Random simulation for reliability of solid rocket motor

The random simulation methods can reduce the expense and yet give satisfactory accuracy for reliability of solid rocket motor. The basic idea is to replace most testing with computer trials. For this purpose, the information obtained from very limited testing is used to construct simulation models with high accuracy. Four goodness-of-fit statistical tests are used. It is beneficial to mix these tests into three combined goodness-of-fit tests. When the number of samples is 8, the confidence level is 0.05. This paper uses four alternative distributions: normal, lognormal, Weibull and limit values and select four parameters: combustion coefficient, pressure exponent, propellant density and characteristic rate of four kinds of SRM. For each of the four characteristic parameters, all three combined goodness-of-fit tests give the same conclusion about the type of distribution, indicating that these three combined tests are all reliable.