Birla Institute of Technology

A Neural Model for SPM Concentration Prediction in Coal Mines

- By: ,

Courtesy of Birla Institute of Technology

Coal mines are the major sources of suspended particulate matter. To implement any control techniques or to install any control equipment around the coal mines, ambient concentration of suspended particulate matter should be known. In this paper a model is developed to find the concentration of suspended particulate matter at various locations away from the source. The model is developed with the help of artificial neural networks. Multilayer Perception Network is used and learning is done by back-propagation algorithm. The data for training the network is collected from the field work done in Northkaranpura Coal Mines in Jharkhand, India. It is found that neural model yields better results than linear model.

KEYWORDS: Suspended particulate matter (SPM), wind speed, stability class, dispersion coefficients, Gaussian plume method, neural networks.

Air pollution is a major health problem affecting the developing and developed countries alike. Substances altering physical or chemical properties of the air added in sufficient concentration to produce a measurable effect on man or vegetation are considered as pollutants [1]. It is seen from studies that critical concentration of pollutants can seriously affect air quality. Air quality models are important tools for managing air quality. A model can predict concentration of pollutants in the atmosphere. A model may show that a threshold concentration of a particular pollutant can trigger an air pollution episode. 

The Industrial development of India depends on the availability of commercial energy, at affordable price. Coal is one of the cheap and abundantly available sources for commercial energy. New mines shall be opened and existing mines shall be expanded for the increase in production of coal. In both cases environmental pollution particularly, air pollution results.

Open cast mining is more severe in air pollution problem compared to under ground mining. Air pollution in coal mines is mainly due to fugitive emission of particulate matter and gases including methane, oxides of nitrogen etc. When a new open cast mine is opened, large amounts of overburden material has to be removed to reach the mineral deposit. Particulate matter is also released during excavation, screening, size reduction, waste removal, transportation, loading stock-piling, handling of coal etc. Vehicular traffic on haul roads can contribute as much as 80% of the dust emitted [2-3].

Models are one of the methods to calculate concentration of suspended particulate matter. Many of the adopted models have been based on physical principles. An Eulerian model which uses input fields of meteorology and emissions, for fine particulate matter, known as RPM (Regional Particulate Model) was designed for NAAQS. It predicts the composition and size distribution of air borne aerosol particulate [4]. The SNAP (Severe Nuclear Accident Program) model developed at Norwegian Meteorological Institute after the Chernobyl nuclear accident is in the Lagrangian frame work. It is used to predict the transport and deposition of radioactive nuclides in the form of gas or aerosol [5]. Standard Gaussian dispersion models are usually used to study the transport and diffusion of particulate matter.

Customer comments

No comments were found for A Neural Model for SPM Concentration Prediction in Coal Mines. Be the first to comment!