改性干水灭火剂灭木垛火试验研究

新型干水灭火剂因其高含水量和特殊核壳结构,具有良好的灭火效果。为探究干水对木垛火的灭火效果,自行制备磷酸二氢铵改性干水并开展小尺寸木垛火灭火试验。结果表明：干水灭火剂能够扑灭小尺寸木垛火且不发生复燃,改性干水的控火时间仅为8 s,控火过程中火焰高度持续快速下降,而干粉灭火剂的控火时间则为20 s,喷撒结束39 s后发生复燃;改性干水对火焰区及木垛的温度抑制效果均优于干粉灭火剂,控火时间内,改性干水作用下木垛表面的平均温降速率高达17.00℃/s,是干粉灭火剂作用下平均温降速率的1.94倍;改性干水能够有效降低木垛内部温度,在喷撒50 s内,木垛中心的平均温降速率为8.78℃/s,而干粉灭火剂缺乏冷却作用,木垛中心的平均温降速率仅为6.10℃/s,无法有效抑制阴燃。     

Almost periodic solutions for impulsive fractional differential equations

In this paper, we consider the existence of almost periodic solutions for impulsive fractional evolution equations involving Caputo fractional derivative. The main results are obtained by means of the theory of operators semi-group, probability density functions, fixed point theorems and the techniques based on fractional calculus. An example is also discussed to illustrate the theory. Some known results are improved and generalized.     

On practical construction of electronically controllable compact current amplifier based on commercially available elements and its application

This paper presents a new concept of a current amplifier (CA) with electronically adjustable features (input resistance and current gain) implemented using commercially available devices. Several variations of basic structure are studied, described and the selected solutions verified by PSpice simulations and, moreover, also by lab experiments. Saving one discrete active device in comparison to already known similar topologies of single-input and single-output version (CA-SISO) is significant benefit whereas all other qualitative features remain very similar to previously reported ones. Simple modification of CA concept to multiple-output version (CA-SIMO), having independent adjustable current gain in each of output branches, is easily available. Experimental tests of features of discussed concepts yield adjustability range of input resistance from 100 Ω to more than 10 kΩ and current gain adjusting from 0.1 to 3.5 with bandwidth (−3 dB) of CA transfer response up to 16 MHz. As an example of application, proposed CA was successfully implemented in quadrature oscillator with a very wide tunability range from 29 kHz to 4.94 MHz. It was achieved by electronic control of several adjustable parameters of proposed CA.     

Improved error performance of a 3/4‐Sezginer space‐time block codes

The Alamouti space‐time block code (STBC) achieves full diversity gain at a rate of 1/2. However, the Alamouti scheme does not provide multiplexing gain. The Silver code offers both diversity and multiplexing gain. It has a minimum normalization determinant of . The Golden code is another STBC that offers both diversity and multiplexing gain. The Golden code is ranked higher than the Silver code because of its lower minimum normalization determinant of , however, the golden code suffers from a high detection complexity in the modulation order of M⁴. The 3/4‐Sezginer code is another STBC, which compromises between the Alamouti scheme and the Golden code in terms of diversity gain and multiplexing gain. The 3/4‐Sezginer code achieves full diversity and half of multiplexing gain. The uncoded space‐time labeling diversity (USTLD) is a recent scheme that improves the error performance when applied to the STBC in multiple‐input multiple‐output (MIMO) systems and will be applied to the 3/4‐Sezginer STBC to improve the error performance in this paper. The theoretical error probability for both the 3/4‐Sezginer STBC and the improved system is formulated using the union bound in this paper. The theoretical error probabilities of both 16‐QAM and 64‐QAM are validated through Monte Carlo simulation. The simulation and theoretical results show that the proposed system with 4 N_R can achieve an SNR gain of 1 dB for 16‐QAM and 1.2 dB 64‐QAM at a bit error rate (BER) of 10^?6.     

Optical study of twin‐tanked ICS solar heaters combined with asymmetrical CPC‐type reflectors

The distribution of solar irradiance on the absorbing surface of a typical integrated collector storage (ICS) system combined with reflector troughs is commonly studied by means of ray tracing techniques. A conceptually different alternative is offered by the method of the average number of reflections (ANR). In the present work, the latter is employed for the systematic optical study of realistic ICS models. In all cases, the solar devices consist of twin cylindrical storage tanks which are mounted on top of stationary asymmetrical CPC‐type reflectors. The emphasis of the current research is mainly placed on the evaluation of the ANR reliability for the calculation of the optical efficiency of the related twin‐tanked devices. Additionally, useful operational parameters, such as the optical performance of the proposed geometries, are also determined. The behavior of the tested ICS systems reveals that the optical efficiency may vary in the range of 0.75 to 0.91, exhibiting a strong dependence on the geometric parameters of the solar devices. The highest efficiency is achieved by the systems which combine large reflecting area and storage tanks in close proximity.     

连续时间完全服务与门限服务两级轮询系统性能研究

在信息分组以连续时间规律到达系统的基础上，对于轮询系统中不同优先级的业务问题，提出区分优先级的两级轮询服务模型。首先，在该模型中，低优先级站点采用门限服务，高优先级站点采用完全服务；然后，在高优先级转低优先级时，将传输服务与转移查询并行处理来降低服务器在查询转换期间所耗费的时间，提高轮询系统的效率；最后，运用马尔可夫链和概率母函数的方法建立了系统的数学模型，通过对数学模型精确解析，得到了连续时间两级服务系统每个站点的平均排队队长和平均等待时间的表达式，精确解析出平均排队队长和平均等待时间的值。仿真实验结果表明：理论计算值与实验仿真值近似相等，说明理论分析正确合理。该模型既能保障低优先级站点服务质量，又能为高优先级站点提供优质服务。     

High gain observer for a class of nonlinear systems with coupled structure and sampled output measurements: application to a quadrotor

This paper proposes a new high gain observer for a class of non-uniformly observable nonlinear systems with coupled structure driven by sampled outputs. The considered class of systems is particularly constituted by several subsystems where each subsystem is associated to a subset of the output variables. The observer design is carried out through two steps. First, a high-gain observer is proposed in the continuous-time output case under the assumption that an adequate persistent excitation condition is satisfied by each subsystem. Then, the proposed observer is redesigned to handle the case of sampled outputs leading thereby to a continuous-discrete time observer. The latter property is achieved thanks to the approach pursued along the convergence analysis. The effectiveness of the proposed observer is emphasised in a realistic simulation framework involving a mathematical model of a quadrotor which is diffeomorphic to the proposed class of considered systems.     

结合引导滤波的自适应多曝光图像融合

针对引导滤波产生的光晕、梯度反转现象，以及图像融合边缘细节丢失的现象，提出一种改进引导滤波的自适应多曝光图像融合算法。在引导滤波中根据梯度信息设定权重函数，并结合图像像素点和一定区域的均值创建函数，共同实现不同区域的纹理特性自适应；利用平均亮度与对比度、饱和度及曝光适中度的关系，设置权值函数，使加权平均融合过程中的权重值不再是固定的数值，而能够根据不同的图像亮度自适应调整，权重值也不同，使得融合后的图像质量更好；将原序列图的细节信息叠加到改进的引导滤波图像中，构建纹理细节层。实验结果削弱了光晕及梯度反转现象，使图像更加真实，细节更加清晰，并且对有小光源的图像处理效果更好。算法结果明显优于多曝光融合算法及引导滤波的多曝光图像融合，在信息熵、互信息和边缘信息评价中分别取得最高2.5%、30%和30%左右的质量提升。