神经网络BP算法预测可靠性增长

叙述可靠性增长的基本概念和神经网络BP算法，着重探讨BP算法用于可靠性增长预测的可能性。通过两个实例的预测结果，与常用的可靠性增长参数模型Duane，Gompertz的结果相比较。实例表明，该方法是可行的，而且是有效的，适应性强。可以预料，它不但能用于仪器设备试制过程中预测可靠性增长，而且可用于武器装备在储存过程之中可靠性下降的预估。

Neural Network BP Algorithm Approach for Prediction of Reliability Growth

In this paper, we introduce the basic concept of the reliability growth and BP algorithm at first, then con- sider emphatically BP algorithm application in the reliablity - prediction , finally, examine the algorithm by some instances and compare with results of Duane model, Gompertz model . Results cleary show tha the approach is feasible and the effective. Further more, it can be effectively apply in the precliction of reliability storage in the weapon system.