面向阶段任务的携行器材品种确定方法

Method for determining for carrying material varieties of stage task

维修器材是有效实施维修保障的物质基础,携行器材品种确定是开展维修器材携行决策的关键。针对执行阶段任务武器装备维修器材品种多、影响因素复杂且关联关系不明确造成的携行器材品种确定困难的现实问题,提出了一种将改进稀疏核主成分分析(sparse kernel principal component analysis,SKPCA)算法与长短时记忆(long short-term memory,LSTM)神经网络模型相结合的阶段任务携行器材品种确定方法。在分析与任务阶段时序相关的携行器材影响因素及特征指标的基础上,运用基于弹性惩罚的SKPCA降维算法,对器材特征进行降维分析并得到低维稀疏特征向量,通过缩减数据容量增强特征指标的可解释性;运用混沌序列改进花授粉算法(flower pollination algorithm,FPA)优化LSTM超参数,构建混沌FPA-LSTM预测模型,精准进行携行器材品种确定。通过对演习携行器材品种确定算例分析验证了所提方法的科学性和可行性。

Maintenance material is the basis for effective implementation of maintenance support.It is the key to make decision of carrying material variety.In order to solve the problem which is difficult to determine the carrying material varieties due to various types of material in the stage task,the complex influencing factors and unclear association relationchip,a method for determining carrying material varieties of stage task combined with the improved sparse kernel principal component analysis(SKPCA)and long short-term memory(LSTM)neural network model is proposed.On the basis of analyzing the influencing factors and characteristic indicators of stage task carrying material,an improved SKPCA dimension reduction method based on elastic penalty is proposed which can reduce the material features dimensionality and obtain low-dimensional sparse feature vectors to enhance the data interpretability.The chaotic sequence is used to improve the flower pollination algorithm(FPA)which optimizing the LSTM hyperparameters and the chaotic FPA-LSTM prediction model is constructed.Through the example analysis by the exercise,the scientificity and feasibility of the proposed method are verified.