Optimal Ship Design and Simulator Development Using Systems Identification Techniques

使用系统识别技术优化船舶设计和模拟器开发

• 批准号: 9362135
• 负责人: Ramani Duraiswami
• 金额: $ 6.5万
• 依托单位: Dynaflow Inc
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1994
• 资助国家: 美国
• 起止时间: 1994-04-01 至 1995-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9362135&HistoricalAwards=false
• 关键词: Optimal; Ship; Design; Simulator; Development

9362135 Duraiswami Optimization in ship system design is a problem which, if properly formulated and solved, would have great economic impact. However, the problem cannot be easily cast mathematically as a classical single objective optimization problem. In ship system design a considerable number of criteria must be satisfied of which some can only be formulated in fuzzy terms. This is particularly true in modeling/simulating ship maneuverability, where the adequateness of a simulator can only be judged by consulting a ship master/pilot. To build a successful ship simulator, the equations of ship motion must be identified. This systems identification problem is ill-posed. In the proposed research we tackle the problem of developing a framework for optimizing design, by incorporating in the procedure the mathematical simulation and identification of ship dynamics in real time. The ill-posed problem will be formulated using multiple physical constraints/objectives, and solved via pareto- optimization. Efficient computer programs implementing our theory and algorithms will be developed. Phase I research will develop an inexpensive prototype ship maneuvering simulator, based on simple PC architectures. In future phases the developed tools will be used to develop procedures for optimal ship system design, and in assisting shipboard decision making.

小行星9362135 船舶系统优化设计是一个问题，如果适当地制定和解决，将有很大的经济影响。 然而，这个问题不能很容易地在数学上作为一个经典的单目标优化问题。 在船舶系统设计中，必须满足相当多的准则，其中有些准则只能用模糊术语表示。 这在船舶操纵性建模/仿真中尤其如此，其中模拟器的适当性只能通过咨询船舶船长/驾驶员来判断。 为了建立一个成功的船舶模拟器，必须确定船舶运动方程。 这个系统辨识问题是不适定的。 在拟议的研究中，我们解决的问题，开发一个优化设计的框架，通过将船舶动力学的数学模拟和识别过程中的真实的时间。 不适定问题将使用多个物理约束/目标来制定，并通过帕累托优化来解决。 有效的计算机程序实现我们的理论和算法将被开发。 第一阶段的研究将开发一个廉价的原型船舶操纵模拟器，基于简单的PC架构。 在未来的阶段中，开发的工具将用于开发最佳船舶系统设计程序，并协助船上决策。