Synergy Study On Ash Fusion And Combustion Characteristics Of Mixed Biomass

The increasing shortage of coal can be solved by replacing power generation coal with biomass. It’s hard to keep the using system of biomass efficient and economic continuously because of the disperse distribution and limited single supply of biomass, which can be solved by using mixed biomass. There is no research of mixed biomass yet, which worth it.The research of power fuel mostly focuses on a certain aspect of the characteristics, such as pyrolysis, combustion, ash fusion and pollutant emissions. After the researching of combustion and ash fusion characteristics here, a combustion-ash fusion characteristics synergy evaluation index was built, which could evaluate the mixed biomass in two aspects.First of all, combustion characteristics of mixed biomass was researched by TGA/SDTA851e comprehensive thermal analyzer, predicted by support vector machine model and regression formula. It was found that, the TG/DTG curve and combustion characteristics got near to a certain kind of biomass with its content increasing; the mixed biomass combustion characteristics could be predicted by support vector machine model optimized by cross validation and genetic algorithm, whose accuracy rate was100%and running time was4.4s, taking quality percentage of proximate analysis components as input; the mixed biomass combustion characteristics index could be predicted by regression formula, with average absolute error1.68.Secondly, single biomass ash components was tested by EDS;(mixed) biomass ash fusion points were measured by ash fusion temperature pyrometer; mixed biomass ash fusion points were predicted by support vector machine model. It was found that, mixed biomass ash fusion points increased with the contents of biomass with high ash fusion points increasing; the mixed biomass ash fusion points could be predicted by support vector machine model optimized by particle swarm optimization algorithm and cross validation, taking ash components as input, with average relative error2.33%and running time2.33s; the model optimized by particle swarm optimization algorithm or genetic algorithm could predict well in less time compared with grid search, and the model optimized by cross validation could do better.Finally, combustion-ash fusion characteristics synergy evaluation index was built with concentration and multiplication, predicted by regression formula. It was found that, the combustion-ash fusion characteristics synergy evaluation index was effective and could evaluate the mixed biomass in two aspects, with C≥35as category Ⅰ,15≤C<35Ⅱ and C<15Ⅲ, decreasing; the index could be predicted by regression formula with the absolute error4.26; the index of biomass mixed with any proportions could be predicted by regression formula with the ash compositions, ash fusion points, proximate analysis components, and combustion characteristics index.