基于人工智能的预测模型用于硫回收装置中的焚烧炉,以预测SO2排放量
Muhammed Thameem1, Abhijeet Raj2, Abdallah Berrouk3
1Department of Chemical Engineering, Khalifa University of Science and Technology, P.O. Box 127788, Abu Dhabi, United Arab Emirates; Center for Catalysis and Separations, Khalifa University of Science and Technology, P.O. Box 127788, Abu Dhabi, United Arab Emirates.
Environmental research
|February 7, 2024
概括
准确预测化学工厂的二氧化硫 (SO2) 排放对于环境安全至关重要. 一个CNN-LSTM编码器-解码器模型在提前数小时预测SO2排放方面表现出卓越的性能.
科学领域:
- 环境科学 环境科学
- 化学工程是化学工程的重要组成部分.
- 人工智能的人工智能
背景情况:
- 化学工厂的污染物排放带来了重大的环境风险.
- 优化流程和环境性能需要可靠的排放预测.
- 硫回收单元 (SRU) 涉及复杂的反应,使传统建模难以实现.
研究的目的:
- 开发和评估基于人工智能的模型,用于预测SRU的SO2排放.
- 优化输入特征和超参数,以提高模型性能.
- 为了比较各种机器学习模型在排放预测方面的有效性.
主要方法:
- 测试了标准的机器学习 (ML) 算法,多层感知器 (MLP),长期短期记忆 (LSTM),一维卷积 (1D-CNN) 和CNN-LSTM模型.
- 优化模型输入特征和超参数,以获得最大的预测准确度.
- 使用平均平方误差 (MSE) 和平均绝对百分比误差 (MAPE) 评估模型性能,用于1,3和5小时前的预测.
主要成果:
- 与其他测试模型相比,CNN-LSTM编码器解码器模型表现出卓越的性能.
- 随着更长的预测时间 (长达5小时),模型的准确性得到了改善.
- 对于5小时的预测,CNN-LSTM模型比1D-CNN,Deep LSTM和单层LSTM模型取得了显著的MAPE优势.
结论:
- 基于人工智能的模型,特别是CNN-LSTM编码解码器,对于SRU中的SO2排放预测是有效的.
- 拟议的模型为优化化学过程和提高环境安全提供了可靠的工具.
- 先进的深度学习架构为复杂的工业排放提供了卓越的预测能力.
相关概念视频
Steps in Outbreak Investigation
130
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
130
Atomic Emission Spectroscopy: Interference
188
In atomic emission spectroscopy (AES), high-temperature atomizers excite a broad range of elements and molecules that generate complex emissions from sources such as oxides, hydroxides, and flame combustion products in the flame or plasma. Several strategies can be employed to minimize spectral interferences caused by overlapping emission lines or bands. These include increasing instrument resolution, choosing alternative emission lines, optimally placing the detector in low-background regions,...
188


