Rongxiao Wang
ABSTRACT
Air pollution caused by industrial production has become a serious problem for public health. This challenging problem promotes the development of the research in the air contaminant dispersion (ADS) prediction, for the management of the emission and leak accident. However, conventional ADS models can hardly meet the requirement of both accuracy and efficiency. The data model, like the artificial neural network (ANN) provides a feasible way of forecasting the dispersion with high accuracy and efficiency. However, the construction of the ANN for prediction needs plenty of data, which is impractical to obtain in most emission cases. To address this problem, an ADS simulation software AermodSystem is applied to build the simulated dispersion scenarios and provide synthetic dataset for the model training and test. Based on the synthetic data set, the ANN prediction model is established, and evaluated on the test set, as well as the Gaussian model. Further, these two models are served as the forward dispersion model and combined with the Particle Swarm Optimization (PSO) for source estimation. The results verify the effectiveness of the proposed model and indicate that the ANN together with the AermodSystem as the data generator is feasible in the air contaminant dispersion forecast and the source estimation of a particular case.
- BING, W., BINGZHEN, C., and JINSONG, Z., 2015. The real-time estimation of hazardous gas dispersion by the integration of gas detectors, neural network and gas dispersion models. Journal of Hazardous Materials 300, 20 (12/30/), 433--442.Google Scholar
- BOZNAR, M., LESJAK, M., and MLAKAR, P., 1993. A neural network-based method for short-term predictions of ambient SO2 concentrations in highly polluted industrial areas of complex terrain. Atmospheric Environment. Part B. Urban Atmosphere 27, 2 (1993/06/01/), 221--230.Google Scholar
- BRIGGS, G.A., 1973. Diffusion estimation for small emissions. Preliminary report. Atmospheric Turbulence and Diffusion Laboratory; NOAA.Google Scholar
- D. WILSON, J. and L. SAWFORD, B., 1996. Review of Lagrangian Stochastic Models for Trajectories in the Turbulent Atmosphere. Boundary-Layer Meteorology 78, 191--210.Google ScholarCross Ref
- FENG, X., LI, Q., ZHU, Y., HOU, J., JIN, L., and WANG, J., 2015. Artificial neural networks forecasting of PM2.5 pollution using air mass trajectory based geographic model and wavelet transformation. Atmospheric Environment 107(2015/04/01/), 118--128.Google Scholar
- FLESCH, T.K., WILSON, J.D., and YEE, E., 1995. Backward-time Lagrangian stochastic dipsersion models and their application to estimate gaseous emissions(1995-06-01).Google Scholar
- G. PERRY, S., CIMORELLI, A., PAINE, R., W. BRODE, R., C. WEIL, J., VENKATRAM, A., B. WILSON, R., F. LEE, R., AND, and D. PETERS, W., 2005. AERMOD: A Dispersion Model for Industrial Source Applications. Part II: Model Performance against 17 Field Study Databases.Google ScholarCross Ref
- GARDNER, M.W. and DORLING, S.R., 1998. Artificial neural networks (the multilayer perceptron)---a review of applications in the atmospheric sciences. Atmospheric Environment 32, 14 (1998/08/01/), 2627--2636.Google Scholar
- HANNA, S.R., BRIGGS, G.A., HOSKER, R.P., SMITH, J.S., UNITED STATES. DEPARTMENT OF ENERGY. OFFICE OF ENERGY, R., UNITED STATES. DEPARTMENT OF ENERGY. OFFICE OF, H., and ENVIRONMENTAL, R., 1982. Handbook on atmospheric diffusion. U.S. department of energy.Google Scholar
- HUTCHINSON, M., OH, H., and CHEN, W.-H., 2017. A review of source term estimation methods for atmospheric dispersion events using static or mobile sensors. Information Fusion 36(7//), 130--148. Google ScholarDigital Library
- KENNEDY, J., 1995. Particle swarm optimization. Proc. of 1995 IEEE Int. Conf. Neural Networks, (Perth, Australia), Nov. 27-Dec. 4(1995), 1942--1948.Google ScholarCross Ref
- L. BARAD, M., 1958. Project Prairie Grass, a field program in diffusion. Air Force Cambridge Center.Google Scholar
- LAURET, P., HEYMES, F., APRIN, L., and JOHANNET, A., 2016. Atmospheric dispersion modeling using Artificial Neural Network based cellular automata. Environmental Modelling & Software 85(2016/11/01/), 56--69. Google ScholarDigital Library
- LTD., H.S.A.T., 2018. Guide of the AermodSystem.Google Scholar
- MA, D., DENG, J., and ZHANG, Z., 2013. Comparison and improvements of optimization methods for gas emission source identification. Atmospheric Environment 81, Supplement C (2013/12/01/), 188--198.Google Scholar
- MA, D. and ZHANG, Z., 2016. Contaminant dispersion prediction and source estimation with integrated Gaussian-machine learning network model for point source emission in atmosphere. Journal of Hazardous Materials 311(2016/07/05/), 237--245.Google Scholar
- PODNAR, D., KORAČIN, D., and PANORSKA, A., 2002. Application of artificial neural networks to modeling the transport and dispersion of tracers in complex terrain. Atmospheric Environment 36, 3 (2002/01/01/), 561--570.Google Scholar
- PONTIGGIA, M., DERUDI, M., BUSINI, V., and ROTA, R., 2009. Hazardous gas dispersion: A CFD model accounting for atmospheric stability classes. Journal of Hazardous Materials 171, 1 (2009/11/15/), 739--747.Google ScholarCross Ref
- ROJAS, R., 1996. Neural networks: a systematic introduction. Springer-Verlag. Google ScholarDigital Library
- SO, W., KOO, J., SHIN, D., and SUP, Y.E., 2010. Optical Sensor and Neural Networks for Real-Time Monitoring and Estimation of Hazardous Gas Release Rate. JOURNAL OF CHEMICAL ENGINEERING OF JAPAN 43, 8 (2010/12/01), 682--690.Google Scholar
- STEVEN HANNA, J.C., HELGE R.OLESEN, 2005. Indianapolis Tracer Data and Meteorological Data National Environmental Research Institute.Google Scholar
- WANG, R., CHEN, B., QIU, S., MA, L., ZHU, Z., WANG, Y., and QIU, X., 2018. Hazardous Source Estimation Using an Artificial Neural Network, Particle Swarm Optimization and a Simulated Annealing Algorithm. Atmosphere 9, 4, 119.Google ScholarCross Ref
- WANG, R., CHEN, B., QIU, S., ZHU, Z., and QIU, X., 2017. Hazard Source Estimation Based on the Integration of Atmospheric Dispersion Simulation and UAV Sensory System. In Modeling, Design and Simulation of Systems: 17th Asia Simulation Conference, AsiaSim 2017, Melaka, Malaysia, August 27 --- 29, 2017, Proceedings, Part I, M.S. MOHAMED ALI, H. WAHID, N.A. MOHD SUBHA, S. SAHLAN, M.A. MD. YUNUS and A.R. WAHAP Eds. Springer Singapore, Singapore, 494--504.Google Scholar
- WANG, R., CHEN, B., QIU, S., ZHU, Z., WANG, Y., WANG, Y., and QIU, X., 2018. Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases. International journal of environmental research and public health 15, 7, 1450.Google Scholar
- XING, J., LIU, Z., HUANG, P., FENG, C., ZHOU, Y., ZHANG, D., and WANG, F., 2013. Experimental and numerical study of the dispersion of carbon dioxide plume. Journal of Hazardous Materials 256-257(2013/07//), 40--48.Google Scholar
- YEGANEH, B., MOTLAGH, M.S.P., RASHIDI, Y., and KAMALAN, H., 2012. Prediction of CO concentrations based on a hybrid Partial Least Square and Support Vector Machine model. Atmospheric Environment 55(2012/08/01/), 357--365.Google Scholar
Index Terms
- The Air Contaminant Dispersion Prediction by the Integration of the Neural Network and AermodSystem
Recommendations
Augmentation of Elman Recurrent Network Learning with Particle Swarm Optimization
AMS '08: Proceedings of the 2008 Second Asia International Conference on Modelling & Simulation (AMS)Despite a variety of Artificial Neural Network (ANN) categories, Backpropagation Network (BP) and Elman Recurrent Network (ERN) are the widespread modus operandi in real applications. However, there are many drawbacks in BP network, for instance, ...
Multiplicative neuron model artificial neural network based on Gaussian activation function
Multiplicative neuron model-based artificial neural networks are one of the artificial neural network types which have been proposed recently and have produced successful forecasting results. Sigmoid activation function was used in multiplicative neuron ...
Artificial Neural Network trained by Particle Swarm Optimization for non-linear channel equalization
Development of a learning method for optimization of network topology of ANN.Use of PSO trained ANN in channel equalization.PSO trained ANN equalizers performs better than PSO based equalizers as well as ANN based equalizers. In this paper, we apply ...
Comments