摘要
由于生态工业园的结构和外界影响因素均存在很多不确定性,给系统的运行带来了很多风险,定量化分析生态工业园风险成为一个研究难题。提出了一种应用GA-BP神经网络的生态工业园风险评价方法。首先提出一套生态工业园风险评价指标体系,并对指标体系的二级指标进行了分析;然后考虑各个评价指标的非线性关联性,提出一种基于遗传算法优化BP神经网络的生态工业园风险评价方法,该方法利用遗传算法优化BP神经网络的初始权值,提高了BP神经网络的求解精度;最后在Matlab R2011b环境下对30组样本进行案例研究。仿真结果表明:应用GA-BP神经网络的评估方法能较好地评估生态工业园风险水平,与标准BP网络相比收敛速度更快,具有更好的全局收敛性。
Since the uncertainty of Eco-industrial parks in form of structure and interaction with the environment,there are many risks during the operation of the parks,and it is difficult to quantify the risks.A risk assessment method of Eco-industrial parks based on GA-BP neural network was proposed.Firstly,the index system of risk assessment of Eco-industrial parks was built up and the secondary indexes were analyzed as well.And then,considering the nonlinear relationship among the assessing indexes,Genetic Algorithm （GA） and BP hybrid model were used as the evaluation method,where GA was used to optimize the initial weights of BP neural network to enhance the accuracy of the evaluating result.Finally,30 samples were simulated with Matlab R2011b,and simulation results showed that the assessing method was valid for decision support and it had characteristics of larger scope search.faster convergence and higher precision than standard BP neural network.
出处
《重庆理工大学学报（自然科学）》
CAS
2013年第10期75-79,86,共6页
Journal of Chongqing University of Technology：Natural Science
基金
江西省高校人文社会科学研究2011年度规划项目(GL1154)
关键词
生态工业园
风险评价
BP神经网络
遗传算法
eco-industrial parks
risk assessment
BP neural network
Genetic Algorithm
