长输管线风险评价随机方法研究

针对目前长输管线的风险评价主要采用基于专家意见的多因素评估方法，由于评价者主观判断上的差异，对不确定因素和随机事件进行因素识别和风险评价，往往产生不同的结果，可信度较差。现场统计分析认为，许多安全事故都是由于不确定随机因素和结构构件失效引起。采用随机方法，将工艺、环境温度、风荷载、随机荷载等因素作为随机变量，按照随机过程，对其进行正态化处理后抽样，再根据可靠度理论，确定出不同随机事件对该结构失效概率的影响。仿真分析结果表明，该方法能够有效提高风险评价结果的精度和准确性。

STUDY ON RISK ASSESSMENT FOR LONG DISTANCE PIPELINE WITH STOCHASTIC METHOD

.2,173-177,2010（ISSN 1674-5086，in Chinese）Abstract：At present，the mainly applyed method of pipeline risk assessment is according to the advice of experts to score the risk factors of long distance pipeline，also this method is used to identify and assess theuncertainties factors and random incidents.Because of the differences of evaluator′s subjective judgment，different results are often produced.So，all the results are low in credibility.According to field statistics，a number of safety incidents are caused by the uncertain random factors and the failure of structural component.On this account，when the structure failure of crossing over pipeline is analyzed，the stochastic method is applied.The random variables include the factors of process，environment，temperature，wind load，random load etc.，in accordance with the random process，after these factors are normal treated and sampled，according to reliability theory，the impact of the different random events on the structure failure probability is determined.The simulation results of the analysis show that the method can effectively improve the accuracy of risk assessment results.