陈亮, 付冬梅. 低合金钢海水腐蚀监测中的双率数据处理与建模[J]. 工程科学学报, 2022, 44(1): 95-103. DOI: 10.13374/j.issn2095-9389.2020.06.17.003
引用本文: 陈亮, 付冬梅. 低合金钢海水腐蚀监测中的双率数据处理与建模[J]. 工程科学学报, 2022, 44(1): 95-103. DOI: 10.13374/j.issn2095-9389.2020.06.17.003
CHEN Liang, FU Dong-mei. Processing and modeling dual-rate sampled data in seawater corrosion monitoring of low alloy steels[J]. Chinese Journal of Engineering, 2022, 44(1): 95-103. DOI: 10.13374/j.issn2095-9389.2020.06.17.003
Citation: CHEN Liang, FU Dong-mei. Processing and modeling dual-rate sampled data in seawater corrosion monitoring of low alloy steels[J]. Chinese Journal of Engineering, 2022, 44(1): 95-103. DOI: 10.13374/j.issn2095-9389.2020.06.17.003

低合金钢海水腐蚀监测中的双率数据处理与建模

Processing and modeling dual-rate sampled data in seawater corrosion monitoring of low alloy steels

  • 摘要: 随着物联网技术的发展,前端传感器的使用使得低合金钢的海水腐蚀监测成为了现实,从而获得了大量的腐蚀数据。针对传统均值法处理双率腐蚀数据带来的数据信息损失以及建模精度下降问题,提出了一种基于综合指标值(CIV)和改进相关向量回归(IRVR)的双率腐蚀数据处理和建模算法(CIV-IRVR)。首先,通过构建CIV表征输入数据的综合影响并采用天牛须搜索(BAS)算法对其参数进行寻优;然后,建立最优CIV序列与输出数据间的线性回归模型将双率数据转化为建模用的单率数据,能够更多地保留原始数据信息;最后,给出了一种BAS算法优化的具有组合核函数的改进相关向量回归建模方法(IRVR),并建立了针对低合金钢海水腐蚀双率数据的CIV-IRVR预测模型。结果表明:相比于均值方法处理双率腐蚀数据,所提方法将建模样本数量由196提升到了1834;相比于海水腐蚀建模领域常用的人工神经网络(ANN)和支持向量回归(SVR)建模方法,所提模型的平均绝对误差(MAE)、均方根误差(RMSE)和决定系数(CD)分别为1.1914 mV、1.5729 mV以及0.9963,在各项指标上均优于对比算法,说明所提模型不仅减少了信息损失还提高了建模精度,对于双率海水腐蚀数据建模具有一定现实意义。

     

    Abstract: With the rapid development of Internet of Things technology, the use of front-end sensors realizes the corrosion potential online detection of low alloy steels in a marine environment, thereby obtaining multitudes of corrosion data. Concerning the problems of data information loss and modeling accuracy reduction caused by the use of the traditional mean value method when processing dual-rate corrosion data, a new dual-rate data processing and modeling algorithm combining the comprehensive index value (CIV) and improved relevance vector regression (IRVR) was proposed. First, the CIV was constructed to characterize the comprehensive influence of the input data, and the beetle antennae search (BAS) algorithm was applied to optimize its parameters. Then, linear regression models between the best CIV sequence and the output data were established to convert the dual-rate corrosion data into single-rate data for modeling, which retained more information of the original corrosion data. Finally, the IRVR method based on BAS optimization of compounding kernels was given to establish the prediction model for dual-rate seawater corrosion data of low alloy steels. The results show that the proposed model CIV-IRVR increases the number of modeling samples from 196 for the mean value method to 1834. Moreover, the mean absolute error, root mean square error, and coefficient of determination of the CIV-IRVR model are 1.1914 mV, 1.5729 mV, and 0.9963, respectively, which outperforms commonly used comparison algorithms, such as the artificial neural network (ANN) and support vector regression (SVR). Moreover, the CIV-IRVR model can help obtain the prediction results with error bars, and it has the absolute error distribution closest to 0, which highlights its excellent predictive performance on the seawater corrosion potential of low alloy steels. Thus, the proposed model not only reduces the information loss and improves the modeling accuracy but also has practical significance for modeling dual-rate seawater corrosion data.

     

/

返回文章
返回