In this paper, numerical inverse analysis is used to predict properties of heat generating material by measuring temperature at outer boundary. Accuracy and efficiency of the method is enhanced by using accurate sensitivity information by use of Semi-Analytical Complex Variable Method(CVSAM). Steady state heat transfer analysis with axi-symmetric model was carried out using finite volume method. Temperature obtained from this analysis, is used as input to inverse method. Objective function for the optimization is difference between computed and measured temperature. This function was minimized with Conjugate Gradient Method (CGM). Coefficients obtained from CVSAM were used in gradient based optimization method. The robustness of developed approach was evaluated by adding Gaussian noise these temperature values. Material properties predicted from this method show close agreement with actual values.
展开▼
