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。     

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．     

用单层分化法进行胚胎干细胞神经分化的研究

用单层分化法研究了 ES细胞从 0～ 316 h(第 14天 )的神经分化过程。无饲养层培养的 ES细胞在无血清的N2 B2 7培养基中开始神经分化。从 72 h开始 ,神经外胚层的特异性转录因子 Sox1表达 ,在克隆的边缘逐渐出现向外迁移的细胞 ,形状呈圆形或椭圆形 ,有 1～ 2个细胞突起。在 191h(第 8天 )之后 ,细胞可迁移至克隆边缘的 2 0 0 μm之外 ,这些有细长突起的细胞表达神经元特异性蛋白 Tau。以后 ,迁移细胞的突起开始交叠成网状。 316 h(第 14天 )时 ,细长的纤维可跨越在不同的克隆之间 ,长达上千微米。此时 ,免疫染色的结果表明 ,在克隆的中央有较多的圆形 Nestin阳性细胞 ,而克隆的边缘有大量的 β- tubulin 阳性细胞 ,它们交织成复杂的网状图像。阶段性表达基因的表达时序研究表明 ,ES细胞的特异性转录因子 Oct4开始减弱时 ,Sox1表达开始 ,此时细胞由全能状态进入早期神经分化状态 ;Tau的表达是 Sox1表达的下游事件。这一时序与胚胎发育中神经发生的时序相符。单层分化法能直观地展示 ES细胞的神经分化过程 ,可作为研究细胞分化、胚胎发育及药物筛选的体外模型     

电流模式神经网络

把电流模式电路引进了神经网络理论，提出了一种用电流模式电路构成的神经网络阈值处理单元，并由此构成了电流模式Ｈｏｐｆｉｅｌｄ神经网络，此网络中信号都是以电流的形式出现的。     

晚稻稻瘟病BP神经网络分区预报

应用相关分析方法分析了浙江省19个县1988～1999年晚稻稻瘟病发病与有关环境因子的关系,筛选了8个气象因子用于晚稻稻瘟病发生程度长期预报。根据各预报因子与稻瘟病发病程度相关性,采用邻接二维图论聚类分析法,将19个点(县)划分为4个生态区。每个生态区内运用BP神经网络技术建立模型,并进行拟合和试报。1997～1999年试报验证,在划分稻瘟病生态区的基础上,应用BP神经网络模型对稻瘟病进行长期预测预报是可行的,3年试报成功率分别是78.95％、84.21％和78.95％。文中还对该方法与过去常用的预报方法的试报结果作了比较。     

高寒地区燕麦根际联合固氮菌研究 Ι固氮菌分离及鉴定

结合气相色谱仪,利用乙炔还原等方法对高寒地区重要饲用植物——燕麦(Avenasativa)根际联合固氮菌进行了分离和鉴定。结果表明,该地区燕麦根际联合固氮菌株较少(8株),菌株分布以根系表面(RP)最多,根表土壤(RS)次之,距根系较远的土壤(NRS)和根内(HP)最少,即:RP>RS>NRS≥HP;菌株固氮酶活性相差较大(C2H4112.5～1147.9nmol/mL·h),具有较高固氮酶活性的菌株较少(C2H4大于500nmol/mL·h只有2株);菌株均为革兰氏阴性菌,经鉴定分属Azotobacter(3株)、Pseudomonas(3株)和Azospirillus(2株)3个属。相对于其它生境和植物,高寒地区燕麦根际联合固氮菌种类较为单一。     

基于Lasso和BP神经网络的组合预测及其应用——以居民消费支出预测为例

在变量选择的基础上，构建基于Lasso方法和BP神经网络的预测模型，并对我国城乡居民的消费支出进行预测，结果显示：基于Lasso方法和BP神经网络的组合预测精度要明显高于BP神经网络、Lasso方法的预测精度；在2014~2020年，我国农村居民消费增长率有所提升，城镇居民消费增长率减缓，城乡居民消费增长率之间的差距呈下降趋势，但短期内城乡居民消费差距依然难以缓和。     

Transplantation of human amnion epithelial cells improves learning and memory function in Alzheimer's disease-like pathology rat model

AIM: To observe the treatment effect and its immune regulation of human amnion epithelial cells (hAECs) on Alzheimer's disease (AD)-like pathology rat model. METHODS: The hAECs were isolated from amnion with trypsin digestion, and the phenotype of hAECs was analyzed by flow cytometry. SD rats (n=48) were randomly divided into sham control group, model group, medium group and hAECs group. AD-like pathology rat model was induced by bilateral intraventricular injection of lipopolysaccharide (LPS). hAECs (5×10⁵) were injected into the hippocampus of the AD-like pathology rats. At 2 weeks after transplantation, the animals were tested by Morris water maze to observe the function of learning and memory. The pathological change of the brain was observed by HE staining. The expression of amyloid β-protein 42(Aβ42) and Tau protein and the level of acetylcholine (ACh) in the injury brain were determined by immunohistochemistry. The survival and differentiation of hAECs in the hippocampus were measured by immunofluorescent technique. The percentages of lymphocyte subsets in the peripheral blood mononuclear cells were analyzed by flow cytometry. The contents of serum cytokines were detected by cytometric bead array. RESULTS: Compared with model group and medium group, hAECs group showed shortened escape latency (P<0.01), increased frequency of going through the platform (P<0.05), reduced loss of hippocampal neurons, decreased expression of Tau protein and Aβ42 in the hippocampus (P<0.05), increased ACh level in the hippocampus (P<0.05), decreased percentages of Th1 and Th17 subsets, increased percentages of Th2 and Treg cells (P<0.05), decreased concentrations of IFN-γ and IL-2 in the serum, and increased concentration of IL-4(P<0.05). CONCLUSION: hAECs improve the cognitive learning and memory function and alleviate pathologic damage of hippocampus through immune regulation in AD-like pathology rats.     

基于无线传感网络隧道亮度信号采集系统的研究

亮度信号是高速公路隧道节能照明控制系统必须检测的物理量。本文基于zigbee的技术特点，设计了一款基于无线传感网络的隧道亮度信号采集系统。系统的传感器节点负责亮度信号的采集，经CC2530处理发送，通过逐级跳转的方式把数据传输到协调器。协调器通过RS232串口与灯具控制器通信，以此实现对亮度信号的实时采集和可靠性传输。通过实验表明，该系统达到了设计的目的，并适用于高速公路隧道节能照明控制系统。