STATISTICAL SIGNATURE ANALYSIS METHOD USING OCEAN CLIMATIC SAMPLE DATA BY COMPUTER

摘要

The present invention relates to marine weather sampling performed by a computer and a statistical signal analyzing method comprising a step for calculating an accumulated distribution function for each weather variable of the temperature, the water temperature, the relative humidity, the wind speed, and the wind direction of the collection group and dividing the accumulated distribution function into slots of the number of specimens to be sampled (4-1); a step for calculating a generation probability of weather variables which are combined to each other with the correlation by using a main component analysis method (4-2); a step for dividing a section between the weather variables and sampling a specimen according to priority for each section (4-3); a step for setting a weather condition of marine according to the extracted specimens (5-1); a step for setting a operation condition of a vessel operated in the weather condition (5-2); a step for analyzing the signals of the vessel for each weather condition and each operation condition and arranging a result (5-3); and a step for arranging a signal number of the vessel for each section and obtaining a probability density function and a cumulative distribution function graph (5-4). The present invention is provided to predict the signals of the vessel according to various operation conditions by using the extracted specimen weather conditions and statistically analyze a result, thereby accurately predicting a signal change feature of the vessel according to the change of the operation condition of the vessel.