摘要
采用模拟退火算法与神经网络相结合的方法建立了地下水水质评价的SA-BP神经网络模型,并对阜新新邱露天煤矿地下水水质进行了评价。结果表明,模拟退火算法具有快速学习网络权重和全局搜索的超强能力,有效地解决了BP算法的局部收敛的问题。应用此方法评价地下水水质简便可靠,预测精度高,具有通用性和客观性等优点。
This paper designs a method combining the neural network with simulated annealing algorithm. By this method, SA-BP neural network model of groundwater quality appraisal is set up, and the groundwater quality of Xinqiu Open Pit Coal Mine in Fuxin had been appraised. The research result indicated that simulated annealing algorithm can study fast the neural weights and search the overall situation, and that simulated annealing algorithm effectively solves the local convergence of BP algorithm. This method is simple, convenient and reliable to appraise groundwater quality, and has many advantages such as high accuracy of prediction and objectivity, etc.
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2005年第z1期244-246,共3页
Journal of Liaoning Technical University (Natural Science)
基金
辽宁工程技术大学校基金项目(02-36).
