优化算法对船舶可靠性的影响分析
|
摘要
通过构建径向基(Radial Basis Function,RBF)神经网络近似模型,分别采用自适应模拟退火算法(Adaptive Simulated Annealing,ASA)、多岛遗传算法(Multi-Island Genetic Algorithm,MIGA)、粒子群法(Particle Swarm Optimization,PSO)3种优化算法,以重量最轻为目标,对油船整体舱段进行基于CCS规范的优化设计;之后利用Abaqus非线性有限元法计算了轻量化舱段的极限强度,并通过建立极限状态方程,进一步研究优化算法对船体极限强度可靠性的影响;最后,为直观、量化地反映船舶的风险变化程度,通过提出失效概率折减率μ,发现PSO-RBF算法下减重单位舱段重量,船体梁失效概率增加最少。
In order to evaluate the effect of optimization algorithm on the ultimate strength reliability of the oil tanker,this paper combines iSIGHT optimization software and MSC finite element software, using parametric finite element modeling technology, optimized Latin experimental design method, radial basis function neural network theory, adaptive simulated annealing algorithm and other technical methods to optimize the design of the oil tanker section. Then, the paper uses the non-linear finite element method to assess the ultimate strength of the lightweight oil tanker section. According to the principle of the reliability, the first-order second-moment method is adopted to calculate the reliability and the failure probability of different optimization algorithms. Finally, the results indicate that if per unit weight reduced, the hull girder failure probability would increase the least under the PSO-RBF algorithm.
In order to evaluate the effect of optimization algorithm on the ultimate strength reliability of the oil tanker,this paper combines iSIGHT optimization software and MSC finite element software, using parametric finite element modeling technology, optimized Latin experimental design method, radial basis function neural network theory, adaptive simulated annealing algorithm and other technical methods to optimize the design of the oil tanker section. Then, the paper uses the non-linear finite element method to assess the ultimate strength of the lightweight oil tanker section. According to the principle of the reliability, the first-order second-moment method is adopted to calculate the reliability and the failure probability of different optimization algorithms. Finally, the results indicate that if per unit weight reduced, the hull girder failure probability would increase the least under the PSO-RBF algorithm.
引文
[1]INGBER L.Adaptive simulated annealing(ASA):Lessons learned[J].2000,25(1):33-54.
[2]丁泉惠,王森,黄修长,等.基于有限元法和多岛遗传算法的飞轮结构参数优化设计[J].噪声与振动控制,2016,36(2):56-60.
[3]KENNED Y J,EBERHART R.Particle swarm optimization[A].Perth,Australia,1995:1942-1948.
[4]EBERHART R,KENNEDY J.A new optimizer using particle swarm theory[A].Nagoya,Japan,1995:39-43.
[5]卢金娜.基于优化算法的径向基神经网络模型的改进及应用[D].太原:中北大学,2015.
[6]中国船级社.油船结构直接计算分析指南[M].北京:人民交通出版社,2003.
[7]庄茁,由小川,廖剑晖.基于ABAQUS的有限元分析和应用[M].北京:清华大学出版社,2009.
[8]HUSSEIN A W,GUEDES S C.Reliability and residual strength of double hull tankers designed according to the new iacs common structural rules[J].Ocean Engineering,2009,36(17):1446-1459.
[9]祁恩荣,崔维成.损伤船舶极限强度可靠性分析[C]//船舶力学学术会议暨船舶力学创刊十周年纪念学术会议.2007.
[10]江晓俐.基于船舶总纵极限强度的可靠性研究[D].武汉:武汉理工大学,2005.
[11]WIRSCHING P H,FERENSIC J,THAYAMBALLI A.Reliability with respect to ultimate strength of a corroding ship hull[J].Marine Structures,1997,10(7):501-518.
[12]段明昕.船体极限强度及其可靠性研究[D].大连:大连理工大学,2011.
[2]丁泉惠,王森,黄修长,等.基于有限元法和多岛遗传算法的飞轮结构参数优化设计[J].噪声与振动控制,2016,36(2):56-60.
[3]KENNED Y J,EBERHART R.Particle swarm optimization[A].Perth,Australia,1995:1942-1948.
[4]EBERHART R,KENNEDY J.A new optimizer using particle swarm theory[A].Nagoya,Japan,1995:39-43.
[5]卢金娜.基于优化算法的径向基神经网络模型的改进及应用[D].太原:中北大学,2015.
[6]中国船级社.油船结构直接计算分析指南[M].北京:人民交通出版社,2003.
[7]庄茁,由小川,廖剑晖.基于ABAQUS的有限元分析和应用[M].北京:清华大学出版社,2009.
[8]HUSSEIN A W,GUEDES S C.Reliability and residual strength of double hull tankers designed according to the new iacs common structural rules[J].Ocean Engineering,2009,36(17):1446-1459.
[9]祁恩荣,崔维成.损伤船舶极限强度可靠性分析[C]//船舶力学学术会议暨船舶力学创刊十周年纪念学术会议.2007.
[10]江晓俐.基于船舶总纵极限强度的可靠性研究[D].武汉:武汉理工大学,2005.
[11]WIRSCHING P H,FERENSIC J,THAYAMBALLI A.Reliability with respect to ultimate strength of a corroding ship hull[J].Marine Structures,1997,10(7):501-518.
[12]段明昕.船体极限强度及其可靠性研究[D].大连:大连理工大学,2011.