基于人工神经网络技术的蔬菜施肥决策研究

阐述神经网络的概念和特点，着重介绍BP神经网络并在Maflab环境下建立基于BP神经网络的施肥模型，以土壤养分（氮、磷、钾)和实际产量为模型输入，以氮肥、磷肥和钾肥的施用量为模型数输出．进行网络训练，直至获得满意的施肥模型。结合投入产出物的价格对模型功能进行扩展，按利润最大化目标进行施肥决策。     

机械设备更新决策专家系统的设计

着重论述了机械设备更新决策专家系统的设计过程,包括知识库的设计、推理机的推理方向和方法、知识获取方法的介绍、综合数据库的规划、人机接口等内容。     

位山灌区渠系优化配水决策支持系

根据位山灌区位于黄河下游、引水含沙量高、测流断面变化大的特点，编制了位山灌区渠系优化配水决策支持系统，主要内容包括来水管理、需水管理、水情资料管理、实际配水统计、配水方案拟定和系统设置等，创新性地进行了多站水量平差计算。     

基于模糊理想点法的多区域多目标

水利工程项目建设次序对水资源方案实施的效果影响较大。因此研究中选用防洪、供水、农村水利水电发展、生态环境保护、工程投资作为水资源方案实施效果的评价目标。考虑到不同区域水利建设的重点不同，本文利用模糊决策分析理论确定出各评价目标在不同区域的权重，基于模糊理想点法建立评价模型，并以湖南省水利工程建设排序为实例进行分析，得出理想的工程排序结果。     

多目标系统模糊决策法在导流标准

施工导流工程是施工组织设计中关键环节，而施工导流标准的选择是施工导流工程的核心，选择合适的施工导流标准可使水电工程施工达到安全可靠和经济可行相统一的目的。利用多目标系统模糊决策法进行导流标准的比选，综合考虑工程的导流投资、经济损失、施工强度等决策指标，建立导流标准的多目标系统模糊决策模型，计算各指标的相对隶属度，根据隶属度最大为最优的原则，选择最优的导流标准方案。实例分析说明，导流标准的多目标系统模糊决策分析模型是可靠的、合适的。     

土壤污染治理决策系统的研究

本文在简单介绍了各类土壤污染治理方法特点的基础上，讨论了污染物类型、污染程度、土壤条件等因素对确定土壤污染治理方法的影响，从环境经济角度，对土壤污染治理方法进行了效果费用分析。为进行土壤污染治理方法的优化选择，建立了一套土壤污染治理决策系统的计算机软件，并对其进行了系统分析。     

广东省海洋捕捞产量灰色预测

为摸清今后海洋捕捞业的发展动态,以广东省海洋与渔业局提供的1996～2003年广东省海洋捕捞业统计数据为基础,分别建立灰色预测模型,通过模型对广东省海洋捕捞业发展趋势做出了近5年预测。预测结果,2010年海洋捕捞产量在167×104t,南海的捕捞产量为156.78×104t,远洋渔业产量为19.82×104t。通过对预测结果的可信度进行分析,进一步确定这种分析结果的合理性,从而为确定今后广东省海洋捕捞业发展方向,渔业生产部门制订投资决策和合理安排生产提供科学依据。     

Mitigating nitrous oxide emissions from agricultural soils by precision management

Nitrous oxide (N₂O) emissions make up a significant part of agricultural greenhouse gas emissions. There is an urgent need to identify new approaches to the mitigation of these emissions with emerging technology. In this short review four approaches to precision managements of agricultural systems are described based on examples of work being undertaken in the UK and New Zealand. They offer the opportunity for N₂O mitigation without any reduction in productivity. These approaches depend upon new sensor technology, modeling and spatial information with which to make management decisions and interventions that can both improve agricultural productivity and environmental protection.     

Marine-Derived Chitosan Nanoparticles Improved the Intestinal Histo-Morphometrical Features in Association with the Health and Immune Response of Grey Mullet (Liza ramada)

Marine-derived substances are known for their beneficial influences on aquatic animals’ performances and are recommended to improve intestinal health, immunity, and anti-oxidative status. The present study investigates the role of chitosan nanoparticles on the intestinal histo-morphometrical features in association with the health and immune response of Grey Mullet (Liza ramada). Chitosan nanoparticles are included in the diets at 0, 0.5, 1, and 2 g/kg and introduced to fish in a successive feeding trial for eight weeks. The final body weight (FBW), weight gain (WG), and specific growth rate (SGR) parameters are significantly increased while feed conversion ratio (FCR) decreases by chitosan nanoparticles compared to the control (p < 0.05). The morphometric analysis of the intestines reveals a significant improvement in villus height, villus width, and the number of goblet cells in chitosan-treated groups in a dose-dependent manner. Additionally, there is a positive correlation between the thickness of the enterocyte brush border and the chitosan dose, referring to an increasing absorptive activity. Histologically, the intestinal wall of Grey Mullet consists of four layers; mucosa, sub-mucosa, tunica muscularis (muscular layers), and serosa. The histological examination of the L. ramada intestine shows a normal histo-morphology. The epithelial layer of intestinal mucosa is thrown into elongated finger-like projections, the intestinal villi. The values of hemoglobin, hematocrit, red blood cells (RBCs), total protein (TP), albumin, and globulin are significantly increased in fish fed 1, and 2 g/kg of chitosan nanoparticles compared to fish fed 0 and 0.5 g/kg (p < 0.05). The highest levels of TP and albumin are observed in fish fed 1 g/kg diet (p < 0.05). The lysozyme activity and phagocytic index are significantly enhanced by feeding chitosan nanoparticles at 0.5, 1, and 2 g/kg, whereas the phagocytic activity is improved in fish fed 1 and 2 g/kg (p < 0.05). The highest lysozyme activity and phagocytic index are observed in fish fed 1 g/kg. SOD is significantly activated by feeding chitosan nanoparticles at 1 g/kg. Simultaneously, glutathione peroxidase (GPx) and catalase (CAT) activities also are enhanced by feeding chitosan at 1 and 2 g/kg, compared to fish fed 0 and 0.5 g/kg (p < 0.05). The highest GPx and CAT activities are observed in fish fed 1 g/kg (p < 0.05). Conversely, the malondialdehyde (MDA) levels are decreased by feeding chitosan at 1 and 2 g/kg, with the lowest being in fish fed 1 g/kg (p < 0.05). To summarize, the results elucidate that L. ramada fed dietary chitosan nanoparticles have a marked growth rate, immune response, and anti-oxidative response. These improvements are attributed to the potential role of chitosan nanoparticles in enhancing intestinal histo-morphometry and intestinal health. These results soundly support the possibility of using chitosan nanoparticles at 1–2 g/kg as a feasible functional supplement for aquatic animals.     

山地电力架空线路运维质量的智能评估

山地电力架空线路大多沿山架设,穿越树竹林,周边的树竹生长缺陷引发短路故障频繁发生,造成用电客户的大量投诉,用户体验满意度下降,抢修抢险费用大幅度攀升,严重影响供电公司的信誉和售电收益.为了提升用电客户的满意度和供电公司的售电受益,采用主成分分析方法提取了山地电力架空线路通道运维质量的3个主要影响因素,它们分别是树竹生长缺陷数、缺陷的消除数和通道运维资金使用合理度.对树竹的生长规律建立一个预测模型,从而得到树竹生长缺陷数的一个预测,根据预测结果划拨架空线路所在供电所的运维资金,消除树竹生长的缺陷数,再用Logistic回归模型,构建架空线路通道运维质量智能评估模型,使供电公司对架空线路运维实现基于数据决策的模式,尽可能保证供电线路零故障运行.