基于平行图像与深度学习的选矿厂设备群升级辅助决策技术

Auxiliary decision-making technology for upgrading concentrator equipment group based on parallel image and deep learning

现有选矿厂设备升级辅助决策技术对神经网络的反向微调结果精度较差,导致选矿厂设备的升级优化效果较差。基于平行图像与深度学习算法,设计选矿厂设备群升级辅助决策技术。使用平行图像的方式采集与渲染选矿厂设备信息,标记语义精度样式,计算修正后图像损失函数。应用深度学习训练辅助决策模型参数获取随机单元激活函数,获取循环次数与误差的重构系数。建立选矿厂设备群升级辅助决策模型。实验结果表明,该技术能以最快的速度得到决策结果,且误差较小,选矿厂设备群升级过程的数据精度与完整性可得到保证。

The existing auxiliary decision-making technology for concentrator equipment upgrading has poor precision in the reverse fine-tuning result of neural network,resulting in poor upgrading and optimization effect of concentrator equipment.Therefore,the auxiliary decision-making technology for concentrator equipment group upgrading is designed based on parallel image and deep learning algorithm.The equipment information of the concentrator is collected and rendered by using images parallel,the semantic accuracy style is marked,and the corrected image loss function is calculated.The activation function of random unit is obtained by using the parameters of the deep learning training aided decision-model,and the reconstruction coefficient of cycle times and error is obtained.An auxiliary decision-making model for upgrading of concentrator equipment group is established.The experimental results show that this method can get the decision results at the fastest speed with less error,and the data accuracy and integrity of the upgrading process of concentrator equipment group can be guaranteed.