Font Size: a A A

Experimental Study On Combustion And Nitrogen Conversion Characteristics During Oxy-CO2 And Oxy-H2O Combustion Of Coal/Biomass And Blend Type Identification

Posted on:2019-12-12Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y LiFull Text:PDF
GTID:1362330548985994Subject:Thermal Engineering
Abstract/Summary:PDF Full Text Request
CO2-based greenhouse gases emissions cause the serious global warming.The application of advanced combustion technologies and the utilization of renewable energy sources are the important ways to reduce CO2 emission.Oxy-CO2 combustion is regarded to be an economically feasible CCS technology in coal-fired power plants.Oxy-H2O combustion,using H2O vapor as the diluent instead of CO2 to moderate the combustion temperature,overcomes some disadvantages in oxy-CO2 combustion.Oxy-H2O combustion is considered to be a promising next-generation combustion technology.However,the studies on oxy-CO2 and oxy-H2O combustion are still not perfect at present.To simulate the oxy-CO2 and oxy-H2O combustion atmospheres,a flow tube system of controlled atmosphere was designed.Meanwhile,the experimental system can synchronously capture combustion images and measure the particle mass and the components of flue gas.The ignition and combustion characteristic of single coal and biomass particles in oxy-CO2 and oxy-H2O combustion were investigated in the experimental system,and the NO and N2O emissions during devolatilization and char combustion stages were studied.Meanwhile,the NO emission during co-firing coal and biomass were also investigated.To optimize the combustion state in power plants,an online blend type identification system was developed based on SVM and flame spectral characteristics.The main research is as follows:Firstly,the ignition and combustion characteristics of single coal and biomass particles were studied in the flow tube reactor in O2/N2,O2/CO2 and O2/H2O.The ignition model,ignition delay time,volatile burnout time and char burnout time were directly obtained by combustion image sequences.The volatile flame temperature and char combustion temperature were calculated by combining the combustion image with two-color pyrometry.The effects of O2 concentration,temperature and fuel type were considered.The ignition and combustion mechanisms of single coal and biomass particles were obtained in O2/N2,O2/CO2 and O2/H2O atmospheres.Secondly,the NO and N2O emissions during devolatilization and char combustion of single coal and biomass particles under air and O2/CO2 conditions were studied in the flow tube reactor.To obtain the nitrogen conversion during the two stages of a single fuel particle,the method of isothermal TGA combining with the fuel properties was proposed to distinguish the two stages.The effects of temperature,CO2 concentration,atmosphere and O2 concentration,H2O concentration,and fuel type were considered.The conversion mechanisms of volatile-N and char-N to NO and N2O of a single fuel particle were obtained under air and O2/CO2 conditions.Thirdly,the NO and N2O emissions during devolatilization and char combustion of single coal and biomass particles in O2/N2 and O2/H2O were investigated in the flow tube reactor.Similarly,the method of isothermal TGA combining with the fuel properties was used to distinguish the devolatilization and char combustion stages.The effects of temperature,O2 concentration,and fuel type were considered.The conversion mechanisms of volatile-N and char-N to NO and N2O of a single fuel particle were obtained in O2/N2 and O2/H2O atmospheres.Fourthly,the NO emissions during co-firing coal and biomass in O2/N2,O2/CO2 and O2/H2O were studied in the flow tube reactor.The effects of blending ratio,coal type,biomass type,and O2 concentration were considered.The conversion mechanisms of fuel-N to NO during co-combustion were obtained in O2/N2,O2/CO2 and O2/H2O atmospheres.Finally,the flame spectral characteristics during co-firing coal and biomass were captured using a flame monitoring system in the 0.3 MW furnace,and SVM was applied to identify the blend types.A total of 22 flame features were extracted from the flame spectral data,and"ReliefF+SVM" was used to obtain the optimal feature vector.The optimal sample number was also acquired.Under the optimal feature vector and the optimal sample number,the average success rate of 5 blends can reach 99.67%.
Keywords/Search Tags:Oxy-CO2, Oxy-H2O, Single particle, Ignition and combustion characteristics, Nitrogen conversion characteristics, Co-combustion, Blend type identification
PDF Full Text Request
Related items