Nonlinear identification of unsteady heat transfer of a cylinder in pulsating cross flow

Unsteady heat transfer of a cylinder in pulsating cross flow is investigated. The heat transfer process is considered as a nonlinear single-input, single-output system, with large amplitude velocity perturbations as "input" and total heat transfer rate from the cylinder surface to the fluid as "output". Quantitatively accurate, low-order models of the heat source dynamics are obtained with a variety of nonlinear system identification methods from time series data generated with unsteady CFD computations. A polynomial type, equation error identification scheme is found to yield very accurate results. In order to obtain the heat source response function in the frequency domain, the equation error model is converted into the frequency domain using harmonic balance as well as harmonic probing approaches. In the harmonic balance approach, a set of nonlinear algebraic equations is solved for the coefficients of the harmonic ansatz. In the harmonic probing method, a recursive relation is obtained to deliver the higher order transfer functions of the nonlinear heat source. Alternatively, a nonlinear black box identification technique is used to identify the heat source dynamics. It is found to capture the amplitudes of higher harmonics with increased accuracy.