摘要
在数据挖掘的理论上,使用主成分分析对样本数据进行降维和预处理,然后以工业循环水的腐蚀速率为研究对象,建立基于改进的T-S模糊神经网络的腐蚀速率预测模型。应用于某石化实际生产数据,进行模型验证,并将该模型与BP神经网络模型进行比较,仿真结果证实了改进T-S模糊神经网络模型的有效性和优越性。
Basing on the data-mining theory, having the principal component analysis adopted to reduce the dimension of sample data and then to preproeess them was implemented, including taking the corrosion rate of industrial circulating water as the research object to establish a prediction model of the corrosion rate based on the improved T-S fuzzy neural network. Applying this prediction model to simulate production data of a petro- chemical company and then comparing it with BP neural network model show that, this improved T-S fuzzy neural network model is effective and has superiority.
出处
《化工自动化及仪表》
CAS
2018年第1期51-55,共5页
Control and Instruments in Chemical Industry
关键词
改进的T-S模糊神经网络
工业循环水
腐蚀预测
主成分分析
遗传算法
模型优化
improved T-S fuzzy neural network, industrial circulating water, corrosion prediction, principal component analysis, genetic algorithm, model optimization
