Uncertainty Quantification in Impulse Thruster Performance Using Polynomial Chaos Expansion

Due to its short operating time and precise attitude control, the miniature impulse thruster has significant application prospects in certain military fields. This study establishes an interior ballistic model based on dynamic mesh technology to compute the impulse thruster's three-dimensional transient internal flow field and investigate the performance characteristics that change over time. To more correctly predict the distribution of thruster performance, the stochastic uncertainty for significant internal ballistic parameters (burning rate index, specific heat ratio, gas constant, explosion temperature, initial temperature) is quantitatively investigated. First, we use Latin hypercube sampling based on the maximin distance criterion to create a sample design with better space-filling characteristics. Then we propose the non-intrusive polynomial chaos expansion method to study the uncertainty quantification of the system, and employ compressed sensing technology to calculate the unknown polynomial coefficients. The uncertainty quantification method adopted in study can create an efficient surrogate model based on small simulations, and then obtain moments and Sobol’ indices conveniently. The results show that burning rate index and initial temperature are the stochastic parameters that have the greatest impact on pressure, thrust, and working time. The specific heat ratio and gas constant have the biggest effects on the total impulse.