某航改燃机氢燃料燃烧室污染排放特性研究

Pollution emission characteristics of hydrogen-fueled combustor of an aero-engine conversion gas turbine

关键词:  燃气轮机  氢燃料  低污染燃烧室  喷射单元  排放特性

Key words:  Gas turbine  Hydrogen fuel  Low emission combustor  Injection module  Emission characteristics

摘要:

为了对氢燃料燃烧室的低污染燃烧组织方案提供技术支撑，根据某航改燃机燃烧室的工况及结构参数，设计了三种氢燃料喷射单元（贫油直接喷射（LDI）单元以及两种贫油预混喷射（LPIA和LPIB）单元），通过数值计算方法研究了其流场特性，燃料-空气混合特性及污染排放特性，然后采用粒子群优化算法（PSO）建立了三种单元的NO_x排放回归模型。研究表明：氢气喷射方式对流场结构影响较大。LDI单元的中心回流区（CRZ）脱离喷嘴一段距离，LPIA单元的CRZ径向尺寸最大，LPIB单元的下游流场轴向速度最快；LPIA单元燃烧室内燃料-空气混合相对于其它两个单元更加均匀；在设计工况下，LDI单元的NO_x排放量不能满足目前新建燃机的排放标准（@15% O₂，≤30 mg/m³），而LPI单元的NO_x排放能够达标，且LPIA相对于LDI单元可降低81.6%；所建立三个单元NO_x排放回归模型考虑了燃烧室进口压力、温度、当量比及气流速度的影响，其决定系数R²分别为0.972，0.969和0.953。本文所采用的单元及其性能数据和建模方法可以为氢燃料燃烧室设计提供有效参考。

Abstract:

In order to provide technical support for the low-pollution combustion organization scheme of the hydrogen-fueled combustor， three hydrogen fuel injection modules （lean direct injection （LDI） module and two lean premixed injection （LPIA and LPIB） modules） were designed according to the working conditions and structural parameters of an aero-combustion engine combustor， and their flow field characteristics， fuel-air mixing characteristics and pollution emission characteristics were studied by numerical calculation methods， the NO_x emission regression models of the three modules were developed using the particle swarm optimization （PSO） algorithm afterwards. The study shows that the hydrogen injection mode has a large impact on the flow field structure： the central recirculation zone （CRZ） of the LDI module is some distance away from the nozzle， the radial size of the CRZ of the LPIA module is the largest， and the downstream flow field of the LPIB module has the fastest flow velocity； more uniform fuel-air mixing in the combustor assembling the LPIA module compared to the other two modules； under the design condition， the NO_x emissions of the LDI module cannot meet the current emission standards of new gas turbines （@15% O₂， ≤30 mg/m³）， while the NO_x emissions of the LPI module can meet the standards， and the NO_x emissions of the LPIA can reduce 81.6% compared with the LDI module； the NO_x emission regression models established for the three modules take into account the effects of combustor inlet pressure， temperature， equivalent ratio and airflow velocity， and their determination coefficients R² are 0.972， 0.969 and 0.953， respectively. The modules and their performance data and modeling method adopted in this paper can provide effective reference for hydrogen-fueled combustor design.