基于神经网络的长输管道综合可靠度评估方法

The Evaluation Method Based on Artificial Neural Network for Integrated Re-liability of Oil Pipeline.

对长输管道可靠性进行评估的传统方法是,将管道结构视为一个串联或并联系统,运用可靠性的理论概念和各种失效概率假设,以故障率的统计资料为计算依据来进行可靠度的计算分析,可靠度也采用故障率等参数来表达。该方法表达方式简单,含义通俗易懂,计算过程简便。然而,该可靠度的结果存在实用性不足的明显缺陷,不便于将其结论直接应用于管道设计、在线检测评估、管道维护等实际工作。为此从6个方面进行了分析,提出了构建综合可靠度的理论设想。运用神经网络的基本原理和模型进行分析建模,并提供了分析思路和步骤。

The evaluation for reliability of long-distance oil pipelines is urgent requirement for pipeline's owner. To analyze the reliability of oil pipeline,the traditional methods of the evaluation for reliability,based on the theory of reliability and supposal of failure rate,regard the pipeline as a system of series connection or parallel connection. The failure rate is used to represent pipeline's reliability. The reliability based on statistic data is unpractical for pipeline's design,inspection and maintenance. The paper shows six imperfections through analyzing above method. The paper puts forward a new concept of integrated reliability. The new concept takes the more factors related pipeline's design,operation and maintenance into consideration. The new method based on artificial neural network is more practical than the traditional ones. The approach and steps to establish analytic model have been given in the paper.