摘要：

循环冷却水在工业生产中占有很大比重, 而腐蚀是循环冷却水系统中常见的水质故障, 严重影响工业生产中设备的运行。通过对某石化公司水质数据的分析, 选取对腐蚀速率影响较大的水质参数, 凭借神经网络良好的非线性能力, 基于NARX神经网络建立了腐蚀速率预测模型。利用该模型对循环冷却水系统腐蚀速率进行预测, 结果较好, 说明该模型可行, 且具有良好的应用前景。

关键词: 循环冷却水, 腐蚀速率, 非线性自回归模型, 预测模型

Abstract:

Circulating cooling water takes up a large proportion in industrial production, and corrosion is the common trouble in water quality which affects the operation of equipment seriously in industrial production. Through the data analysis on the water quality of a petrochemical company, the parameters of water quality having serious influence on the corrosion rate are accessed. By virtue of the nonlinear ability of neural network, and based on NARX neural network, the corrosion rate prediction model has been established. The corrosion rate of circulating cooling water systems can thus be predicted by making use of the model. The results are good, indicating that the model is feasible, and has a good application prospect.

Key words: circulating cooling water, corrosion rate, nonlinear autoregressive models, prediction model

中图分类号: 

• TP183