Virtual Sensor Data Generation for Internal Conditions of Missile Canisters Using External Environmental Information

Purpose: This study aims to address the difficulty of long-term measurement of missile canister internal temperature by proposing a method to generate virtual sensing data using ambient temperature records. The goal is to produce continuous internal temperature data for degradation prediction and condition assessment.Methods: Statistical correction parameters, including the scaling factor (standard deviation ratio) and bias (phase difference), were derived by comparing missile canister internal temperature samples with ambient temperature data. For unmeasured dates, the average and standard deviation of ambient temperature were calculated, and the most statistically similar date within the measured period was identified using Euclidean distance. The corresponding parameters were then applied to estimate the missile canister internal temperature.Results: The generated data showed high similarity with actual measurements in both numerical values and overall trends. The method enables internal temperature estimation without additional physical sensors, contributing to reduced maintenance costs and improved operational efficiency.Conclusion: This study presents a temperature estimation approach that generates virtual sensing data for missile canisters using ambient temperature records. It offers a practical alternative in environments where direct measurement is limited and supports enhanced degradation modeling. The method is also applicable to other weapon systems and industrial settings with similar constraints.