基于改进粒子群算法的径向基人工神经网络淀粉基发泡复合材料性能预测  

Prediction of properties of starch matrix foam composites by radial basis function artificial neural network based on improved particle swarm optimization

作　　者：李孟山[1] 吴维[1] 陈炳生[1] 管立新[1] 武燕[2] LI Mengshan;WU Wei;CHEN Bingsheng;GUAN Lixin;WU Yan(College of Physics and Electronic Information, Gannan Normal University, Ganzhou 341000, China;College of Mathematics and Computer Science, Gannan Normal University, Ganzhou 341000, China)

机构地区：[1]赣南师范大学物理与电子信息学院,赣州341000 [2]赣南师范大学数学与计算机科学学院,赣州341000

出　　处：《复合材料学报》2017年第12期2882-2889,共8页Acta Materiae Compositae Sinica

基　　金：国家自然科学基金(51663001;51463015)

摘　　要：以乙烯-醋酸乙烯酯(EVA)和淀粉质量比、甘油质量分数和NaHCO_3质量分数为输入,以拉伸强度和回弹率为输出,建立基于种群熵多样性评估和收敛、发散策略的粒子群改进算法的径向基人工神经网络(RBF ANN)的淀粉基发泡复合材料性能预测模型。结果表明,该模型的预测效果较好,预测均方差和相关系数分别为0.0160和0.9890。预测发现,淀粉基发泡复合材料的拉伸强度随甘油含量的增加而缓慢降低,随NaHCO_3含量的增加先减少后增加;回弹率随甘油含量的增加而递增,随NaHCO_3含量的增加而先增加后减少。A prediction model of starch matrix foam composites by radial basis function artificial neural network（RBF ANN）based on chaotic self-adaptive particle swarm optimization algorithm with population entropy diversity and convergence divergence strategy was established.The input variables of this model included ethylene-vinyl acetate（EVA）/starch mass ratio,glycerin content and NaHCO3 content,and the output variables were tensile strength and rebound rate.The results show that the proposed model has a good performance.The root mean square error of prediction and correlation coefficient are 0.0160 and 0.9890,respectively.The prediction results show that the tensile strength of starch matrix foam composites reduces slowly with the increase of glycerin content,and it reduces firstly and then increases with the increase of NaHCO3 content.The rebound rate increases with the increase of glycerin content,and it increases firstly and then decreases with the increase of NaHCO3 content in the starch matrix composites.

关 键 词：性能预测 发泡材料 神经网络 粒子群 混合智能 

分 类 号：TB324[一般工业技术—材料科学与工程] TB322