New Numerical Analysis Method for Creep Lifetime of Gas Turbine Materials by Using the Discrete Cosine Transform

Abstract

Abstract Recently the Discrete Cosine Transform[1], [2], [3] which is a modified Fourier Transform has begun to be used to express coefficients of creep equations using the power law or the exponential law such as Bailey-Norton law[4], [5] and θ Projection[6], [7], [8], [9], [10]. In addition, the Discrete Cosine Transform has begun to be used to express a creep equation itself. We have already found that the Discrete Cosine Transform can express the temperature and stress dependence property of the coefficients of the creep equations at the same time by the two-dimensional Discrete Cosine Transform using 8 × 8 discrete signals[11]. Furthermore, we have already found that the Discrete Cosine Transform can fit measured creep strain values very well from the primary creep region to the tertiary creep region using 8 discrete signals and it can estimate creep strain values between the measured points by interpolation very well[12]. However it has not been known if the Discrete Cosine Transform can predict the long term creep curve by using the short term creep data yet. Therefore, as a next stage, we tried to estimate the long term creep curve from the short term creep data of gas turbine materials by extrapolation using the Discrete Cosine Transform. As a result, we were able to obtain a useful numerical analysis method by utilizing the Discrete Cosine Transform Coefficients and others as a new extrapolation method. It is found that this new numerical method would be able to predict the configuration of 150,000-hour creep curve by using 500-hour to 13,000-hour short term creep data.