A new LeNet-5 gas identification convolutional neural network structure for electronic noses is proposed and developed in this paper. Inspired by the tremendous achievements made by convolutional neural networks in the field of computer vision, the LeNet-5 was adopted and improved for a 12-sensor array based electronic nose system. Response data of the electronic nose to different concentrations of CO, CH4 and their mixtures were acquired by an automated gas distribution and test system. By adjusting the parameters of the CNN structure, the gas LeNet-5 was improved to recognize the three categories of CO, CH4 and their mixtures omitting the concentration influences. The final gas identification accuracy rate reached 98.67% with the unused data as test set by the improved gas LeNet-5. Comparison with results of Multiple Layer Perceptron neural networks and Probabilistic Neural Network verifies the improvement of recognition rate while with the same level of time cost, which proved the effectiveness of the proposed approach.
Titanium
carbide (Ti3C2T
x
) with a distinctive structure, abundant surface chemical groups,
and good electrical conductivity has shown great potential in fabricating
superior gas sensors, but several challenges, such as low response
kinetics, poor reversibility, and serious baseline drift, still remain.
In this work, γ-poly(l-glutamic acid) (γ-PGA)
with a blocking effect is exploited to modify Ti3C2T
x
, thereby stimulating the positive
response behavior of Ti3C2T
x
and improving its gas sensing performance. On account of the
unique synergetic interaction between Ti3C2T
x
and γ-PGA, the response of the flexible
Ti3C2T
x
/γ-PGA
gas sensor to 50 ppm NO2has been improved to a large extent
(average 1127.3%), which is 85 times that of Ti3C2T
x
(only 13.2%). Moreover, the as-fabricated
Ti3C2T
x
/γ-PGA
sensor not only exhibits a shorter response/recovery time (average
43.4/3 s) compared with the Ti3C2T
x
-based sensor (∼18.5/18.3 min) but also shows
good reversibility and repeatability (relative standard deviation
(RSD) <1%) at room temperature within 50% relative humidity (RH).
The improved gas sensing properties of the Ti3C2T
x
/γ-PGA sensor can be attributed
to the enhancement of effective adsorption and the blocking effect
assisted by water molecules. Furthermore, the gas sensing response
of the Ti3C2T
x
/γ-PGA
sensor is studied at different RHs, and humidity compensation of the
sensor is carried out using the multiple regression method. This work
demonstrates a novel strategy to enhance the gas sensing properties
of Ti3C2T
x
by γ-PGA
modification and provides a new way to realize highly responsive gas
detection at room temperature.
Low-concentration formaldehyde (HCHO) together with ethanol/toluene/acetone/α-pinene (as an interference gas of HCHO) is detected with a micro gas sensor array, composed of eight tin oxide (SnO2) thin film gas sensors with Au, Cu, Pt or Pd metal catalysts. The characteristics of the multi-dimensional signals from the eight sensors are evaluated. A multilayer neural network with an error backpropagation (BP) learning algorithm, plus the principal component analysis (PCA) technique, is implemented to recognize these indoor volatile organic compounds (VOC). The results show that the micro gas sensor array, plus the multilayer neural network, is very effective in recognizing 0.06 ppm HCHO in single gas component and in binary gas mixtures, toluene/ethanol/α-pinene with small relative error.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.