Research of dissolved oxygen prediction in recirculating aquaculture systems based on deep belief network

Recirculating aquaculture has received more and more attention because of its high efficiency of treatment and recycling of aquaculture wastewater. The content of dissolved oxygen is an important indicator of control in recirculating aquaculture, its content and dynamic changes have great impact on the healthy growth of fish. However, changes of dissolved oxygen content are affected by many factors, and there is an obvious time lag between control regulation and effects of dissolved oxygen. To ensure the aquaculture production safety, it is necessary to predict the dissolved oxygen content in advance. The prediction model based on deep belief network has been proposed in this paper to realize the dissolved oxygen content prediction. A variational mode decomposition (VMD) data processing method has been adopted to evaluate the original data space, it takes the data which has been decomposed by the VMD as the input of deep belief network (DBN) to realize the prediction. The VMD method can effectively separate and denoise the raw data, highlight the relations among data features, and effectively improve the quality of the neural network input. The proposed model can quickly and accurately predict the dissolved oxygen content in time series, and the prediction performance meets the needs of actual production. When compared with bagging, AdaBoost, decision tree and convolutional neural network, the VMD-DBN model produces higher prediction accuracy and stability.     

刺参响应灿烂弧菌侵染差异microRNAs鉴定及靶基因分析

microRNA参与基因的转录后调控,在真核生物的生长发育、细胞分化和免疫防御等过程中发挥重要作用。刺参(Apostichopus japonicus)病害问题已成为产业发展的主要限制因素之一,而其病害发生的分子机制尚待进一步完善。本研究以刺参重大疾病“腐皮综合征”的重要致病原灿烂弧菌(Vibrios splendidus)为侵染菌株,通过人工侵染实验制备患病刺参样本,采用miRNA-seq技术对侵染组(PT16S)和对照组(PT10H)各3头刺参的体壁组织进行miRNA测序,通过相关生物信息学软件对miRNAs进行鉴定和分析,筛选差异表达miRNAs (DEmiRNAs)并预测其靶基因,构建关键调控途径的miRNA-mRNA调控网络。结果显示,PT10H组平均得到5 902 588条有效序列,194个已知miRNA和19个新的miRNA;PT16S组平均得到5 053 529条有效序列,182个已知miRNA和42个新的miRNA。对2组鉴定到的miRNA进行差异表达分析,共筛选到2个上调和11个下调的具有显著差异的DEmiRNAs (P≤0.05),上调的DEmiRNAs靶基因预测结合到3010个靶基因,注释到585个GO terms及24条信号通路(P≤0.05),下调的DEmiRNAs靶基因预测到19 072个靶基因,注释到514个GO terms以及22条信号通路(P≤0.05)。对筛选到的DEmiRNAs进行实时荧光定量PCR (qRT-PCR)验证,显示miRNA-seq与qRT-PCR的一致率达到70%。根据KEGG分析结果构建泛素介导的蛋白水解途径和Notch信号通路的miRNA-mRNA调控网络,结果显示,13个DEmiRNAs分别靶向结合134个与泛素介导的蛋白水解相关的mRNAs和109个与Notch信号通路相关的mRNAs,Aja-miR-184、Aja-miR-2478和Aja-miR-9277p等DEmiRNAs可能参与对Notch信号通路和对泛素介导的蛋白水解的调控。相关研究结果将为刺参疾病发生调控网络建立和机制解析提供依据。     

降低农村电网线损的措施

从改善电力网络功率分布、合理安排经济运行方式、加强农村电网改造等角度,论述了降低技术线损的措施;从制定线损指标计划、加强计量管理和抄收管理、建立管理线损考核制度和约束机制等方面入手,提出降低管理线损的具体措施。     

以优化的BP神经网络模型为基础研究山东近海口虾蛄空间分布与环境因子的关系

BP神经网络模型作为一种常用的机器学习方法,被广泛应用于物种分布模型,以解析生物分布与环境因子的关系。与传统回归模型相比,该模型可以灵活处理变量间的非线性关系,但其结构复杂,在参数设置方面存在不确定性,从而影响模型的预测与应用。根据2016—2017年山东近海口虾蛄渔业资源调查与环境数据,利用BP神经网络模型构建口虾蛄资源分布模型,同时利用数据分组处理算法(group method of data handling, GMDH)、遗传算法(genetic algorithm, GA)和自适应算法(adaptive algorithm)分别对模型输入变量、初始权值和隐节点数目3方面进行优化,构建7种不同组合优化模型。结果显示,7种模型的优化效果存在明显差异,单方面和两方面组合优化模型预测性能基本保持一致;而三方面共同优化其均方根误差与残差平方和分别为0.35和1.94,较初始模型的0.52和2.40更小,且相关系数最大为0.45,表明模型优化效果最好。对比优化前后发现,口虾蛄资源密度随纬度和底层盐度变化趋势基本保持一致,而随底层温度的升高,口虾蛄资源密度存在较大差异。此外,最优模型较初始模型增加水深为关键环境因子,对口虾蛄的资源密度具有重要影响。本研究进一步开发了BP神经网络模型参数优化的方法,证明了参数优化对BP模型的预测性能具有重要影响,模型优化对于分析口虾蛄资源密度与环境因子的关系具有重要意义。     

四川省农产品溯源服务体系和管理机制研究

分析了国内外农产品溯源发展状况,并结合目前四川省气象局开展的同气候品质认证相结合的溯源服务本身的特殊性,提出了完善农产品溯源服务体系和管理机制的相关建议.     

神经网络在养殖水质精准预测方面的研究进展

目前神经网络研究文献成果较多,虽然在水质精准预测方面起到了一定的参考,但由于文献缺少科学分类,使用率不高,导致学者难以找到研究切入点。针对这一问题,本文将神经网络方法在养殖区水质精准预测方面的文献按照海水和淡水两大领域进行分类,主要对每个领域所应用的预测模型从正反馈架构、循环架构和混合架构三个方向对海水时空序列文献进行分类研究和综述,发现混合架构模型的预测性能优于正反馈模型和循环架构模型,有利于提升不同深度水质预测模型的精度。另外,本文对基于神经网络方法的三维水质预测模型进行了初步探讨,发现学者的研究成果更多地集中在水表层和水中层的不同位置水质参数的变化方面,而神经网络方法对水表层水质预测精度比水中层和水深层水质预测精度高。     

基于北斗船位数据的流刺网网次和方向提取方法研究

渔船监测系统(vessel monitoring system,VMS)在渔业管理中发挥着重要作用,一直以来都受到各国及渔业组织的重视,我国近海渔船主要采用基于北斗的VMS系统。刺网是我国主要的捕捞作业方式之一,提出一种使用北斗数据提取流刺网作业网次以及网长和方向的方法。该方法使用阈值综合判别的方法判断捕捞作业的状态,通过航速、空间距离、时间间隔和航向变化的阈值判别作业时收网状态的船位点,然后根据收网状态的起始点判定放网状态的起始点。根据范例中提取的516组网次,使用程序判别的船位点和人工判断的船位点有74%相同,表明有较好的一致性,具有精度高、处理快速、实时程度高的特点。该方法可为我国渔业管理和相关研究提供新的手段,并可应用于捕捞控制、海上执法等多个领域。     

信息检索中提取检索概念之分析

在信息检索的研究和实践中．由于查询与文件集中词的不匹配现象导致一些相关的文件不能被成功地检索出来,这是影响检索效果的一个关键的问题。本文提出通过了解课题的背景知识和分析提取检索概念,提高检索效果。     

个人信息管理与风险规避研究

信息时代,人们面对海量的信息,如何获取、存储、组织使用个人信息,如何保护个人信息安全、规避风险,已变得刻不容缓,个人信息管理效果的好坏深刻影响着人们的生活和工作效率。失去控制和无组织的信息不仅不构成人们所需要的资源,相反,它会混淆视听、阻碍人们对信息的吸收和利用。因此,要克服个人信息的分散化和碎片化现象,加强对个人信息资源的管理、开发和利用,规避个人信息安全风险,是信息时代面对的主要任务。     

Is it a hindrance for an invasive aquatic species to spread across scattered habitat patches?

1. Despite their importance for aquatic biodiversity, ponds are among the most vulnerable freshwater habitats. Owing to their isolation in terrestrial environments, ponds are expected to be relatively well protected from biological invasions, but this depends on many factors. 2. The purpose of this study was to examine the influence of a range of variables (water quality, habitat, and landscape attributes) on the colonization of discrete ponds by a widespread aquatic invader, the red swamp crayfish Procambarus clarkii, which can disperse overland. 3. Investigations were conducted in two networks of ponds, each located in close proximity to a large invaded marsh. The two marshes under study differed in the length of time since the crayfish were introduced. 4. The proportions of colonized ponds and crayfish abundances were moderate in both networks, but higher in the network that had been invaded first. In both networks the distance to the marsh was the main predictor of pond colonization, considering similar energy costs to cross aquatic and terrestrial habitats for the recently invaded network, but assuming that dispersal was 25 times costlier across the terrestrial matrix than via streams for the earlier invaded network. Pond characteristics had no influence on crayfish occurrence in either network. Furthermore, predictions of pond invasion were lower for the recently invaded network. 5. The importance of the distance to the marsh indicates that natural dispersal was the main process of pond colonization by crayfish. Findings also suggested that overland dispersal was rare and costly. By contrast, streams were probably significant in facilitating crayfish dispersal. Differences between the two networks might arise from an invasion process still in progress in the recently invaded network. 6. From a management viewpoint, local actions are encouraged to prevent the spread of crayfish via streams. In addition, broader‐scale actions to mitigate other human disturbances would improve the outlook for pond biodiversity.