Application of D-ELM in quantitative analysis of FTIR spectrum of mine gas
-
摘要: 为了对D-ELM的矿井气体定量分析效果进行比较和分析，建立了3种矿井气体定量分析模型，分别是SVM模型、极限学习机（ELM）模型和动态极限学习机（D-ELM）模型。由于每个模型每次预测结果在一定范围内变化，所以取每个模型运行10次预测结果平均均方根误差（ARMSE）、平均相关系数（AR）、平均模型运行时间（AT）来评价各模型对气体定量分析的性能。PSO-SVM模型的结果为：[0.054 2，0.998，200.38 ]；ELM模型的结果为:[1.042 1，0.989 4，0.26]；DELM模型结果为：[0.043 8，1，2.01]。综合预测精度和预测速度表明，DELM模型要优于另外2种模型。Abstract: Three quantitative analysis models for mine gas were established, the support vector machine (SVM) model, the extreme learning machine (ELM) model and the dynamic extreme learning machine (D-ELM) model, to analysis and compare with the result of D-ELM. Since the results of each model were changed in a certain range, taking the average root mean square error (ARMSE), the average correlation coefficient (AR) and the average model running time (AT), all of them were the average values of 10 results of each model, as the standards to evaluate the performance of the model in the quantitative analysis of the mine gas. The results of particle swarm optimization(PSO)-SVM model, ELM model and D-ELM model were as follow: [0.054 2,0.998,200.38], [1.042 1,0.989 4,0.26], [0.043 8,1,2.01]. Considering the predicted accuracy and predicted speed, D-ELM is better than the other two models.
-
Keywords:
- D-ELM /
- quantitative analysis /
- infrared spectrum /
- PSO-SVM /
- mine gas
-
-
期刊类型引用(8)
1. 周会娟，余尚江，陈晋央，陈显，孟晓洁. 一种双面感压式光纤土压力传感器. 兵工学报. 2023(S1): 132-137 . 
百度学术
2. 吕欢祝，钟文博，秦亮，张克非. 聚合物封装的光纤光栅压力传感器的特性研究. 激光杂志. 2020(08): 63-67 . 百度学术
3. 杨洋，赵勇，吕日清，刘兵，郑洪坤，杨洋，王孟军，崔盟军，杨华丽. 多参量一体化光纤传感器及标校系统的研制与开发. 红外与激光工程. 2019(10): 185-191 . 百度学术
4. 吴国军，何少灵，桑卫兵. 温度精度补偿的光纤光栅土压力传感器. 机电工程技术. 2018(06): 47-49 . 百度学术
5. 郭红英，王召巴. 基于光纤光栅的高压固体压力传感器研究. 分析化学. 2017(07): 980-986 . 百度学术
6. 袁斌. 光纤Bragg光栅传感器的通信系统设计与实现. 激光杂志. 2017(12): 67-70 . 百度学术
7. 李凯，赵振刚，李英娜，蔡陈，彭庆军，李川. FBG可变灵敏度压力传感器设计. 传感器与微系统. 2016(06): 69-71 . 百度学术
8. 孙搏，隋青美，王静，曹帅帅，王宁，李海燕，刘斌. 微型布拉格光栅土压力传感器的设计与试验. 仪表技术与传感器. 2016(10): 20-23+27 . 百度学术
其他类型引用(2)
计量
- 文章访问数: 557
- HTML全文浏览量: 36
- PDF下载量: 41
- 被引次数: 10
下载:
陕公网安备 61011302001501号 