论文部分内容阅读
由于真空玻璃需求受到多重因素的影响,传统的数据模型无法准确找寻订单变化规律,导致预测精度较低。为了提高预测精度,结合灰色神经网络模型,利用思维进化算法优化灰色神经网络,确定灰色神经网络的最优初始参数。分别利用灰色神经网络(GNN)模型和思维进化-灰色神经网络(MEC-GNN)模型,用训练好的网络预测某真空玻璃制造商订单,预测结果表明:改进的MEC-GNN模型明显地提高了预测结果的精度。
As the demand for vacuum glass is affected by multiple factors, the traditional data model can not accurately find the order variation and lead to lower prediction accuracy. In order to improve the prediction accuracy, combined with gray neural network model, the use of thinking evolutionary algorithm to optimize gray neural network to determine the optimal initial parameters of gray neural network. The orders of a manufacturer of vacuum glass are predicted with the trained network by using the gray neural network (GNN) model and the evolutionary evolution-gray neural network (MEC-GNN) model respectively. The results show that the improved MEC-GNN model obviously improves Predict the accuracy of the result.