Effects of hydraulic retention time,temperature, and effluent recycling on efficiency of anaerobic filter in treating rural domestic wastewater

With rural population expansion and improvement of the socio-economic standard of living, treatment of rural domestic wastewater has rapidly become a major aspect of environmental concern. Selection of a suitable method for treatment of rural domestic wastewater depends on its efficiency, simplicity, and cost-effectiveness. This study investigated the effects of hydraulic retention time(HRT), temperature, and effluent recycling on the treatment efficiency of an anaerobic filter(AF) reactor. The first round of experimental operations was run for three months with HRTs of one, two, and three days, temperatures of 18℃, 21℃, and 24℃, and no effluent recycling. The second round of experimental operations was conducted for another three months with HRTs of three and four days; temperatures of 30.67℃, 30.57℃, and 26.91℃; and three effluent recycling ratios of 1:1, 1:2, and 2:1. The first round of operations showed removal rates of 32% to 44% for COD, 30% to 35% for TN, 32% to 36% for +4NH-N, 19% to 23% for 3NO-N-, and 12% to 22% for TP. In the second round of operations, the removal rates varied from 75% to 81% for COD, 35% to 41% for TN, 31% to 39% for +4NH-N, 30% to 34% for 3NO-N-, and 41% to 48% for TP. The average gas production rates were 6.72 L/d and 7.26 L/d for the first and second rounds of operations, respectively. The gas production rate increased in the second round of operations as a result of applied effluent recycling. The best removal efficiency was obtained for an optimum HRT of three days, a temperature of 30℃, and an effluent recycling ratio of 2:1. The results show that the removal efficiency of the AF reactor was affected by HRT, temperature, and effluent recycling.     

On consensus algorithms design for double integrator dynamics

This paper considers the consensus problem of double integrator multi-agent systems where: (i) each agent is subject to input saturations, and (ii) the velocity (second state) of each agent is not available for feedback. We present new consensus algorithms that handle simultaneously the above mentioned situations. Sufficient conditions are derived such that consensus algorithms developed for first- and second-order multi-agent systems in ideal situations can be used to account for input saturations and remove the requirement of velocity measurements. To illustrate the effectiveness of the proposed approach, we propose solutions to two different second-order consensus problems in the case where the input is saturated and the velocity states are not available for feedback and simulation results are provided in each case.     

Finite-time tracking control for multiple non-holonomic mobile robots based on visual servoing

This paper investigates finite-time tracking control problem of multiple non-holonomic mobile robots via visual servoing. It is assumed that the pinhole camera is fixed to the ceiling, and camera parameters are unknown. The desired reference trajectory is represented by a virtual leader whose states are available to only a subset of the followers, and the followers have only interaction. First, the camera-objective visual kinematic model is introduced by utilising the pinhole camera model for each mobile robot. Second, a unified tracking error system between camera-objective visual servoing model and desired reference trajectory is introduced. Third, based on the neighbour rule and by using finite-time control method, continuous distributed cooperative finite-time tracking control laws are designed for each mobile robot with unknown camera parameters, where the communication topology among the multiple mobile robots is assumed to be a directed graph. Rigorous proof shows that the group of mobile robots converges to the desired reference trajectory in finite time. Simulation example illustrates the effectiveness of our method.     

Global decentralised stabilisation for a class of uncertain large-scale feedforward nonlinear systems

In this paper, the problem of global decentralised stabilisation for a class of uncertain large-scale feedforward nonlinear systems is investigated. The system under consideration is allowed to contain unknown non-Lipschitz continuous nonlinear terms. The design of the global decentralised controllers takes a two steps procedure. First of all, based on the adding a power integrator technique and the homogeneous domination approach a local homogeneous decentralised controller is proposed for each subsystem of the large-scale feedforward nonlinear system. Then, we integrate a series of nested saturation functions with the homogeneous decentralised controllers and adjust the saturation levels to ensure globally asymptotic stability of the closed-loop system. Simulation studies are conducted to illustrate the effectiveness of the proposed control method.     

General parallel structure digital repetitive control

This article presents a parallel structure digital repetitive control (PSDRC) scheme, where the internal models of all harmonics are decomposed into multiple parallel connected groups. Compared with conventional repetitive controller, the proposed parallel structure enables repetitive controller to regulate the error convergence rate at each group of harmonic frequencies independently and offers faster total error convergence rate. Moreover, PSDRC can achieve zero-error tracking or perfect disturbance rejection for all harmonics without occupying more data memory. A stability criterion with rigorous proof for PSDRC system is addressed, which is compatible with those existing stability criteria for existing RC schemes. An application example of three-phase pulse-width modulation inverter is provided to demonstrate the effectiveness of the proposed PSDRC scheme.     

一种用于AFS的半实物仿真平台

为了给自适应前照灯系统（Adaptive Front-Lighting System,AFS）提供一个控制算法验证平台,开发了一套基于汽车驾驶模拟器的AFS半实物硬件仿真平台。在该仿真平台中,PC机采集驾驶模拟器的档位、油门、离合、刹车、方向盘等驾驶信息,由车辆动力学模型模拟实际车辆行为,采用AFS控制模型计算车灯转角控制量,并发送至电机驱动模块进而控制车灯转动,同时将车灯转角信息反馈至PC机。实验表明,该系统能实时记录和显示AFS工作过程中的多种参数,便于进行实验观察和数据分析,从而为AFS系统的控制算法提供一个检验和修正的平台。     

基于粒子群优化神经网络的语音情感识别

提出了一种基于粒子群优化算法的人工神经网络,并把它应用到语音情感识别系统中。依据情感的维度空间模型,分别提取了韵律特征与音质特征,研究了谐波噪声比特征随情感类别的变化。利用粒子群优化算法(PSO)训练随机产生的初始数据,优化神经网络的连接权值和阈值,快速地实现网络的收敛。在实验中比较了BP神经网络、RBF神经网络与PSO神经网络分别用于语音情感识别的识别率,PSO神经网络的平均识别率高于BP神经网络6.7%,高于RBF神经网络5.4%。结果显示,粒子群优化神经网络用于语音情感识别提高了识别性能。     

协同OFDM系统联合中继选择、子载波配对和功率分配算法

研究了协同OFDM系统中的资源分配问题,提出了一种联合中继选择、子载波配对和功率分配的算法.首先在假定等功率分配的情况下,对中继节点进行选择并对子载波进行配对,然后在此基础上对选定的子载波对进行功率分配.分析结果显示所提算法是一种计算复杂度比较低,且系统容量接近最优容量的算法.     

基于多尺度-多形状HOG特征的行人检测方法

提出一种图像中人体快速自动检测方法。提取图像的多尺度-多形状方向梯度直方图（HOG）特征向量,用于描述人体的形状特征,结合Adaboost机器学习法训练级联型分类器,以加速人体的检测过程。相比较传统算法,该方法没有采用静态背景模型,也不是仅仅依赖于易受外部环境因素干扰的颜色信息,从而一定程度地适应了人体姿态变化,以及非结构化环境下常见的光照波动、背景杂乱等不良因素所带来的干扰。实验验证了该方法的准确性和较高的计算效率。     

Integration of transparent glass window with LTCC technology for μTAS application

The LTCC substrate makes it possible to build various microsystems which integrate not only passive components such as resistors, capacitors and inductors but also 3D structures such as cavities and channels. Nevertheless non-transparency is a main limitation of the LTCC-based microfluidic systems. The goal of this paper is to present technology which allows an optical transparent element to integrate with LTCC co-firing process. A micrototal analysis system (μTAS), which is based on the LTCC–glass technology, enables optical measurements. The study shows that integration of sodium glass material is feasible not only with zero-shrinkage LTCC (HL 2000, HL 800) but also with a standard one (DP 951). A FEA (finite element analysis) is used to calculate stress inside the LTCC–glass structure. A series of LTCC–glass windows with different sizes and shapes is investigated to observe size limitation of the integration method. The example ceramic–glass structures (chambers, mixer) with glass windows are made in order to present the possibilities of this new technology.