摘要
为提高人因失误风险评价的精度和效率,考虑人因失误对系统的影响,用人因失误概率(HEP)、失误影响概率(EEP)以及失误后果严重度(ECS)3个参数,对人因失误风险进行度量,以满足概率风险评价(PSA)的最终目的。将神经网络和模糊逻辑推理有机地结合,利用获得的120组人因失误风险数据,建立基于自适应神经-模糊推理系统(ANFIS)的人因失误风险评价模型,用于人因失误风险重要性识别。结果表明,用该模型可克服专家判断的主观性、模糊性和不确定性等缺陷,使人因失误风险评价结果更符合实际。
In order to improve the accuracy and efficiency of risk assessment of human error, effects of human errors on the system are considered, and the three parameters namely HEP, EEP and ECS are used to measure the risk of human error to satisfy the objective of probability risk assessment （PSA）. Using the obtained 120 sets of human error risk data, a risk assessment model of human error is built to assess the importance of human error risk based on ANFIS. The results show that risk assessment model of human er- ror can overcome the defects such as subjectivity, fuzzyness and uncertainty of expert judgment, and make assessment results of human error risk more realistic.
出处
《中国安全科学学报》
CAS
CSCD
北大核心
2014年第1期72-77,共6页
China Safety Science Journal
基金
国家自然科学基金资助(71071051
71371070
71301069)
岭东核电公司科研项目(KR70543)
