Prediction of baking results from farinograph measurements by using stepwise linear regression and artificial neuronal networks

In an effort to extract additional data from farinograph experiments a model was developed to simulate the measurements and correlate the parameters of the model with results from baking tests. This additional information can be used in bakeries to predict the baking properties of the flours and adjust the recipes to maintain a constant product quality. For this eight different flours were characterized with a farinograph and 13 different results from baking experiments. An approach with five nonlinear differential equations was able to model the farinograph measurements very well (average R² = 0.995 ± 0.005). While a stepwise multilinear regression only showed weak correlations in cross validation between a single parameter of the model and the baking volume (R² = 0.745) and the volume yield (R² = 0.796) respectively, the artificial neuronal network was more successful. For the baking weight (R² = 0.926), the dough yield gross (R² = 0.909) and net (R² = 0.913) strong correlations were found. A good correlation for the baking volume (R² = 0.853) was also determined, while the volume yield showed comparable results to the linear regression (R² = 0.792).     

基于卷积神经网络的拖拉机工况识别

农机工况识别在细化农机作业状态和帮助掌握区域污染物排放趋势方面有着重要的研究价值。基于拖拉机不同运行状态下的行驶速度、发动机转速以及实时油耗等时间序列,首次提出将图像识别方法引入到拖拉机工况识别中的思路,并分别应用参数优化的支持向量机与卷积神经网络对实际作业拖拉机工况进行研究。结果表明:(1)基于参数优化的支持向量机可以较好地实现样本点的工况识别且识别准确度达到99.851 9%,但无法实现农机工况的连续性识别,同时无法对农机工况转换阶段进行有效识别。(2)以拖拉机运行速度与发动机转速等信息构建样本图像来描述农机工况变化的数据表达,并在此基础上应用卷积神经网络可以有效实现农机工况的连续性识别,且识别准确率可以达到93.3%。本研究在农机工况识别方面具有一定参考价值,并为后续农机不同工况下区域污染物排放研究提供技术支持。     

面向养殖水体分布差异的COD光谱法检测研究

面向养殖水体，传统光谱法对化学需氧量（Chemical Oxygen Demand，COD）检测模型构建的基础：源域（现有样本库）与目标域（检测地水体）间光谱数据独立同分布。但是当源域与目标域分布间存在差异时，由源域得到的低误差模型常在目标域上表现下滑。针对该问题，提出面向UV Vis光谱的域对抗训练网络（DAUVwpNet），将分布不同的源域和目标域数据映射至相同分布的特征空间中，使其在该空间的分布距离尽可能接近，从而在特征空间中对源域训练的目标函数也可以迁移至目标域上，以降低模型在目标域的误差。试验表明：面向同一批测试数据，DAUVwpNet的预测误差为0.78，要低于传统模型的预测误差（0.85）；DAUVwpNet预测值与实测值间相关系数为0.95，要高于传统模型的相关系数（0.89）。表明了该网络能够较好对齐两域特征空间数据分布，降低因分布差异带来的COD检测误差。     

“互联网+"背景下农村网络扶贫新思路的探讨

本文通过分析“互联网+”时代下网络扶贫的三种新模式，指出了政府、企业、农户三方力量 各自在网络扶贫中的定位和职责，进一步夯实网络扶贫基础，从而提高扶贫治理成效。     

The effects of density and artificial substrate on intensive shrimp Litopenaeus vannamei nursery production

Recirculating aquaculture systems (RAS) can be installed indoors, allowing year-round production of tropical animals in nearly any climate. A nursery phase is commonly used in Litopenaeus vannamei production since it allows for enhanced biosecurity and better quantification of animals while reducing space requirements. However, it is unclear whether animal density and inclusion of artificial substrate may improve shrimp performance during the nursery phase. In this experiment, we compared shrimp production parameters in two stocking densities with or without the use of an artificial substrate by creating four treatments: low-density LD; 1500 PL/m⁻³, low-density with substrate LDS, high-density HD; 3000 PL/m⁻³), and high-density with substrate (HDS). The LDS and HDS treatments included 0.46-m² of high-density polyethylene 2.5-cm mesh as a substrate, which increased the tank surface area by 21 %. Each treatment was randomly assigned to four 160-l culture tanks, each with a biofilter. The shrimp had an initial weight of 4 mg and were grown for 50 days. The low-density treatments had significantly higher dissolved oxygen (DO) and pH than the high-density treatments (P ≤ 0.001). Specifically, LDS had the highest DO and pH followed by the LD, HD, and HDS treatments, respectively. High-density treatments had significantly higher NO₂-N levels than low-density treatments during week 2 of the experiment when an unusually high concentration of nitrite was observed. FCR was significantly lower in both low-density treatments than in high-density treatments. At harvest, the total biomass (kg m⁻³) was significantly higher in high-density treatments than in low-density treatments (P ≤ 0.001), and the HDS treatment had a significantly greater biomass output than HD. Producers should consider artificial substrate and higher densities during nursery production to maximize shrimp production; however, the effects on water quality should also be taken into account.     

人工神经网络在电力系统短期负荷预测中的应用

根据电力系统短期负荷变化的特性，提出BP模型在实际负荷预测应用中的方法和步骤．对BP网络结构、样本空间、收敛性等作了有针对性的研究．结果表明：多层神经网络应用于电力系统短期负荷预测是可行和有效的．其预报结果比传统的负荷预测方法更准确、经济、效果更好．     

论电子阅览室的知识导航与信息服务

本文立足于知识导航与信息服务，对网络环境下电子阅览室的服务内容、服务保障、目前面临的现状进行了论述，并提出了服务对策。     

扁穗牛鞭草人工草地中空心莲子草杂草防除效果研究

利用水花生净除莠剂可有效地防除扁穗牛鞭草人工草地中的杂草空心莲子草,平均株防效达26.7%～99.8%,并且在较高防效下药剂对扁穗牛鞭草较为安全。在空心莲子草杂草生长盛期,14%的水花生净乳油以浓度为675ml/hm2～900ml/hm2、各处理对水600ml的防除效果最佳,可达到100%。综合考虑,以525ml/hm2～675ml/hm2的浓度为宜。     

岭西电网经济分析

本文对岭西电网的经济运行调整进行了分析、总结。并应用《岭西电网经济运行计算》系统的计算结果，提出了岭西电网的经济运行调整办法。     

落叶松人工林林分径级结构的研究

本文描述了落叶松人工林抚育后径级结构特征。试验表明，抚育后的林分在一定密度条件下，其密度遵从Ｗｅｉｂｕｌｌ分布，并能保持相当长的时间不变；其密度较大，Ｗｅｉｂｕｌｌ参数值在２～３间变动，ｂ值在６～８间变动；稀疏时不能用Ｗｅｉｂｕｌｌ分布模拟。