旱区地膜覆盖技术的研究进展及发展前景

综述了国内外旱区地膜覆盖技术在应用与发展、应用机理与效果、主要技术模式等方面取得的主要研究进展,指出了地膜覆盖应用研究中存在的问题,在此基础上展望了旱区地膜覆盖技术研究与发展的前景。     

基于SHAP重要性排序和机器学习算法的灌区渠道调度流量预测

渠道泄水闸能够快速排除灌区入渠洪水,避免渠道漫顶。研究以淠史杭灌区灌口集泄水闸为例,以闸门调度流量为目标变量,以不同时段过去和未来降雨量、泄水闸闸上实时水位及其变化量为特征变量,比较8种机器学习算法的预测精度,同时采用Shapley Additive exPlanations（SHAP）法分析特征变量重要性。结果表明：1）集成学习算法预测评价指标优于传统回归算法,8种机器学习算法中随机森林回归（random forest regression, RFR）算法预测精度最高（训练集均方根误差、平均绝对误差、均方误差及决定系数分别为 0.146 m³/s、0.094 m³/s、0.021 m³/s、0.976;测试集分别为0.306 m³/s、0.197 m³/s、0.093 m³/s、0.931）;2）采用SHAP法确定的特征变量重要性排序表明灌口集泄水闸闸上水位对于泄水闸调度流量的预测结果影响最大,占特征重要性值总和的34.6%;3）以过去6 h降雨量、过去9 h降雨量、未来6 h降雨量、灌口集泄水闸闸上水位作为输入变量的RFR算法预测灌口集泄水闸调度流量效果最佳,模型误差指标为（训练集均方根误差、平均绝对误差、均方误差及决定系数分别为0.126 m³/s、0.080 m³/s、0.016 m³/s、0.982;测试集分别为0.263 m³/s、0.164 m³/s、0.069 m³/s、0.950）,研究结果对灌区防洪调度决策具有重要参考价值。     

用分布式水循环模型与机器学习预测内蒙古河套灌区节水潜力

为探究内蒙古河套灌区真实节水潜力,该研究构建河套灌区分布式水循环模型与基于机器学习的盐分模型,设置节水方案集,定量分析各方案下的灌区引、耗水量、地下水埋深、积盐量变化等。结果表明：1）水面蒸发的纳什系数均不低于0.654,相对误差绝对值不高于分别为4.82%,相关关系为0.88,排水过程纳什系数均不低于0.600,相对误差绝对值不高于分别为5.11%,相关关系为0.82,地下水埋深的纳什系数均不低于0.628,相对误差绝对值不高于分别为5.12%,相关关系为0.86,满足灌区水循环满足精度要求。本文选择采用土壤盐分模型,得到土壤积盐量与实测值的纳什系数均不低于0.76,满足精度要求。2）渠道砌衬方案S1、田间节水调控方案S2、种植结构调整方案S3的耗水节水量分别为2.93亿、3.02亿和2.54亿m³。S1+S2+S3组合方案灌区耗水节水量最多,为9.11亿m³,S2+S3方案组合次之。3）渠系水利用系数提高,将引起地下水水位下降,不利于排盐,S1方案下地下水埋深大于3 m的面积比例较基准方案增加了7.59%,不利于灌区排盐。田间工程措施使得相应的农田入渗量减少,地下水位下降,有利于灌区脱盐,S2方案下地下水入渗补给量较基准方案减少2.57亿m³,灌区地下水位下降较为明显,S2方案有利于灌区脱盐。S3方案下地下水入渗补给量略微减少,地下水位变化不大,有利于灌区脱盐。不同方案组合,S1+S2、S1+S2+S3方案下对地下水埋深影响较大,尤其是S1+S2+S3方案在灌区西北部、山前、乌拉特前旗、乌梁素海东部的形成连片埋深高值区,影响区域生育期农田作物与林草地植被生长。S1S2方案下不利于灌区脱盐,自然植被生育期平均埋深超过2.5 m的比例较基准方案增加了5.46%。在综合考虑生态环境的约束下,推荐耗水节水量最大的方案S2+S3,即灌区适宜的耗水节水潜力为5.69亿m³。该方案下虽然也会引起地下水位略有下降、进乌梁素海排入水量略微减少,但最有有利于灌区排盐。研究可为引黄灌区节水方案制定与灌溉管理提供技术支撑。     

Coping with imbalanced data problem in digital mapping of soil classes

An unsolved problem in the digital mapping of categorical soil variables and soil types is the imbalanced number of observations, which leads to reduced accuracy and the loss of the minority class (the class with a significantly lower number of observations compared to other classes) in the final map. So far, synthetic over- and under-sampling techniques have been explored in soil science; however, more efficient approaches that do not have the drawbacks of these techniques and guarantee retention of the minority classes in the produced map are essentially required. Such approaches suggested in the present study for digital mapping of soil classes include machine learning models of ensemble gradient boosting, cost-sensitive learning and one-class classification (OCC) of the minority class combined with multi-class classification. In this regard, extreme gradient boosting (XGB) as an ensemble gradient learner, a cost-sensitive decision tree (CSDT) within the C5.0 algorithm, and a one-class support vector machine combined with multi-class classification (OCCM) were investigated to map eight soil great groups with a naturally imbalanced frequency of observations in northwest Iran. A total of 453 profile data points were used for mapping the soil great groups of the study area. A data split was done manually for each class separately, which resulted in an overall 70% of the data for calibration and 30% for validation. The bootstrapping approach of calibration (with 10 runs) was performed to produce multiple maps for each model. The 10 bootstraps were evaluated against the hold-out validation dataset. The average values of accuracy measures, including Kappa (K), overall accuracy (OA), producer's accuracy (PA) and user's accuracy (UA), were explored. In addition, the results of this study were compared with a previous study in the same area, in which resampling techniques were used to deal with imbalanced data for digital soil class mapping. The findings show that all three suggested methods can deal well with the imbalanced classification problem, with OCCM showing the highest K (= 0.76) and OA (= 82) in the validation stage. Also, this model can guarantee the retention of the minority classes in the final map. Comparing the present approaches with the previous study approach demonstrates that the three newly suggested methods can remarkably increase both overall and individual class accuracy for mapping.     

旱地冬小麦膜侧宽幅匀播增产效应研究

小麦膜侧宽幅匀播技术是密植作物高产栽培新技术,其抗旱增产效果显著。为了给该技术在旱地小麦生产上的推广应用提供理论支持,以冬小麦普冰151为指示品种,采用田间小区试验,比较了冬小麦旱地膜侧宽幅匀播、膜侧沟播、常规条播三种不同种植模式的增产效应。结果表明,旱地膜侧宽幅匀播冬小麦产量极显著高于膜侧沟播和常规条播,折合产量为5 902.8 kg/hm2较对照的增产率为24.54%。表明冬小麦膜侧宽幅匀播集雨保墒效果好、抗旱作用显著,又利于密植和高产。     

全膜双垄沟播一膜两茬栽培技术

为了实现节本增效,在灵台县采用了一膜两年用免耕栽培的方法,开展了全膜双垄沟播玉米收获后穴播冬油菜的生产模式研究,即在玉米成熟后,采用人工收获,清除玉米秸秆,保护好上茬地膜,在原地膜上穴播点种冬油菜,冬油菜产量可达3 080 kg/hm2,是一种轻简方便,高产高效的模式。     

汽车塑料零部件VOC的产生原因分析及解决措施

工程塑料代替金属材料,使汽车更轻便、更节能、更环保,同时也带来了VOC污染.介绍了汽车中使用的几种最主要塑料、VOC的来源及危害,提出了降低VOC浓度的措施,为汽车塑料零部件生产企业如何降低VOC提供借鉴,保障驾乘人员身体健康.     

不同播期玉米覆膜直播对产量的影响

玉米覆膜直播省工,增加土壤温度,保蓄水分,增加玉米抗旱能力。通过不同播期试验得出,在福泉市4月4日播种获得的产量最高。     

平塘县玉米地膜覆盖栽培技术要点及成效

平塘县属典型的喀斯特山区,土地破碎,水源条件较差,实施玉米半膜覆盖保墒技术项目能增温保墒,弥补温、光、水资源的不足,从而提高玉米产量。在农村劳动力少、常年干旱情况下,实行玉米地膜覆盖栽培,既减少劳动力投入又起到增收目的。对平塘县2012年实施玉米地膜覆盖主要栽培技术及应用效果进行分析总结,可供贵州山区实施地膜覆盖提供参考。     

Analysis on rollover index and rollover risk factors of vehicles

The influence of forepoles on the patterns of a single tunnel plastic collapse was discussed using Acutronic 661. A series of plane strain centrifuge model tests in over-consolidated clay were conducted. Tests were carried out in a circular cavity with stiff resin inclusions embedded around the tunnel vault. These inclusions simulated elements of a secant pipe or forepole arch system. An image processing system was used to measure displacements below surface level and displacement transducers at the surface level of the models. The aim of the experimental work was to study the ground response due to simulated tunnel construction and to analyse how the vault stiff inclusions affect tunnel stability. According to the principle of the upper bound theorem of plastic limit analysis, a collapse mechanism of the single circular tunnels with embedded forepoles in cohesive soil was established. An upper bound equation of stability ratio was obtained. The upper bound solutions of stability ratio under different soil cover above the tunnel, undrained shear strength at tunnel axis level and embedded forepoles range were discussed from the equation. In the end,the correctness of the upper bound solutions was proved by the results of centrifugal model tests.