基于PCA-GA-BP 神经网络的状态评估算法

针对传统评估方法主观性强的缺点及BP神经网络自身缺陷,提出基于数据知识的PCA-GA-BP状态评估组合算法。采用主成分分析对样本数据进行降维处理,利用遗传算法对BP神经网络的初始权值阈值进行优化,将历史数据作为学习样本训练神经网络,处理实时信息得到评估结果,并通过实例进行算法验证分析。结果表明,该算法是可行的,适用于复杂武器装备的状态评估。

Condition Estimating Algorithm Based on PCA-GA-BP Neural Network

Aiming at the traditional evaluation methods has disadvantage of strong subjectivity and defects of BP neural network, the combinational algorithm PCA-GA-BP based on data is established. Sample data dimensions are reduced by principal component analysis, the initial weights and threshold of BP neural network are optimized by genetic algorithm. The neural network is trained by historical data and can be used to evaluate real-time information, and algorithm is validated through the case analysis. The results show that, the algorithm is feasible, which is suitable to condition evaluation for complex weapon equipment.