Admissible consensus and consensualization of high-order linear time-invariant singular swarm systems

Admissible consensus analysis and consensualizing controller design problems for high-order linear time-invariant singular swarm systems are investigated. Firstly, by projecting the state of a singular swarm system onto a consensus subspace and a complement consensus subspace, a necessary and sufficient condition for admissible consensus is presented in terms of linear matrix inequalities (LMIs). An approach to decrease the calculation complexity is proposed, by which only three LMIs independent of the number of agents need to be checked. Then, by using the changing variable method, sufficient conditions for admissible consensualization are shown. An explicit expression of the consensus function is given, and it is shown that the modes of the consensus function can be arbitrarily placed if each agent is R$R$-controllable and impulse controllable and the interaction topology has a spanning tree. Finally, theoretical results are applied to deal with cooperative control problems of multi-agent supporting systems.     

Accuracy analysis and stabilization design of the Fabry–Perot frequency discriminator in double-edge wind lidars

The measurement error of a double-edge wind lidar caused by a disturbed Fabry–Perot interferometer (FPI) is analyzed. Several error sources such as air pressure variations, temperature changes, and mechanical vibrations are considered in the measurement-error model. The simulation results show that a double-edge wind lidar is so sensitive to environmental variations that the measurement error reaches ±60 m/s if the FPI is not stabilized. In order to compensate the external disturbance acting on the FPI, a nonlinear proportional–integral–derivative (PID) control scheme is designed based on the transmission measurement of the calibration channel. An arc tangent function is used to improve the feedback gain of the usual PID control design. The results show that with the new controller the measurement accuracy of the wind lidar can be improved 4–5 times in comparison with the usual control design, and the range of the measurement error is only ±3 m/s.     

利用同步荧光光谱快速鉴别潲水油

为快速鉴别潲水油,采用三维同步荧光光谱结合平行因子法解析潲水油的特征波长差(Δλ),并利用支持向量机建立潲水油鉴别模型。结果表明,潲水油的特征Δλ为60 nm;特征Δλ下的样品原始同步荧光光谱经过主成分分析提取5个主成分,以径向基函数(RBF)为核函数,利用网格搜索和6-fold交叉验证优化建模参数,得到惩罚因子C=512、核参数g=0.5,该条件下建立的模型对训练集和预测集的判别率均达到100%。采用同步荧光光谱可以快速、准确地鉴别潲水油。     

基于多项式模型的混沌系统平方和算法脉冲控制

针对一大类混沌系统,提出一种新颖的基于多项式模型的脉冲控制方法.首先,建立系统的多项式模型.其状态方程由系统状态的多项式矩阵与其单项式组成列向量构成.与其他建模方法相比,该方法不必使用任何预制的假设.其次,提出基于平方和优化算法的脉冲控制方法,使得混沌系统的状态能够实现渐近稳定.基于该算法的脉冲控制与基于线性矩阵不等式凸优化算法的结果相比,能得到更大的脉冲间距,从而可以使用较少的控制能量实现同样的控制效果.最后,仿真实验结果验证了本方法的有效性.     

Bifurcations and chaos control in discrete small-world networks

An impulsive delayed feedback control strategy to control period-doubling bifurcations and chaos is proposed. The control method is then applied to a discrete small-world network model. Qualitative analyses and simulations show that under a generic condition, the bifurcations and the chaos can be delayed or eliminated completely. In addition, the periodic orbits embedded in the chaotic attractor can be stabilized.     

HIRFL数字化高频低电平控制系统研究

在HIRFL加速器系统中, 需要对射频加速电压的幅度和相位进行精确控制,以实现对重离子的精确俘获、 加速和引出。传统的幅度、相位稳定控制系统采用幅度和相位两个反馈闭合环路来分别稳定腔体电压的幅度和相位。 数字化高频低电平控制系统(LLRF) 基于可编程逻辑门阵列（FPGA）和数字信号处理（DSP）, 采用直接数字频率合成（DDS）与数字正交调制解调（I/Q）技术来实现对高频功率源的控制。 相位控制精度更高, 系统更加稳定。 目前控制系统在假负载上通过了长期稳定性的实验和高功率实验, 幅度偏差小于或等于±1％, 相位偏差小于或等于±0.5°。 In order to ensure that the beam quality is well enough, we need to precisely control the frequency, amplitude and phase of cavity electric field. Traditional control system consists of amplitude loop and phase loop. And these two loops control amplitude and phase stability respectively. The digital low level radio frequency (LLRF) system, which uses advanced digital control technology, needs only one feed back loop to control amplitude and phase stability. The phase control precision and stability of the system are higher than the traditional control system. The LLRF system is based on field programmable gate array (FPGA) and digital signal processing (DSP), and implemented by direct digital frequency synthesis (DDS) and digital orthogonal modulation and demodulation (I/Q) technology. The digital LLRF system has been tested in a long term stability and high power experiments. The amplitude deviation is lower than ±1%, and phase control accuracy is within ±1°.      

制冷系统的广义预测控制方法研究

制冷系统的蒸发器过热度的控制对于制冷系统稳定、高效运行具有重要作用。文中通过实验辨识了基于过热度和蒸发温度的制冷系统多输入多输出模型,并采用广义模型预测控制方法,针对制冷系统的变负荷运行工况下进行了实验研究。结果表明,该方法很好的适应了制冷系统变过热度、变蒸发温度的控制要求,相比常规的单输入单输出PID控制,跟踪性能良好、动态控制精度高,可以适应制冷系统负荷大范围变动的控制需求。     

微重力下低温贮箱压力控制技术进展

针对微重力下低温贮箱压力升高的问题,简要介绍了几种常用的低温贮箱压力控制的方法及原理,讨论了它们各自存在的问题,并简要展望未来低温贮箱压力控制技术的发展方向。文章指出,未来研究重点在于将主动制冷与被动绝热相结合,实现零蒸发贮存的热转移技术。     

基于LabVIEW的高功率微波驱动源监控系统

针对高功率微波源对前级驱动电源的计算机自动控制要求,设计了一套1 200 kV高功率微波驱动电源监控系统。系统采用多功能数据采集卡控制大功率调压器产生连续可调的工频电压,再经过环氧高压变压器和整流硅堆等转换为直流高压,对储能电容进行充电控制;采用高速数据输入输出卡控制触发系统按时序进行工作;监测计算机通过RS-485串口方式对直流高压、闸流管阳极电压、闸流管对地电流等进行实时状态监测;控制主机通过以太网与中央控制计算机实现通讯,可以单独控制电源工作,也可以通过中央计算机统一协调工作;采用LabVIEW作为软件开发平台,利用图形控件完成整个电源系统的控制监测等功能的设计;为了解决因电磁干扰强而引起的地电位抬高、高压采集不正常等问题,系统的软硬件都融入了可靠性设计。实验结果表明,该系统工作可靠稳定,实时性强,界面友好,操作简单,具有良好的可扩展性和移植性。     

火花通道长度对激光触发开关特性的影响

利用超高速相机,在曝光时间100 ns时,对不同能量下激光在氮气气体开关中形成的火花通道进行拍照,得到了不同焦距下激光波长266 nm时激光火花通道长度与激光能量的关系。开展了激光触发气体火花开关的实验研究,激光触发开关延时、抖动随激光能量的增加而减小。将火花通道长度与激光触发开关的特性进行了分析,气体开关的抖动随着激光火花通道长度的增加而减小,当火花通道沿电极间轴向长度达到开关电极间距40%时,开关的抖动为亚ns量级。