METHOD FOR DESIGN PROPELLER OF VESSEL BY IMAGE ANALYSIS OF CAVITATION BASED ON MACHINE LEARNING AND COMPUTER-READABLE RECORDING MEDIUM INCLUDING THE SAME

Disclosed are a method for designing a propeller of a vessel by image analysis of cavitation based on machine learning and a computer-readable recording medium recorded with a computer program for executing the method on a computer. The method includes: a step (S110) of deriving the optimal main dim...

Full description

Saved in:
Bibliographic Details
Main Author LEE YOUNG BUM
Format Patent
LanguageEnglish
Korean
Published 08.11.2022
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Disclosed are a method for designing a propeller of a vessel by image analysis of cavitation based on machine learning and a computer-readable recording medium recorded with a computer program for executing the method on a computer. The method includes: a step (S110) of deriving the optimal main dimensions of a propeller in a range that meets a specific conditional expression by an optimization technique when prior information including horsepower of the main engine and number of revolutions of the propeller are given after guaranteeing a derating point of the main engine of a vessel to be designed; a step (S120) of performing a cavitation tunnel test for the propeller manufactured according to the derived optimal main dimensions; a step (S130) of photographing cavitation image in the propeller in the cavitation tunnel test; a step (S140) of analyzing whether there is cavitation from the cavitation image through image-based machine learning; and a step (S150) of determining the optimal main dimensions, in which cavitation is not generated, as the final optimal main dimensions by changing the optimal main dimensions, redesigning the propeller, and repeatedly performing the cavitation tunnel test, photographing the cavitation image, and analysis on whether there is cavitation when cavitation is generated. Accordingly, the present invention may quickly determine whether cavitation is generated after the cavitation tunnel test and determine by machine learning the optimal main dimensions, in which cavitation is not generated. 본 발명은, 설계선의 주기관 디레이팅 포인트(derating point) 확정 후, 주기관의 마력과 프로펠러의 회전수를 포함하는 선행정보가 주어진 경우 최적화기법에 의한 특정 조건식을 충족하는 범위 내에서의 프로펠러의 최적 주요치수를 도출하는 단계(S110), 도출된 최적 주요치수에 따라 제작된 프로펠러에 대해 공동수조 실험을 수행하는 단계(S120), 공동수조 실험에서 프로펠러에서의 공동영상을 촬영하는 단계(S130), 영상기반 기계학습을 통해 공동영상으로부터 캐비테이션 발생유무를 분석하는 단계(S140), 및 캐비테이션이 발생하면, 최적 주요치수를 변경하여 프로펠러를 재설계하여서, 공동수조 실험과 공동영상 촬영과 캐비테이션 발생유무 분석을 반복수행하여 캐비테이션이 발생하지 않는 최적 주요치수를 최종 최적 주요치수로 결정하는 단계(S150)를 포함하여, 공동수조 실험 후 캐비테이션 발생여부를 신속하게 판단하고, 기계학습에 의해 캐비테이션이 발생하지 않는 최적 주요치수를 결정할 수 있는, 기계학습기반 캐비테이션 영상분석에 의한 선박 프로펠러 설계 방법 및 동 방법을 컴퓨터에서 실행하기 위한 컴퓨터 프로그램이 기록된, 컴퓨터 판독 가능한 기록 매체를 개시한다.
Bibliography:Application Number: KR20210056458