Parameters Identification of Ship Heave and Pitch Motion Model Based on Improved Firefly Algorithm

• Published in: 2024 43rd Chinese Control Conference (CCC) pp. 1334 - 1339
• Main Authors: Zhang, Xiufeng, Gao, Xiaopeng, Chen, Yunong, Yu, Yingjie
• Format: Conference Proceeding
• Language: English
• Published: Technical Committee on Control Theory, Chinese Association of Automation 28.07.2024
• Subjects: chaotic theory; Improved Firefly Algorithm; Mathematical models; Parameter estimation; parameter identification; Prediction algorithms; Roulette Wheel Selection Strategy; Ship heave and pitch motion; sigmoid function; Simulation; Training; Transportation; Wheels
• DOI: 10.23919/CCC63176.2024.10661701

Ships oscillate periodically while navigating at sea. To study the impact of the ocean environment on ship transportation, it is necessary to build accurate mathematical models of ship motion. Based on the theory of rigid body motion, build the mathematical models for ship heave and pitch, and discretize them; By introducing sigmoid function, roulette wheel strategy, and chaos theory, an improved Chaotic Firefly Algorithm Based on Roulette Wheel Selection Strategy (CRSFA) is proposed, which utilizes representative unimodal and multimodal test functions to optimize the algorithm through different dimensions. Compared with the original algorithm, the improved Firefly Algorithm based on Roulette Wheel Selection Strategy (CRSFA) proposed effectively avoids algorithm oscillations in this paper, accelerates algorithm convergence speed, and improves the algorithm's global searching ability. Based on the CFD numerical simulation data of the KCS ship model in calm water and 2 to 3 level sea conditions, parameter identification are carried out for the heave and pitch motion of the KCS ship model. The identified parameters are used to simulate and predict the heave and pitch motion of the ship using CRSFA and original FA separately. The simulation results show that fitted with the real time motion trend, verifying the feasibility of the identification scheme proposed in this paper.