Seismic Data Analysis Regression Model on Reactor Pressure Vessel using Fast Fourier Transform and Machine Learning.

The paper presents a way for data analysis of seismic data in order to predict stress intensity data on reactor pressure vessel because it is important to investigate the integrity of the reactor pressure vessel. As the seismic waveform data are time-series data, fast Fourier Transform is implemented for data processing. After feature extraction using fast Fourier Transform, machine learning algorithms were used to analyze and predict the stress intensity data for regression. We applied Support Vector Regression, Random Forest Regression, K-nearest Neighbor Regression and Gradient Boosting Regressor and compared these algorithms in order to improve good accuracy on the regression. This research shows that it is possible to make the correlation between the seismic waveform data and the stress intensity for reliability on the reactor pressure vessel.