基于改进多目标决策模型的大桥水库灌区渠系自适应规划方法

提出基于改进多目标决策模型的大桥水库灌区渠系自适应规划方法,合理规划大桥水库灌区渠系配水,提高水资源利用率,降低无效弃水以及对生态环境的影响。以配水结束后灌区各渠系缺水量、渠道输水损失量以及农作物生产带来的灰水足迹三者最小为大桥水库灌区渠系规划多目标决策函数,以渠道输水能力、水量和时间为约束条件,构建大桥水库灌区渠系规划的多目标决策模型;以栅格法构建路径规划的运行环境,通过移动几率优化、信息素挥发系数自适应调整两方面改进蚁群算法,通过改进蚁群算法寻优获取渠道缺水量、输水损失量、灰水足迹最小的水库灌区渠系规划结果。实验证明：该方法可以有效规划大桥水库灌区渠系的水资源,规划后的灌区渠系在缺水量、渠道输水损失量以及灰水足迹方面都有较好的表现,且效率高、应用性强。     

除草机器人机械臂的逆向求解与控制

设计了一种基于机器视觉导航和杂草识别的除草机器人模型,该机器人能沿作物行间自主行走并能准确地识别和“清除”杂草。设计了除草机器人的机械臂除草执行系统,求取了机械臂运动学逆解,用VC++开发了控制程序。试验显示,图像处理算法所需时间少,能够适应户外自然光线在一定范围的变化,机械臂能够平稳动作并精确定位杂草目标。     

基于大系统递阶理论的区域抗旱应急调水方案

为了解决旱情紧急情况下关中西部各灌区和各用水部门之间的用水矛盾,该文以大系统递阶理论为基础,采用动态规划方法建立了以总用水量最大为目标的区域水量优化调配模型,根据需水与来水的不同频率组合,拟定了3种一级备选配水方案（第3方案有27种二级备选配水方案）,通过对各方案调度结果分析,并考虑实际调度运行的可操作性,推荐采用第3种方案以夏灌调水量作为主要控制因素进行归类合并后形成4灌区11种调水方案。该调水方案在2005－2007年的实际使用中,取得了很好的效果,对该区域的抗旱应急调水提供了科学依据。     

渭北西部丘陵区小流域土地生态规划与设计

改善和优化土地利用方式是综合治理黄土高原水土流失的关键,以黄土高原小流域的恢复生态、提高土地生产力、实现经济可持续发展为目的,以渭北西部丘陵沟壑区的典型流域为例,对陕西千阳县冉家沟小流域进行生态规划与设计。主要包括小流域土地利用现状分析、土地生态类型的评价、土地利用系统的结构优化及土地利用的生态设计与规划等。     

土法炼锌区大气沉降Pb、Zn、Cd及其对土壤质量的影响

Dust emissions from smelters, as a major contributor to heavy metal contamination in soils, could severely influence soil quality. Downwind surface soils within 1.5 km of a zinc smelter, which was active for 10 years but ceased in 2000, in Magu Town, Guizhou Province, China were selected to examine Pb, Zn, and Cd concentrations and their fractionation along a distance gradient from a zinc smelter, and to study the possible effects of Pb, Zn, and Cd accumulation on soil microorganisms by comparing with a reference soil located at a downwind distance of 10 km from the zinc smelter. Soils within 1.5 km of the zinc smelter accumulated high levels of heavy metals Zn (508 mg kg-1), Pb (95.6 mg kg-1), and Cd (5.98 mg kg-1) with low ratios of Zn/Cd (59.1--115) and Pb/Cd (12.4--23.4). Composite pollution indices (CPIs) of surface soils (2.52--15.2) were 3 to 13 times higher than the reference soils. In metal accumulated soils, exchangeable plus carbonate-bound fractions accounted for more than 10% of the total Zn, Pb, and Cd. The saturation degree of metals (SDM) in soils within 1.5 km of the smelter (averaging 1.25) was six times higher than that of the reference soils (0.209). A smaller soil microbial biomass was found more frequently in metal accumulated soils (85.1--438 μg C g-1) than in reference soils (497 μg C g-1), and a negative correlation (P < 0.01) of soil microbial biomass carbon to organic carbon ratio (Cmic/Corg) with SDM was observed. Microbial consumption of carbon sources was more rapid in contaminated soils than in reference soils, and a shift in the substrate utilization pattern was apparent and was negatively correlated with SDM (R = -0.773, P < 0.01). Consequently, dust deposited Pb, Zn, and Cd in soils from zinc smelting were readily mobilized, and weredetrimental to soil quality mainly in respect of microbial biomass.     

从沉积物中分离重金属不同提取方法的比较

Three extraction methods, ultrasonic assisted extraction (USE), microwave assisted extraction (MSE), and conventional single extraction (CSE), in conjunction with the modified three-stage BCR sequential extraction procedure (SEP) were applied to examine the contents of Cd, Cu, Cr, Ni, Pb and Zn from lake sediment samples, to know whether these techniques can reduce extraction time and improve reproducibility. The SEP and developed alternative single extraction methods were validated by the analysis of certified reference material BCR 601. By the use of optimized sonication and microwave conditions, steps 1, 2 and 3 of the BCR sequential extraction methods (excluding the hydrogen peroxide digestion in step 3, which was not performed with sonication and microwave) could be completed in 15-30 min and 60- 150 s, respectively. The recoveries of total extractable metal contents in BCR 601, obtained by three single extractions ranged from 93.3%-102%, 88.9%-104% and 81.2%-96.2% for CSE, USE and MSE, respectively. The precision of the single extraction methods was found in the range of 3.7%-9.4% for all metals (n = 6).     

兼顾碳贮量和木材生产目标的森林经营规划研究

以吉林省汪清林业局金沟岭林场为对象,基于小班调查数据,以木材生产和碳增量净现值最大为规划目标,考虑采伐量不大于生长量、均衡采伐、生长模型等约束,建立了多目标规划模型。通过LINGO软件求解,得到了50a规划期的最优经营方案,并进行了灵敏度分析。结果表明:各个森林类型在不同分期间伐强度在1%～15%之间,择伐强度在1%～35%之间。规划期木材总采伐量为182.33万m3,规划期末地上碳贮增量为47.92万t;规划期总收益为95 434.40万元,其中木材净现值91 895.10万元,碳贮增量净现值3 539.30万元。多目标经营方案可以同时满足对木材生产和碳贮增量的需求,是一个折衷的方案。与木材生产经营方案相比,多目标经营方案和碳贮增量经营方案在规划期内的木材净现值分别减少2.67%、45.43%,但地上碳贮增量净现值分别增加29.88%、50.42%。因此增加碳贮增量要以减少木材采伐量为代价。分析了低、中、高3种碳价格对多目标经营方案的影响,表明整个规划期内木材采伐量随着碳价格的增加而减少,而碳贮增量随着碳价格的增加而增加。     

遗传规划多类图像分类算法研究

基于树结构的遗传规划图像分类器,仅能从树根节点输出单一实数值,要完成多类图像分类任务比较困难。提出遗传规划多类图像分类算法,等差权值中心动态边界确定分类算法和权值快速下降中心动态边界确定分类算法。在遗传规划分类器进化过程中,动态确定训练样本各类的边界,通过不同的权重设置策略,使类的边界值设定更趋合理,以提高分类器分类的准确度。图形分类测试表明,这2种算法,用于难度大的多类图像分类问题,能提高分类的准确度,分类效果理想。     

A simulation-based two-stage interval-stochastic programming model for water resources management in Kaidu-Konqi watershed, China

This study presented a simulation-based two-stage interval-stochastic programming(STIP) model to support water resources management in the Kaidu-Konqi watershed in Northwest China.The modeling system coupled a distributed hydrological model with an interval two-stage stochastic programing(ITSP).The distributed hydrological model was used for establishing a rainfall-runoff forecast system,while random parameters were provided by the statistical analysis of simulation outcomes.The developed STIP model was applied to a real case of water resources management planning in Kaidu-Konqi watershed,where three scenarios with different water resources management policies were analyzed.The results indicated that water shortage mainly occurred in agriculture,ecology and forestry sectors.In comparison,the water demand from municipality,industry and stockbreeding sectors can be satisfied due to their lower consumptions and higher economic values.Different policies for ecological water allocation can result in varied system benefits,and can help to identify desired water allocation plans with a maximum economic benefit and a minimum risk of system disruption under uncertainty.     

超声-微波协同萃取桂花叶总黄酮工艺的优化

运用二次通用旋转组合设计法,研究了采用超声-微波协同萃取法提取桂花(Osmanthus fragrans Lour.)叶总黄酮工艺中乙醇体积分数、液料比、提取功率、提取时间对桂花叶总黄酮提取率的影响,建立了具有良好预测性能的提取条件数学模型,确定了最佳提取工艺条件.结果表明,最佳工艺条件为乙醇体积分数59％,提取功率140W,提取时间13.2 min,液料比23∶1( V/m,mL∶g),在最佳工艺条件下总黄酮提取率可达4.4210％.