The Huangtai Plant Lean Coal And Fuel Mix Of Coal Sulfur Release Experimental Study

The coal is particularly important in the first source energy, and the tinpot high-sulfur coal accounts for major proportion. The sulfurous pollutant resulting from the coal combustion not only destroys environment, but also damages the security and the economical efficiency of the power plant. Although it is researched much, the sulfur release characteristic of the inferior coal and the blended coal is studied less. And in the staged combustion and the closing-to-wall air, it is hardly done. So it is important to study the sulfur release characteristic of the high-sulfur lean coal and the blended coal which is composed of the lean coal and the other coal.To begin with, the SO₂ and H₂S release characteristic of HT coal which is the dominating coal fired in Huangtai power plant is studied with the operating condition changed. The operating condition included the alterable excess air coefficient, the alterable granularity of the pulverized coal, the alterable ratio of the primarily air and the second air in the staged combustion, and the alterable air volume of the closing-to-wall air. And it was studied with the temperature-changed in the tube furnace of static combustion. The result was showed that the SO₂ and H₂S release characteristic was obviously disciplinarian with the excess aircoefficient, granularity of the pulverized coal and temperature changed. And mixing of the primarily air and the second air was delayed to affect the sulfur release characteristic. At the initial stages of the combustion, with the proportion of the primarily air, the SO₂ concentration increased and the H₂S concentration decreased, and when the proportion of the primarily air was little, the release of SO₂ was brought forward. Then the closing-to-wall air the concentration of H₂S decreased, with the closing-to-wall air added. However, it did not affect the SO₂ release.Then, the sulfur release of characteristic and combustion characteristic of blended, which stemed from putting the lean coal, the bituminous coal and the anthracite into the HT coal, was respectively in the one dimensional flame test-bed and the thermogravimetric analysis apartment. The result was showed that the combustion characteristic of the blended coal was different with the coal quality characteristic and the blend ratio changed in the blended coal. The burn out efficiency of blended which was mensurated in the one dimensional flame test-bed was close correlative with the burn out efficient of the single coal and the blend ratio of the single coal. The sulfur release characteristic of the different blended coal was complicated. And the SO₂ release concentration of the different blendedcoal along the hearth was among that of single coal with the blend ratio of the single coal changed. The caculated SO2 concentration of blended coal according to the blend ratio of the single coal was different to the SO2 concentration from the experiment, with the coal rank and the blend ratio of the single coal changing. The H2S release concentration of the blended coal depended the combustion opertator to a considerable degree. The disciplinarian of the H2S release concentration in the blended coal, which was not among the single coal sometime, combustion was intricate.At last, applying to the BP artificial neural network which was typically provided with three layer structure forecasted the SO2 and H2S concentration of the different blended coals, that other coal was put in HT coal. The input variable included the anthrax characteristic and sulfur content of the each coal, burnout characteristic, blend ratio of the single coal, the pulverized coal concentration, and temperature of the furnace. And the cutput variable was made of the H2S concentration at the NO.3 measuring point and SO2 concentration at the trail of the furnace. And there were thirteen nerve cells in the latent lay by the calculation. Applying to the BP net arithmetic possessing the momentum gene, trained the twenty-one sample data from the expriment and the three sample data was used to check up the validity of the BP artificial neural network. Through comparing the predicted value with the metrical value it was showed the minimal relative error was 3.7% and the maximal relative error was 21%, and the error met with the technical need. Therefore the artificial neural network could forecast the sulfurous pollutant concentration.The SO2 and H2S release characteristic of the different blended coals was particularly studied in this paper, and the artificial neural network model was put up in order to forcast the sulfur release of the blended coal. So it was hoped that there was some basal action on sulfur capture during combustion, preventing high temperature corrosion of water-wall tubes and optimizing coal blending technology.