基于深度学习的回转窑燃烧状态监测系统

Combustion State Monitoring System of Rotary Kiln Based on Deep Learning

有效地识别与预测回转窑内燃烧状态是工业生产中非常重要和具有挑战性的问题,然而,传统的基于图像处理的方法前期需要对数据进行大量的预处理,并且精度不高。针对这个问题,提出了一种基于深度学习的回转窑燃烧状态监测系统,通过端到端网络,免去传统的基于特征提取方法繁杂的程序。另外,提出的卷积循环神经网络(CRNN)能有效地提取火焰图像序列特征来预测回转窑内的燃烧状态。在真实数据集上进行大量重复实验,结果表明,所提出的方法中,卷积神经网络(CNN)能快速准确地识别回转窑燃烧状态,同时CRNN也能有效地预测窑内燃烧状态。数据表明该方法不但有效且鲁棒性强,具有很大的工业应用前景。

Effectively identifying and predicting the combustion state in a rotary kiln is a very important and challenging problem in industrial production.However,traditional image-based methods require a large amount of data pre-processing in the early stages,and their accuracy is limited.To solve this problems,a combustion state monitoring system for rotary kiln based on deep learning is proposed.The end-to-end network eliminates the complicated procedures of traditional feature-based extraction methods.In addition,the proposed convolutional recurrent neural network(CRNN)can effectively extract the flame image sequence features to predict the combustion state in the rotary kiln.A large number of repeated experiments are performed on the real data set.The experimental results show that in the proposed method,the convolutional neural network(CNN)can quickly and accurately identify the combustion state in the rotary kiln.At the same time,CRNN can also effectively predict the combustion state in the kiln.The data show that the proposed method is effective and robust,and has great industrial application potential.