远洋船舶经济航速优化预测算法研究

发布时间：2018-09-08 09:15

【摘要】：20世纪80年代以来，随着经济全球化的高速发展，世界各区域经济联系日益紧密，不同商业区域之间的货运需求极其旺盛，引领世界航运业不断发展，由于加入世界贸易组织开放市场，我国的航运企业面临着与全世界的对手的激烈竞争，而且当前中东局势动荡，燃油价格变幻莫测和再攀新高，燃油成本已经成为船舶运输成本中最大的支出项，并且所占比例不断增加。因此，，为了在激烈的市场竞争中取得先机，保持自身的竞争力，控制燃油成本已经成为航运企业成本控制的重中之重。其次，节约资源、保护环境是我国的基本国策。党的十八大明确指出：“要节约集约利用资源，推动资源利用方式根本转变，加强全过程节约管理，大幅降低能源、水、土地消耗强度，提高利用效率和效益。推动能源生产和消费革命，控制能源消费总量，加强节能降耗，支持节能低碳产业和新能源、可再生能源发展，确保国家能源安全[1]”。船舶运输所属的交通运输行业是建设资源节约型、环境友好型社会的重要领域，节约资源、保护环境是发展现代化交通运输业的重要内容[2]。另外，由于船舶驱动的特点，螺旋桨消耗的功率与主机的转速成平方的关系，所以航速的优化和降低可以节省大量的燃油消耗。综上所述，船舶经济航速预测算法的研究与开发，对于我国资源节约环境友好行交通运输业的建立，对于航运企业的企业经营、生存发展以及对于环境友好型社会的建立都具有重要的意义，是现代航运业发展中不可或缺的一个环节。船舶经济航速预测系统的研究与开发具有重要的经济和社会意义，能够实现经济节能与环境保护的宏伟目标。本课题主要从节能经济这一指导指标出发，期望实现最大的经济效益和最小的燃油消耗的良好结合的目标，以远洋航行的舰船作为研究对象，通过前期的在真实远洋状态下实验航行所得的船舶现场数据，利用数据挖掘技术以及数据预处理技术对真实数据进行处理筛选，利用统计机器学习工具通过机器学习算法训练建立此舰船模型，在此基础上通过新型演化算法对关系模型的结构参数进行寻优。最终能够给舰船在大洋航行时，针对不同的海况和航行状态、船舶参数提供实时的经济航速，保证船舶航行的经济可靠性，实现舰船燃油消耗的降低。
[Abstract]:Since the 1980s, with the rapid development of economic globalization, the economic ties between different regions of the world are increasingly close, the demand for freight transport between different commercial areas is extremely strong, leading the development of the world shipping industry. As a result of China's entry into the World Trade Organization (WTO), our shipping enterprises are facing fierce competition with their competitors from all over the world. Moreover, the current situation in the Middle East is volatile, fuel prices are unpredictable and high again. Fuel cost has become the largest expenditure item in shipping cost, and the proportion is increasing. Therefore, in order to obtain the first opportunity in the fierce market competition and maintain its own competitiveness, controlling the fuel cost has become the most important part of the cost control of shipping enterprises. Second, saving resources and protecting the environment is the basic national policy of our country. The 18th National Congress of the CPC pointed out clearly: "it is necessary to economize on intensive utilization of resources, promote the fundamental transformation of the way of using resources, strengthen the whole process of economizing management, substantially reduce the intensity of energy, water and land consumption, and improve the efficiency and efficiency of utilization." To promote the revolution in energy production and consumption, to control the total amount of energy consumption, to strengthen energy conservation and reduce consumption, to support energy-saving and low-carbon industries and new energy sources, to develop renewable energy sources, and to ensure national energy security [1]. The shipping industry is an important field in the construction of a resource-saving and environment-friendly society. Saving resources and protecting the environment is an important part of the development of modern transportation [2]. In addition due to the characteristics of ship driving the power consumed by the propeller is square to the engine speed so the optimization and reduction of the speed can save a lot of fuel consumption. To sum up, the research and development of ship economic speed prediction algorithm is beneficial to the establishment of resource saving and environmental friendly transportation industry in our country, and to the management of shipping enterprises. Survival and development as well as the establishment of an environmentally friendly society are of great significance and are an indispensable link in the development of modern shipping industry. The research and development of ship economic speed prediction system has important economic and social significance and can realize the grand goal of economic energy saving and environmental protection. Based on the guiding index of energy saving and economy, this paper hopes to realize the goal of combining the maximum economic benefit with the minimum fuel consumption, and takes the ocean-going ships as the research object. The actual data are processed and screened by using data mining technology and data preprocessing technology through the previous ship site data obtained from the experimental voyage under the real ocean-going condition, and the data mining technology and the data preprocessing technology are used to process and screen the real data. Statistical machine learning tools are used to build the ship model through machine learning algorithm training. Based on this, the structural parameters of the relational model are optimized by a new evolutionary algorithm. Finally, it can provide real-time economic speed for different sea conditions and navigational states, ensure the economic reliability of ship navigation, and achieve the reduction of fuel consumption of the ship when the ship is sailing in the ocean.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：U695.2

【相似文献】