基于径向基神经网络杯形件拉深成形变压边力预测技术研究

分析了RBF神经网络的预测策略和方法,并建立了板料拉深成形的变压边力预测神经网络模型.采用正交设计法进行样本参数的制定,利用板材成形CAE软件Dynaform获得训练数据,利用被训练好的神经网络对薄板成形过程中变压边力的预测技术进行了研究.数值模拟结果表明,此方法对拉深成形变压边力的预测是可行的.

A study on the prediction technology of variable blank-holding force for deep drawing forming of cup shaped parts based on radial basis neural network

The prediction strategy and method of radial basis force(RBF) neural network were analyzed,and the neural network model for the prediction of variable blank-holding force in the deep drawing forming of sheet materials was established.The constitution of sample book parameters was carried out by adopting the orthogonal designing method and by the use of sheet material forming CAE software Dynaform to obtain the training data.The study on the prediction technology of variable blank holding force in the forming process of sheet metal was carried out by utilizing the being trained neural network.The result of numerical simulation showed that this method is feasible for the prediction of variable blank holding force of deep drawing formation.