Genetic algorithm based optimization method for kentledge laying of submersibles

Abstract

Objectives Typically, the fixed kentledge laying scheme of a submersible requires a significant amount of work and can often produce unpleasant results in engineering practice. Intelligent algorithms are considered for application in order to optimize the scheme. This paper proposes a genetic algorithm based method to solve the problem. MethodsFirst, by studying typical transverse sections of kentledge laying in a submersible and reverse-thinking the finite element method, a simplified equivalent mathematic model is constructed. Next, constraint functions and objective functions are extracted from the model. By using a genetic algorithm, improved fixed kentledge laying schemes are acquired with lower centers of gravity. By computing different examples, this method is proven to be effective in gaining good gravity center results with a much smaller workload.ResultsThe result shows that the gravity center of the improved scheme can be 23% lower compared to the ordinary scheme. This method is also effective in balancing longitudinal moment and lateral moment at the same time. ConclusionsThis study shows that the equivalent mathematical model and optimization method are feasible, and can improve both work efficiency and gravity results. Several general principles are also concluded, which can be helpful in engineering practice.

Keywords

• submersible
• fixed ballast
• mathematic model
• genetic algorithm
• general principle of kentledge laying