中图分类号：TP212；TP183 文献标志码：A DOI: 10.16157/j.issn.0258-7998.233821
中文引用格式：朱梓涵，陶洋，梁志芳. 基于改进OS-ELM的电子鼻在线气体浓度检测[J]. 电子技术应用，2023，49(10)：71-75.
英文引用格式：Zhu Zihan，Tao Yang，Liang Zhifang. Online gas concentration detection of electronic nose based on improved OS-ELM[J]. Application of Electronic Technique，2023，49(10)：71-75.

Online gas concentration detection of electronic nose based on improved OS-ELM

Abstract：Electronic nose is a bionic sensing system, which can identify many gases at the same time, so it is used in many fields. The gas concentration detection algorithm is the core part of the gas quantitative analysis by electronic nose. In order to improve the accuracy of the electronic nose concentration detection algorithm, a prediction model based on online sequential-extreme learning machine (OS-ELM) is proposed. The model uses one-dimensional convolutional neural network (1DCNN) to extract features, uses OS-ELM to predict gas concentration, and proposes an improved Particle Swarm Optimization(PSO) algorithm to overcome the problem that OS-ELM needs to manually adjust model parameters. The theoretical analysis shows that the improved algorithm has stronger search ability than the traditional PSO algorithm. Finally, the experimental results show that the proposed model has higher prediction accuracy and generalization ability compared with the traditional prediction model.

Key words :electronic nose；concentration detection；one-dimensional convolution neural network；online sequential-extreme learning machine；particle swarm optimization