An optimisation method for anti-blast performance of corrugated sandwich plate structure based on neural network and sparrow search algorithm

This paper presents an optimization method for the design of marine metal sandwich plates to enhance the anti-blast performance of ship structures. The proposed method combines a neural network with the sparrow search algorithm to efficiently optimize the structure. The orthogonal design method is utilized to select appropriate sample points for the surrogate model. Dynamic responses of the ship under blast loading are obtained using the ABAQUS finite element software. By establishing an anti-blast surrogate model, the optimization problem is simplified. To overcome the limitations of the traditional backpropagation neural network, genetic algorithm and Adam algorithm are employed. Finally, the optimal solution is obtained based on the sparrow search algorithm. The results enable us to determine the structural dimensions with the best blast resistance. The proposed optimisation method can provide inspiration for improving the safety of ships.