首页|基于直接吸收光谱深度学习神经网络模型的CO2浓度检测研究

基于直接吸收光谱深度学习神经网络模型的CO2浓度检测研究

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
CO2是温室气体的主要成分之一，其对全球气候变化和环境质量有重大影响，燃煤电厂作为我国最大的CO2排放源，面临着严峻考验。为准确、快速、低成本地检测燃煤电厂CO2浓度，促进燃煤电厂低碳发展，本文利用分布反馈半导体激光器构建了一种高灵敏度的CO2气体检测系统，同时采用HITRAN数据库作为深度学习模型的数据集，建立了一维卷积神经网络(one-dimensional convolutional neural network，1D-CNN)模型，采用反向传播神经网络算法检测CO2浓度并与直接吸收技术进行了对比，并通过K折交叉验证法和调整模型参数来提升1D-CNN模型的性能。结果表明，1D-CNN模型的决定系数R2 值可达到0。999 7，相对误差为1。07%，绝对误差为7。88 mg/m3，模型建立较为符合要求;利用1D-CNN模型调用最优参数，对比预测数据和真实数据，平均相对误差为6。06%，平均绝对误差为17。97 mg/m3，决定系数R2=0。999 41，可得模型的预测结果具有较高的精度。这种基于直接吸收光谱深度学习神经网络模型的气体浓度检测模型在测量CO2气体浓度方面具有较高的准确性和可靠性，可为电力行业的环保监测和节能减排提供有力的技术支持。
Research on CO2 concentration detection based on deep learning neural net-work model of direct absorption spectroscopy
Carbon dioxide(CO2)is a principal component of greenhouse gases,significantly impacting global climate change and environmental quality.Coal-fired power plants,being the largest source of CO2 emissions in China,face se-vere challenges.Therefore,to accurately,rapidly,and cost-effectively monitor the CO2concentration in coal-fired power plants and promote their low-carbon development,this paper utilizes a distributed feedback semiconductor laser to con-struct a high-sensitivity CO2 gas detection system.It also validates and employs the HITRAN database as the dataset for a deep learning model,establishes a one-dimensional convolutional neural network(1D-CNN)model,and a back-propagation neural network for CO2 concentration detection.These models are compared with direct absorption spec-troscopy technique,and the performance of the 1D-CNN model is enhanced through K-fold cross-validation and parame-ter adjustment.The results show that the determination coefficient(R2)of the 1D-CNN model can reach 0.9997,with a relative error of 1.07%and an absolute error of 7.88 mg/m3,indicating the model's suitability.By utilizing the optimal pa-rameters of the 1D-CNN model,a comparison between predicted and actual data reveals an average relative error of 6.06%,an average absolute error of 17.97 mg/m3,and an R2 of 0.99941,demonstrating high accuracy in the model's predictions.This gas concentration detection model based on direct absorption spectroscopy and a deep learning neural network exhibits high accuracy and reliability in measuring CO2concentrations,offering robust technical support for envi-ronmental monitoring and energy conservation and emission reduction within the power industry.

coal fired power plantscarbon emissionsdeep learningtunable diode laser absorption spectroscopyar-bon dioxideHITRAN database

史文辉、李凯亮、宫廷、田亚莉、孙小聪、郭古青、邱选兵、李传亮

展开 >

山西省精密测量与在线检测装备工程研究中心,山西省光场调控与融合应用技术创新中心,太原科技大学应用科学学院,山西 太原 030024

燃煤电厂 碳排放 深度学习 吸收光谱 CO2 HITRAN数据库

国家重点研发计划国家自然科学基金山西省留学人员科技活动项目山西省省筹资金资助回国留学人员科研资助项目山西省基础研究计划太原科技大学科研启动基金

2023YFF071810052076145&12304403202300312023-151202203021222204&20230302121222420222121&20232033

2024

10.19944/j.eplep.1674-8069.2024.01.006

电力科技与环保
国电科学技术研究院

电力科技与环保

影响因子：0.653
ISSN：1674-8069
年,卷(期)：2024.40(1)
  • 38