基于工业机理模型和人工智能算法的零件周期生产量化预测研究

Quantitative Prediction of Part Cycle Production Based on Industrial Mechanism Model and Artificial Intelligence Algorithm

为了通过精准的预测,提高零件周期生产能力,课题组研究了基于工业机理模型和人工智能算法的零件周期生产量化预测方法。分析零件周期生产能力影响因素的工业机理,将设备作业周期时间、资源单耗以及系统正常运行时间3个量化指标作为卷积神经网络的输入,零件周期生产能力作为卷积神经网络的输出,使用基于卷积神经网络的零件周期生产量化预测方法,完成零件周期生产能力的有效预测。实验结果表明:该方法的预测结果与真实结果基本一致,具有较高的零件周期生产量化预测精度;将卷积神经网络的学习速率设置为0.005 Mibit/s,卷积核大小设置为5×5时,可提高零件周期生产量化预测效果。

In order to improve the capacity of parts cycle production through accurate prediction,a quantitative prediction method of parts cycle production based on industrial mechanism model and artificial intelligence algorithm was studied.The industrial mechanism of the influencing factors of the periodic production capacity of parts was analyzed.Three quantitative indexes of the equipment operation cycle time,the unit consumption of resources and the normal operation time of the system were taken as the input of the convolution neural network.The periodic production capacity of the parts was taken as the output of the convolution neural network.The quantitative prediction method of part cycle production based on convolution neural network was used to complete the effective prediction of part cycle production capacity.The experimental results show that the prediction results of the method are basically consistent with the real results,and the method has high precision of quantitative prediction of part cycle production;the learning rate of convolution neural network is set to 0.005 Mibit/s,and the convolution kernel size is set to 5×5,it can improve the quantitative prediction effect of part cycle production.