融合混合神经网络多工况船烃耗率预测方法、系统和装置
By integrating hybrid neural networks and adaptive piecewise fitting algorithms, and utilizing multi-source quantitative information obtained from sensors on LNG ships to train the Lstm-Attention network, the problem of insufficient accuracy in traditional LNG ship hydrocarbon consumption rate prediction methods is solved. This achieves more efficient and accurate fuel consumption management, adapts to complex sea conditions, and extends the service life of LNG ships.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 中海油能源发展股份有限公司采油服务分公司
- Filing Date
- 2026-05-12
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional methods for predicting hydrocarbon consumption rates in LNG carriers rely on static tests and empirical formulas, which are costly and fail to account for the effects of dynamic sea conditions, resulting in insufficient accuracy of the prediction results.
A hybrid neural network is used to acquire multi-source quantitative information from sensors on the LNG ship, train the Lstm-Attention network, and combine it with an adaptive piecewise fitting algorithm to generate a dynamic curve of driving speed-hydrocarbon consumption rate, which can dynamically adapt to complex environmental changes.
It improves the training efficiency and accuracy of prediction models, reduces energy consumption, extends the service life of LNG ships, adapts to different ship types and sea conditions, optimizes fuel consumption, and conforms to the trend of green shipping development.
Smart Images
Figure CN122198258B_ABST