检索结果-内蒙古大学图书馆

Chinese Automation Congress (CAC)

作者： Wang, Qingfeng Xie, Taiyang Wang, Shuaiqi Jilin Univ Coll Elect Sci & Engineer State Key Lab Integrated Optoelect Changchun Jilin Peoples R China Jilin Univ Coll Elect Sci & Engn Changchun Jilin Peoples R China

ISBN: (纸本)9781728113128

The integration of sensors array and pattern recognition to replace large scale analytical instruments is an important and feasible method for accurate and rapid measurement of atmospheric pollution gases. This paper proposed a quantitative detection method of mixed gases based on long short-term memory (LSTM) recurrent neural network, including data pre-processing, network structure design, model training and prediction process. This method can extract the deep characteristics of sensors array's responses automatically and match the complex nonlinear characteristics more accurately. It is proved that the proposed LSTM prediction model has a strong applicability and higher accuracy in the concentration identification of gas mixtures by comparing with the experiments of existing models.

关键词： gas sensors array pattern recognition air pollution gases deep learning recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

gas sensors array as a device to classify mold threat of the buildings 5th

Gas sensors array as a device to classify mold threat of the...

引用

5th National Congress of Environmental Engineering

作者： Suchorab, Z. Sobczuk, H. Guz, L. Lagod, G. Lublin Univ Technol Fac Environm Engn Lublin Poland

ISBN: (纸本)9781138031630

Mold threat is an important problem in building industry. It is one of the most important features that cause Sick Building Syndrome occurrence. In many circumstances this problem is caused by the increase of moisture in the building envelopes but also excessive indoor air humidity. Problem of mold threat is often connected with introduction of new technologies and materials applied for the external walls of the buildings and improper performance of natural ventilation. These factors create suitable conditions for growth of mold. E-nose technique is currently often applied to evaluate air quality in environmental measurements. This paper presents the possibility of classifying mold threat in indoor air of different rooms using low-selective Metal Oxide Semiconductors (MOS) gas sensor array being the element of an e-nose device. Conducted investigation confirmed that e-nose supported by statistical data processing enables to distinguish air from different rooms affected with mold at a satisfactory level.

关键词： e-nose MOS sensors gas sensors array mold threat buildings

来源：评论

学校读者我要写书评

暂无评论

A Movie Should Be Forever: Monitoring the Degradation Pathway of Photographic Films

A Movie Should Be Forever: Monitoring the Degradation Pathwa...

引用

IEEE International Symposium on Olfaction and Electronic Nose (ISOEN)

作者： Di Zazzo, Lorena Magna, Gabriele Capuano, Rosamaria Prati, Silvia Sciutto, Giorgia Lozano, Marco Chavez Mazzeo, Rocco Di Natale, Corrado Paolesse, Roberto Univ Roma Tor Vergata Dept Elect Engn Rome Italy Univ Roma Tor Vergata Dept Chem Sci Rome Italy Univ Bologna Dept Chem G Ciamicin Ravenna Italy

This contribution study aims at investigating the emission of volatile organic compounds due to degradation mechanism in photographic film. gas chromatography-mass spectrometry (GC-MS) and an Electronic Nose based on ... 详细信息

ISBN: (纸本)9781665458603

关键词： cellulose films gas sensors array GC-MS porphyrinoids

来源：评论

学校读者我要写书评

暂无评论

Application of Machine Learning Methods for an Analysis of E-Nose Multidimensional Signals in Wastewater Treatment

引用

sensors 2023年第1期23卷 487页

作者： Pilat-Rozek, Magdalena Lazuka, Ewa Majerek, Dariusz Szelag, Bartosz Duda-Saternus, Sylwia Lagod, Grzegorz Lublin Univ Technol Fac Technol Fundamentals PL-20618 Lublin Poland Kielce Univ Technol Fac Environm Geomat & Energy Engn PL-25314 Kielce Poland Inst Rural Hlth Lublin PL-20090 Lublin Poland Lublin Univ Technol Fac Environm Engn PL-20618 Lublin Poland

The work represents a successful attempt to combine a gas sensors array with instrumentation (hardware), and machine learning methods as the basis for creating numerical codes (software), together constituting an electronic nose, to correct the classification of the various stages of the wastewater treatment process. To evaluate the multidimensional measurement derived from the gas sensors array, dimensionality reduction was performed using the t-SNE method, which (unlike the commonly used PCA method) preserves the local structure of the data by minimizing the Kullback-Leibler divergence between the two distributions with respect to the location of points on the map. The k-median method was used to evaluate the discretization potential of the collected multidimensional data. It showed that observations from different stages of the wastewater treatment process have varying chemical fingerprints. In the final stage of data analysis, a supervised machine learning method, in the form of a random forest, was used to classify observations based on the measurements from the sensors array. The quality of the resulting model was assessed based on several measures commonly used in classification tasks. All the measures used confirmed that the classification model perfectly assigned classes to the observations from the test set, which also confirmed the absence of model overfitting.

关键词： machine learning multidimensional signals analysis t-SNE method k-median method random forest electronic nose gas sensors array wastewater treatment wastewater quality

来源：评论

学校读者我要写书评

暂无评论

Application of Machine Learning Algorithms to Classify Peruvian Pisco Varieties Using an Electronic Nose

引用

sensors 2023年第13期23卷 5864页

作者： De-La-Cruz, Celso Trevejo-Pinedo, Jorge Bravo, Fabiola Visurraga, Karina Pena-Echevarria, Joseph Pinedo, Angela Rojas, Freddy Sun-Kou, Maria R. Pontif Catholic Univ Peru Dept Engn Lima 15088 Peru Pontif Catholic Univ Peru Dept Sci Lima 15088 Peru

Pisco is an alcoholic beverage obtained from grape juice distillation. Considered the flagship drink of Peru, it is produced following strict and specific quality standards. In this work, sensing results for volatile compounds in pisco, obtained with an electronic nose, were analyzed through the application of machine learning algorithms for the differentiation of pisco varieties. This differentiation aids in verifying beverage quality, considering the parameters established in its Designation of Origin". For signal processing, neural networks, multiclass support vector machines and random forest machine learning algorithms were implemented in MATLAB. In addition, data augmentation was performed using a proposed procedure based on interpolation-extrapolation. All algorithms trained with augmented data showed an increase in performance and more reliable predictions compared to those trained with raw data. From the comparison of these results, it was found that the best performance was achieved with neural networks.

关键词： artificial neural network random forest support vector machine electronic nose gas sensors array beverage quality

来源：评论

学校读者我要写书评

暂无评论

Identification of thermal fault states in cable insulation sheaths based on gas sensor arrays

引用

IEEE Transactions on Dielectrics and Electrical Insulation 2025年

作者： Meng, Xiaobo Lin, Yu Li, Kailin Pang, Zhichao Cheng, Lefeng Zheng, Canfeng Guangzhou University School of Mechanical and Electrical Engineering Guangdong Province Guangzhou 510006 China

As one of the most commonly used electrical equipment materials, low and medium voltage cables inevitably suffer from overheating faults and cause fires during operation due to harsh working environments and overloaded operation. However, there is a lack of economical and reliable cable fire detection methods. In the early stage of fire, gas volatiles are often produced earlier than visible smoke. Therefore, this paper focuses on gas volatiles in early cable fires. Taking polyvinyl chloride (PVC), a commonly used outer sheath material for lowand medium-voltage cables, as the research object, gas chromatography-mass spectrometry was used to study its gaseous volatile products and their temperature characteristics when overheating, and it was found that di(2-ethylhexyl) phthalate (DEHP) and dibutyl phthalate (DBP) produced at low temperatures had positive temperature characteristics under the area-normalized quantification method. Based on these two main gas products and the broad-spectrum characteristics of the gas sensor, the gas sensor array was selected and constructed, and the gas response data of PVC cables under six temperature conditions were designed. The samples were classified into three states, namely, normal, early warning, and alarm, according to the sample state, and the Extra Trees algorithm was used to categorize the sample state, and the recognition accuracy rate reached 97%. During the whole experiment, the smoke alarm did not send out any alarm signal. © 1994-2012 IEEE.

关键词： cable pyrolysis fire warning gas sensors array

来源：评论

学校读者我要写书评

暂无评论

Design of an Odor Search Robot System Based on Open Sampling System 33

Design of an Odor Search Robot System Based on Open Sampling...

引用

33rd Chinese Control and Decision Conference (CCDC)

作者： Wang, Qingfeng Wang, Shuaiqi Ni, Huaixu Jilin Univ Coll Elect Sci & Engn Changchun 130012 Peoples R China

ISBN: (纸本)9781665440899

An odor source search robot system based on open sampling system is designed in this paper. First, a tracked vehicle equipped with five sensor arrays and cameras which positions can be adjusted arbitrarily is designed for intelligent and fast gas sensing and odor source localization. Then, the circuit system is designed based on MCU of Raspberry Pi. The gases information from the olfactory system is uploaded to the host computer via Wi-Fi and the video information is captured on the web page using OpenCV. If needed the operator can judge the direction of the odor source through the information displayed on the control panel and controls the robot to search for the odor source. Finally, the particle swarm optimization algorithm is proposed for the odor source location of multi-robots and the simulation is realized by Python and Fluent. The robot system proposed in this paper can detect the concentration of gases accurately and find the odor source quickly.

关键词： Odor search robot Intelligent perception gas sensors array Particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

Design of an Odor Search Robot System Based on Open Sampling System

Design of an Odor Search Robot System Based on Open Sampling...

引用

第33届中国控制与决策会议

作者： Qingfeng Wang Shuaiqi Wang Huaixu Ni College of Electronic Science and Engineering Jilin University

An odor source search robot system based on open sampling system is designed in this ***,a tracked vehicle equipped with five sensor arrays and cameras which positions can be adjusted arbitrarily is designed for intelligent and fast gas sensing and odor source ***,the circuit system is designed based on MCU of Raspberry *** gases information from the olfactory system is uploaded to the host computer via Wi-Fi and the video information is captured on the web page using *** needed the operator can judge the direction of the odor source through the information displayed on the control panel and controls the robot to search for the odor ***,the particle swarm optimization algorithm is proposed for the odor source location of multi-robots and the simulation is realized by Python and *** robot system proposed in this paper can detect the concentration of gases accurately and find the odor source quickly.

关键词： Odor search robot Intelligent perception gas sensors array Particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

Quantification of Multiple gases Mixture Based on Fuzzy Partition and Model Ensemble 38

Quantification of Multiple Gases Mixture Based on Fuzzy Part...

引用

38th Chinese Control Conference (CCC)

作者： Wang, Qingfeng Min, Jiaxing Cai, Mao Jilin Univ Coll Elect Sci & Engn Changchun 130012 Peoples R China

ISBN: (纸本)9789881563972

gases concentration estimation based on the integration of fuzzy partition and model ensemble with drift compensation for metal oxide gas sensors array is proposed in this paper. The sensors response data is divided into several sub-data sets newly according to the drift variation of sensors baseline over time using fuzzy clustering method, and then regression models of these sub-data sets are established based on weighted multi-output support vector regression. The optimal weights of sub-regression models are obtained by traversing search during training process. Based on least squares support vector regression, the optimal weight functions in model ensemble are fitted by training the clustering center and the optimal weights of sub-regression models. In the testing stage, the weights of ensemble models are calculated from the optimal weight function and clustering center. This method can adaptively change the recognition models so that it can cope with the change of sensors drift and ensure the long-term accuracy of concentration quantification.

关键词： gas sensors array Fuzzy clustering Support vector regression Model ensemble

来源：评论

学校读者我要写书评

暂无评论

Miniaturized gas Sensor Platform for Selective Vapors Detection at Room Temperature 34

Miniaturized Gas Sensor Platform for Selective Vapors Detect...

引用

34th Symposium on Microelectronics Technology and Devices (SBMicro)

作者： Peres, Henrique E. M. Galeazzo, Elisabete Dantas, Michel O. S. Sparvoli, Marina Santos, Francislei S. A. Falla, Maria P. H. Univ Sao Paulo Dept Engn Sistemas Eletron Escola Politecn Sao Paulo Brazil Univ Fed ABC Ctr Engn Modelagem & Ciencias Sociais Santo Andre SP Brazil Inst Fed Bahia IFBA Salvador BA Brazil Univ Brasilia UNB Gama Brasilia DF Brazil

ISBN: (纸本)9781728119090

gas sensors are usually made of metallic oxides and need to operate at relatively high temperatures (about hundreds of Celsius degrees). Besides, they generally present drawbacks: low selectivity, elevated response time, and degradation. Novel nanostructured materials such as carbon nanotubes and graphene, among others, have been proposed to overcome these problems due to their unique electronic properties. However, these materials also present low selectivity for many gas species. In order to take advantage of the benefits of using carbon materials for gas sensing at room temperature, this work presents a proposal of humidity and volatile organic compounds (VOCs) sensing based on a miniaturized platform composed of four sensors. The lack of selectivity of individual materials is compensated by crossing the responses of each sensor. Results showed very fast response and recovery times (just few seconds), as well as good sensitivity (response variation up to four orders of magnitude). With this approach, the selectivity to discriminate the vapors can be improved, since each substance has an individual response pattern in the gas sensor array.

关键词： gas sensors array carbon-based materials humidity sensor VOC sensor room temperature detection

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：