Determination of Ignition Delay Time for Solid Propellants Based on Multimodal Data Fusion

ABSTRACT Solid propellant ignition delay time is a critical parameter for evaluating rocket engine combustion stability, thrust response, and combustion efficiency. Conventional single‐sensor measurement methods, constrained by their respective physical principles and operational characteristics, often yield inconsistent results, limiting reliable assessment of ignition behavior. To address this, we introduce a multi‐sensor data fusion framework based on Dempster‐Shafer (D‐S) evidence theory, integrating four diagnostic approaches: imaging, temperature, spectroscopy, and pressure. Experimental results reveal systematic deviations among the individual methods. While imaging and temperature measurements produced relatively consistent averages (132.7 and 157.6 ms, respectively), spectral and pressure methods returned significantly higher values (223.3 and 398.6 ms). The corresponding within‐group standard deviations reached 118.7 and 155.9 ms, highlighting the limited repeatability of any single technique. The fusion framework substantially improved measurement coherence. The average within‐group standard deviation decreased by 86.8%, from 127.5 to 16.8 ms. Notably, the mean absolute error of the pressure method dropped by 94.6% (from 265.9 to 14.4 ms), while temperature and spectral methods achieved error reductions of 95.2% and 85.8%, respectively. These results validate the capability of D‐S evidence theory to mitigate systematic bias and enhance reliability in multimodal combustion diagnostics. This study demonstrates that data fusion not only resolves methodological inconsistencies in ignition delay characterization, but also generates a consistent, physically meaningful dataset. The approach supports improved solid propellant ignition analysis, combustion model calibration, and advanced propulsion system design.