Chinese Journal of Ship Research, Volume. 17, Issue 2, 156(2022)
Ship structural optimization method based on daptive mutation particle swarm algorithm
Figures & Tables(14)
![[in Chinese]](/Images/icon/loading.gif){kind=icon}
Table 1. Testing functions
View tableView in ArticleTable 1. Testing functions
非线性函数 数学表达式 搜索区间 最小值 Ackley $f(x) = - 8\exp \left( - 0.2\sqrt {\dfrac{1}{n}\displaystyle\sum\limits_{i = 1}^n { {x} _i^2} } \right) - \exp \left[\dfrac{1}{n}\displaystyle\sum\limits_{i = 1}^n {\cos (2{\text{π} } {x_i})} \right] + 8 + {\text{e} }$ n=30，[−5,5] 0 Rastrigin $f(x) = \displaystyle\sum\limits_{i = 1}^n {[x_i^2 - 10 \times \cos (2{\text{π} } {x_i}) + 10]}$ n =10，[−10,10] 0 Sphere $ f(x) = \displaystyle\sum\limits_{i = 1}^n {x_i^2} $ n =30，[−5,5] 0 Rosenbrock $ f(x) = \displaystyle\sum\limits_{i = 1}^n {[100 \times {{(x_i^2 - {x_{i + 1}})}^2} + {{({x_i} - 1)}^2}]} $ n =15，[−30,30] 0
Table 2. Test results by different algorithms
View tableView in ArticleTable 2. Test results by different algorithms
函数名称 寻优结果 最小值 GA算法 PSO算法 AMPSO算法 Ackley 0.670 0.760 0.620 0 Rastrigin 0.030 4.978 0.010 0 Sphere 0.056 0.041 0.019 0 Rosenbrock 29.171 15.590 7.433 0
Table 3. Comparison of calculated errors using test samples for training different BP neural networks
View tableView in ArticleTable 3. Comparison of calculated errors using test samples for training different BP neural networks
训练次数 BP神经网络 GA-BP神经网络 PSO-BP神经网络 AMPSO-BP神经网络 平均误差/% 最大误差/% 平均误差/% 最大误差/% 平均误差/% 最大误差/% 平均误差/% 最大误差/% 1 0.21 1.60 0.12 0.52 0.11 0.52 0.08 0.42 2 0.33 1.26 0.13 0.63 0.12 0.41 0.09 0.47 3 0.30 2.08 0.13 0.93 0.15 0.49 0.08 0.30 4 0.66 3.36 0.15 0.86 0.15 0.93 0.08 0.30 5 0.77 4.59 0.17 2.29 0.11 0.70 0.08 0.45
Table 4. Comparison of a cross-bar truss optimization results
View tableView in ArticleTable 4. Comparison of a cross-bar truss optimization results
参数 OPTDYN[ 7 ]CONMIN[ 7 ]可行方 向法[ 9 ]退火模 拟法[ 9 ]文献[ 8 ]本文 方法 S1/cm2 165.82 162.59 194.95 197.68 196.87 185.74 S2/cm2 0.65 12.19 0.65 0.68 0.645 0.85 S3/cm2 162.01 160.46 153.09 150.90 159.46 152.40 S4/cm2 125.10 102.14 96.64 93.22 96.63 91.75 S5/cm2 0.65 0.65 0.65 0.70 0.645 0.74 S6/cm2 0.65 11.29 0.65 4.09 3.09 3.27 S7/cm2 99.36 108.14 54.97 48.09 45.858 52.58 S8/cm2 131.10 127.30 134.07 137.69 138.387 135.31 S9/cm2 133.81 135.36 134.80 136.23 131.935 138.61 S10/cm2 7.36 16.19 0.65 0.67 0.645 0.86 W/kg 2 480.96 2 522.22 2 292.00 2 286.00 2 300.68 2 272.10
Table 5. Comparison of a gangboard optimization results
View tableView in ArticleTable 5. Comparison of a gangboard optimization results
优化设计变量 优化前 优化后 GA-BP-GA PSO-BP-GA AMPSO-BP-GA 底板L型骨材 腹板高度/mm 50.0 31.0 31.1 41.0 底板L型骨材− 翼板厚度/mm 4.0 2.8 2.9 2.6 底板L型骨材− 腹板厚度/mm 4.0 2.2 3.1 2.6 底板L型骨材− 翼板宽度/mm 16.0 13.3 12.6 13.0 面板L型骨材 腹板高度/mm 80.0 75.6 77.9 66.0 面板L型骨材− 翼板厚度/mm 6.7 6.4 6.4 6.4 面板L型骨材− 腹板厚度/mm 5.0 4.3 4.0 3.0 面板L型骨材− 翼板宽度/mm 22.0 21.5 20.8 20.7 非车道面板/mm 5.0 3.4 4.3 3.0 肋板/mm 5.0 3.4 4.3 3.0 底板/mm 5.0 3.4 4.3 3.0 内车道面板/mm 7.0 5.6 5.6 5.0 外车道面板/mm 10.0 8.0 7.8 7.9 后端纵桁/mm 8.0 7.2 6.5 6.1 前段纵桁/mm 16.0 10.0 10.2 10.4 底板/mm 5.0 3.4 4.3 3.0 车道底板/mm 8.0 7.2 6.5 6.1 跳板重量/kg 2 281.4 1 702.5 1 872.8 1 520.7
Table 6. Structural check after optimization
View tableView in ArticleTable 6. Structural check after optimization
板材 梁 变形/mm 许用变形/mm 校验结果 应力/MPa 许用应力/MPa 应力/MPa 许用应力/MPa 62.90 176.25 171.00 171.55 1.71 7.50 合格
Tools
Get Citation
Copy Citation Text
Yijing WANG, Guang'en LUO, Chenyang WANG, Shuang LI. Ship structural optimization method based on daptive mutation particle swarm algorithm[J]. Chinese Journal of Ship Research, 2022, 17(2): 156
Paper Information
Category: Ship Structure and Fittings
Received: Feb. 23, 2021
Accepted: --
Published Online: Mar. 24, 2025
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20