Artificial neural network for the identification of unknown air pollution sources |
| |
| Institution: | 1. Escuela de Ciencia y Tecnolog??a, Universidad Nacional de General San Mart??n, Calle Alem 3901, 1653 Buenos Aires, Argentina;2. Unidad de Actividad Qu??mica, Centro Atómico Constituyentes, Comisión Nacional de Energ??a Atómica, Avda. Libertador 8250, 1429 Buenos Aires, Argentina;1. Meteorological Research Institute, 1-1 Nagamine, Tsukuba, Ibaraki 3050052, Japan;2. Center for Regional Environment Research, National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 3058506, Japan;3. Fukushima Project Office, National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 3058506, Japan;4. Department of Chemistry, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji, Tokyo 1920397, Japan;5. Atmosphere and Ocean Research Institute, University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 2778564, Japan;1. Department of Building Science, School of Architecture, Tsinghua University, Beijing, China;2. Institute of Industrial Science, University of Tokyo, Tokyo, Japan;3. Beijing Key Lab of Indoor Air Quality Evaluation and Control, Beijing 100084, China;1. Civil and Environmental Engineering School, University of Science & Technology Beijing, Beijing 100083, China;2. Beijing Key Laboratory of Information Service Engineering, Beijing Union University, Beijing 100101, China;1. Guangzhou Institute of Industrial Technology, Guangzhou, 511458, China;2. Guangdong Technology Center of Work Safety Co., Ltd., China |
| |
| Abstract: | Artificial neural networks (ANN), whose performances to deal with pattern recognition problems is well known, are proposed to identify air pollution sources. The problem that is addressed is the apportionment of a small number of sources from a data set of ambient concentrations of a given pollutant. Three layers feed-forward ANN trained with a back-propagation algorithm are selected. A test case is built, based on a Gaussian dispersion model. A subset of hourly meteorological conditions and measured concentrations constitute the input patterns to the network that is wired to recover relevant emission parameters of unknown sources as outputs. The rest of the model data are corrupted adding noise to some meteorological parameters and we test the effectiveness of the method to recover the correct answer. The ANN is applied to a realistic case where 24 h SO2 concentrations were previously measured. Some of the limitations of the ANN approach, together with its capabilities, are discussed in this paper. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
