前馈多层神经网络是应用最广泛的一种人工神经网络 ,在各门学科领域中都具有很重要的实用价值 ,其学习能力和容错能力对不确定性模式识别问题的求解具有独到之处。前馈多层神经网络所使用的 BP算法虽然简单且易于实现 ,但收敛速度缓慢。因此基于收敛速度及存储空间的考虑 ,文中着重探讨了一种较实用的 BP网络优化算法——向量序列的 ε-加速算法 ,并对相关的收敛性和稳定性作了理论上的分析和研究。实验结果表明这种方法有很好的效果 Feedforward multi-layer neural network is the most widely used artificial neural network, which has very important practical value in all disciplines. Its learning ability and fault-tolerant ability are unique to the solution of the uncertainty pattern recognition problem . The BP algorithm used in feedforward multi-layer neural network is simple and easy to implement, but the convergence speed is slow. Therefore, based on the convergence speed and storage space considerations, this paper focuses on a more practical BP network optimization algorithm - vector sequence ε-acceleration algorithm, and the related convergence and stability of a theoretical analysis and research . Experimental results show that this method has a good effect