基于船舶操纵性的船舶海上航行安全研究

发布时间：2018-10-24 21:44

【摘要】：船舶操纵性作为船舶重要的航行性能之一,一直以来与船舶海上航行安全有着紧密的联系,受到了国际海事组织、各国政府以及专家学者们的广泛关注。根据统计,近十年来与船舶操纵性有关的海上交通事故类型的事故数占海上交通事故总数超过30%,凸显出船舶操纵性在船舶海上航行安全中的重要性。同时,随着未来船舶向着智能化和无人化的趋势发展,降低了人为因素的影响,船舶操纵性在船舶海上航行安全中的重要性将逐渐增加。基于上述原因,为了减少与船舶操纵性相关的海上交通安全事故的发生数量,降低相应的海上人员和财产损失,保护好海洋生态环境,本文基于船舶操纵性,从以下三部分开展了船舶海上航行安全研究工作,首先利用实船操纵性试验获得的静态特征参数和指标开展船舶操纵性评价研究,然后基于动态的船舶运动数学模型开展船舶操纵运动控制设计与分析研究,最后同时利用动态的船舶操纵运动过程和静态的操纵性试验特征参数开展紧迫局面下船舶操纵避碰研究,详情如下:1.基于IMO和ABS《船舶操纵性标准(准则)》,利用模糊网络分析法提出了一种新颖的基于等级的船舶操纵性评价模型。该评价模型能够初步解决船舶操纵性评价中各操纵性因素间相互关联、相互影响的问题,并改进现有船舶操纵性评价中各操纵性因素权重不合理的缺陷。基于该评价模型,使得船舶操纵性评价的过程及结果更为合理,并对各操纵性因素赋予了合理的权重值。最后,利用三艘船舶实船操纵性试验数据确定船舶操纵性等级,并进行分析和对比研究,结果表明该评价模型的合理性和有效性。通过确定船舶操纵性等级,能直观、准确的反映船舶操纵性是否满足海上航行安全的最低要求,并能够为进一步提升船舶操纵性和海上航行安全水平提供理论支撑。2.针对一类增益已知的不确定严格反馈离散时间非线性系统,提出了一种新颖的自适应神经网络控制设计,并进一步对增益未知的船舶操纵运动离散时间非线性系统进行控制设计与分析。通过将单一神经网络控制技术和最少学习参数算法相结合,所提出控制算法能够较好的解决常规自适应后推控制设计所存在的"复杂性爆炸"问题和"维数灾难"问题,并在控制设计中通过引入辅助设计系统以有效的克服输入饱和限制问题。根据所提方法设计的控制器结构简单,仅包含一个实际控制律与一个自适应律,在线学习的神经网络输入变量和权重都大为减少,最终整个系统需要在线学习的可调参数减少至一个。与现有控制算法相比,本文所提出的控制结构更为简单,学习参数最少,并且计算量也最小。在船舶操纵运动系统进行控制设计过程中,通过考虑由实船船舶操纵性试验结果分析得到的船舶操纵性指数,使得最终的船舶操纵运动控制系统能较为准确反映出船舶在海上实际操纵过程中的操纵性,为进一步开展船舶避碰研究提供坚实的理论基础。3.开展处于紧迫局面会遇态势下的船舶操纵避碰研究,首先,针对紧迫局面定义下的避碰态势,通过将船舶操纵运动数学模型、船舶操纵运动控制系统、以及船舶动态避碰要素计算模型相结合,提出了一种基于操纵性的船舶操纵避碰动态辅助系统。该辅助系统能基本上消除过去在船舶避碰过程中常常忽略船舶操纵性的缺陷,并利用该辅助系统对处于紧迫局面内两船的三种会遇态势进行了仿真研究,仿真结果证明了该辅助系统的合理性和有效性,也表明在避碰过程中考虑船舶操纵性的必要性和重要性。随后,基于船舶操纵性和协同学理论提出了一种新颖的协同船舶领域模型。基于协同学理论和分析大量实船操纵旋回试验数据,所提出的协同船舶领域模型初步解决了常规船舶领域中未考虑会遇船舶间相互影响的问题以及未充分考虑船舶操纵性的不足。与常规船舶领域相比,所提出的协同船舶领域更为合理,为减少由于船舶操纵性因素所导致的船舶碰撞事故提供有效的理论基础,以切实提供船舶海上航行的安全水平。通过开展上述有关船舶操纵性的研究工作,本文不仅能为当前船舶海上航行安全,同时也为未来智能化和无人化的船舶海上航行安全提供了理论支撑,对提高当前和未来的船舶海上航行安全水平具有一定的理论和现实意义。
[Abstract]:As one of the important navigation performances of the ship, the ship maneuverability has been closely related to the maritime safety of ships, and has been paid close attention by the International Maritime Organization (IMO), the governments of various countries and experts and scholars. According to the statistics, the accidents of maritime traffic accidents related to ship maneuverability over the last decade accounted for more than 30% of total maritime traffic accidents, which highlight the importance of ship maneuverability in marine navigation safety. At the same time, as the future ships develop towards the intelligent and unmanned trend, the influence of human factors is reduced, and the importance of ship maneuverability in marine navigation safety will gradually increase. Based on the above reasons, in order to reduce the occurrence quantity of maritime traffic safety accidents related to ship maneuverability, the corresponding marine personnel and property losses are reduced, and the marine ecological environment is protected, From the following three parts, the safety research work of marine navigation is carried out, firstly, the ship maneuverability evaluation is carried out by using the static characteristic parameters and indexes obtained by the actual ship maneuverability test, and then the ship steering motion control design and analysis research is carried out on the basis of the dynamic ship motion mathematical model, At last, the dynamic ship maneuvering motion process and the static maneuverability test characteristic parameters are used to carry out collision avoidance study under the urgent situation, and the details are as follows: 1. In this paper, a novel class-based ship maneuverability evaluation model is proposed based on fuzzy network analysis. The evaluation model can initially solve the problems of mutual correlation and mutual influence among maneuverability factors in ship maneuverability evaluation, and improve the defects of unreasonable weight of various maneuverability factors in the existing ship maneuverability evaluation. Based on the evaluation model, the process and result of ship maneuverability evaluation are more reasonable, and the reasonable weight value is given to the handling factors. Finally, the ship maneuverability grade is determined by using the ship maneuverability test data of three ships, and the analysis and comparison are carried out. The results show that the evaluation model is reasonable and effective. By determining the ship maneuverability level, it is possible to intuitively and accurately reflect whether the ship's maneuverability meets the minimum requirements of maritime navigation safety, and can provide theoretical support for further improving the ship maneuverability and the safety level of sea navigation. In this paper, a novel adaptive neural network control design is proposed for a class of gain-known discrete-time nonlinear systems that do not determine strict feedback, and the control design and analysis of discrete-time nonlinear systems with unknown gain are presented. By combining the single neural network control technique with the minimum learning parameter algorithm, the proposed control algorithm can solve the problems existing in the conventional self-adaptive push control design. "complexity explosion" Question and "dimension disaster" the problem is solved and the input saturation limit problem is effectively overcome by introducing a design aid system in the control design. According to the method designed by the method, the structure of the controller is simple, only one actual control law and an adaptive law are contained, the input variables and the weights of the neural network input online learning are greatly reduced, and the whole system needs to reduce the adjustable parameters of online learning to one. Compared with the existing control algorithm, the control structure proposed in this paper is simpler, the learning parameters are minimum, and the calculation amount is also the minimum. in that control design proces of the ship steer motion system, the ship maneuverability index obtained by the analysis of the ship maneuverability test result of the real ship is taken into account, so that the final ship steer motion control system can accurately reflect the maneuverability of the ship in the actual operation of the sea, It provides a solid theoretical foundation for further research on ship collision avoidance. carrying out collision avoidance research on ships under the situation of urgent situations, firstly, aiming at the collision avoidance situation under the definition of an urgent situation, combining the ship steering motion mathematical model, the ship steering motion control system and the ship dynamic collision avoidance element calculation model, A dynamic auxiliary system for ship handling collision avoidance based on maneuverability is proposed. the auxiliary system can basically eliminate the defects that the ship maneuverability is often neglected during the collision avoidance process of the ship, and the simulation results prove that the auxiliary system is reasonable and effective, It also shows the necessity and importance of considering ship maneuverability during collision avoidance. Then, a novel cooperative ship domain model is proposed based on ship maneuverability and cooperative student theory. Based on the theory of Synergetics and the analysis of the experimental data of large-scale real-ship maneuvering cycle, the proposed cooperative ship domain model has initially solved the problem of mutual influence among ships in the field of conventional ships, as well as the lack of full consideration of the ship's handling ability. Compared with the conventional ship field, the proposed cooperative ship field is more reasonable, and provides an effective theoretical basis for reducing ship collision accidents caused by ship maneuverability, so as to effectively provide the safety level of ship navigation at sea. Through carrying out the above research work on ship maneuverability, this paper not only provides theoretical support for the navigation safety of the ship at sea, but also provides theoretical support for the future intelligent and unmanned marine navigation safety. It has some theoretical and practical significance for improving the safety level of marine navigation at present and in the future.
【学位授予单位】：大连海事大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U661.33;U676.1

【相似文献】