Abstract

The performance of turbomachines is often dependent on the unsteady flow fields they naturally produce, owing primarily to row-to-row interactions from both moving and stationary components. When it comes to computational fluid dynamics, a disparity exists between steady state and transient simulation as far as accuracy is concerned, albeit the computational cost of transient simulation on fully complex industrial hardware can be overwhelming.  This study bridges the gap by presenting a harmonic balance conjugate heat transfer simulation approach in Simcenter STAR-CCM+, to model the unsteady flow phenomena while also providing accurate temperature predictions throughout the gas turbine blade solid bodies.  The harmonic balance method used is a mixed time domain and frequency domain technique, which is suitable for periodic unsteady flows and is much less expensive than transient simulation. With this method, the impact of capturing these unsteady flow structures, such as the wake interactions and secondary cooling flows, is quantified on the resulting metal temperature distribution. Such is investigated and characterized throughout an industrial gas turbine blade with fully complex internal cooling passages, as well as film cooling for the external blade surface. Comparisons to steady simulation and transient multi-time scale simulation are also made to quantify the relative computational cost and fidelity of each approach.