Nanowire-Based Sensor Array for Detection of Cross-Sensitive Gases Using PCA and Machine Learning Algorithms

Self Cite

In the last two decades, GaN nanostructures of various forms like nanowires (NWs), nanotubes (NTs), nanofibers (NFs), nanoparticles (NPs) and nanonetworks (NNs) have been reported for gas sensing applications. In this paper, we have reviewed our group’s work and the works published by other groups on the advances in GaN nanostructures-based sensors for detection of gases such as hydrogen (H2), alcohols (R-OH), methane (CH4), benzene and its derivatives, nitric oxide (NO), nitrogen dioxide (NO2), sulfur-dioxide (SO2), ammonia (NH3), hydrogen sulfide (H2S) and carbon dioxide (CO2). The important sensing performance parameters like limit of detection, response/recovery time and operating temperature for different type of sensors have been summarized and tabulated to provide a thorough performance comparison. A novel metric, the product of response time and limit of detection, has been established, to quantify and compare the overall sensing performance of GaN nanostructure-based devices reported so far. According to this metric, it was found that the InGaN/GaN NW-based sensor exhibits superior overall sensing performance for H2 gas sensing, whereas the GaN/(TiO2–Pt) nanowire-nanoclusters (NWNCs)-based sensor is better for ethanol sensing. The GaN/TiO2 NWNC-based sensor is also well suited for TNT sensing. This paper has also reviewed density-functional theory (DFT)-based first principle studies on the interaction between gas molecules and GaN. The implementation of machine learning algorithms on GaN nanostructured sensors and sensor array has been analyzed as well. Finally, gas sensing mechanism on GaN nanostructure-based sensors at room temperature has been discussed.

Section: Machine Learning Algorithms On Gan Nanostructured Sensorsmentioning

confidence: 99%

“…( B ) Score plot from principal component analysis (PCA) analysis for various concentrations of NO 2 , ethanol, SO 2 , H 2 , O 2 and H 2 O, which includes up to 95.1% of the total variance. Figures adapted with permission from [ 71 ], Copyright 2020 IEEE.…”

Section: Figurementioning

confidence: 99%

Gallium Nitride (GaN) Nanostructures and Their Gas Sensing Properties: A Review

Khan

Rao

2020

Self Cite

“…The rapid advancement of computational intelligence technology and increasing computing power provide vital supports to achieve efficient and reliable analyte identification and mixture recognition using sensor arrays. A number of parametric or non-parametric machine learning and other optimization techniques have been adopted to classify types of analytes/environments or quantitatively determine concentrations of analytes in a mixture, such as decision tree (DT), support vector machine (SVM), naive bayes (kernel), K -nearest neighbor (KNN), random forest (RF), and artificial neural network (ANN) [ 10 , 19 , 20 , 21 , 22 , 23 ]. While many studies have demonstrated certain level of success for offline analysis, it remains challenging to deploy the chemical sensor arrays for real-time applications.…”

Section: Introductionmentioning

confidence: 99%

Early-Stage Gas Identification Using Convolutional Long Short-Term Neural Network with Sensor Array Time Series Data

Zhou

Liu

2021

Gas identification/classification through pattern recognition techniques based on gas sensor arrays often requires the equilibrium responses or the full traces of time-series data of the sensor array. Leveraging upon the diverse gas sensing kinetics behaviors measured via the sensor array, a computational intelligence- based meta-model is proposed to automatically conduct the feature extraction and subsequent gas identification using time-series data during the transitional phase before reaching equilibrium. The time-series data contains implicit temporal dependency/correlation that is worth being characterized to enhance the gas identification performance and reliability. In this context, a tailored approach so-called convolutional long short-term memory (CLSTM) neural network is developed to perform the identification task incorporating temporal characteristics within time-series data. This novel approach shows the enhanced accuracy and robustness as compared to the baseline models, i.e., multilayer perceptron (MLP) and support vector machine (SVM) through the comprehensive statistical examination. Specifically, the classification accuracy of CLSTM reaches as high as 96%, regardless of the operating condition specified. More importantly, the excellent gas identification performance of CLSTM at early stages of gas exposure indicates its practical significance in future real-time applications. The promise of the proposed method has been clearly illustrated through both the internal and external validations in the systematic case investigation.

“…Multivariate methods such as PCA, PCR, and machine learning have been employed for this purpose [ 43 , 44 , 45 , 46 , 47 ]. The combined use of electronic noses that employ arrays of sensors with the aforementioned methods has been applied to discriminate and quantify gases (e.g., NO 2 , ammonia, ethanol, acetone) [ 43 , 47 , 48 , 49 , 50 , 51 , 52 , 53 ]. Most of these works implement the mentioned data analysis by using sensor response feature vectors as input for the multivariate and machine learning approaches.…”

Section: Introductionmentioning

confidence: 99%

On the Use of Pulsed UV or Visible Light Activated Gas Sensing of Reducing and Oxidising Species with WO3 and WS2 Nanomaterials

González

Casanova-Cháfer

Alagh

et al. 2021

This paper presents a methodology to quantify oxidizing and reducing gases using n-type and p-type chemiresistive sensors, respectively. Low temperature sensor heating with pulsed UV or visible light modulation is used together with the application of the fast Fourier transform (FFT) to extract sensor response features. These features are further processed via principal component analysis (PCA) and principal component regression (PCR) for achieving gas discrimination and building concentration prediction models with R2 values up to 98% and RMSE values as low as 5% for the total gas concentration range studied. UV and visible light were used to study the influence of the light wavelength in the prediction model performance. We demonstrate that n-type and p-type sensors need to be used together for achieving good quantification of oxidizing and reducing species, respectively, since the semiconductor type defines the prediction model's effectiveness towards an oxidizing or reducing gas. The presented method reduces considerably the total time needed to quantify the gas concentration compared with the results obtained in a previous work. The use of visible light LEDs for performing pulsed light modulation enhances system performance and considerably reduces cost in comparison to previously reported UV light-based approaches.