Fault-alarm-threshold optimization method based on interval evidence reasoning

In order to use IER method to update interval threshold, it is necessary to define a set of evaluation grades for interval threshold and observation data. There is a group of evaluation grades defined by experts H = (H1, H2, H3),where H1,H2 and H3 denote normal, fault, severe fault respectively. It is noted that the reference value setting for evaluation grades H1, H2, H3 are based on the technical description and expert knowledge of the equipment or system. Suppose that [ β− n,1, β + n,1 ] and [ β− n,2, β + n,2 ] stands for interval belief degrees of the interval threshold a1 and observation data a2 relative to the evaluation grades Hn , then the interval belief structures can be expressed as: