摘要
针对BP神经网络存在易陷入局部极小值的缺点,利用遗传算法全局搜索能力强的特点,对BP神经网络的权值和阈值进行优化,建立了基于GA-BP网络的某型装备作战效能评估模型,给出了实际数据的评估结果.实验结果表明:GA-BP神经网络的泛化能力、评估准确性比BP神经网络的效果更好,这为评估某型装备的作战效能提供了一种新的思路.
Aimed at the drawback that the BP neural network is likely to fall into the local minimum,in this paper,the weight and threshold of the neural network are optimized by utilizing the powerful global search capability of genetic algorithm,a model of operational effectiveness evaluation for a certain equipment is set up based on GA-BP network,and the evaluation result is given with real data.Experimental results illustrate that the generalization ability and evaluation accuracy of GA-BP neural network are even better than those of BP neural network,thus,paving a new way for assessing the operational effectiveness of a certain type of equipment.
出处
《空军雷达学院学报》
2012年第5期346-348,共3页
Journal of Air Force Radar Academy
关键词
BP神经网络
遗传算法
装备作战效能
效能评估
BP neural network
genetic algorithm（GA）
equipment operational effectiveness
effectiveness evaluation
