The invention discloses a ship attitude prediction method based on an SCA-HHO-GRU model, and belongs to the technical field of deep learning and computer vision. The method mainly comprises the following steps: S1, acquiring to-be-processed three-degree-of-freedom data of a ship motion attitude, and preprocessing the data, which mainly comprises the following steps: sorting and acquiring the data, intercepting the data size, and performing normalization processing; s2, the best automatic optimization method is provided by using the advantages of a Harris Hawks Optimizer (HHO) and a Sine Cosine Algorithm (SCA), and the best automatic optimization method mainly comprises the following steps: designing a cross parallel optimization method framework, and initializing parameters of the optimization algorithm; s3, performing automatic optimization on parameters of the GRU neural network by adopting an SCA-HHO optimization algorithm, and fully developing the processing capability of the neural network, which mainly comprises the following steps: initializing a GRU network structure, and searching optimal parameters through an HHO-SCA optimization algorithm framework; and S4, assigning an optimal parameter corresponding to the obtained optimal fitness value to the GRU network, and predicting three-degree-of-freedom data of the ship after training. The SCA-HHO-GRU model framework provided by the invention can be used for solving the problems of low precision and large error during ship attitude prediction. By adopting the model provided by the invention, the motion attitude of the ship can be accurately predicted, and the normal take-off and landing of a shipboard aircraft and the safety of workers on the ship can be guaranteed.
本发明公开了一种基于SCA‑HHO‑GRU模型的船舶姿态预测方法,属于深度学习和计算机视觉技术领域。该方法步骤主要包括:S1.获取船舶运动姿态的待处理三自由度数据,对数据进行预处理,主要包括以下部分:整理获取数据,截取数据大小,规范化处理;S2.利用哈里斯鹰优化算法(Harris Hawks Optimizer,HHO)和正余弦优化算法(Sine Cosine Algorithm,SCA)的优点,提出一种最好的自动寻优方法,其中主要包括以下部分:设计交叉并行的优化方法框架,初始化优化算法的参数;S3.采用SCA‑HHO优化算法对GRU神经网络的参数进行自动寻优,充分开发神经网络的处理能力,主要包括以下部分:初始化GRU网络结构,通过HHO‑SCA优化算法框架搜索最优参数;S4.将得到的最佳适应度值对应的最佳参数赋值给GRU网络,进行训练后预测船舶三个自由度数据。利用本文提出的SCA‑HHO‑GRU模型框架能够解决对船舶姿态预测时精度不高,误差较大的问题。采用本文提出的模型能够精准的预测船舶的运动姿态,保障舰载机的正常起降和船上工作人员的安全等。
Ship attitude prediction method based on SCA-HHO-GRU model
一种基于SCA-HHO-GRU模型的船舶姿态预测方法
2023-09-22
Patent
Elektronische Ressource
Chinesisch
IPC: | B63B Schiffe oder sonstige Wasserfahrzeuge , SHIPS OR OTHER WATERBORNE VESSELS |
Ship motion attitude prediction method based on multiple combinations
Europäisches Patentamt | 2023
|Ship anti-overturning self-righting control method based on attitude fitting prediction
Europäisches Patentamt | 2023
|Online Contents | 1994