Abstract: Ship operational data are crucial for assessing vessel performance, predicting energy efficiency, and enabling predictive maintenance, thereby facilitating safe, efficient, and intelligent operations in the maritime industry. However, the effective utilization of such data is often hindered by poor data quality, high uncertainty, and complex processing workflows. To address these challenges, this paper proposes a robust framework for analyzing ship operational data. The framework enhances data usability by optimizing the procedures for data processing and analysis. Specifically, it incorporates a practical method for identifying steady-state operating conditions, establishes rational data filtering strategies, and applies internationally standardized methods for correcting environmental influences. The proposed framework is validated using real operational data from a container ship. Results demonstrate that the processed data can clearly reveal the vessel's long-term performance trends, providing a reliable basis for predictive maintenance decisions. This study offers a practical and effective solution for improving the quality and usability of ship operational data, thereby supporting the advancement of big-data applications towards smart shipping.

Key words: ship operational data, data analysis framework, steady-state identification, environmental influence correction, smart shipping

摘要： 实船营运数据可有效支撑实船性能评估、能效预测与前瞻性维护,助力航运业实现安全、高效、节能的智能化运营。然而,当前实船营运数据分析面临原始数据质量差、不确定性高、处理流程复杂等挑战,这些挑战制约了其在船舶性能评估与运营决策中的有效应用。该文提出了一种鲁棒的实船营运数据分析框架,通过优化数据处理与分析流程,提升了数据可用性。该框架通过引入实用的稳态数据段识别方法、设置合理的数据筛选策略,以及应用国际标准组织认可的环境影响修正方法,可有效提升实船数据的可靠性与可用性。基于某集装箱船的实际营运数据对该方法进行了验证,经处理后的数据能够清晰反映船舶的长期性能变化趋势,可为前瞻性维护决策提供支撑。该研究为提升实船营运数据的品质和可用性提供了一种实用且有效的解决方案,有助于推动面向智能船舶的大数据应用。

关键词: 实船营运数据, 数据分析框架, 稳态识别, 环境影响修正, 智能船舶