基于Dropout_LSTM_LEC模型的采煤机高度趋势预测
DOI:
作者:
作者单位:

河南理工大学 计算机科学与技术学院

作者简介:

通讯作者:

中图分类号:

TP391.9

基金项目:

国家自然科学(61872126);河南省高等学校重点研究项目(16A520052)


High trend forecast of shearer based on Dropout_LSTM_LEC model
Author:
Affiliation:

College of Computer Science and Technology,Henan Polytechnic University,Jiaozuo

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    提高采煤机高度趋势预测精度对于实现采煤机自动化生产具有重要意义。针对目前传统采煤机记忆截割存在精度低的问题,考虑到截割轨迹数据具有非线性、局部突变等特征,提出一种基于Dropout优化算法和深度长短期记忆神经网络(Long Short-Term Memory, LSTM)耦合局部误差修正(Local Error Correct, LEC)的采煤机高度预模型(Dropout-LSTM-LEC)。该模型中,以LSTM为基础,建立多层级LSTM预测模型提升预测精度;同时使用Dropout优化算法完成模型的训练,缓解模型过拟合问题;最后结合局部突变修正方法,校正预测结果,以减少截割轨迹局部突变带来的预测误差。实际验证表明,改进模型相较于梯度提升回归树和支持向量回归算法在平均绝对误差、平均绝对百分误差、均方根误差方面均具有更好的表现。

    Abstract:

    To improve the prediction accuracy of shearer height trend is of great significance to realize automatic production of shearers. Due to the low precision of traditional miner memory truncation, and considering the non-linear and local mutation of the cutting track data, a miner height pre-model (Dropout-LSTM-LEC) based on Dropout optimization algorithm and deep long-short-term memory neural network (LSTM) coupled with local error correction (LEC) is presented. In this model, based on LSTM, a multilevel LSTM prediction model is established to improve the prediction accuracy. At the same time, Dropout optimization algorithm is used to complete the training of the model and alleviate the problem of model overfitting. Finally, with the local mutation correction method, the prediction results are corrected to reduce the prediction error caused by the local mutation of the truncation track. The actual validation shows that the improved model performs better in average absolute error, average absolute percentage error and root mean square error than the gradient lifting regression tree and the support vector regression algorithm

    参考文献
    相似文献
    引证文献
引用本文

安葳鹏,高 枫,邵一帆,等. 基于Dropout_LSTM_LEC模型的采煤机高度趋势预测[J]. 科学技术与工程, , ():

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-11-04
  • 最后修改日期:2022-05-16
  • 录用日期:2022-06-11
  • 在线发布日期:
  • 出版日期:
×
亟待确认版面费归属稿件,敬请作者关注