碎甘薯秧离散元仿真参数测量与标定*

针对甘薯秧蔓机械化回收过程中离散元仿真研究缺乏准确参数值的问题,采用直接测量和虚拟标定相结合的方法对碎甘薯茎秆和叶片离散元仿真参数进行研究。采用物理试验法获得碎甘薯秧的本征参数、碰撞恢复系数等参数值及碎甘薯秧颗粒的静摩擦系数参数范围,并为离散元法仿真设计了不同的参数组合。通过堆积角优化仿真试验确定甘薯叶片本征参数及其他不易直接测量的离散元仿真参数。Plackett-Burman试验表明,甘薯茎秆—甘薯茎秆和甘薯茎秆—45钢的静摩擦系数、甘薯茎秆—甘薯茎秆和甘薯茎秆—甘薯叶的滚动摩擦系数均显著影响堆积角。运用最陡爬坡试验和Box-Behnken优化试验标定了对碎甘薯秧堆积角有显著影响的参数值,以得到的参数进行颗粒堆积仿真试验,测得堆积角平均值为40.51°,与实测值相对误差为0.972%,说明物理试验加优化仿真试验来标定离散元参数是可行的,标定所得的参数可作为甘薯秧茎叶离散元仿真参数。     

丛生竹林下竹荪富硒菌棒覆土栽培技术

在丛生竹林下对不同基质配方和硒浓度菌棒进行埋棒栽培以及覆土方式栽培试验，结果表明，菌棒不同基质配方对竹荪产量及其营养成分有显著影响，而且三列浅沟形和双列龟背形2种覆土方式有利于提高竹荪产量。改良基质配方菌棒竹荪产量比常规配方提高约50%；在添加外源硒质量分数为0~2.0 mg/kg的浓度范围内，竹荪子实体产量先随着硒浓度的增加而逐渐增加后再有所降低，基质中添加1.5 mg/kg硒肥比不添加产量提高了195.30%。基质中添加硒质量分数为1.0~2.0 mg/kg的硒肥可以较显著地提高竹荪花的硒含量，其干物质中硒含量平均值从约2.50 mg/kg提高到8.05~13.30 mg/kg，外源硒肥利用率达到9.70%~15.36%。菌棒不同基质配方对竹荪子实体的粗蛋白及粗多糖含量有明显影响，改良配方竹荪蛋与竹荪花的粗蛋白含量是常规配方的1.21和1.29倍，其粗多糖含量是常规配方的4.81和1.35倍；基质中添加硒肥与不添加硒肥相比，竹荪蛋与竹荪花中粗蛋白含量分别增加了40.90%和14.30%。在硒质量分数为1.0 mg/kg的竹荪菌棒林下覆土栽培试验中，三列浅沟形覆土方式单位面积鲜竹荪蛋产量最高，为10.27 kg/m²；双列龟背形覆土方式单个菌棒的鲜竹荪蛋产量最高，达1.40 kg。三列浅沟形和双列龟背形2种覆土方式基质生物转化率分别为93.00%和94.14%，达到了较高水平。     

农村污水治理PPP项目风险因素分析--基于ISM-MICMAC模型

当前我国在探索将PPP模式应用于农村污水治理领域时面临诸多挑战,准确识别与分析项目风险因素具有实际意义。基于全寿命周期视角,采用文献研究与德尔菲法识别出农村污水治理PPP项目面临的21个风险因素,运用解释结构模型(ISM)与交叉影响矩阵相乘法(MICMAC)构建了风险因素的六级层次结构模型并分类,以揭示了风险因素间的内在联系,并提出政策建议。研究表明,项目缔约与运营阶段是关键阶段,合作单位选择是最根本的风险因素。此研究可为我国农村污水治理PPP项目风险分担与控制研究提供新思路,同时为政府部门提供投资决策与风险管理的依据。     

基于距离相关系数和KNN回归模型的森林蓄积量估测研究

【目的】探究Landsat8 OLI数据和KNN算法在森林蓄积量估测中的潜力。【方法】以湖南省湘潭县为研究区,采用Landsat8 OLI数据和同时期的二类调查数据,通过距离相关系数筛选特征,分别采用线性回归模型(MLR)、K-近邻模型(KNN)、距离加权KNN模型(DW-KNN)和优化欧式KNN模型(FW-KNN)对森林蓄积量进行估测。使用十折交叉方法进行精度检验,对检验结果进行对比分析。【结果】3种KNN模型的估测结果均高于传统的线性模型,并且在3种KNN模型中,FW-KNN算法效果最好,决定系数达到0.69,为3种模型中最高;3种KNN模型中,本研究优化欧氏距离KNN模型的估测精度最高,其均方根误差为30.3%,相比于传统KNN模型的均方根误差降低了5.1%,相比于DW-KNN模型降低了3.3%。【结论】采用DW-KNN蓄积量估测结果明显优于其他两种模型,说明通过特征与蓄积量的相关性优化样本间的距离是一种可行的KNN优化方法。     

草鱼肠系膜脂肪组织高质量总RNA提取方法的研究

基于传统的酸性酚—异硫氰酸胍—氯仿一步提取法,比较分析多种优化操作步骤,摸索出一种高质量提取体质量为80~150 g草鱼肠系膜脂肪组织总RNA的改良方法。试验结果显示,相较于肝脏、脾脏、肠道等脏器组织,草鱼肠系膜脂肪组织RNA丰度低,且极易在样品前处理阶段出现顽固性降解问题。探索发现,将取样量增至约30 mg,可提升RNA产量以满足常规试验需求。针对降解难题,改良常规的样品前处理技术流程,采用鲜样液氮速冻,冻样直接放入TRIzol试剂中裂解,并即刻进行长时间机械匀浆,时长约3 min等核心操作步骤,可显著降低脂肪组织样品RNA的降解。Agilent生物分析仪检测结果显示,改良方法提取的草鱼肠系膜脂肪组织RNA完整度高,关键RIN值为8.7~9.0。研究推测,长时间机械匀浆所形成的持续剪切冲击力或许有助于TRIzol试剂中的异硫氰酸胍等成分突破油滴阻碍而有效抑制内源性RNA酶。本方法提取的草鱼脂肪组织总RNA质量可满足高通量转录组测序要求。     

基于BP神经网络的人机联合率定SWMM研究

为解决雨洪管理模型(SWMM)在率定过程中参数复杂、过程繁琐等问题。以西宁市某地块为例建立SWMM模型,利用Morris筛选法进行参数灵敏度分析,并根据灵敏度分析的结果进行人工率定;另外利用BP神经网络对模型进行率定,并结合参数灵敏度对其进行优化。对3种率定方案进行分析,结果表明:水文水力模块参数的相对灵敏度基本一致,其中灵敏度较大的参数为子汇水区面积(Area)、不透水率(Imperv)和不透水区洼地蓄积量(Destore-Imperv),并且不同降雨条件下模型参数的灵敏度存在差异。经过优化后的BP神经网络参数率定方法的模型模拟效果最好,纳什系数最大。结合灵敏度优化BP神经网络的人机联合率定方法一方面能提高BP神经网络率定的准确性,另一方面又能提高传统人工率定的效率。     

Age estimation of different tree species using a special kind of an electrically recording resistance drill

The determination of tree age is an important issue for urban green planning, forestry and dendrology; finding non-destructive and quasi-non-destructive methods for this purpose is of great theoretical and practical importance. The resistance drilling method is quasi-non-destructive because the average diameter of an opening that remains after drilling does not exceed 3 mm. Do electrically recording resistance drills allow precise assessment of tree age? The aim of this study is to assess the accuracy of determining the number of tree rings based on an examination of this special kind of drilling resistance profiles for three tree species, the pine Pinus sylvestris L., the oak Quercus robur L., and the birch Betula pendula ROTH. In 2015 and 2016, 15 pine trees, 15 oak trees, and 15 birch trees were randomly selected. For each studied tree, a measurement was conducted using the electrically recording resistance drill IML-Resi E400 with a flat-tipped 1.5/3 mm steel needle (research sample), and an increment core was taken (reference sample). The drill used was not a real Resistograph^®. The analysis of the E400-profiles underestimated the number of tree rings; the mean bias error (MBE) values were –6.5, –2.5, and –6.0 years for pine, oak, and birch, respectively. The proportion of investigated trees with less than five years difference between the research and reference samples varied from 38.4 (birch) to 66.7 (oak) percent. The accuracy of tree age determination was lowest for birch and highest for oak. The binomial generalised linear model (GLM) revealed that the most accurate tree age assessments were obtained from tree rings wider than 2 mm. The measurements clearly showed that the electrically recording resistance drill IML-Resi E400 enables a quick, although approximate, tree age assessment. Future research should concentrate on electronically regulating and recording drills, providing a higher spatial and signal resolution, and a stronger correlation to wood density.     

菠萝蜜种子淀粉制备香草兰微胶囊的工艺研究

以菠萝蜜种子淀粉为壁材,采用饱和水溶液法制备香草兰精油微胶囊,以包埋产率为指标,采用响应面分析法对微胶囊的包埋条件进行优化探讨。结果表明:5个单因素中,影响最显著的因素为壁芯材比例、包埋温度和包埋时间。响应面优化得到香草兰精油微胶囊的最佳工艺条件为壁芯材比例为7.5∶1;包埋时间72 min;包埋温度54℃,此条件下的包埋产率为(95.46±0.2)%,包埋率为(76.35±0.6)%,载油量为(27.73±0.3)%。试验证明,此条件结果与模型预测值相吻合,此工艺条件可为菠萝蜜种子淀粉包埋香草兰精油微胶囊工艺提供理论依据。     

沉淀法制备的紫杉醇白蛋白微粒包封率测定

采用沉淀法制备紫杉醇白蛋白微粒,采用高效液相色谱法(HPLC)测定紫杉醇质量浓度,以二氯甲烷为有机相,采用萃取法分离白蛋白微粒与游离药物,测定其包封率。结果表明:紫杉醇在10~100 mg·L-1范围内线性良好,平均回收率可达到97.6%~101.0%;有机溶剂萃取法测定紫杉醇白蛋白微粒包封率的平均回收率,达94.80%~101.67%。这种方法测定紫杉醇白蛋白微粒回收率准确可靠,适用于沉淀法制备的白蛋白微粒的包封率测定。     

Selection efficiency of high inorganic phosphorus-low stachyose lines from soybean breeding populations of different genetic backgrounds

Reducing the antinutritional components of soybean (Glycine max) seed is necessary for producing animal feed with increased metabolizable energy and nutrient availability. The objective of this study was to evaluate marker-assisted and phenotypic selection for the high inorganic phosphorus (Pi) and low stachyose traits controlled by the Lpa and MIPS1 mutations in six breeding populations. Phenotypic selection for low phytate individuals carrying Lpa mutations using Pi was reliable, but the assay should include parental genotypes to construct thresholds before performing selections, as the effect of Lpa on Pi depends on genetic background. Phenotypic selection for low phytate individuals carrying MIPS1 using Pi should be complemented with sugar data because selection based solely on Pi may be inefficient, as this trait is not environmentally stable in MIPS1 germplasm. Marker-assisted selection for simple sequence repeat (SSR) linked to Lpa2 (Satt527, Satt561) depended on the pedigree of the population suggesting that perfect markers (e.g., single nucleotide polymorphism) should become the molecular tool for screening a large number of progenies with high efficiency.