Variance component analysis of parthenocarpy in elite U.S. processing type cucumber (Cucumis sativus L.) lines

Summary Parthenocarpic (seedless) U.S. processing type cucumber (Cucumis sativus L.) germplasm can bear more high quality fruit when compared to their seeded counterparts. Knowledge of genetic components of variation for parthenocarpy would assist cucumber breeders to incorporate this economically important trait into commercial varieties. The inheritance of parthenocarpy in elite U.S. processing type cucumber was, therefore, investigated by examining the single-harvest yield of F₃ progeny derived from a mating between line 2A (P₁, parthenocarpic) and line Gy8 (P₂, non-parthenocarpic) grown in two fields (E-block and G-block at Hancock, Wisc.) in the summer of 2000. Environmental variance accounted for about 90% of total phenotypic variance in both locations. The degree of dominance genetic variance was 0.3 and 2.2 at G-block and E-block, respectively, and the minimum number of effective factors controlling parthenocarpy was estimated as 5 (G-block) to 13 (E-block). Estimates of heritability were significantly lower when based on individual plants within an F₃ family and on F₃ family mean performances. While narrow-sense and broad-sense heritability of individual plants within F₃ family was always < 0.1, narrow-sense heritability for F₃ family mean performance ranged between 0.33 (E-block) and 0.62 (G-block), and broad-sense heritability ranged between 0.53 (E-block) and 0.67 (G-block). Thus, in this population, advanced generation selection for parthenocarpy based on F₃ family mean performance will be more effective than selection of individual plants within F₃ family.     

Simulating of Asymptotic Distributions of Extreme Wind Speed in ChongQing Using Short-term Wind Speed Records

In order to reduce the potential statistical error from the short sample of wind speed(the data of wind speed between 1990 and 1999 in ChongQing district),a method is introduced that simulates the asymptotic distributions of annul maximum wind speed according to the monthly maximum wind speed data.The simulated results of three extreme distributions(Gumbel,Frechet and reverse Weibull distributions) and Generalized Parato Distribution are contrasted and analysed.The reverse Weibull distribution is evaluated as the best asymptotic one for short-term wind speed records in ChongQing.And the optimal value of annul maximum wind speed is estimated.     

Improved Modal Capacity Spectrum Method for Estimating Seismic Performance of Structures

Performance-based seismic design involves how to confirm inelastic displacement demand of structures under earthquake in concise and reasonable way. Two shortages need to improve in analyzing the elastic-plastic seismic responses of structures if the traditional capacity methods are employed. The shortages are the uncertainty of equivalent high damp of elastic-plastic structures and the limitation of the higher order vibration mode effect being not taken into account. To solve these problems, the authors propose the concept and analyzing process of improved capacity spectrum method based on modal pushover analysis and the elastic seismic responses spectrum modified by strength reduction coefficient and ductile coefficient. The example of analyzing the performance of frame structure is shown which may be a simple and effective method used in engineering. This method has application potential in Performance-based seismic design.     

基于PLC的甘蔗预切种切割平台控制系统设计

甘蔗切种是甘蔗预切种种植模式下的极为重要工作过程,但在实际工作中,大部分仍然依靠人工手动切种,切种质量不高,蔗芽损失大,且劳动强度大,生产效率低。为解决此问题,自主开发预切式横向甘蔗智能切种机的数控切割工作台,通过PLC控制多把切刀在切割平台的横向移动,通过液压系统控制多把切刀的切割,经过对切刀定位精度试验分析,分析表明,该控制系统控制切刀定位的精度较高,每把切刀相对定位精度误差均小于0.6%,重复定位误差小于0.5%。研究结果对推进甘蔗预切种的开发,对促进甘蔗种植技术的发展有重要的应用价值。     

基于M-估计的线性化稳健配准算法研究

复杂曲面在制造中的广泛应用对曲面配准技术提出了新的要求,特别是不同区域精度存在差异的复杂曲面配准问题日益突出。为了稳健估计思想推广到不同区域精度存在差异的复杂曲面配准,给出了基于M-估计的一种稳健配准算法。该算法利用M-估计子削弱复杂曲面低精度数据对配准结果的影响,但是这一模型目标函数是高度非线性的分段函数,求解效率不高。现有配准方法已能够迅速获得较好初始位置,因此利用Taylor展式线性逼近偏差函数,得到配准问题M-估计的线性化模型,提高了配准模型估计效率。每步迭代利用F-范数最小逼近旋转矩阵。对仿真数据和实测叶片数据进行试验,结果证明,对精度存在差异的复杂曲面所提算法比最近点迭代算法更加合理。     

加工精度动态损失趋势预测的研究

在大批量的机械加工过程中，由于影响加工精度的因素较多，目前尚难以建立加工精度动态损失的精确数学模型。本研究以柱塞套内孔加工为例，分别探讨了应用时间序列分析，灰色系统理论及人工神经网络原理等手段建立加工精度动态损失数学模型的方法，并通过实例对各种建模精度进行了分析比较。     

采用不规则三角网插值进行土地整理项目前期平整土方量的计算

土地整理项目前期，要求快速高效地计算平整土方量。该文在一般现有地形图等高线与高程点的资料基础上，采用不规则三角网插值的方式进行等值线插值，基于插值后的等值线，采用不规则三角网剖分插值的方式进行数字高程模型的建立，并计算土方量。根据实测数据，数字高程模型精度符合要求，据此进行土方量的计算具有一定的精度，这种方法具有较高的可行性。     

超级电容动态容值测量与容值函数参数估计

针对目前的超级电容参数在线识别方法动态参数测量困难,通用性差,精度低的缺点,提出一种超级电容动态容值测量与参数识别方法。该方法中提出了动态电容值测量模型,通过电荷关系式与能量关系式联立推导动态电容值测量方程组,并采用限定记忆最小二乘法进行参数识别,应用时域仿真对该方法进行了验证。仿真结果表明,该文提出的电容值测量与参数识别方法适用于多种不同的容值函数。当测量信号存在噪声时仍可对电容测量,平均误差低于1%,参数估计相对误差低于20%,可以应用于超级电容动态容值测量。     

基于3S技术的草地估产系统的研究与应用

该研究旨在利用全球定位系统GPS、借助GIS系统和遥感技术RS的统计和结合功能，通过多年地面观测工作和卫星影像的印证和草地专家系统分析，通过在GPS相关接收的芯片实现中改进m序列发生器的结构，实现草地信息获取，建立草地估产系统。利用该系统对草原进行估产的结果表明：所设计的各数据接口工作正常，估产结果与实测结果有很好的吻合度，且估产精度有所提高，使小范围内所获研究结果和草地视图信息得以外延和扩展。     

基于遥感抽样的国家尺度农作物面积统计方法评估

国外从20世纪70年代已将遥感用于农业统计,中国从起步至今也有20多年的历史。遥感以其准确、及时、客观的优势推动了农业统计的进步,但对于遥感统计的精度评估一直不甚明确,影响了遥感监测的可信度和遥感技术优势的充分发挥。该文首先分析了目前国内外农业遥感统计的主要方法及其评估方式,针对中国大尺度农作物面积遥感统计方法的实际,采用多种指标对其进行评估,目的是对遥感统计调查的准确度予以明确的定义和计算。采用亚米级差分GPS采样获取足量样本的方法,对L andsat TM提取的水稻面积进行验证,基于原始的误差矩阵计算得出生产者精度为89.53%,用户精度为95.37%,总体精度为87.02%,地面实测数据和图像解译结果的相关系数为0.96;引入误差指标σ来直接反映分类结果较真值的总体误差,当σ〉0时,遥感解译结果较实际偏小,反之偏大,计算得到水稻遥感解译面积的σ值平均为0.084。