改进根轨迹的快速反射镜控制参量设计

为了保证快速反射镜控制系统在时域和频域同时具有良好性能,采用了一种改进根轨迹的控制参量设计方法。首先,根据辨识出的模型,建立音圈电机驱动的快速反射镜传递函数的根轨迹;其次,根据对控制系统性能指标的要求,导出闭环控制系统预期主导极点位置;再次,通过配置控制器的零点位置,使系统的根轨迹具有预期的主导极点;最后,根据模值公式和相角公式,确定系统增益和控制器极点位置。结果表明,采用改进根轨迹法设计的控制系统具有良好的动态性能,还可以有效抑制闭环系统的机械谐振,而且简单易行、通用性强。     

无人机俯仰角增益调参控制律设计

针对传统增益调参控制系统设计过程中离线工作量大的问题,研究设计了无人机俯仰角增益调参控制系统。首先对无人机模型进行线性化和简化处理,得到俯仰角系统模型;其次针对不同特征点处的子模型,利用根轨迹法设计得到阻尼器,利用Ziegler-Nichols以及Signal Constraint方法整定得到PID参数,得到增益调度表;然后利用sftool工具箱生成全局控制器;最后对得到的无人机俯仰角控制系统进行仿真。仿真结果表明,所设计的无人机俯仰角控制系统在全局范围内具有良好的动态响应性能。     

无人机俯仰角增益调参控制律设计

针对传统增益调参控制系统设计过程中离线工作量大的问题,研究设计了无人机俯仰角增益调参控制系统。首先对无人机模型进行线性化和简化处理,得到俯仰角系统模型;其次针对不同特征点处的子模型,利用根轨迹法设计得到阻尼器,利用Ziegler-Nichols以及Signal Constraint方法整定得到PID参数,得到增益调度表;然后利用sftool工具箱生成全局控制器;最后对得到的无人机俯仰角控制系统进行仿真。仿真结果表明,所设计的无人机俯仰角控制系统在全局范围内具有良好的动态响应性能。     

高阶加时滞模型的系统频域辨识

针对工业控制系统中带有时滞环节的高阶过程对象,由于设计控制器比较困难,文中采用二阶加纯滞后模型的频域辨识方法,通过采样对象的输入和输出数据,分析其频率响应并确定重要频率段。选取若干个重要频率响应点,利用幅频特性估算出系统模型参数,达到模型降阶目的。MATLAB仿真结果表明,文中方法适用于广泛的高阶时滞对象,并具有较好的准确性和鲁棒性。     

高阶效应下与扰动频率相关的调制不稳定性

为了探讨光纤中高阶效应下与扰动频率相关的交叉相位调制不稳定性条件和增益谱特点,从包含5阶非线性和4阶色散的耦合非线性薛定谔方程出发,采用线性稳定性分析,计算分析了两光波扰动频率不等时的不稳定性条件和增益谱。结果表明,当2阶、4阶色散同号时,增益谱有两种可能的形式,其中一种的谱范围包含了另一种的。被包含谱的不稳定条件与通常情况不同,且谱特性受色散和5阶非线性的影响小。随着某一扰动频率增大,增益谱的谱区数及变化规律将随两光波所处的光纤色散区的不同而不同。而正(负)5阶非线性可使谱峰增加(减小)、谱宽变宽(窄)。该研究对高重复率脉冲串产生具有一定意义。     

基于模型降阶的分数阶鲁棒控制器

针对复杂高阶被控对象控制器设计及参数整定困难问题,提出了一种基于模型降阶的分数阶鲁棒控制器设计方法。首先将复杂高阶模型近似为含时滞环节的降阶分数阶模型,根据原模型的奈奎斯特曲线特征,结合序列二次规划法,得到降阶近似模型的参数值。在此基础上完成分数阶控制器的结构设计,通过公式推导,给出了最大灵敏度鲁棒性指标与控制器整定参数的新的计算方法,最后结合复合时域性能指标整定控制器参数。仿真结果表明,所设计模型降阶参数求解无需全局寻优,收敛速度快,且降阶模型很好地逼近原系统,设计的鲁棒控制器使原系统具有良好的控制品质。     

自动增益控制系统的非线性分析

自动增益控制(AGC)系统中的非线性是一个突出的问题。和一般的控制系统不一样,它的非线性包含在执行元件中。对于一阶和阶跃输入的情况,本文以状态转移法分析了这种非线性离散系统,并以简单的图解法描绘了状态转移过程。对于高阶和信号幅度任意起伏的情形,则把AGC系统等效为一个易于分析的“标准”的控制系统。这个等效无论对离散系统或连续系统都是严格成立的。文中提出了使AGC系统线性化的方法,它用对数放大器作非线性校正,使具有严重影响的非线性AGC系统变为线性系统。还进而提出了两个互为对偶的应用:一个是利用增益控制原理求出信号幅度的对数值并使之数字化,另一个是用模拟式对数放大器作开环增益控制。     

高阶色散下随入纤功率变化的不稳定性增益谱

为了探讨不同入纤功率下高阶色散对交叉相位调制不稳定性增益谱的影响,从光纤中包含高阶色散的耦合非线性薛定谔方程出发,采用线性稳定性分析法,分二阶、四阶色散系数同号、异号和二阶色散系数为0等几种情形,计算了双光束的交叉相位调制不稳定性增益谱随入纤光功率的变化规律,并对各种谱特性的产生机制作了分析。结果表明,当二阶、四阶色散系数同号时,随着入纤功率的增大,增益谱由开始的两个分离谱区逐渐变宽并合成1个谱区;当二阶、四阶色散系数异号和二阶色散系数为0时,增益谱只有靠近零点的第1谱区,且谱宽和谱峰随着入纤功率的增大而增大。此研究对高重复率的超短光脉冲串的产生有一定的理论指导意义。     

基于系统矩阵有序实Schur分解的模型降阶

对复杂高阶系统进行模型简化，为研究与设计系统提供了方便的条件。对系统模型的降阶是控制系统设计与仿真工作者的一个重要研究课题，对高阶复杂系统的分析、设计、仿真等具有重要的理论意义和工程实用价值。针对目前比较成熟的基于系统矩阵的有序实Schur分解方法，提出了采用时矩输出拟合来改善降阶系统输出响应逼近程度的改进方法，从而使简化模型具有输出误差更小、计算简便等优点，通过给出的实例仿真证实了这一点。     

交流电机变频调速讲座第七讲按转子磁链定向的异步电动机矢量控制系统

前讲表明，异步电动机的动态数学模型是一个高阶、非线性、强耦合的多变量系统，虽然通过坐标变换，可以使之降阶并化简，但并没有改变其非线性、多变量的本质。因此，需要异步电动机调速系统具有高动态性能时，必须面向这样一个动态模型的难题。经过许多学者的潜心研究和实践，首先获得成功和应用的是按转子磁链定向的矢量控制系统。     

An active interference filter as an optical maser amplifier

Approximate theoretical expressions for gain, bandwidth, and root gain-bandwidth are derived by introducing negative absorption into equations for a Fabry-Perot interference filter. Resuits of gain vs frequency calculations are given for a 10-cm maser with high single-pass gain and a 100-cm maser with low single-pass gain. Root gain-bandwidth is shown to be a constant for a given cavity as long as the cavity bandwidth is much smaller than the Doppler or fluorescent linewidth of the maser transition. Frequency drift due to temperature variations and the necessity for precise control of single-pass gain are two practical problems. These must be solved in order to make a stable amplifier with high gain.     

A computational technique for the robust root locus

In this paper, the authors present a modified method for generating multiparameter root loci of both continuous and discrete time systems. A more practical computational method for plotting the robust root locus is developed. The graph of the locations of the poles of the transfer function of the closed-loop system corresponding to each gain can be plotted readily and accurately with this method. The sensitivity of poles to the coefficients of the characteristic polynomial can thus be examined, and the appropriate tuning gain selected to achieve a better robustness. Also, a stricter bound of the zeros of the characteristic polynomial is given to further reduce the complexity of the computation of the robust root locus. The technique is applied to the design of robotic manipulators. Simulations of controller design for the PUMA 762 robot disk grinding process are included     

一种用于1V助听器的低功耗增益控制系统

文中提出了一种用于助听器的低功耗增益控制系统.与传统增益控制系统相比,利用两个MRC电路模块同时实现了自动增益控制和指数增益控制功能,有效地降低了系统功耗.同时为了解决传统设计方法在声音压缩工作状态下功耗增加的问题,提出了一种高效增益控制电路,实现了系统从非压缩状态转到压缩状态时,系统功耗的显著降低.该系统在特许半导体公司0.13μm标准CMOS工艺下流片实现,芯片在1V电源电压下的测试结果表明,芯片的功耗控制在45μW以内,且在600mVp-p输出摆幅下的总谐波失真仅为0.3%.     

Stabilizing High Order Sigma Delta Modulators by Time Delay Feedback Technique

Second and high order sigma delta modulators (SDM) have been extensively used in data conversion processes since their high resolution. However it is difficult to stabilize second and high order SDM over a wide signal range because of the integrals output saturation. A time delay compensation has been proposed in this paper in order to stabilize high order SDM by keeping the integrators output signal levels below saturation limits most of time. The proposed compensation technique is very simple and contains two design parameters; time delay element and compensation gain. It has been shown that high mode jumping leads integrator clipping can be prevented by tuning the compensation design parameters. Particularly w-domain describing function has been employed to tune compensation gain for desired mode/modes in the closed loop. The effect of the compensation gain has been graphically monitored for a constant delay time. Furthermore, the effect of compensation on overall signal-to-noise ratio (SNR) has also been discussed. The proposed compensation technique can be implemented to various configurations of oversampling converters; single-loop and multiple-loop.     

A soft-switched, high-frequency resonant rectifier andcharacteristics of the controlled system

In this paper, a high frequency and soft-switched AC/DC rectifier employing a series-type resonant circuit is proposed to obtain the sinusoidal AC line current. Using the proposed rectifier, the high power factor and low harmonic currents are obtained in the AC line. Furthermore, several advantages such as the high power density and wide output voltage range can also be available. To regulate the output voltage of the proposed circuit, the digital proportional-integral (PI) controller is used whose sampling time corresponds to the zero-crossing point of the AC line voltage. Including this controller, the model for the closed-loop system is developed in order to examine the stability and dynamic characteristics. With this model, the gain of the PI-controller is determined by the pole assignment technique and the closed-loop dynamics is investigated by the root locus plots. A good agreement is made between the simulated and experimental results     

自平衡机器人的机械建模与自平衡算法

针对高阶次、多变量、非线性、欠驱动的单轮机器人系统,提出了一种最优控制算法。首先采用机械方程推导出力学模型,通过将其在平衡位置处进行线性化,得到系统的线性化标称方程,然后用对称根轨迹（SRL）配置极点的控制方法实现了独轮机器人的动态稳定和平衡控制。仿真的结果验证了力学模型的正确性和反馈线性化控制算法的有效性。     

基于FB-CFAR的多径信道阶检测算法

多径信道阶的检测是无线通信参数化信道估计的首要步骤,对系统性能有着重要的影响.然而,目前使用的检测方法大多是基于Akaike信息理论准则(AIC)或最小描述长度准则(MDL)提出的.这些准则在高信噪比条件下性能较好,但在低信噪比时检测性能不好.本文提出了一种基于前后平滑恒虚警率准则(FB-CFAR)的算法来提高信道阶的检测性能.OFDM系统中的仿真结果表明该算法能有效提高低信噪比时多径信道阶的正确检测概率.     

Application of the root locus technique to the closed-loop SO2 pacemaker-cardiovascular system

A previously developed nonlinear model of the pacemaker-cardiovascular system contained a term of the form 1/CO in the forward loop. CO, the cardiac output, is a linear function of heart rate and exercise level. To evaluate the dynamic behavior of the system over a range of pacemaker controller gains, a piecewise linear model was developed. The model was linearized by taking the first two terms of the Taylor's series expansion of 1/CO about the steady state heart rate. A root locus algorithm, with the unique feature of a time delay element in the forward loop, was used to calculate the linear model closed-loop poles as a function of pacemaker controller gain for exercise levels of 25, 50, and 100 W. Step response simulation of the complete nonlinear model showed dynamics close to that predicted from the pole location in the root locus plot. Thus, the root locus technique can be used to quantify systematically the dynamics of the pacemaker-cardiovascular system for different exercise levels and over a wide range of controller gain values.