摘要
为了进一步提升综采面瓦斯浓度预测的准确率,设计了一种基于GA-LSTM的瓦斯浓度预测模型。该模型利用GA算法来优化LSTM网络的参数,以解决传统LSTM网络预测不平衡以及易陷入局部极值的问题,确保网络的预测性能。通过实际测量的方式构建瓦斯浓度数据集,并在keras+Tensorflow平台上进行训练。结果表明:相较于传统的RNN、BP瓦斯浓度预测模型,GA-LSTM模型寻优速度较快,预测均方误差最小,预测值与真实值更为接近,预测准确率与精度更高,满足实际需求。
To further improve the accuracy of gas concentration prediction at the comprehensive mining face,a GA-LSTM-based gas concentration prediction model is designed.The model uses the GA algorithm to optimize the parameters of the LSTM network to solve the problems of unbalanced prediction and the tendency to fall into local extremes of the traditional LSTM network and to ensure the prediction performance of the network.The gas concentration dataset is constructed by actual measurements and trained on the keras+Tensorflow platform.The results show that compared with the traditional RNN and BP gas concentration prediction models,the GA-LSTM model is faster in finding the best prediction,has the most minor mean square error,the prediction value is closer to the actual value,and the prediction accuracy and precision are higher to meet the practical needs.
作者
王德忠
朱国宏
王禹
王神虎
WANG Dezhong;ZHU Guohong;WANG Yu;WANG Shenhu(Shanxi Engineering Vocational College,Taiyuan 030031,China)
出处
《煤炭技术》
CAS
北大核心
2023年第1期219-221,共3页
Coal Technology
基金
山西省教育科学“十四五”规划课题(PJ-21057)。
