基于深度学习的航拍光伏板红外图像热斑检测方法研究

针对光伏电站光伏板热斑故障难以检测的问题,结合无人机巡检技术,提出一种基于深度卷积神经网络的光伏板热斑快速检测方法。首先设计了光伏板识别模型,将Yolov4主干特征提取网络替换成轻量级网络MobileNetV2,并将PAnet网络中标准3×3卷积替换为深度可分离卷积,实现了将光伏板快速从红外图像中识别出来。为快速识别热斑并解决光伏板反光噪声问题,将MobileNetV2网络引入DeeplabV3+模型中,改进由于下采样造成的目标缺失,并将交叉熵损失函数修改为Dice损失函数来进一步提高分割精度。试验结果表明,该方法能够准确识别光伏板热斑,光伏板识别准确率为99. 56%,检测速度为22. 1帧/秒。光伏板识别后的热斑分割准确度达到95. 99%,交并比mIou达到85. 58,检测速度为24. 5帧/秒,该方法能够满足光伏板故障检测的需要。     

独立储能电站参与电力现货市场机制及试运行分析

随着风光新能源装机容量迅速增长,加快健全储能设施参与电力现货市场交易机制有利于促进新能源消纳。为提高系统调节能力,以山东省独立电储能参与现货市场交易机制为研究对象,提出了独立储能以自调度模式参与现货市场的运行机制。仿真运行数据表明,储能装置参与现货市场对促进新能源消纳和提升电力供应保障能力具有重要作用。     

位置敏感探测器光斑脱靶误差补偿方法研究

为了提高位置敏感探测器(PSD)的位置检测范围,解决光斑在探测器光敏面脱靶时无法准确定位光斑的问题,提出一种光斑脱靶误差补偿方法,分析了脱靶前后PSD检测光斑能量重心变化规律,建立了PSD光强信号与位置检测误差间的函数关系。实验结果表明:当PSD光敏面尺寸为12mm×12mm时,对于半径为5mm的高斯光斑,通过所提出的光斑补偿方案补偿后PSD的X轴检测范围提高了66.7%,位置检测平均相对误差不超过5%,该方法对提高PSD位置检测性能具有重要的意义。     

发用两侧参与调峰的现货市场联合出清模式设计

针对新型电力系统下新能源装机容量大,发电侧调峰资源紧缺,弃风、弃光现象严重的问题,文中设计了一种考虑发用两侧参与调峰的现货市场联合出清模式,将发电侧深度调峰、柔性负荷调峰、新能源消纳与现货市场相结合,充分发掘发用两侧的调峰能力。首先,将独立储能和用户作为用户侧调峰资源,并根据用户是否具有实时跟踪调度指令的能力,将其分为灵活调节用户和非灵活调节用户,在此基础上设计了发用两侧参与调峰的日前市场和实时市场出清模式并建立出清模型。其次,综合考虑各市场主体的贡献和受益程度,设计了调峰辅助服务费用的结算与分摊机制。最后,通过算例验证了所提市场模式对于新能源消纳的有效性和合理性。算例分析中,文中所提市场模式相比仅发电侧参与调峰的模式可降低17.3%的弃风弃光量。     

Theoretical analysis of shunting effect in resistance spot welding (RSW) of AA2219

Shunting effect is the loss of electrical current via the secondary circuit provided due to existence of previous nugget in a series of welding spots. This phenomenon is important for products containing intermittent spots. In this study, a theoretical model is developed for shunting effect in resistance spot welding for aluminum alloy 2219. Welding distance together with welding current and time is included in the model to analyze the shunting effect on final nugget quality. Thermal and electrical interactions are considered in the model as well as geometrical aspects of the process. Temperature dependence of material properties, integration through the whole volume of the nugget and heat-affected zone, and assuming a simple cooling mechanism are the most important modifications designated in the proposed theoretical model. Predicted results of nugget diameter are compatible with experimental and finite element analysis results.     

6082-T6铝合金回填式搅拌摩擦点焊接头的疲劳性能的有限元分析

利用有限元分析软件ANSYS模拟了6082-T6铝合金回填式搅拌摩擦点焊接头在不同载荷作用下的应力场,进而分别运用名义应力法和局部应力-应变法计算得到当前载荷水平下的疲劳寿命。结果表明:在较大的载荷水平下,应力法分析结果与试样实际寿命接近,在较小的载荷水平下,局部应力-应变法分析结果与试样实际寿命接近;试样疲劳寿命随焊点间距的增加而减小。通过分析6082-T6铝合金回填式搅拌摩擦点焊接头疲劳断口形貌,发现疲劳裂纹起裂于焊点根部的热影响区和热力影响区,同时沿上板焊点边缘和下板母材处横向扩展。     

一种同步输出的稀疏树前导零检测电路

为了降低前导零检测电路的延时和提高端口输出的同步性,提出了一种稀疏树前导零检测结构及动态电路的实现方法。通过递归前导零算法表达式定义了两种新的布尔运算逻辑,以构成稀疏树中的结点。精确控制动态电路中预充时钟的偏移量,在抑制电路漏电的同时控制输出端口延时差。该结构能够以最少的逻辑级数和均衡的运算单元负载实现检测前导零功能,对于不同数量的待测信号,通过扩展"点操作"和"块操作"单元互连网络来生成新电路。采用伪随机激励向量验证了电路功能的正确性,大幅缩短了验证时间,在SMIC 40nm工艺下仿真显示,各端口输出延时差小于1ps,数据路径长度可降低20%。     

A study on the effect of polishing fluid volume in ball end magnetorheological finishing process

Ball end magnetorheological finishing is a unique process that utilizes a magnetically controlled ball of polishing fluid at the tip of the rotating tool to finish workpiece of different materials and shapes. The aim of this research is to study the effect of polishing fluid volume on finishing spot size and the surface finish associated with it. A magnetostatic simulation is done to find the variation of flux density in the working gap and on the workpiece surface. The maximum limit of the polishing fluid volume is selected on the basis of area of threshold magnetic flux density (minimum value required for finishing) region on the workpiece surface. The surface characteristics and the diameter of the finished spot are analyzed by varying the fluid volume. The surface obtained with high fluid volume is poorly finished and has scratch marks as the excess fluid flows out from the working gap and forms a thick ring at the periphery of the tool tip. Contrary to this, if the fluid volume is too less, then it merely rotates over the workpiece surface without causing any finishing action. An optimum range of fluid volume produces a good quality surface finish with constant finished spot size.     

Oil spill detection from SAR intensity imagery using a marked point process

This paper presents a new algorithm for the detection of oil spill from SAR intensity images. The proposed algorithm combines the marked point process, Bayesian inference and Markov Chain Monte Carlo (MCMC) technique. In this paper, the candidates of oil spills or dark spots in a SAR intensity image are characterized by a Poisson marked point process. The marked point process is formed by a group of random points (as a point process modelling the locations of oil spills) and a set of parameters including geometric parameters of windows centred at the random points and gamma distribution parameters (as the marks attaching to each point). As a result, the candidates of oil spills are represented by a group of windows, in which the intensities of pixels follow independent and identical gamma distribution with lower mean than that for the identical gamma distribution of the pixels out of windows. Following the Bayesian paradigm, the posterior distribution, which characterizes the locations and statistical distributions of oil spills, can be obtained up to a normalizing constant. In order to simulate from the posterior distribution and to estimate the parameters of the posterior distribution, the Revisable Jump MCMC (RJMCMC) algorithm is used. The optimal locations and sizes of dark spots are obtained by a maximum a posteriori (MAP) algorithm. The proposed approach is tested using Radarsat-1 SAR images with oil spills indicated by human analysts. The results show that the proposed approach works well and is very promising.     

以课程知识点为基础的专家网络考试系统设计

目前计算机网络考试系统已经成为各种教育、培训机构进行学业考核的一种重要方式;但组织方式大多存在自培自考,考试内容随意性大,管理松散。如何实现学院范围内,甚至是跨校选修课程在教务管理侧面的教考分离及统计管理,解决考试内容的专业性、公正公平性,解决由于考生众多,机房不足带来的考试多时段、多场次性及试卷保密性,防范考试过程中的作弊、代考及考试中特殊意外处理。通过课程专家组以课程知识点为基础的题库建设,以教务考试工作流程为核心管理,及以学生考试过程为核心控制的三方综合网络考试系统是解决上述问题的一种方法。