Facility On/Off Inference by Fusing Multiple Effluence Measurements

We consider the problem of inferring the on/off operational state of a reactor facility by using effluence mea- surements of three noble gases, namely, Ar-41, Cs-138, and Xe-138, which are collected on the facilitys ventilation stack. We first present classifiers based on thresholding measurements of individual effluence types, and then present methods that combine their outputs or measurements. We develop sample- based implementations of five fusers based on a simple majority rule, Chow's recognition function, physics-based radiation counts model, correlation-coefficient method, and Fisher's combined probability test. We apply the latter four fusers to pairs and all three gas effluence types. Our results show that: (i) these gas effluence measurements are effective in inferring the on/off status of a reactor facility, for example, best fusers achieve 97% detection at 1% false alarm rate, and (ii) performance depends on the data and classification method, and in particular, fusers that combine three effluence types based on physics-based models, correlation-coefficients and Fisher's method outperform majority rule and Chow's fusers as well as individual and pairs of effluence types, thereby illustrating the importance of fuser choice.