Loading...

Allah Yari, Mahdi | 2009

729 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 39856 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Soltanieh, Mohammad; Moslehi Moslehabadi, Parivash
  7. Abstract:
  8. Air pollution caused by industrialization is the problem which adversely affects human life. Among air pollutants suspended particles, especially particles smaller than 10 microns (PM10), for their high concentration in air in large cities are the major index as air pollutant. Due to their small size, PM10 can penetrate into the aspiration organs causing harmful effects. The objective of this work is to develop an Artificial Neural Network (ANN) model for prediction of short-term concentration of PM10 in the city of Tehran. Complex mechanism of reactions, numerous types of pollutant materials produced from transportation and industrial activities, variety of sources, difficulties in data collection and complexity of transformation processes of these particles in the atmosphere make the relationship between PM10 concentration and determinant parameters highly nonlinear and makes the prediction of concentration of particles a difficult task. In this project, the data collected from three stations in Tehran, Bazar, Fatemi and Aghdasiye for the period 2004-6 were used to develop a multi-layer perceptrone (MLP) neural network model. The results are compared with actual data and the results of traditional multiple regression (MLR). Root mean square error (RMSE) of MLP are for Bazar 35.28, 30.77, 49.03; for Fatemi 21.92, 23.45, 30.75 and for Aghdasiyeh 46.26, 54.36, 54.70 for 2006, 2005-6, 2004-5-6, respectively. These amounts for MLR are for Bazar 42.31, 37.19, 61.96, for Fatemi 24.74, 26.12, 33.43 and for Aghdasiye 61.99, 62.53, 53.60, respectively, which indicate that the results of MLP are so much better than MLR. In addition, the correlation coefficient for MLP, in comparison with actual data are for Bazar 0.641, 0.684, 0.7, for Fatemi 0.619, 0.58, 0.52 and for Aghdasiyeh 0.575, 0.546, 0.598, respectively which are reasonable results.

  9. Keywords:
  10. Neural Network ; Air Pollution ; Multiple Regression Analysis ; Multi-Layer Perceptron (MLP) ; Air Particulate Matter

 Digital Object List

  • محتواي پايان نامه
  •   view

 Bookmark

No TOC