Welcome to Acta Armamentarii ! Today is Share:

Acta Armamentarii ›› 2022, Vol. 43 ›› Issue (5): 1012-1022.doi: 10.12382/bgxb.2021.0240

• Paper • Previous Articles     Next Articles

An Ammunition Quality Evaluation Method Based on Least Squares Support Vector Machine

YANG Jianxin1, LAN Xiaoping1, FENG Yadong1, YANG Yiming1, GUO Zhiming2   

  1. (1.Information Center of China North Industries Group Corporation, Beijing 100089, China;2.Ordnance Science and Research Academy of China, Beijing 100089, China)
  • Online:2022-03-17

Abstract: For the problems about the less sample data,large test consumption and ineffective use of manufacturing process quality data, in order to solve these problems,an ammunition quality evaluation method is proposed based on least squares support vector machine(LSSVM)optimized by improved salp swarm algorithm.The firing success rate of batches of new ammunition is estimated based on Bayesian by using target test data as input.On this basis,an evaluation model of the relationship between ammunition batch manufacturing quality data and ammunition firing success rate is developed using LSSVM.The LSSVM is optimized with an elite center of mass and a salp swarm algorithm improved by backward learning strategy,which effectively improves the accuracy of the evaluation model.And the validity of the evaluation model was verified by using a new type of ammunition as an example.The validated results show that the model has higher accuracy and stronger robustnesscompared with the traditional LSSVM,LSSVM improved by particle swarm algorithm and LSSVM improved bysalp swarm algorithm.

Key words: ammunition, qualityevaluation, salpswarmalgorithm, leastsquaressupportvectormachine, Bayesianmethod

CLC Number: