초록

용어보기논문에서 용어와 풀이말을 자동 추출한 결과로, 시범 서비스 중입니다. AI-Helper

Abstract ▼ AI-Helper

The prediction of F10.7, which is an indicator showing the solar activity, influencing seriously on atomic oxygen erosion of satellite surface coating materials is one of the key issues in the satellite design. It has been actively conducted recently using not only statistical but machine learning ...

The prediction of F10.7, which is an indicator showing the solar activity, influencing seriously on atomic oxygen erosion of satellite surface coating materials is one of the key issues in the satellite design. It has been actively conducted recently using not only statistical but machine learning techniques. In this paper, we study a method for using a long short-term memory (LSTM) network in predicting F10.7 data. We suggest a method of model validation to solve the over-fitting problem without extracting a validation sample out of the whole data set which does not reduce the size of the training data set. Also we show that predicting 12 points together at a time provides less forecasting error (RMSE, MAPE) than predicting one point at a time. The results from using LSTM network show more stable and competitively better forecasting performance than those from parametric long range dependent time series model autoregressive integrated moving average (ARIMA) model.

주제어