Development and Application of a Simulation-Based Design Process for Fuel-Flexible Burners using Additive Manufacturing

为了向未来的可持续能源系统转型，迫切需要燃料柔性燃烧器来应对从天然气向全氢燃料运行的逐步过渡。由于氢基燃料与传统燃料具有根本不同的燃烧特性，在火焰稳定性和回火、排放和混合气形成方面，这类燃烧器的开发存在重大挑战。这一建议的最终目标是使工业燃烧器系统的设计能够特别强调在确保可制造性和安全性的同时，使用不同的氢气/甲烷混合物进行稳健和高效的操作。在拟议的工作中，这些挑战将通过将模型开发和基于模拟的设计过程与实验研究和创新的添加剂制造(AM)相结合来解决。后者允许在设计上有很大的自由度，以创建紧凑、灵活和本质稳定的燃烧器。在这个项目中，将开发一种逆向设计过程，并将其应用于工业多射流燃烧器的特定目标量的优化，其中包括火焰特性，如稳定性(火焰回火和喷出)和排放(NOx，CO)，以及设计参数，如内部压力损失和耐热性。一个挑战是需要准确和成本效益高的氢基燃料模拟技术，这将作为该项目的一部分进行开发。这包括雷诺平均纳维斯托克斯(RANS)模拟，它将通过使用大涡模拟结果和实验数据应用基于集合卡尔曼滤波的数据同化来提高预报能力。该实验提供了在RANS模型中包含热扩散不稳定性的影响所必需的湍流火焰的消耗速度和特性。然后应用同化RANS模型，在考虑AM施加的设计约束的同时，使用伴随灵敏度来优化燃烧器的几何形状和运行策略AM将使用燃烧器的参数模型来促进设计过程。进一步的发展将提高AM制造的燃烧器的质量，包括优化薄壁、丝状特征和可实现的悬臂角度，特别是耐高温材料。该方案中包括燃烧工程学和AM在内的强跨学科方法将为燃料柔性工业燃烧器的基于模拟的设计过程提供方法学框架。

For the transformation towards future sustainable energy systems, fuel-flexible burners are urgently needed to cope with the gradual transition from natural gas to fully hydrogen-fueled operation. Significant challenges exist for the development of such burners in the areas of flame stability and flashback, emissions, and mixture formation, as hydrogen-based fuels have fundamentally different combustion properties compared to conventional fuels. The ultimate objective of this proposal is to enable the design of industrial burner systems with special emphasis on robust and efficient operation with different hydrogen/methane blends while ensuring manufacturability and safety. In the proposed work, these challenges will be tackled by combining model development and a simulation-based design process with experimental investigations and innovative additive manufacturing (AM). The latter allows for a large degree of freedom in design needed to create compact, flexible, and intrinsically stable burners. In this project, an inverse design process will be developed and, as a demonstration, applied to optimize an industrial multi-jet burner with respect to specific target quantities including flame properties, such as stability (flame flashback and blow-off) and emissions (NOx, CO), as well as design parameters, such as internal pressure loss and heat resistance. One challenge is the need for accurate and cost-efficient simulation techniques for hydrogen-based fuels, which will be developed as part of the project. This comprises Reynolds-averaged Navier Stokes (RANS) simulations, which will be improved in predictiveness by applying data assimilation based on the Ensemble Kalman Filter using results from large-eddy simulations and experimental data. The experiment provides consumption speeds and properties of turbulent flames necessary to include effects of thermodiffusive instabilities in the RANS model. The assimilated RANS model will then be applied to optimize the burner geometry and operating strategies using adjoint sensitivities while considering design constraints imposed by AM. The design process will be facilitated by AM using parametric models of the burner. Further developments will be made to improve the quality of AM-fabricated burners, which includes optimizations regarding thin walls, filigree features, and achievable overhang angles especially for high-temperature resistant materials. The strongly interdisciplinary approach including combustion engineering and AM in this proposal will enable the methodological framework for a simulation-based design process of fuel-flexible industrial burners.