Zongran Dong

For the ship pipe routing design (SPRD) problem, previous studies have mainly employed bio-inspired algorithms such as multi-objective ant colony optimization (MOACO), non-dominated sorting genetic algorithm II (NSGA-II), and multi-objective particle swarm optimization (MOPSO). This paper proposes a novel approach based on the multi-objective A* (MOA*) algorithm to solve the SPRD. First, the optimization objectives and constraints of the SPRD problem are defined, and then an MOA*-based routing framework is developed. The time and space complexities of the approach are analyzed, and key components such as the cost functions, the solution dominance relationship, dynamic probability-based pruning, and neighbor node exploration strategy are designed to enhance solution diversity and search efficiency. Additionally, a space cascade expansion method is proposed to improve the computational efficiency of the MOA* in large-scale grid spaces. Comparative studies with MOACO, NSGA-II, GA-A*, and gray wolf optimization (GWO) on simulated cases of varying complexities and practical piping scenarios demonstrate the effectiveness of the MOA*. Furthermore, the applicability of the MOA* is validated against practical piping requirements, including the rapid generation of sub-optimal solutions, non-orthogonal routing, and partitioned pipe layouts. Experimental results, supported by a C++/OpenGL-based prototype software, show that the MOA* requires no extensive parameter tuning, exhibits stable computational efficiency and optimization capability, and demonstrates competitive performance in Pareto-optimal diversity compared with other algorithms.

With the continuous growth of user reviews, identifying underlying sentiment across multi-source texts efficiently and accurately has become a significant challenge in NLP. Traditional single models in cross-domain sentiment analysis often exhibit insufficient stability, limited generalization capabilities, and sensitivity to class imbalance. Existing ensemble methods predominantly rely on static weighting or voting strategies among homogeneous models, failing to fully leverage the complementary advantages between models. To address these issues, this study proposes a heterogeneous ensemble sentiment classification model integrating multi-view features and dynamic weighting. At the feature learning layer, the model constructs three complementary base learners, a RoBERTa-FC for extracting global semantic features, a BERT-BiGRU for capturing temporal dependencies, and a TextCNN-Attention for focusing on local semantic features, thereby achieving multi-level text representation. At the decision layer, a meta-learner is used to fuse multi-view features, and dynamic uncertainty weighting and attention weighting strategies are employed to adaptively adjust outputs from different base learners. Experimental results across multiple domains demonstrate that the proposed model consistently outperforms single learners and comparison methods in terms of Accuracy, Precision, Recall, F1 Score, and Macro-AUC. On average, the ensemble model achieves a Macro-AUC of 0.9582 ± 0.023 across five datasets, with an Accuracy of 0.9423, an F1 Score of 0.9590, and a Macro-AUC of 0.9797 on the AlY_ds dataset. Moreover, in cross-dataset ranking evaluation based on equally weighted metrics, the model consistently ranks within the top two, confirming its superior cross-domain adaptability and robustness. These findings highlight the effectiveness of the proposed framework in enhancing sentiment classification performance and provide valuable insights for future research on lightweight dynamic ensembles, multilingual, and multimodal applications.

Pipe route design is one of the most complex and time-consuming steps in the ship detailed design. A large number of scholars have studied the optimization of ship pipe routes, but there is still a lack of a fast and effective method for the automatic design of complex pipe routes. For this reason, this article proposes the auto-routing methods of single pipes, branch pipes, and multipipes. The auto-routing methods are based on an Improved A* algorithm. Compared with the traditional A* algorithm, the Improved A* algorithm considers the constraints of ship pipe route design. And by introducing “Blocking Iteration” and dynamic weight, the accuracy and efficiency of the algorithm are improved. On this basis, this article uses Python and OpenGL to develop the automatic pipe route design system. The system supports the automatic pipe route design, visualization of optimization results, and two-way interaction with three-dimensional (3-D) design software. Finally, the practicability of the proposed methods and system is verified by the simulation on a comparison case and an actual fuel filling and transfer pipe system.

goal of ship pipe route design (SPRD) is to seek the near-optimal paths that meet various constraints and objectives. Due to the complex construction of routing space, diverse piping constraints, and the large number of pipes, SPRD is one of the most difficult and time-consuming tasks even to a skilled pipe designer. This paper proposes automatic approaches for solving SPRD with A* algorithm and genetic algorithm (GA). Firstly, by simplifying the equipment and decomposing the routing space into grids, the mathematical model of SPRD is created. Then, the improved A* algorithm (A*-Router) for single pipe routing is introduced. The evaluation function, auxiliary tables and algorithm framework of A*-Router are presented. To obtain high-quality and diverse layouts, the improved GA (A*-GA-Router) is formulated by A*-Router and the connection-points strategy. Several new genetic operators of A*-GA-Router are de-signed to improve the routing performance. For multiple pipes routing, the novel algorithm (Multi-Pipes-Router) which calls A*-GA-Router internally is put forward. It arranges pipes according to the specified routing sequence and can produce parallel layout under the function of GA optimization and connection-points strategy. To cope with branch-pipe routing widely existing in engineering, a new pipe router (Branch-Pipe-Router) is put forward using a modified Steiner Tree framework in combination with the pro-posed single pipe routing algorithms. Compared with the traditional methods based on coevolution, it is more versatile and can effectively balance the layout quality and time efficiency. Finally, the feasibility and effectiveness of the proposed algorithms are demonstrated by the experiments on the designed and actual cases.

Ship pipe route design (SPRD) is to search the near optimal pipe routes that meet various constraints and objectives in a constrained ship space, which is one of the most time-consuming and difficult process in ship production. This paper proposes an