基于粒子群优化模糊神经网络的高技术知识创新评价  

Evaluation method of high-tech knowledge innovation based on particle swarm optimization fuzzy neural networks

作　　者：张海峰[1,2] 梁工谦[1] 张晶[1] 

机构地区：[1]西北工业大学管理学院,陕西西安710072 [2]北京航天情报与信息研究所,北京100854

出　　处：《系统工程与电子技术》2012年第5期973-976,共4页Systems Engineering and Electronics

基　　金：国家自然科学基金(70771089);陕西省软科学项目(2011KRM34)资助课题

摘　　要：针对高技术知识创新非线性、不确定性、时变性的特点,建立了评价指标体系;结合粒子群优化算法,提出了一种改进的模糊神经网络评价模型。该模型能够进行多个并行时变模糊神经网络组合算法,这些算法通过进化预置网络的连接权值、阈值和补偿参数,实现网络的学习和精确推理。通过仿真应用,证明了此种模型结构与算法适用性好,便于计算机实现,且全局收敛能力、收敛速度和泛化精度等性能均优于原先的学习算法。According to the characteristic of nonlinearity,uncertainty,time variation,this paper presents high-tech knowledge innovation capacity evaluation index system,and puts forward an improved fuzzy neural network evaluation model combined with particle swarm optimization.This model can combine multiple concurrent time-varying fuzzy neural network algorithm and realize network of learning and accurate reasoning,by evolution preset network connection weights,threshold and compensation parameters with particle swarm optimization.Through simulating application,it has been proved that this model structure and the algorithm are feasible and facilitate for computer implementation,and get the overall convergence speed and generalization ability,convergence precision of superior original learning algorithm.

关 键 词：高技术知识创新 模糊神经网络 粒子群优化算法 评价方法 

分 类 号：C93[经济管理—管理学]