| 摘要 |
The present invention provides marine weather sample extraction performed by a computer, and a statistical signal analysis method using the same and comprising the steps of: (4-1) obtaining an accumulative distribution function for each of weather variables of atmospheric temperature, water temperature, relative humidity, wind speed, and wind direction, which form a population, and dividing the accumulative distribution function into slots corresponding to the number of samples to be extracted; (4-2) obtaining an occurrence probability of the weather variables combined and correlated with each other by using a principal component analysis method, and setting a high priority for sample extraction to data of which the occurrence probability is low; (4-3) partitioning zones between the weather variables and extracting samples according to the priority for each of the zones; (5-1) setting marine weather conditions according to the extracted samples; (5-2) setting operating conditions of a vessel to be operated under the weather conditions; (5-3) performing a signal analysis of the vessel for each of the weather conditions and the operating conditions and sorting the results thereof; and (5-4) obtaining probability density function and accumulative distribution function graphs by arranging a signal collection of the vessel for each interval. According to the present invention, a signal change characteristic of the vessel according to a change in the operating conditions of the vessel can be more accurately predicted by using the extracted sample weather conditions to predict signals of the vessel according to various operating conditions and statistically analyze the results thereof. |
