基于神经网络的港口装卸矿石物料含水率预测研究

打开文本图片集

中图分类号：U653.7;X736.1 文献标志码：A 文章编号：1003-5168（2025）22-0016-07

DOI： 10.19968/j.cnki.hnkj.1003-5168.2025.22.003

Research on Predicting the Moisture Content of Port Loading andUnloading OreMaterialsBased on Neural Networks

ZHANG Zuming HE Dasi ZHAO Qiangqiang （Schoolof Smart Energy andEnvironment，Zhongyuan Universityof Technology，Zhengzhou 45ooo7，China）

Abstract： [Purposes] To address the current situation where port dust control primarily relies on water spray dust suppression technology，which lacks inteligence and automation and fails to establish a connection between spray water control and ore moisture content，this paper proposes the use of neural network models to achieve short-term prediction of moisture content，offering improvement ideas for spray control strategies.[Methods] Three prediction models-namely，the Random Forest model，the Particle Swarm Optimization algorithm model，and the Convolutional Neural Network model based on the decision tree mechanism-were proposed and established.Relevant data from the site and the laboratory were used to train and test these models，and the prediction data from the diffrent models were compared.[Findings] The prediction accuracy of the three models，from highest to lowest，was found to be： Convolutional Neural Network based ontheDecision TreeMechanism > Particle Swarm Optimization algorithm > Random Forest model. Among them，the Convolutional Neural Network based on the decision tree mechanism achieved the highest prediction accuracy，with the moisture content model reaching a prediction accuracy of 91.8% . [Conclusions] The results indicate that the convolutional neural network based on the decision tree mechanism can efectively predict the moisture content of ore for dust control， providing theoretical support for port dust pollution control. Keywords： port; ore material; moisture content; neural network prediction; dust controls

0 引言

随着经济全球化的快速发展，世界正在以一种全新的模式重塑货物贸易格局]。（剩余6392字）