Loads Equalization Based on Optimization Principle and the Hierarchical Model for Distribution Networks

Loss reduction is important for distribution networks and it is available to reduce loss by loads equalization. With increasing numbers of vertexes in distribution networks, compounding of loads are rapid increasing. It is hard to achieve the global optimization for load balancing by traditional model of distribution networks. Base on analyzing character of loads equalization for distribution networks, this paper provides a new method which conjugate the optimization principle and the hierarchical model of the distribution networks, can transfer the optimization issues into multi stage decision making problems,and can also achieve the optimization equalization of the distribution networks loads.     

Summarization of Current Research on Application of Durability of Structures

This paper summarized the recent research results about the durability of structures, introduced their current application in durability design, appraisal, estimation of rest life and strategy of maintenance and strengthening as well as in other fields.     

Seismic Behavior of Asymmetric Building under Pushover Analysis

Pushover analysis is a tool to study the inelastic behavior of structures under strong seismic action. The 3D-pushover analysis was performed to analyze the inelastic lateral-torsional coupled response of multistory asymmetric structures. The symmetric frame and frame-wall systems were studied in this paper, and the capability curve of the top displacement and base shear,the story drift and the position of rotation center were compared with reference to corresponding symmetric structures. The characters of inelastic response of multistory asymmetric structures were found. The conclusion provides reference for performance-based seismic design of asymmetric structures.     

基于随机森林回归算法的小麦叶片SPAD值遥感估算

使用机器学习中的随机森林(RF)回归算法构建小麦叶片SPAD值遥感反演模型。以2010—2013年江苏地区试验点稻茬小麦3个生育期(拔节、孕穗、开花)的叶片为材料,结合我国自主研发的环境减灾卫星HJ-1对研究区域进行同步监测,分析了各生育期叶片SPAD值与8种植被指数间的相关性;以0.01水平下显著相关的植被指数作为输入参数,使用RF回归算法构建了每个生育期的小麦SPAD反演算法模型,即RF-SPAD模型,以支持向量回归(SVR)和反向传播(BP)神经网络算法构建的SVR-SPAD模型和BP-SPAD模型作为比较模型,以R2和均方根误差(RMSE)为指标,分析了每个生育期3个模型的学习能力和回归预测能力,结果表明:RF-SPAD模型在3个生育期都表现出最强的学习能力,R2和RMSE在拔节期分别为0.89和1.54,孕穗期分别为0.85和1.49,开花期分别为0.80和1.71;RF-SPAD模型在3个生育期的回归预测能力都高于BP-SPAD模型,高于或接近于SVR-SPAD模型,R2和RMSE在拔节期分别为0.55和2.11,孕穗期分别为0.72和2.20,开花期分别为0.60和3.16。     

高压旋芯喷嘴的雾化特性

对直射式喷嘴进行加旋芯改进，改进后喷嘴的喷雾情况符合不稳定性理论规律。利用三维激光多普勒粒子测速仪（3D—PDPA），测量了单相流细水雾喷嘴在1m处截面的雾化特性参数；研究了喷嘴直径和喷射压力对雾化特性的影响。试验结果表明，喷雾压力大于2MPa后，1m处截面的喷雾锥角、雾滴索太尔平均直径（SMD）和轴向速度的变化不大。所测试的系列喷嘴的最佳压力范围为7～8MPa；所形成的细水雾雾滴SMD均小于100μm，属于超细水雾范畴。     

基于角位移控制的功率分流式自动变速器速比控制研究

阐述了功率分流式自动变速器中电机控制式无级变速单元的速比控制原理。通过对速比控制电机目标角位移、角位移滑模变结构控制器的建模,来确定整个控制器的设计参数。在对角位移目标轨迹跟踪的仿真中,控制器能够较好地实现在有外在负载干扰下对角位移目标轨迹的跟踪。     

装载机发动机与液力变矩器功率匹配优化

装载机发动机与液力变矩器之间的匹配对各自性能的发挥及整机牵引特性均有重要的影响。在拟合装载机发动机与液力变矩器有关特性曲线的基础上,以装载机的动力性为目标,建立发动机与液力变矩器匹配的优化模型,优化装载机发动机与液力变矩器的功率匹配,给出了算例并分析了优化结果。     

神经网络方法在车内噪声信号预测中的应用

针对以往车内噪声有源控制过程中存在的次级声源在初级声场采样传声器上的声反馈问题，提出了车内噪声信号识别和预测的神经网络方法。对被试面包车在稳态和非稳态两种工况下的试验研究表明，利用车身悬置点和发动机的振动信号，通过BP神经网络来识别和预测车内驾驶员耳旁噪声是可行的，可以解决次级声源在初级声场采样传声器上的声反馈问题。     

铰链密封式传动滚子链的道路试验研究

通过30000km道路行驶试验研究表明，密封滚子链的磨损形式，对销轴和套筒而言是以疲劳磨损为主,伴随有粘着磨损与磨粒磨损；对链板和O形圈而言主要是疲劳磨损，并且O形圈的磨损要比链板的磨损严重得多。销轴和套筒表层磨损后产生了循环硬化特性，有利于提高其铰链副的耐磨性。     

用转轮内三元流动计算对泵叶轮进行改型设计

在有厚翼栅奇点解法的基础上，利用一种考虑叶片厚度影响的叶片式水力机械转轮内准三元流动新的分析方法、准确计算了泵叶轮叶片表面上的速度与压力变化分布规律，获得了许多水力性能评估参数；通过评价这些参数，对叶片型线进行了修改，获得了比较满意的结果。为叶轮的改型设计提供了一条较有价值的新途径。