现有舱体布局建模方法存在灵活性差、计算复杂度高、易陷入局部最优等问题，因此研究基于空间散乱数据点的舱体布局模型具有重要价值。本文在传统模拟退火算法基础上进行改进，采用非线性降温策略、引入空间拓扑关系约束并增加记忆功能，结合深度学习中自动编码器进行数据预处理，运用DBSCAN算法划分功能区域，以空间利用率、通行效率、功能分区合理性为目标函数，考虑船舶重心平衡、设备安装规范等约束条件，构建了基于空间散乱数据点的舱体布局模型。验证改进模拟退火算法在舱体布局中多目标优化中的有效性，构建的模型在空间利用率、通行效率、功能冲突点数等指标上相较于传统方法有显著提升，为船舶舱体布局优化提供了更优的解决方案。
The existing cabin layout modeling methods have problems such as poor flexibility, high computational complexity, and being prone to fall into local optimum. Therefore, it is of great value to study the cabin layout model based on spatially scattered data points. This study improves on the basis of the traditional simulated annealing algorithm, adopts a nonlinear cooling strategy, introduces spatial topological relationship constraints and adds a memory function. Combined with the autoencoder in deep learning for data preprocessing, the DBSCAN algorithm is used to divide the functional areas, and the objective functions are space utilization rate, passage efficiency and the rationality of functional zoning. Considering the constraints such as the center of gravity balance of the ship and the equipment installation specifications, a cabin layout model based on spatially scattered data points was constructed. The effectiveness of the improved simulated annealing algorithm in multi-objective optimization of cabin layout was verified. The constructed model has significant improvements in indicators such as space utilization rate, passage efficiency, and the number of functional conflict points compared with traditional methods, providing a better solution for the optimization of ship cabin layout.
2025,47(11): 170-174 收稿日期：2025-1-12
DOI：10.3404/j.issn.1672-7649.2025.11.030
分类号：U667.65
作者简介：王秋璐(1989-),女,硕士,讲师,研究方向为设计学、计算机三维及计算机图像设计等
参考文献：
[1] 徐伟哲, 曹俊, 季红涛. 基于人机工程学的通透型载人潜水器舱室布局设计[J]. 中国机械, 2024(35): 36-39.
[2] 李洲君, 李月. 云环境下多舱室船体设计信息快速查询方法[J]. 舰船科学技术, 2024, 46(21): 178-181.
LI Z J, LI Y. Fast query method of multi-cabin hull design information in cloud environment[J]. Ship Science and Technology, 2024, 46(21): 178-181.
[3] 刘浏, 伍玉宙. 耦合因素特征提取下船舶舱室通风设计方法[J]. 舰船科学技术, 2024, 46(19): 147-150.
LIU L, WU Y Z. Ventilation design method of ship cabin under coupling factor feature extraction[J]. Ship Science and Technology, 2024, 46(19): 147-150.
[4] 马林, 陈登凯, 施红旗, 等. 基于多种群遗传算法的舰艇生活舱室布局优化设计方法研究[J]. 西北工业大学学报, 2025, 43(1): 189-199.
[5] 张维, 杨康宁, 张民. 一种求解不等圆Packing问题的改进遗传模拟退火算法[J]. 西北工业大学学报, 2017, 35(6): 1033-1039.
[6] 张冰清, 万众. 基于模糊聚类分析的货运物流枢纽宏观布局层次研究[J]. 广东交通职业技术学院学报, 2023, 22(1): 12-15+31.
[7] 齐继阳, 竺长安. 改进型模拟退火算法在设备布局设计中的应用[J]. 计算机工程, 2007(1): 241-243.
