With the widespread adoption and benefits of the Internet of Things (IoT), this technology has expanded its reach into the realm of automation and control, giving rise to a branch known as industrial networks. The use of IoT technology is on the rise due to its daily increasing advantages and ease of use. However, it has also introduced new challenges in control systems, prompting researchers to conduct numerous studies to address these challenges and propose innovative solutions. One of these challenges, for example, is the limited bandwidth of the network, which leads to challenging issues such as variable random time delays and packet loss. In light of these challenges, using networked control systems instead of many traditional control methods can significantly improve the performance and stability of closed-loop systems. In this context, this paper aims to enhance system tolerance and reliability, along with permanent state error correction, by presenting methods for designing networked controllers that are influenced by random delays. To achieve this goal, we first propose a predictive control model with reduced prediction horizons, assuming constant delays in the input. This controller not only ensures stability but also tracks the reference input. Simulation results demonstrate that this controller can tolerate some variations in delay despite being designed for constant delays. However, it is not robust against significant variations in delay inherent to network environments. Therefore, in the second approach, a controller based on the Lyapunov function and linear matrix inequalities is designed for a system with time-varying input delays. This controller, in addition to stabilizing the system, tracks the reference input by minimizing the state error.

In order to improve the efficiency of resource allocation in the power Internet of Things, a micro-service model for power Internet of Things was proposed by edge computing Business data was divided into 4 different types, and resource matching calculation for aggregated services was completed by edge computing. Computing resource allocation model was used to obtain the computing resource allocation and resource supply of the edge computing terminals. In the test experiment, the CPU resource occupancy curve was obtained in the condition of different resource scaling ratios. It verified the feasibility of the algorithm. The resource utilization ratio of simultaneous trigger service and non-simultaneous trigger service was compared and analyzed. The results show that it verifies the optimization effect of this algorithm, which can reduce the total cost.

针对油井巡检机器人与障碍物的接触率高,造成设备故障率高增加石油生产成本问题,提出基于地图加权的遗传算法.首先将地图进行栅格化,建立栅格地图模型,并进行加权设置.其次引入遗传算法模型进行路径规划,将每次路径规划结果存入染色体中并计算路径长度,最后筛选最大权值中的路径最短染色体,并绘制路线.在参数设定相同的条件下,采用基于地图加权的遗传算法、经典遗传算法进行比对实验,仿真结果表明,基于地图加权的遗传算法优先选择了不靠近障碍物的栅格的情况下完成了路径规划任务,机器人与障碍物的接触率下降了74.91％,时间和路程仅增加0.3179 s与32％.

In order to improve the state monitoring capability of power devices and simultaneously obtain temperature and dynamic and static stress responses, a state monitoring system was designed based on optical fiber sensor network. The chirped grating modulation technique was used on the Fiber Bragg Grating sensor package to make the temperature and strain response of the test position different, so that the simultaneous solution of the two parameters was realized by the decoupling algorithm. On the basis of constructing a two-parameter decoupling model, the trends of temperature and strain with respect to wavelength were simulated and analyzed. In the experiment, the state of the power device was tested and analyzed from three aspects. In the temperature calibration experiment, the temperature responds linearity reached 0.9986, and the deviation of the comparison test result with the FOT2300 temperature detector was better than 1.3%. In the static strain response experiment, the system sensitivity in the range of 0.1kN-1kN was 0.0036nm/N, and the system sensitivity in the range of 1kN-5kN was 0.0018nm/N. In the dynamic response experiment, the impact of heavy objects at different heights had obvious peak differences. The wavelength offset was inversely proportional to the time half-width, and the spectral line shape had obvious characteristics. It can be seen that the system has good distinguishability when obtaining the state information of power device temperature and static and dynamic stress.

现有的步进电机运动可逆式隔爆机构结构复杂、功耗大,基于滑块继续运动的可恢复隔爆机构无法再次恢复为待发状态.针对上述问题,提出一种单向轴承电机驱动的可逆式引信隔爆机构.利用单向轴承的运动特性,使步进电机正转驱动导爆筒直线运动,反转驱动导爆筒旋转运动;直线运动确定导爆筒的位置,旋转运动确定导爆筒的角度,根据指令信息即可完成爆炸序列的对正或使爆炸序列处于隔爆状态.验证结果表明,该隔爆机构可以实现引信延迟解除隔离,可以实现引信在待发状态与安全状态之间转换,可显著提升隔爆机构的测试性,并且具有成本低、结构简单等优点.

Today, due to increasing environmental considerations and energy demand, hydrogen fuel can be a suitable alternative energy carrier. This fuel can be produced from technologies based on fossil fuels (such as gasification and fossil fuel reforming), and/ or renewable sources (such as water electrolysis). The method of using water electrolysis technology to produce hydrogen energy along with the use of clean energy sources can be the most popular method due to the non-emission of polluting gases. In addition, such an approach can be fruitful in order to achieve development and sustainable society. The proposed energy system, which includes solar photovoltaic, solid oxide electrolyzer, dish collector, and solid oxide fuel cell, has not yet been evaluated in the literature. In other words, the components of the introduced hybrid system have new relationships and structures. The introduced energy system can produce hydrogen fuel and thermal and electrical energies. The main feature of the present paper is the use of clean solar energy to supply both the electrical and thermal energy required by the electrolyzer process to produce hydrogen fuel. The findings indicate that the introduced hybrid system can generate 217.3 kWh of electricity per day with an efficiency of approximately 50.5%. Furthermore, the energy cycle is capable of producing 4.34 kg/day of hydrogen fuel.

At present, the emergency search Unmanned Aerial Vehicle (UAV) cluster has problems such as low search efficiency, low coverage integrity, and poor stability of multi-unit network. In this regard, a terminalrouting UAV area search task planning strategy based on Optimized Fuzzy C-Means Algorithm (O-FCMA) combined with Optimize-Hybrid Particle Swarm Optimization (O-HPSO) algorithm is proposed. Based on the scope of UAV monitoring area, by establishing the spatial model of the search area, this paper further uses 0FCMA to divide the area geometrically, and uses O-HPSO to realize the path planning in the divided area, so as to realize the planning of the overall task of multi-UAV cluster search. The simulation experiment is performed. The use of O-HPSO combined with O-FCMA for passive UAV area search task is compared with ACO or simulated annealing algorithm combined with K clustering algorithm or FCMA, under the condition of ensuring the full coverage of the search area, active search compared with non-active UAV, the UAV decision time is reduced by 7%similar to 21%, 16%similar to 31%, and the search efficiency is increased by 7%similar to 13%, 3%similar to 7%. The method reduces effectively the UAV cluster decision time and improves the search efficiency.

传统水田粗放型灌溉不仅对作物生长带来不利影响,而且不能够充分利用自然降水,在很大程度上造成了水资源的浪费.该研究设计了基于智慧农业技术的网络化水田作物精准灌溉系统.建立了通信节点最优部署模型、作物耗水预测模型、降水预测模型、最优化灌溉决策模型以及基于模糊控制理论的精准灌溉决策系统;提出了基于维诺图改进的飞蛾扑火优化算法(Voronoi Moth Flame Optimization,VI-MFO)的灌溉网络通信节点优化部署方法,以提升灌溉网络通信效率并降低通信能耗;最后,将水田状态信息及气象参数作为精准灌溉决策系统输入,经过系统决策,自适应控制水田灌溉设备进行精准灌溉.对江苏地区水稻田进行仿真,结果表明,所提出的智能灌溉系统与传统非智能决策系统相比,灌溉设备动作频次降低26.67％,灌溉量减少40.82％,排水量减少33.89％.所提出的智能灌溉决策系统节约了水资源.

针对青少年长时间使用电子产品导致视力下降的问题,提出一种基于模糊控制的智能感控视力保护仪;该保护仪的控制系统综合运用了传感器技术、数字信息处理技术、模拟数据处理技术以及智能算法,实现了软硬件的结合;针对检测过程中距离和光照强度难调节的问题,引入了模糊控制算法,通过模糊控制规则来调节距离和光照,并进行了仿真实验,将采集到的光照强度和检测距离作为输入量,定时时间作为输出量,达到提醒学生的目的 ;对比实验结果表明,模糊控制比PID控制的定时时间增加了10.6 min,相对时间增大了18.563％,能够为学生创造更良好的学习环境.

针对超宽带(UWB)技术在AGV室内定位方面数据传输受多径效应干扰大且非视距影响严重、传统测距算法测量得出的结果无法避免算法运行和硬件响应时间带来的误差等问题,提出一种基于神经网络预测的双边定位算法.首先在单边双向测距模型的基础上引入双边双向测距模型,其次采集非视距条件下的测距误差序列,并输入反向传输神经网络预测模型中,得到定位标签的误差改正值,最后引入回归方程系数和中位值滤波算法对改正值进行修正.通过对比实验可知,引入回归方程系数和中位值滤波算法后,视距测距可以将平均误差控制在5 cm以内,非视距中基于神经网络预测方法得出的定位坐标的相对欧氏距离比基于多面函数拟合模型方法缩小了10 cm左右,精度更高.

文章对钢琴的实时检测进行了研究.采用物体遮挡光电传感器的方式,对琴键进行检测,可以使检测系统更加实时地对琴键信息进行反馈.利用单片机进行数据的存储和记录,将学习者的演奏琴键顺序音符传输给评估系统,使数据通过串口通信传递到LED屏幕上,使钢琴学习者能够更加方便、便捷地进行钢琴学习.

对于一个农田占地面积大,灌溉用水量少的农业国而言,高效用水影响着农业的稳步发展;针对这一现象,提出一种基于模糊RBF神经网络的灌溉策略,将隐层的输出函数构成一组基来逼近目标函数,预测出灌溉的需水量,并用模糊控制计算出具体的灌溉时间;以山西省忻州地区种植的玉米利合328号为例,对气象条件、土壤条件以及作物特性进行分析,可以根据不同环境计算出灌溉时间;对大田采取高效、实时的灌溉策略,不仅可以提高灌溉用水的利用率,还可以为大田灌溉的发展奠定基础,增加了作物的产量.

为实现MEMS技术加工的微桥膜低能化,设计不同结构和材料的桥膜,仿真研究桥膜电流密度、温度的变化过程;通过研究方形桥区对角线的电热特性曲线,以解决电流与H型微桥膜材料的匹配问题.结果表明:微桥膜两端输入电流为500A时,H型桥膜桥区温度分布最均匀;相比铜微桥膜,金桥膜达到熔点所需的时间短,但是平均升温速率低;铜微桥膜的熔化面积大、分布更均匀.在300～700A输入电流范围内,对于铜、金、铝H型微桥膜,输入电流依次为500A、450A、400A时起爆可以提高它们的能量利用率.

In the paper,in view of the current community residents temperature detection research，an Internet community residents temperature detection system is proposed,for the Arduino board to collect the body temperature information NB-IoTto transport connection cloud server, database building community at the same time，the daily temperature data collected “cloud” and more reduce the cross in fection，which realize detection equipment more convenient，higher，universal better physiological parameters acquisition，achieve the precise control.

针对5G背景下大规模智慧农业传感器网络通信节点在三维空间内部署时,传统的随机节点部署方式存在高能耗、高成本及节点脱节问题,提出了一种基于维诺图(VD)与飞蛾扑火算法(MFO)结合的节点优化部署策略.首先,利用5G通信高带宽的特点,设计了一种分层式通信方案,在此基础上,建立了分层式节点最优部署模型、通信能耗模型以及传感器信息传输网络全连通模型;其次,针对MFO算法种群生成的随机性及搜索无向性的缺陷,设计了基于三维VD的引导式搜索算法(VIMFO);最后,通过对相关模型进行优化求解,从而取得传感器网络通信节点的部署位置.仿真结果表明,与经典MFO算法相比,VIMFO算法能够在提升寻优速度37.89％的基础上,降低传感器通信网络能耗6.9％;在保证农田传感器通信网络节点全连通的前提下,提升了通信网络节点的生存周期.

针对大面积水域内,多水质检测机器人编队通信网络节点优化部署问题,提出基于免疫遗传算法的解决策略.通过建立被监测水域的划分模型、机器人编队通信网络的全联通模型、通信网络节点部署及能耗模型,进而对模型进行免疫遗传优化求解,以获得节点的部署位置.仿真结果表明,上述方法在保证水质监测机器人编队通信网络全联通的前提下,完成了所指定的水域内的网络节点的优化部署,降低了通信网络的能耗,具有实际的应用价值.

针对利用增雨弹进行人工增雨过程中,现有的基于单门限的端点检测算法的方法难以准确检测出声音信号端点,无法对增雨弹进行准确定位的缺陷.提出一种新的双阈值端点检测算法,首先利用巴特沃斯滤波器对声音信号进行去噪,初步去除干扰.然后利用声信号短时特征,设置端点检测阈值参数.最后,将4个感兴趣段进行提取,进而获得增雨弹炸点的准确位置.实验结果表明,所述方法能够有效提取感兴趣段的起始点与结束点,符合实验预期与后续实验要求,具有实际的应用价值.

For the pedestrian sequence collected by the video surveillance system under the side view, the feature recognition technologies such as face recognition and iris recognition are difficult to identify, the gait recognition technology based on human joints location is studied. To locate the joints of the human body, a method based on the contour and skeleton of the human body is proposed. Firstly, the improved ZS refinement algorithm is used to extract the human skeleton, and then the contour is combined with the skeleton to locate the joints. The joint angle is calculated as the gait feature, and the gait features of different individuals are classified and identified by support vector machine. The identity verification and method comparison on CASIA gait database of Chinese Academy of Sciences show that the method has better recognition rate and is robust to the state of wearing coat and backpack.

针对现代无线通信系统的多标准多体制要求,高效、快速并相对低成本地对大量信号和数据进行处理变得越来越重要,软件无线电技术由其易修改和低成本的优势,把无线通信技术提升到一个软件化易扩展的新高度.利用通用的软件无线电平台,在GNU Radio编程环境中完成了基于HackRF的无线通信收发系统.阐述了GNU Radio编程环境及HackRF的功能及构成,随后在对GMSK调制系统建立基于单机的实验仿真并成功完成的基础上,以GNU Radio+ HackRF为模型,进一步设计并实现了无线视频流的传输,从而实现了基于软件无线电平台的无线通信.从实验结果得出,此系统可以较好地发送并接收视频流信号.

针对传统的局部二值模式(LBP)算法在提取表情图像特征时会遗漏掉某些关键的局部信息这一特点,提出了一种改进的LBP算法用于表情识别.首先在传统的LBP算子的基础上用中心像素与邻域像素灰度值做差值运算取其绝对值,并将每个绝对值相加再计算平均值然后把平均值作为阈值记为M,该阈值作为以后LBP编码时的参数.为了获得更多有效的特征信息,决定对表情图像的每个区域进行顺时针和逆时针两次编码,得到不同的两个特征直方图,再进行整合.最后采用支持向量机(SVM)进行分类识别.在JAFFE数据库和CK+数据库上实验所获得的结果表明,所采用的方法优于传统的LBP算法.