基于高分辨率影像的近海养殖区的一种自动提取方法

近海养殖调查在海岸带资源开发利用规划、环境保护与沿岸海上交通规划中处于重要地位。利用遥感技术开展近海养殖调查与分析,是当前的海岸带与湿地遥感的热点方向之一。以福建省宁德市三都澳海区为实验区,在分析近海养殖的渔排和海带/紫菜养殖区光谱特征的基础上,建立用于近海养殖区快速检测的光谱特征指数,在此基础上利用统计均值纹理与阈值检测算法,并结合近海养殖区的几何形状判断,实现了不同类型近海养殖区的高精度自动提取。实验结果表明:该方法能够快速准确地监测不同类型养殖区的分布情况,总体检测精度达到90%以上,可以为管理部门科学地规划管理海岸带养殖区提供有效的信息支持和辅助决策分析。     

半潜式波浪能养殖平台与系泊系统的耦合动力分析

以“澎湖号”半潜式波浪能养殖平台为研究对象,开展养殖平台及其系泊系统在浪、流联合作用下水动力性能的研究。基于三维势流理论,应用一阶泰勒展开边界元方法对其频域水动力进行计算,研究不同浪向下平台的运动响应和二阶波浪漂移力。采用间接时域分析方法和异步耦合方法实现平台和系泊系统的时域耦合动力求解,对不同浪流组合工况下平台的时域运动与锚链张力进行分析。研究表明,在南海常规海况中平台稳定性较好,立柱间水动力干扰尤为明显,顺浪时平台所受的二阶波浪漂移力最小。在与系泊系统的耦合作用下,平台的横摇与纵摇运动受入射波浪向的影响较大,入射波波高对顺浪下的平台横摇运动和横浪与斜浪下平台纵摇运动影响明显。     

不同填料强化生态浮床对淡水养殖水体的净化效果

人工填料强化生态浮床是近年来发展起来的一种新兴生态工程修复技术。选取3种不同类型的人工填料与挺水植物组合,分析其对高磷浓度养殖水体和上海市养殖池塘水体的净化效果。结果表明,在高磷浓度(2.36~3.42 mg/L)养殖水体中,人工填料强化生态浮床对TN、TP、COD的平均去除率分别为34.9%,12.2%,26.1%,且3组系统之间差异不明显。在上海市养殖水体平均浓度条件下,3组系统对TN和TP的去除能力为:立体弹性填料强化生态浮床＞组合填料强化生态浮床＞生物绳填料强化生态浮床,立体弹性填料强化生态浮床对TN和TP的去除率分别为83.9%和69.3%;3组系统对COD和NH₄⁺-N的去除率均在70%以上。因此,立体弹性填料强化生态浮床可以考虑应用于上海市淡水养殖池塘水体的生态修复工程。     

构建设施型精准化循环水养殖系统的技术研究

鉴于当前水资源短缺、环境恶化、水产养殖业高速发展,以及保障食品安全等问题的不断呈现,促使资源型工厂化流水养殖模式急需向高效型循环水养殖方向转变,这是产业发展提升的必然趋势。本文以工业化养殖理念为指导、节能减排为目标,以鲆鲽类产业技术体系构建的循环水养殖模式为样板,按循环水养殖系统(RAS)的设计原理,优化养殖工艺流程,构建一种设施型精准化的RAS,并初步分析探讨了海水鱼类在此集成系统中的应用效果。该系统将为广大水产养殖企业提供一种建设成本低廉、系统运行稳定可控、容易管理的实用新型养殖模式,也可为下一步发展现代水产精准养殖工程体系提供理论和实践依据。     

国内外工厂化循环水养殖模式水质处理研究进展

循环水养殖模式是一种新型的高效养殖模式,以养殖用水净化后循环利用为核心特征。节电、节水、节地。符合当前国家提出的循环经济、节能减排、转变经济增长方式的战略需求。本文以循环水养殖模式工艺流程为主线,结合近两年养殖模式科学研究和产业发展,围绕水质处理和应用,分别对大颗粒物滤除、臭氧消毒、气浮综合处理、生物滤池、紫外线杀菌、PH调节等水质处理环节进行了总结和探讨,为今后适用于中国国情的循环水养殖模式的标准的水处理工艺建立提供相应的参考。     

水产养殖环境因子监控系统

鉴于水质对水产养殖的重要性,研制了水产养殖环境因子监控系统。该系统由上位机和下位机组成,上位机由PC实现,下位机以AT89S51单片机为核心。系统辅以专家系统,可以实现水产养殖环境因子的实时监控、数据的采集、显示、存储等功能。     

基于Modbus协议的Android端工厂化养殖移动监测软件

设计了基于Modbus协议的Android端移动监测软件。该软件运行在Android平台上,通过蓝牙协议套接Modbus协议获取养殖水参数采集卡数据,从而实现养殖现场数据的移动收发。该软件具有高度稳定性和专用性,易于使用、维护和升级。     

具有SE结构的改进的PSPNet用于紫菜养殖区识别

目前深度学习技术被广泛应用于解决实际问题的各个领域,其中在筏式海水养殖区识别方面也取得了很大的进展。但由于海水养殖区环境复杂,导致容易产生误分问题。本文主要利用高分遥感影像,通过人工标注三分类标签,利用具有SE结构的改进的PSPNet模型来进行紫菜养殖区的识别工作,有效的改善了小样本误分问题。PSPNet能同时捕获紫菜养殖区的高级语义特征以及不同尺度的空间信息,SE结构通过学习的方式来自动获取到每个特征通道的重要程度,提升有用的特征并抑制对当前任务用处不大的特征,以提高小样本识别率。实验结果表明,与经典的FCN,U-Net网络相比该方法在紫菜养殖区的识别上有更好的效果,能够实现紫菜养殖区的提取应用。     

Multi-residue and multi-class determination of antibiotics in gilthead sea bream (Sparus aurata) by ultra high-performance liquid chromatography-tandem mass spectrometry

This paper describes a method for the determination of 41 antibiotics from seven different classes in gilthead sea bream (Sparus aurata) by ultra-high-performance liquid chromatography with tandem mass spectrometry (UPLC-MS/MS). Sulfonamides, trimethoprim, tetracyclines, macrolides, quinolones, penicillins and chloramphenicol were simultaneously determined. Fourteen procedures for sample treatment were tested and an extraction with acetonitrile and ethylenediaminetetra acetic acid was found to be the best option. The methodology was validated in accordance with Decision 2002/657/EC. Precision in terms of relative standard deviation (RSD) was under 17% for all compounds, and the recoveries ranged from 92% to 111%. CCα and CCβ were determined according to the maximum residue limit or the minimum required performance limit, when necessary. The validation provided evidence that the method was suitable for application in routine analysis for the detection and confirmation of antibiotics in muscle of gilthead sea bream, an important and intensively produced fish in aquaculture.     

Optimization of microwave-assisted extraction with saponification (MAES) for the determination of polybrominated flame retardants in aquaculture samples

The efficiency of microwave-assisted extraction with saponification (MAES) for the determination of seven polybrominated flame retardants (polybrominated biphenyls, PBBs; and polybrominated diphenyl ethers, PBDEs) in aquaculture samples is described and compared with microwave-assisted extraction (MAE). Chemometric techniques based on experimental designs and desirability functions were used for simultaneous optimization of the operational parameters used in both MAES and MAE processes. Application of MAES to this group of contaminants in aquaculture samples, which had not been previously applied to this type of analytes, was shown to be superior to MAE in terms of extraction efficiency, extraction time and lipid content extracted from complex matrices (0.7% as against 18.0% for MAE extracts). PBBs and PBDEs were determined by gas chromatography with micro-electron capture detection (GC-μECD). The quantification limits for the analytes were 40–750 pg g^?1 (except for BB-15, which was 1.43 ng g^?1). Precision for MAES-GC-μECD (%RSD?<?11%) was significantly better than for MAE-GC-μECD (%RSD?<?20%). The accuracy of both optimized methods was satisfactorily demonstrated by analysis of appropriate certified reference material (CRM), WMF-01.