智能优化模型与智能算法在船舶能效优化管理中的应用主要体现在船舶能效评估

The application of intelligent optimization models and intelligent algorithms in ship energy efficiency optimization management is mainly reflected in ship energy efficiency evaluation and prediction, route intelligent optimization, speed intelligent optimization, and hull and trim optimization for energy efficiency improvement. <br><br>In terms of ship energy efficiency evaluation and prediction, Yan et al. established a neural network model for evaluating ship energy efficiency based on the energy efficiency data collected by actual ships, which can realize the prediction and evaluation of ship energy efficiency. Yuan et al. applied artificial neural network and Gaussian process to ship energy consumption evaluation and conducted experiments. The results showed that speed optimization can effectively reduce ship energy consumption, and route optimization and trim optimization considering weather can also reduce ship energy consumption. Wickramanayake et al. systematically analyzed ship energy consumption prediction based on machine learning methods. They compared the effectiveness of methods based on random forest, gradient enhancement, and neural network for the multivariate time series fleet energy consumption prediction problem. The results showed that the use of random Forest technology can obtain more accurate prediction results. <br><br>Alonso et al. used the method of combining artificial neural network and genetic algorithm to optimize the performance of marine diesel engines, trying to find the parameter configuration that meets the most stringent emission regulations to reduce the fuel consumption of ships. Tillig et al. used Monte Carlo methods and general ship energy system models to analyze the uncertainty of ship fuel consumption forecasts at various stages of the ship’s life cycle. The results showed that the uncertainty of fuel consumption forecasts can be reduced and the fuel consumption monitoring of ships improved. And the predicted level of intelligence has an important role in promoting the reduction of ship energy consumption. <br><br>In terms of ship energy efficiency prediction, Yang et al. proposed a gray box model based on genetic algorithm, which solved the limitation of weather factors, and compared it with the gray box model based on time series parameter estimation. The results showed that the algorithm has higher ship energy efficiency. The accuracy of prediction can effectively reduce the emission of polluting gas. Wang Shengzheng et al. established an alternating sparse self-encoding network model, and predicted the impact of the marine environment on ship navigation by using association rule algorithms to select features of navigation data. The proposed network model can not only reduce training time, but also improve prediction accuracy.

The application of intelligent optimization model and intelligent algorithm in ship energy efficiency optimization management is mainly reflected in ship energy efficiency assessment and prediction, route intelligence optimization, speed optimization, and hull and tilt optimization for energy efficiency improvement.<br><br>In the area of ship energy efficiency assessment and prediction, Yan and other neural network models based on energy efficiency data collected by real ships have been established to assess the ship's energy efficiency level, so that the prediction and evaluation of ship energy efficiency can be realized. Yuan and other artificial neural networks and Gaussus processes are applied to the evaluation of ship energy consumption and experiments are carried out, and the results show that speed optimization can effectively reduce ship energy consumption, and considering weather path optimization and tilt optimization can also reduce ship energy consumption. Wickramanayake and others systematically analyzed the prediction of ship energy consumption based on machine learning methods, compared the effectiveness of fleet energy consumption prediction based on random forests, gradient enhancement and neural network methods for multivariable time series, and showed that more accurate prediction results could be obtained by using random forest techniques.<br><br>Alonso and other methods, using the combination of artificial neural network and genetic algorithms, optimize the performance of the ship's diesel engine, trying to find out the most stringent emission regulations to meet the parameter configuration, in order to reduce the ship's fuel consumption. Using Monte Carlo method and general ship energy system model, Tillig and so on analyze the uncertainty of ship fuel consumption prediction at all stages of the ship's life cycle, and the results show that it can reduce the uncertainty of fuel consumption prediction, improve the intelligent level of ship fuel consumption monitoring and prediction, and promote the reduction of ship's energy consumption.<br><br>In the aspect of ship energy efficiency prediction, Yang and so on put forward the gray box model based on genetic algorithm, solved the limitation of weather factors, and compared with the gray box model based on time series parameter estimation, the results show that the algorithm has higher accuracy of ship energy efficiency prediction, which can effectively reduce the emission of polluting gases. Wang Shengzheng and others have established alternating sparse self-coding network model, which predicts the impact of marine environment on ship navigation by using the correlation rule algorithm, and can not only reduce training time, but also improve prediction accuracy.

The application of intelligent optimization model and intelligent algorithm in ship energy efficiency optimization management is mainly reflected in ship energy efficiency evaluation and prediction, route intelligent optimization, speed intelligent optimization, and hull and trim optimization for energy efficiency improvement.<br>In the aspect of ship energy efficiency evaluation and prediction, Yan et al. Established a neural network model to evaluate the ship energy efficiency level based on the energy efficiency data collected by the real ship, which can realize the prediction and evaluation of ship energy efficiency. Yuan et al. Applied the artificial neural network and Gaussian process to the ship energy consumption evaluation and carried out experiments. The results show that the speed optimization can effectively reduce the ship energy consumption, and the path optimization and trim optimization considering the weather can also reduce the ship energy consumption. Wickramanayake and others systematically analyzed the ship energy consumption prediction based on machine learning method. Aiming at the fleet energy consumption prediction problem of multivariate time series, they compared the effectiveness of the methods based on random forest, gradient enhancement and neural network. The results show that the random forest technology can obtain more accurate prediction results.<br>Alonso et al. Used the method of combining artificial neural network and genetic algorithm to optimize the performance of marine diesel engine, trying to find the parameter configuration to meet the most stringent emission regulations, so as to reduce the fuel consumption of the ship. Tillig et al. Used Monte Carlo method and general ship energy system model to analyze the uncertainty of ship fuel consumption prediction at all stages of the ship's life cycle. The results show that it can reduce the uncertainty of fuel consumption prediction, improve the intelligent level of ship fuel consumption monitoring and prediction, and play an important role in reducing ship energy consumption.<br>In the aspect of ship energy efficiency prediction, Yang et al. Proposed a grey box model based on genetic algorithm to solve the limitation of weather factors, and compared it with the grey box model based on time series parameter estimation. The results show that the algorithm has higher accuracy of ship energy efficiency prediction and can effectively reduce the emission of polluting gases. Wang shengzheng et al. Established the alternating sparse self coding network model, and predicted the impact of marine environment on ship navigation by using association rules algorithm to select the features of navigation data. The proposed network model can not only reduce the training time, but also improve the prediction accuracy.