基于满意度优化的过程报警阈值自适应调整

Satisfaction Optimization Based Self-adaptive Adjustment of Process Alarm Thresholds

过程参数报警阈值的合理设置对报警系统的性能具有重要的影响,针对传统的过程报警阈值静态优化的局限性,提出了一种根据满意度优化的过程报警阈值自适应调整方法。首先,基于综合报警处理率和报警时间恢复率的模糊推理获得满意度指标;然后,根据误报、漏报与满意度的关系计算两者的权重,利用核密度估计法建立误报率和漏报率的加权目标函数;最后,通过数值优化方法求解最优阈值。工业实例数据验证表明,基于此方法的过程报警系统具有较好的环境适应性,性能得以进一步的改善。

The reasonable setting of process parameter alarm thresholds greatly influences the performance of an alarm system. In response to the limitations of traditional static threshold optimization, a novel satisfaction optimization based approach to self-adaptive adjustment of process alarm thresholds is explicitly introduced in this paper. Firstly, based on alarm treatment rates and alarm time recovery rates, a fuzzy reasoning strategy is employed to get the satisfaction index. Subsequently, the relations between the satisfaction index and FAR(false alarm rate), MAR(miss alarm rate) are used to calculate weights of FAR and MAR involved in the optimization objective function, in which kernel density estimations are conducted.Finally, numerical optimization methods are employed to achieve the optimal thresholds. Industrial data simulation results show that the proposed method has better environment adaptability and improved performance for alarm systems.