求。