Abstract: Error back propagation (EBP) is a widely used training algorithm for feedforward neural networks (FFNNs)，but low learning rate limits its applications in the networks with complex topology architecture and large patterns.In this work, two modifications on Levenberg-Marquardt algorithm for FFNNs were made.One modification was made on the objective function, while the other was the evaluation of the initial weights and biases.The modified algorithm gave a better convergence rate compared to the traditional EBP algorithm and it was less computationally intensive and required less memory.The performance of the algorithm was verified separately with polymer and protein systems.The results showed that the BP network based on modified Levenberg-Marquardt algorithm could be used to predict the binodal curve of H₂O/DMAc(N，N-dimethylacetamide)/PSf(polysulfone) system and lysozyme solubility in aqueous salt solution.

Key words: BP网络, EBP算法, Levenberg-Marquardt算法, 大分子体系

摘要： 误差反向传播(EBP)算法目前已广泛应用于Back-propagation (BP)网络的学习和训练，但存在网络收敛速度慢的缺点.从目标函数和网络权值与阈值的初始化两方面对标准的Levenberg-Marquardt算法做了改进，改进的算法可以减少计算的复杂性及对内存的需求，尤其对具有较大样本及复杂拓扑结构的网络效果更为明显.基于改进的Levenberg-Marquardt算法的BP网络对蛋白质体系的溶解度和聚合物成膜体系的液-液相平衡性质进行模拟和预测，结果表明：改进的Levenberg-Marquardt算法较传统的EBP算法的收敛速度大大提高，且能较好地用于预测溶菌酵素在盐溶液中的溶解度和水/二甲基乙酰胺/聚砜成膜体系的双结点曲线.

关键词: BP网络, EBP算法, Levenberg-Marquardt算法, 大分子体系