海上光污染对船舶夜航安全的影响与对策研究

发布时间：2018-05-30 03:13

本文选题：船舶夜航安全 + 船舶号灯　；参考：《大连海事大学》2015年博士论文

【摘要】：随着人类社会工业化与城市化进程的加快,以及人们对海洋资源开发与利用的深入,海上光污染作为一种新型污染形态,对船舶夜航安全产生了严重威胁。从探讨海上光污染的概念出发,利用光学、色度学及视觉工效学等原理,系统分析了海上光污染对船舶夜航安全影响的机理。在此基础上,提出海上光污染对船舶夜航安全的影响因子,并以此为主要评价指标设计了调查问卷；利用SPSS软件对有效问卷的数据进行统计分析,验证了前述机理分析的正确性。采用船舶号灯可识别性作为评价海上光污染程度的指标,建立了基于BP神经网络的船舶号灯识别模型；为了提高船舶号灯神经网络识别模型的识别效率,对克隆选择算法进行了针对性的改进,将灵敏度分析法用于确定寻优尺度因子,并用于BP神经网络的结构优化,提高了寻优的方向性与效率。通过对在海上拍摄的实景照片进行学习和识别的仿真,验证了基于改进克隆优化算法BP网络模型的有效性。为了提高船舶号灯在海上光污染水域的可识别性,提出了一种基于激光二极管阵列的船舶号灯系统设计方法。基于半导体激光器的发光原理及船舶号灯的技术属性要求,以激光二极管作为光源,提出激光船舶号灯的设计方法,并仿真分析了该新型号灯的灯光强度、光弧范围及颜色属性。仿真结果表明,该新型号灯满足《国际海上避碰规则》(下文简称《规则》)对船舶号灯各项技术属性的要求。从法律角度提出防治海上光污染对船舶夜航安全影响的两点措施。其一,在现行《规则》第三条中增加第14款——“高亮度水域”的定义,并修改《规则》第二章第三节的标题及内容,使得《规则》能够更好地指导避碰实践、厘清船舶间的避让责任；其二,在分析认定海上光污染的法律依据及我国对此立法缺位的基础上,提出防治海上光污染的立法构想。综上,本文分析并验证了海上光污染对船舶夜航安全影响的机理；建立了基于改进克隆算法的船舶号灯BP网络识别模型,并取得了良好的仿真效果；从技术与法律两个层面提出了防治海上光污染的措施。
[Abstract]:With the acceleration of the industrialization and urbanization of human society and the development and utilization of marine resources, the marine light pollution, as a new pollution form, poses a serious threat to the safety of ships' night navigation. Based on the concept of light pollution at sea and the principles of optics chromaticity and visual ergonomics the mechanism of the influence of light pollution on the safety of night navigation is analyzed systematically. On this basis, the influence factors of marine light pollution on the safety of ship night navigation are put forward, and the questionnaire is designed as the main evaluation index, and the effective questionnaire data are analyzed by SPSS software. The correctness of the mechanism analysis is verified. A ship light recognition model based on BP neural network is established to improve the recognition efficiency of ship light signal recognition model. The clonal selection algorithm is improved and the sensitivity analysis method is used to determine the optimization scale factor and to optimize the structure of BP neural network to improve the orientation and efficiency of the optimization. The effectiveness of BP neural network model based on the improved clone optimization algorithm is verified by learning and recognizing the real scene photos taken at sea. In order to improve the identifiability of ship lights in light-polluted waters at sea, a design method of ship lights system based on laser diode array is proposed. Based on the luminescence principle of semiconductor laser and the technical property requirement of ship light, the design method of laser ship light is put forward with laser diode as light source, and the light intensity of the new type lamp is simulated and analyzed. Range of light arc and color attribute. The simulation results show that the new model meets the requirements of the International Code for preventing collisions at Sea (hereinafter referred to as "Code") for the technical properties of ship lights. From the point of view of law, this paper puts forward two measures to prevent and control the effects of light pollution on the safety of night navigation of ships. First, to add the definition of "high luminous waters" to Article 3 of the existing rules, and to amend the title and content of section 3 of Chapter II of the rules so that the rules can better guide the practice of collision avoidance. Secondly, based on the analysis of the legal basis for the determination of light pollution at sea and the absence of legislation in our country, the legislative conception of preventing and controlling light pollution at sea is put forward. In summary, this paper analyzes and verifies the mechanism of the influence of marine light pollution on the safety of ship night navigation, establishes the BP network recognition model of ship lights based on improved clone algorithm, and obtains good simulation results. The measures of preventing and controlling light pollution at sea are put forward from two aspects of technology and law.
【学位授予单位】：大连海事大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：U676.1

【参考文献】