论文部分内容阅读
对于正在使用的工业厂房的安全控制,日常维护和监控起着重要的作用。结合既有的监测数据,利用数值分析软件MATLAB和BP神经网络基本原理建立实时分析模型,即以现有监测数据为训练样本,对神经网络进行训练,利用训练后的网络进行预测。以某钢厂500t废酸水处理车间监测数据为例,利用BP神经网络进行预测,预测结果和实际位移数据相吻合,结果表明了神经网络对监测数据进行处理和预测的可行性,对监测工作起到指导和预警作用。
For the safety control of the industrial plant in use, routine maintenance and monitoring play an important role. Combined with the existing monitoring data, the real-time analysis model is established by using the basic principles of numerical analysis software MATLAB and BP neural network. That is, the existing monitoring data is used as a training sample to train the neural network and the network after the training is used to predict. Take the 500t waste acid water treatment plant monitoring data of a steel mill as an example, the prediction is made by using BP neural network. The prediction results are in good agreement with the actual displacement data. The results show that the neural network can process and predict the monitoring data. Play a guiding and early warning role.