Double membership belief rule base for oil pipeline leakage detection

作者：Huimin Guan, Zeyang Si, Jinting Shen, Wei He, Guohui Zhou, Hongyu Li · 发表于：Measurement Science and Technology · 年份：2025 · DOI：10.1088/1361-6501/ae2b25 · 被引用次数：2 · 研究领域：Water Systems and Optimization、Structural Integrity and Reliability Analysis、Infrastructure Maintenance and Monitoring

Abstract Accurate detection of oil pipeline leaks can effectively prevent environmental disasters and reduce economic losses. While the belief rule base (BRB) excels in handling uncertainty and complex reasoning, it is limited by the fixed triangular membership function used to transform input information, which may not accurately match various fuzzy data distributions. Therefore, this paper proposes a double membership belief rule base (DM-BRB) method for leakage detection. A new distance-based membership function is designed to assign membership degrees to all parameters, thus compensating for the inherent defects of the triangular membership function. Input information is processed using double membership degrees, tailored to the characteristics of oil pipeline leak detection. Additionally, a constrained projection covariance matrix adaptive evolutionary strategy is implemented to maintain interpretability during the optimization process. Experimental results show that the new DM-BRB exhibits superior detection performance.