PhD Thesis Defense: Eric Krivitzky

"A Framework and Approach for Leveraging Unsteady Response In Turbocompressor Flowfields"

Turbomachinery is an essential technology in the transfer of mechanical work to or from a fluid stream. Forming a cornerstone component in nearly all electrical energy production and air transportation propulsion systems, turbomachinery also accounts for significant energy transfer due to its omnipresence in fluid handling, including water pumping and process machinery. As modern system designers look towards optimized arrangements that enhance system flexibility to highly variable conditions, increase the density of energy transfer, and reduce the amount of lost work produced, the performance and operability demands on turbomachinery components continues to increase. For turbocompressors, a turbomachinery subtype that transfer work to a fluid, the flowfield can be classified into two flow regimes, demarcated by a stability boundary representing an operational limit. For aerodynamic loadings above this stability limit, the flowfield is highly complex, exhibiting a broad range of temporal and spatial features, limiting work transfer and increasing entropy production. The blade-level instabilities, referred to as rotating stall, are the result of deleterious flowfield features, sensitive to perturbation, which have grown with aerodynamic loading.

Based on the thesis that critical destabilizing flow structures exhibit coherent response to periodic excitation and can be usefully organized via tuned periodic forcing, the work presented herein emphasizes the dynamical behavior of a representative compressor flowfield under periodic transients and the difficulty in extracting useful information on flowfield response in the post-stall regime. A new analysis approach is developed that enables better understanding of the rotating stall process, providing guidance for the use of data-driven tools and new approaches for control development. Emerging decomposition and operator-based analysis approaches are borrowed from dynamical system modeling to aid in deducing the coherent structures, their unforced behaviors, and critical forcing frequencies and locations. In this work, linear stability analysis and resolvent analysis are used to identify the underlying flow structures contributing to the onset of instability. Through a demonstrated surrogate model for compressor stability, a conceptual framework and practical approach are developed, such that the unsteady response of turbomachinery flows can be leveraged to achieve wider operability and enhance the transfer of usable work.

Thesis Committee

• Benoit Cushman-Roisin (Chair)
• Simon Shepherd
• Colin Meyer
• Louis Larosiliere, PhD (Elliot Group)

Contact

For more information, contact Julia Abraham at julia.s.abraham@dartmouth.edu.