Reliability Evaluation Method for Initiating Explosive Device Output Performance Based on SMOTE-Bootstrap Method in Small Sample Sizes

To improve the output reliability evaluation accuracy in small sample sizes for initiating explosive devices, a new method based on the Synthetic Minority Over-Sampling Technique (SMOTE) algorithm with the Bootstrap method is proposed. Firstly, the evaluation principle of the new method is clarified, which applies SMOTE algorithm to generate the optimal expanded sample. Next, the expanded sample is combined with the original sample to reconstruct the evaluation sample. And then the Bootstrap method is used to evaluate the reliability. Secondly, numerical simulations are carried out to compare with the classical Second-order Approximation, Bootstrap, Bayes Approximation, and Two-dimensional Unilateral Tolerance Coefficient method. The results demonstrate that the new method is more accurate and stable in small sample sizes. Finally, taking a certain Stab Detonator as an instance, comparison validation of large and small samples is conducted. The results show that the new method expands the available evaluation sample size and improves the accuracy of the evaluation, which is suitable for initiating explosive devices in small sample sizes.