Calculation for Primary Combustion Characteristics of Boron-based Fuel-rich Propellant Based on GA-BP Neural Network

摘要

With Genetic Algorithm (GA) optimizing weights and biases of Back-Propagation (BP) neural network, a calculation model for primary combustion characteristics of boron-based fuel-rich propellant based on GA-BP neural network was established and validated, and then was used to predict primary combustion characteristics of boron-based fuel-rich propellant. The results show that the calculation error of burning rate is less than ±6.6%; in the formulation range (HTPB 28% -32%, AP 30% -35%, MA 4% -8%, GFP 0% -5%, B 30%), when increasing HTPB content with a corresponding decrease in MA content, burning rate decreases and pressure index first increases, then decreases and finally rises slightly; when increasing AP content with a corresponding decrease in MA content, both burning rate and pressure index increase; when increasing AP particle size, burning rate decreases and pressure index first increases and then decreases; when increasing GFP content with a corresponding decrease in HTPB content, both burning rate and pressure index increase; the variation of the calculation data is consistent with the experimental results.