Session: CS-10-01 Recent Developments in Chinese Codes and Standards-1

Paper Number: 153251

153251 - Research on Data Mining Model of Corrosion Influencing Factors Based on Dual Algorithm-Driven 

Abstract: 

Addressing the issue of unclear correlations among corrosion factors in industrial process pipelines, this paper utilizes a dataset of corrosion-related indicators from the atmospheric and vacuum distillation unit's overhead draw-off gas oil pipeline in a refinery. The dataset is preprocessed to obtain a usable experimental dataset. By integrating the correlation coefficient method with the Apriori association rule mining algorithm, a data mining algorithm is designed to study the correlation between various corrosion influencing factors and the corrosion rate. A data mining algorithm for the influencing factors of petroleum process pipeline corrosion, based on the correlation coefficient method and the Apriori association rule mining algorithm, is proposed. The dataset is encoded to excavate the strength of the correlation between each corrosion influencing factor and the corrosion rate, as well as to mine strong correlation rules. The algorithm is optimized by examining its final implementation effects. Finally, the designed algorithm is applied to perform correlation analysis and association rule mining on various corrosion factors and corrosion rates, obtaining the influence weights of each corrosion factor on pipeline corrosion rates and deriving strong association rules, which are then validated using a verification dataset. This study provides significant practical implications for optimizing corrosion protection measures for process pipelines and offers valuable references for further research in related fields.

Presenting Author: Wei Li Beijing University of Chemical Technology

Presenting Author Biography: Student

Authors:

Zhou Fang Beijing University of Chemical Technology
Wei Li Beijing University of Chemical Technology
Jinkui Feng China Special Equipment Inspection and Research Institute
Jin Deng Special Equipment Inspection and Research Institute
Liangchao Chen Beijing University of Chemical Technology