The inherent complexity of large ships makes it challenging to evaluate ship designs systematically and scientifically. Knowledge-based expert systems can be reasonable solutions. However, this problem needs more rationality and better operability, especially in complicated ship-equipment suitability evaluation problems with numerous indicators and complex structures. This paper presents a hybrid multi-criteria decision-making (MCDM) framework to extend the ship-equipment suitability evaluation to group decision-making settings, where individual consistency and group consensus are thoroughly investigated to improve rationality and operability. As a result, an improved Interpretive Structural Modeling (ISM) method is developed to construct the evaluation index systems. Furthermore, based on an applicability analysis of the selected MCDM methods, an improved Analytical Hierarchy Process (AHP) method is proposed to distribute the index weights, and an applicable Fuzzy Technique for Order Preference by Similarity to Ideal Solution (Fuzzy TOPSIS) method is utilized to evaluate and select appropriate ship designs. Finally, a ship-equipment environmental suitability evaluation case is examined. The results indicate that the proposed framework improves the rationality and operability of the decision-making process and provides practical support to decision-makers for the systematic and scientific evaluation of ship designs. Therefore, it can also be applied to other ship design evaluation and selection problems.

The complex dynamic characteristics and marine environment pose significant challenges to the motion controller design of unmanned underwater vehicles (UUV). However, the controller parameters are usually fixed once manually adjusted, so they cannot adapt to environmental changes. This paper aims to present a machine learning-based parameter adaptive S-plane controller for the motion control of UUVs. The Q-learning-based adaptive control method is adopted to realize the optimization and automatic tuning of controller parameters in different environments. The optimal mapping between the input state and output action is found through the self-learning mechanism of Q-learning. Simulation results indicate that the proposed controller can adjust the controller's parameters in real-time and online and has a good control effect and adaptive ability for the actual environment.

Because the propeller advance coefficient varies widely during ship operation, this study numerically investigates the propulsive performance and wake kinetic-energy transport of an integrated propulsor combining a pre-duct with propeller boss cap fins (PBCF) over J = 0.30–1.10. The bare-propeller, PBCF, pre-duct, and integrated configurations were compared. Axial and circumferential resolved mean-flow kinetic-energy fluxes were evaluated over full-wake and near-hub regions, and the combined response was compared with an additive reference derived from the isolated-device responses. The integrated configuration increased the prescribed-inflow thrust–power ratio by 1.67%–6.29%, with the maximum at J = 0.745. Near the design condition, both flux responses remained close to the additive reference. At J = 1.10, only 6%–45% of the axial-flux increase and 15%–20% of the referenced circumferential-flux reduction were retained. The small design-condition deviation resulted mainly from cancelation between positive and negative local responses. At J = 1.10, the negative axial deviation was maintained mainly by the outer slipstream, whereas the positive circumferential deviation arose jointly from the near-hub and outer regions and persisted downstream mainly through the outer region. These results suggest that the pre-duct-modified inflow and propeller wake alter the local flow environment experienced by the PBCF. The attack angle produced the largest parameter-response ranges, reaching about 11.5% and 17% points for thrust–power ratio and axial flux, respectively, at J = 1.10. An outlet-diameter ratio of Dd/Dd,0=0.9 achieves the highest equal-weight score (about 0.70) and is identified as the most balanced candidate within the investigated numerical design space.