Scholarlink

After the Industrial Revolution, it was observed that the output rates of factories increased exponentially, and prices for goods have decreased. But this has also resulted in increase in pollution caused by humans, harming the environment and posing health risks. It is essential to forecast and control air pollution levels. This study article offers a comprehensive investigation of the PM2.5 level forecast using the Air Quality Index (AQI) for Delhi, India. Where Long Short-Term Memory (LSTM) models are used in the study along with two hybrid decomposition-based prediction models: Empirical Mode Decomposition (EMD) and Ensemble Empirical Mode Decomposition (EEMD) so as to reduce the noise component from the signal caused due to inaccuracies in the acquisition of information. To evaluate these models’ respective performances, a dataset covering the period from November 25, 2020, to January 24, 2023, is used. According to the results, the EEMD-LSTM model outperforms the other two model EMD-LSTM & LSTM models.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Decomposition Methods Based Holistic Approach with LSTM Network on Pollution Prediction

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Gaurav Pandey,
  • Rajneesh Sharma,
  • Shiv Sagar Singh,
  • Priyanshu Jha,
  • Kushagra Kumar

摘要

After the Industrial Revolution, it was observed that the output rates of factories increased exponentially, and prices for goods have decreased. But this has also resulted in increase in pollution caused by humans, harming the environment and posing health risks. It is essential to forecast and control air pollution levels. This study article offers a comprehensive investigation of the PM2.5 level forecast using the Air Quality Index (AQI) for Delhi, India. Where Long Short-Term Memory (LSTM) models are used in the study along with two hybrid decomposition-based prediction models: Empirical Mode Decomposition (EMD) and Ensemble Empirical Mode Decomposition (EEMD) so as to reduce the noise component from the signal caused due to inaccuracies in the acquisition of information. To evaluate these models’ respective performances, a dataset covering the period from November 25, 2020, to January 24, 2023, is used. According to the results, the EEMD-LSTM model outperforms the other two model EMD-LSTM & LSTM models.