基于半边折叠代价算法的数字孪生模型构建

Construction of Digital Twin Model Based on Half Fold Cost Algorithm

为解决数字孪生模型于生产过程中出现的虚实数据交互效率低,虚拟调试效果差等现象,基于数字孪生模型构建的基本准则,首先,对模型最小单元进行建模,再根据各个单元相关约束进行组装;其次,模型构造成功后进行网格化处理,采取半边折叠算法,对复杂的数字孪生模型进行轻量化处理;最后,通过BP神经网络解决运动向量迁移问题,对模型在生产过程中的动作进行刻画。实验结果显示,构建的数字孪生模型精确度提高,轻量化处理后,有效减少了孪生模型的数据量,减少规模达到30%,充分保证在真实生产过程中,数字孪生模型的优异运行。

At present,domestic enterprises all hope to transform to digital with the help of digital twin technology,but they generally pay less attention to the model,resulting in slow data interaction and poor virtual debugging effect in real operation.Based on the basic principle of digital twin model,the minimum unit of the model is modeled,and then assembled according to the constraints;After the successful construction of the model,half-fold algorithm is used to simplify the twin model.Finally,the problem of motion vector migration is solved by BP neural network,and the actions in the whole production process of the model are described.The experimental results show,After simplification,the memory occupation of the model is reduced by 30%,which effectively reduces the amount of data of the twin model and fully ensures the excellence of the twin model in the real production process.