低模数水玻璃对橡胶轻骨料混凝土力学性能的影响

为了了解不同掺量水玻璃对橡胶轻骨料混凝土(RLC)力学性能的影响,在RLC中加入水玻璃,并用20％粉煤灰代替水泥,制作水玻璃掺量分别为0,2％,4％,6％,8％的混凝土试块.通过核磁共振试验研究其微观孔隙变化及强度形成机理;进行对数函数曲线预测和BP神经网络强度预测,判断其可靠性并对比优劣.结果表明:水玻璃的掺入使RL...     

Recording cerebellar neuron activities in swimming goldfish

ABSTRACT:   Neuronal activities were investigated in the cerebellum of immobilized and swimming goldfish Carassius auratus . Extracellularly recorded neural activities of the cerebellum in immobilized goldfish were characterized and classified into five types. Based on the waveforms and recording depths, these five neural activity types were estimated to originate from three identified classes of cerebellar neurons: Purkinje cells, eurydendroid cells, and granule cells. Chronic recording of cerebellar neuron activities in unrestrained goldfish was performed for more than 100 h. During the chronic recordings, a submersible amplifier attached to the goldfish head, and a multielectrode array developed for the present study were used. Neuronal activities in the cerebellum of free-swimming fish could also be classified into five types as in the immobilized condition. Firing patterns of two neurons identified as Purkinje cells and eurydendroid cells were analyzed during turning movements of the goldfish. The firing patterns of these neurons changed in relation to turning movements. Although some improvements are required, the chronic recording method developed in the present study can be applied to further investigations concerning the direct relationship between brain neural activities and certain behavior.     

平原河网河流生态修复体系理论研究

平原河网地区多城市密集,随着城市化水平的提高,城镇建设不断与河争地,河网中大量末端河道被封堵或填埋,河网水系自然形态结构受到影响,河网调蓄功能锐减,河流生态系统的健康状况受到严重的威胁,河网河流生态修复问题的解决迫在眉睫。以河流生态修复理论为研究基础,就如何建立合理的平原河网河流生态修复体系展开研究,以期为平原河网地区的河流生态修复提供参考依据。     

Preliminary study to evaluate the effects of dietary bile acids on growth performance and lipid metabolism of juvenile genetically improved farmed tilapia (Oreochromis niloticus) fed plant ingredient‐based diets

A 9‐week feeding experiment was conducted to investigate the effects of dietary bile acids (BAs) on juvenile genetically improved farmed tilapia (GIFT) (Oreochromis niloticus) based on the evaluations of growth performance and parameters relevant to lipid metabolism. Each of five vegetable protein‐based diets containing BAs at a level of 0, 0.05, 0.15, 0.45 or 1.35 g/kg diet was fed to three replicates with 40 fish (8.2 g per fish). The results showed that weight gain (WG) increased significantly with the increase in BAs from 0 to 0.15 g/kg diet and then decreased significantly at a higher BA supplementation. Dietary BAs significantly reduced the crude lipid content in the whole body, muscle and liver tissue of GIFT. Fish fed diet with 1.35 g BAs/kg diet developed serious nuclear migration and vacuolization in hepatocytes. Gall bladder appeared to contain white solid and has fragile capsules. Dietary BA supplementation had significant effects on serum biochemical indices and activities of lipid metabolism enzymes in liver and intestine. In conclusion, dietary bile acid supplementation (0.15 g/kg) can facilitate the lipid metabolism and therefore promote the growth of tilapia. However, overdosed dietary BAs induced gallstone development, disrupted lipid metabolism and depressed the growth performances of GIFT.     

工厂化水产养殖水质监测系统

工厂化水产养殖的密度高、风险大,养殖对象对pH、溶解氧、温度、氨氮、亚硝酸盐等水质参数的变化敏感,受影响严重,监测水质参数极为重要。本文针对工厂化水产养殖水质监测特点和需求,研发了工厂化水产养殖水质监测系统。分析研究pH、溶解氧、温度、亚硝酸盐等水质参数的阈值,设计水质监测数据无线采集节点和基于Zigbee的无线监测网络,建立水质监测系统软件平台。结果表明,该系统能够实现工厂化水产养殖水质实时监测,保证生产安全,提高水产养殖生产效率。     

基于数字孪生技术的平面渔网破损检测方法研究

为防止渔网破损造成养殖鱼类逃逸,有必要对网衣进行破损检测。为了克服人工检测劳动强度大且效率低下的缺点,实现渔网的精准实时监测,本研究提出了一种基于数字孪生的网衣破损检测方法,可利用传感器代替人工监测。该方法首先从渔网的数值仿真模型获取大量的仿真传感数据,然后将数据用于人工神经网络的训练与测试,最后生成可进行网衣破损识别的数字孪生体。数字孪生体可根据传感器监测到的数据来判断网衣是否发生破损。在数值模拟中,考虑各种波浪条件以及网衣的破损情况。在训练人工神经网络中,将有效波高Hs、谱峰周期Tp以及横纲竖纲的拉力值作为输入变量,将网衣完整状态以及破损状态作为输出。经过测试分析,该识别模型根据传感器数据识别网衣是否破损的平均准确率为94.32%,由此可见,数字孪生技术能准确检测到渔网的损坏,可以作为网衣破损检测的一种新方法。     

印度产紫菜Pyropia chauhanii优良品系的选育与特性分析

为培育出藻体薄且适合在我国南方高水温海区栽培的紫菜新品种,利用~(60)Co-γ射线辐照和高温胁迫处理印度产紫菜Pyropia chauhanii野生型品系(PC-WT)的叶状体,分离出优良品系PC-M,随后通过研究2个品系在耐高温性、生长、主要光合色素含量、单孢子和壳孢子放散量等方面的差异后发现,在18和23°C温度组中,2个品系的壳孢子存活率、分裂率和假根发生率均无显著性差异,但在27和29°C温度组中,PC-M品系的壳孢子存活率比PC-WT品系分别提高了250.7%和305.4%,分裂率分别提高了42.4%和67.1%,假根发生率分别提高了86.6%和175.3%;将在23°C下培养30 d的叶状体分别置于18、23、27和29°C下培养10 d,在18、23和27°C组中,PC-M品系的叶状体绝对生长率分别是PC-WT品系的5.1、5.3和7.5倍,特定生长率分别是PC-WT品系的1.3、1.3和1.8倍;在27°C下培养15 d或在29°C下培养10 d,PC-WT品系的叶状体均放散了大量的单孢子,藻体流失严重,仅剩下基部,而PC-M品系的叶状体均没有放散单孢子、藻体形态完整、光泽好、生长速率快,培养至30 d后才发生轻微的卷曲。与PC-WT品系相比,常温组(23°C)的PC-M品系的3种主要光合色素(chl.a、藻红蛋白和藻蓝蛋白)含量以及壳孢子放散量分别提高了39.4%、209.8%、94.8%和36.7%,但藻体的平均厚度反而减少了31.6%。上述结果证实,与PC-WT品系相比,PC-M品系具有藻体薄、色素含量高、生长快、耐高温、壳孢子放散量大、单孢子不放散等优点,有望被培育成适宜栽培的新品种。     

基于显著性增强和迁移学习的鱼类识别研究

针对水下鱼类无法快速准确识别的难点,提出一种具有图像主体自动增强功能的鱼类迁移学习方法。该方法将鱼类RGB图像转换至Lab颜色空间后,利用中央周边算子计算得到整个输入图像的显著性值,进而提供鱼类目标的潜在区域,并结合GrabCut算法获取鱼类分割图像,最终将融合分割图的原始图像送入优化后的残差网络中进行训练。通过对23种鱼类进行识别试验,结果显示,固定ImageNet数据集上ResNet-50预训练模型的conv1层和conv2层参数,微调高层参数的方法能够取得最好的识别效果,且在公开的Fish4Knowledge数据集上,该模型取得了最高的识别准确率,平均识别精度达到99.63%。与其他卷积神经网络方法的对比结果显示,本方法在Fish4Knowledge和Fish30Image数据集上的识别精度和时间性能均具有较大优势,其中识别准确率至少提升4.98%。多个数据集上的试验验证了模型的有效性。     

Graph theory in action: evaluating planned highway tracks based on connectivity measures

Maintaining connectivity among local populations in a fragmented landscape is crucial for the survival of many species. For isolated habitat patches, stochastic fluctuations and reduced gene flow can lead to high risk of extinction. The connectivity of the landscape is especially crucial for the carabid species living in the fragmented forests of the Bereg plain (NE Hungary and W Ukraine) because a highway will be constructed through the plain. Our purpose is to (1) evaluate the impacts of three possible highway tracks, (2) suggest a solution that is realistic with less impact on connectivity than other plans and (3) discuss how to decrease the disadvantageous effects of each track. Our results, based on a network analysis of landscape graph of patches and ecological corridors, indicate that the intended highway could have deleterious consequences on forest-living carabids. Relatively simple actions, like the establishment of stepping stones, could compensate for the loss of habitat connectivity and promote the survival of carabids, or minor modifications in one possible track could diminish its adverse effects. While many other studies would be needed for a comprehensive assessment of the biotic impact of the highway, we provide an example on the usefulness of network analysis for land use management. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

无线传感器网络技术在养殖业污染监测上的应用

综述了国际上无线传感器网络技术发展的特点和趋势,提出无线传感器网络技术在畜禽和水产养殖业污染监测上应用需要解决的2个技术问题:一是选择合理的IP无线传输方式;二是选择合理的系统结构,达到较高的性价比。设计出适合不同地域、不同生态环境的无线传感器网络的拓扑结构,实现在户外或野外环境下对畜禽和水产养殖业污染的监测。