干旱区枸杞滴灌灌溉制度试验研究

以宁夏大田滴灌枸杞为研究对象,分析了不同水分处理条件下滴灌枸杞土壤水分动态、生长量变化及水分对产量品质的影响,基于多因素考虑提出了合理的枸杞滴灌灌溉制度。主要结论有:干旱区枸杞土壤水分变化敏感区域为距树干水平距离40cm处的20~70cm土层,灌水前后土壤含水率变幅在0.2%~4.3%。随灌水量的增加枸杞地径及株高随之增加,而冠幅的变化规律性不强,西侧及南侧的枝条生长量明显高于东侧及北侧。3 750m3/hm2时鲜果产量、干果产量、百粒重、粒度及枸杞多糖含量均最高,干鲜比为1∶4.66,综合认为其水分利用效率最高且品质相对较好。平水偏丰年型(P=40%)条件下枸杞滴灌灌溉制度为灌溉定额为3 750m3/hm2,灌水次数为14次,其中萌芽期灌水1次,营养生长期2次,盛花期、盛果期各4水,秋季生长期灌水3次。     

考虑未来降水的南方水稻灌溉风险决策节水效果分析

提高降雨利用率是节约灌溉用水的一种有效途径。提出了一种考虑未来降水的灌溉风险决策方法,以广西青狮潭灌区为例验证这一方法的节水效果。收集了桂林站2013-2014两年早稻和晚稻生育期逐日对未来7d的气象预报数据和相应时段的气象观测数据,推求了两年淹灌和间歇灌溉两种灌溉模式下采用常规灌溉决策和风险灌溉决策的灌溉制度,并分析了风险决策的节水效果。结果表明,早稻和晚稻平均降低灌水量23.5和21.9mm,节水分别为38.3%和6.7%。采用风险决策可以避免因灌后遇雨造成的灌水浪费,从而减少灌溉用水量、排水量以及灌水次数。     

遗传算法在节水高效灌溉制度优化中的应用——以冬小麦为例

利用遗传算法对非充分灌溉条件下冬小麦的最优灌溉制度设计进行求解,并提出了与之对应的约束条件及处理方法。以我国华北地区某试验站为例来说明该算法及处理方法的可行性。结果表明,该方法能够实现对冬小麦整个生育期灌溉水量进行合理分配,使冬小麦在缺水敏感期内得到充分灌溉,从而达到节约灌溉用水,提高水分利用效率的目的。     

Long-term growth and yield responses of olive trees to different irrigation regimes

The aim of this work was to evaluate long-term effects of different irrigation regimes on mature olive trees growing under field conditions. A 9-year experiment was carried out. Three irrigation treatments were applied: no irrigation, water application considering soil water content (short irrigation), or irrigation without considering soil water reserves and applying a 20% of extra water as a leaching fraction (long irrigation). Leaf water content, leaf area, vegetative growth, yield and fruit characteristics (fruit size, pulp:stone ratio and oil content) were determined yearly. Results showed that growth parameters did not show significant differences as a consequence of applied water. Yield was increased in irrigated trees compared to non-irrigated ones, but little differences between short and long irrigation were observed, only when accumulated yield from 1998 to 2006 was considered. Irrigation did not cause significant differences in fruit size or pulp:stone ratio either. Irrigation regimes similar to those applied in this experiment, under environmental conditions with relatively high mean annual precipitation, does not increase growth, yield or fruit characteristics when compared to rain-fed treatment, and consequently, the installation of a irrigation system could be not financially profitable.     

改进PSO的多目标约束项目进度计划问题研究

在项目进度计划问题的研究中,为实现在多执行模式和多资源约束条件下,对项目活动的执行顺序和执行模式进行合理安排的目标,以使项目工期最短,建立了多目标进度计划的目标函数,确定了约束条件.结合微粒群算法求解该组合优化问题.并通过对标准微粒群算法的改进,引入了交换子和交换序的概念,提高了算法的寻优速度和寻优能力,并结合具体的实例验证了改进算法的有效性.     

计及机器调整和物料运送时间的Job-Shop调度算法

对考虑机器调整时间和物料运送时间的Job-Shop调度问题进行了研究.优化计算时考虑了物料运送时间的作用及其对车间调度的影响,并给出了其求解方法;利用GA、SA与PSO相结合的混合PSO算法解决了Job-Shop调度问题.仿真实例模拟实际工况,对工件的运送、加工和机器调整时间及其相互间的影响进行了分析和计算,并与传统的忽略机器调整时间、物料运送时间以及直接计入工件加工时间的调度方案进行了对比,结果表明提出的方法可行,更加符合实际生产工况.     

近红外光谱多核并行谱区选择任务调度策略研究

在近红外光谱的谱区选择算法中,滑动窗口偏最小二乘法具有较高的预测精度,但是其程序运行时间很长。在不改变算法预测精度的前提下,首先以分段间隔偏最小二乘法为基础研究了顺序分配法、等间距法和排序法等3种任务调度策略对并行算法性能的影响。在这3种任务调度策略中,排序法具有较好的负载均衡性和较高的并行效率。然后在16核云计算平台下将排序法分配策略分别应用于组合分段偏最小二乘法、反向分段偏最小二乘法和滑动窗口偏最小二乘法等谱区的并行选择算法中。经实验测试,在单核串行算法下反向分段偏最小二乘法和滑动窗口偏最小二乘法的程序运行时间分别为9.22 h和55.51 h,在排序法分配策略下采用2核并行算法时其程序运行时间分别缩短为4.98 h和29.03 h,分别节省了45.99%和47.70%的程序运行时间。实验结果表明:当考虑并行效率和计算成本时,在1~16核的并行算法中,选用2核并行执行以上4种谱区选择算法都具有最高的并行效率和性价比。     

Optimal model on integrated production planning andscheduling of steelmaking and continuous casting process

Considering the uncertainty and complexity of production planning and scheduling in steel making and continuous casting, an optimal simulation model is built. Based on the analysis of the logistics in steelmaking process, the casting plan is regarded as the global goal, and the procedural information is taken into account, as well as the local evolution rule that abstracted from scheduling rule. To simulate transferring of materials in the process, the state of operation sites are updated as a result of the global goals, process information and local rule. The goal directed simulation is carried out based on the self organization evolution mechanism. Finally, an optimal integrated production plan is made by a continuous simulation evolution. A simulation case for eight hours production plan shows that the model is valid and efficient.     

Agronomic and economic response to furrow diking tillage in irrigated and non-irrigated cotton (Gossypium hirsutum L.)

The Southeast U.S. receives an average of 1300 mm annual rainfall, however poor seasonal distribution of rainfall often limits production. Irrigation is used during the growing season to supplement rainfall to sustain profitable crop production. Increased water capture would improve water use efficiency and reduce irrigation requirements. Furrow diking has been proposed as a cost effective management practice that is designed to create a series of storage basins in the furrow between crop rows to catch and retain rainfall and irrigation water. Furrow diking has received much attention in arid and semi-arid regions with mixed results, yet has not been adapted for cotton production in the Southeast U.S. Our objectives were to evaluate the agronomic response and economic feasibility of producing cotton with and without furrow diking in conventional tillage over a range of irrigation rates including no irrigation. Studies were conducted at two research sites each year from 2005 to 2007. Irrigation scheduling was based on Irrigator Pro for Cotton software. The use of furrow diking in these studies periodically reduced water consumption and improved yield and net returns. In 2006 and 2007, when irrigation scheduling was based on soil water status, an average of 76 mm ha⁻¹ of irrigation water was saved by furrow diking, producing similar cotton yield and net returns. Furrow diking improved cotton yield an average of 171 kg ha⁻¹ and net return by $245 ha⁻¹ over multiple irrigation rates, in 1 of 3 years. We conclude that furrow diking has the capability to reduce irrigation requirements and the costs associated with irrigation when rainfall is periodic and drought is not severe.     

The effects of irrigation methods with effluent and irrigation scheduling on water use efficiency and corn yields in an arid region

A great challenge for the agricultural sector is to produce more food from less water, particularly in arid and semi-arid regions which suffer from water scarcity. A study was conducted to evaluate the effect of three irrigation methods, using effluent versus fresh water, on water savings, yields and irrigation water use efficiency (IWUE). The irrigation scheduling was based on soil moisture and rooting depth monitoring. The experimental design was a split plot with three main treatments, namely subsurface drip (SSD), surface drip (SD) and furrow irrigation (FI) and two sub-treatments effluent and fresh water, which were applied with three replications. The experiment was conducted at the Marvdasht city (Southern Iran) wastewater treatment plant during 2005 and 2006. The experimental results indicated that the average water applied in the irrigation treatments with monitoring was much less than that using the conventional irrigation method (using furrows but based on a constant irrigation interval, without moisture monitoring). The maximum water saving was obtained using SSD with 5907 m³ ha⁻¹ water applied, and the minimum water saving was obtained using FI with 6822 m³ ha⁻¹. The predicted irrigation water requirements using the Penman-Monteith equation (considering 85% irrigation efficiency for the FI method) was 10,743 m³ ha⁻¹. The pressure irrigation systems (SSD and SD) led to a greater yield compared to the surface method (FI). The highest yield (12.11 × 10³ kg ha⁻¹) was obtained with SSD and the lowest was obtained with the FI method (9.75 × 10³ kg ha⁻¹). The irrigation methods indicated a highly significant difference in irrigation water use efficiency. The maximum IWUE was obtained with the SSD (2.12 kg m⁻³) and the minimum was obtained with the FI method (1.43 kg m⁻³). Irrigation with effluent led to a greater IWUE compared to fresh water, but the difference was not statistically significant.