An Error Compensation Algorithm for Ship Structural Stress in Ice Fields Based on Temperature-Strain Coupling Tests

Abstract With the global climate warming, glaciers and ice sheets in polar regions are melting, leading to an increasing number of ice-going ships navigating in the Arctic and Antarctic regions. To ensure the safe operation of these ships in polar regions, structural stress monitoring of polar ships is a necessary choice. Stress monitoring equipment typically operates within a temperature range of up to −30°C. To meet the navigational requirements of ice-going ships in extremely low temperatures during winter, this paper investigates structural stress compensation techniques for low-temperature environments in ice fields. Two selected types of fiber optic strain sensors were instrumented on the experimental components. Structural strains under different temperatures ranging from −50°C to 50°C and loading conditions were monitored and recorded. The error characteristics of different sensors under various loading conditions were analyzed. Based on the experimental data, this paper explores error compensation algorithms for structural stress monitoring using both the least squares method and neural network approach. This research provides reference for data processing in ship structural stress monitoring and enhances the assessment accuracy of structural safety monitoring systems.