摘要
目的充分利用海洋腐蚀数据,深入分析数据规律。方法在BP神经网络的基础上引入遗传算法,以克服神经网络模型固有缺陷,提高预测精度和训练速度。结果对GA-BP人工神经网络进行了简要阐述,并以铜合金在海水中的腐蚀数据为例,应用GA-BP人工神经网络建立了海水腐蚀预测模型,对预测结果进行了评价。结论预测结果表明,模型能满足设计要求,具有较好的泛化能力。
This paper aims to take full advantage of corrosion data and deeply analyze corrosion law.The genetic algorithm(GA)was introduced to improve the back propagation(BP)artificial neural network(ANN),with a view to overcome the in-herent defect of ANN and increase prediction accuracy and training speed.In this paper,a brief interpretation of GA-BP artifi-cial neural network was given.And based on the corrosion data of copper in marine environment,the GA-BP artificial neural network method was applied to the building process of marine corrosion prediction model.The experimental results showed that the model can meet the design requirements and had good generalization ability.
作者
张彭辉
肖攸安
赵建仓
丁康康
侯健
ZHANG Peng-hui;XIAO You-an;ZHAO Jian-cang;DING Kang-kang;HOU Jian(State Key Laboratory for Marine Corrosion and Protection,Luoyang Ship Material Research Institute,Qingdao 266101,China;Wuhan University of Technology,Wuhan 430063,China)
出处
《装备环境工程》
CAS
2021年第12期73-78,共6页
Equipment Environmental Engineering