Considering the uncertainty of the customer delivery express quantities, the integer quadratic programming model of location problem of multi-type self-pick-up points was constructed by introducing the customer satisfaction function and aiming at minimizing the operating cost of self-pick-up points and maximizing the customer satisfaction. The real path distance between the demand point and the alternative self-pick-up point was obtained by AutoNavi platform, and the uncertainty of the express delivery quantity was described by the triangular fuzzy number to transform the model into a fuzzy chance-constrained programming model. The example solving was carried out by the Epsilon constraint algorithm combining with Cplex solver, and comparing the NSGA-II algorithm, the fuzzy set theory was used to obtain a compromise solution. Compared with the optimal solution, the customer satisfaction is reduced by 11.4%, and the operating cost is increased by 26.3%. The results show that as the number of customer express delivery increase, the operating cost of self-pick-up points and customer satisfaction show a trend of increasing first and then decreasing,constructing multi-type self-pick-up points can effectively balance cost and customer satisfaction to achieve system optimization.
In order to solve the problem of ship collision avoidance decision-making in open water, a ship collision avoidance decision-making method based on hybrid particle swarm algorithm was proposed. Firstly, in view of the limitation that particle swarm algorithm was prone to fall into local optimization at the later stage of iteration, the concept of Gaussian position mutation was introduced to expand the search range of particles, and the inertia weight was improved by adaptive strategy, so as to improve the local search ability of particles while ensuring the diversity of particles. Secondly, the objective function was constructed based on the multi-objective optimization method, and the ship collision risk model based on the fuzzy comprehensive evaluation strategy was introduced, taking into account the International Regulations for the Preventing Collision at Sea, common practices of seafarers, navigation safety and economy, and realizing the collision avoidance path planning under the situation of multi-ship encounter. Finally, the effectiveness of the proposed algorithm was verified by Matlab simulation experiment. Compared with the standard particle swarm algorithm, the decision-making effect of ship collision avoidance is significantly improved.
Aiming at the problem that the existing ship speed prediction model based on machine learning algorithm was impossible to balance the high calculation accuracy, strong generalization ability and fast calculation speed, a ship speed prediction model based on LightGBM (Light Gradient Boosting Machine) was proposed. Taking an inland river ship equipped with an energy efficiency monitoring system as the research object, the ship speed prediction model with real-time wind speed, wind direction, water depth, water speed, tail shaft speed, shaft power and main engine fuel consumption as inputs was established by using LightGBM algorithm, and compared with the speed prediction results of seven machine learning algorithms as RR (Ridge Region), SVR (Support Vector Region), DT (Decision Tree), BPNN (Back Propagation Neural Network), RF (Random Forest), GBDT (Gradient Boosting Decision Tree) and XGBoost (Extreme Gradient Boosting) at the same time. The results show that the ship speed prediction model based on LightGBM ranks second in accuracy, generalization ability and calculation speed with best comprehensive performance, which can realize fast prediction of ship speed on the premise of ensuring high prediction accuracy and strong generalization ability.
Aiming at the low-resolution problem of sea ice concentration (SIC) re-analysis products, the ordinary Kriging algorithm and the improved co-Kriging algorithm were used to obtain higher resolution data. The Gaussian model and spherical model were selected respectively to perform spatial fine interpolation of SIC, and the accuracy of the interpolation results in Arctic navigable waters was tested and analyzed. The results show that the variogram, covariable and target interpolation precision have great influence on the interpolation performance of ordinary Kriging algorithm and co-Kriging algorithm, and the spherical model still has high interpolation accuracy under extreme data conditions. Kriging algorithm can be used for interpolation prediction of polar SIC with different models according to different actual requirements, so as to provide higher resolution SIC data for polar ships and serve the safe navigation of polar ships.
In order to understand the characteristic of submerged waterjet hydrodynamic noise, the numerical prediction method of waterjet noise was explored by using point source model and doundary element method. The characteristics of fluctuating pressure on the inner wall of duct and the shielding effect of the duct on the acoustic field of the impeller were analyzed. The results show that the interaction between impeller and stator of submerged waterjet thruster results in strong radial force pulsation, that caused the stationary parts sound field contribution the most radially. The duct has shielding effect on the sound field of rotating impeller in a large range, and most prominent in radial. As result, there is a big difference between the sound field directivity of impeller and propeller.
In order to improve the power quality and stability of the dual-electric marine DC propulsion system, an improved distributed collaborative control strategy of the ship-borne hybrid energy storage system composed of supercapacitors and batteries was designed. The battery and supercapacitor were controlled with proportional sag and integral sag respectively to achieve high and low frequency power distribution. Each battery was equipped with a local distributed compensator to enable autonomous power distribution and state-of-charge (SoC) balancing between multiple parallel batteries. At the same time, the DC bus voltage was used to adjust the voltage secondarily to realize the autonomous recovery of the DC bus voltage, and combined with the integrated droop control to achieve fast SoC recovery of the supercapacitor. Finally, a general mathematical model of the shipboard hybrid energy storage system was established, and the dynamic analysis was carried out by using the proposed model, which theoretically verified the effectiveness of the system’s battery SoC dynamic equalization, supercapacitor SoC fast recovery, autonomous power distribution and DC bus voltage autonomous recovery. The results of hardware-in-the-loop experiments are consistent with the theoretical analysis, which effectively proves the multi-function characteristics of the system.
In order to solve the problem that the traditional smoke detectors were easily affected by the environment and location of the ship engine room, resulting in untimely fire response, a smoke recognition model of the ship engine room based on machine vision was proposed. Firstly, the universal database, real ship scenes and smoke images were used to build a graphic knowledge base. Secondly, the fusion model of migration learning and residual network was constructed to realize the migration and learning of smoke characteristics, and the validity of the model was verified by using the validation data set. Finally, the fire photos of a real ship workshop were selected to verify the validity of the model. The results show that compared with the traditional smoke detector, the proposed model can provide 83 s early warning, and compared with other intelligent algorithms, this model can improve the accuracy of smoke recognition. This method can identify ship smoke more quickly and avoid major catastrophic fire accidents, therefore can be used as an intelligent monitoring method for ships.
In order to improve the surface wear resistance and corrosion resistance of TC4 titanium alloy, CoCrW coating was prepared by laser cladding technology on the surface of TC4 titanium alloy, and its process, wear resistance and corrosion resistance were studied. The results show that the CoCrW cladding layer and the TC4 matrix exhibit good metallurgical bonding, and the microstructure of the cladding coating is uniform and dense, mainly composed of dendrites crystals. For the laser power 3000 W, the maximum hardness of the cladding layer is 1160 HV, nearly 4 times harder than TC4 substrate 324 HV under the same process condition including scanning speed, defocus, spot diameter and lap rate, and at this power, the average friction coefficient is as low as 0.2363,the wear amount is minimal as good wear resistance, the wear mechanism is abrasive wear and slight adhesive wear. The average friction coefficient of TC4 substrate is 0.3598,and the wear mechanism is adhesion wear and fatigue spalling wear, in this case, the electrochemical corrosion potential of the cladding layer is higher that result a lower corrosion rate with good corrosion resistance.
