Pollutant Source Localization Based on Siamese Neural Network Similarity Measure

In this paper, we present an optimization methodology for reducing the number of sensors in an existing monitoring network. These sensors measure the concentration of pollutant gas in the air, in order to estimate the position and intensity of a pollutant source. In a Hierarchical Agglomerative Clustering (HAC) framework we aim to regroup sensors of the same behavior, based on similarity measure, then, we keep only one sensor of each cluster. Unlike previous studies that used Pearson correlation coefficient and euclidean distance, our work uses a similarity measure based on Siamese Neural Networks (SNN). The methodology was tested on simulated measurements based on real atmospheric conditions. And Monte Carlo Markov Chain (MCMC) in a Bayesian inference framework was used to identify the source position and intensity.