基于CNN-Bi-LSTM模型的煤含水率预测研究

Research on the prediction method of coal moisture content based on CNN-Bi-LSTM model

煤含水率监测对降低储运煤的碳排放指标意义重大,针对港口煤场堆垛含水率监测需求,提出了一种基于CNN-Bi-LSTM网络的煤炭含水率预测方法,该方法基于卷积神经网络和双向长短期记忆网络,利用其特征提取及时间序列特征记忆能力,通过采集黄骅港煤炭转运堆场的海量煤含水率数据和场区气象数据,对多源数据训练学习和融合分析,实现对港口煤炭含水率预测,并进行了有效性实验验证。实验结果表明,与传统算法模型相比,所提出的CNN-Bi-LSTM混合神经网络模型在预测精度、收敛率和鲁棒性方面表现最优,使用该预测方法建立的洒水管控模型可有效降低煤炭堆场用水量,减少煤炭堆场的起尘概率,环境状况也得到有效改善。

The monitoring of coal moisture content is of great significance for reducing the carbon emission index of coal storage and transportation.Aiming at the monitoring demand of the stacking moisture content of the port coal yard,a prediction method of coal moisture content based on the CNN-Bi-LSTM network is proposed.This method is based on the convolutional neural network(CNN)and the bidirectional long short-term memory(Bi-LSTM)network,uses their feature extraction and time series feature memory ability,conducts the training,learning and fusion analysis of multi-source data by collecting the massive coal moisture content data and the meteorological data from the coal transfer yard of Huanghua Port,realizing the prediction of coal moisture content in ports and effective experimental verification.The results show that compared with the traditional algorithm model,the proposed CNN-Bi-LSTM hybrid neural network model performs optimally in terms of prediction accuracy,convergence rate and robustness.After establishing a watering control model using this prediction method,the water consumption of the coal yard and the probability of dust generation in the coal yard can be effectively reduced,and the environmental conditions can also be improved.