Table 1.
Review of environmental sources of air pollution malodour evaluated using e-nose.
| Source of Malodours | Type of Sensor; Sampling; Measurement | Analysis Method | Description | Reference |
|---|---|---|---|---|
| landfill gas odors | 16 tin oxide sensors; 20 L/min; meas. 90 s | ANN (35 neurons in hid. layer) | high correlation of particular sensors with odour concentration, low prediction network error (MSE) for MLP 0.000410 and RBF 0.000755 in the range of 0 to 200 ouE/m3 | [18] |
| landfill site | 3 × EOS835; 6 thin film MOS; 180 mL/min; 3 min meas./12 min recovery [19] | PCA | quantification of time percentage when the presence of odours is perceived at the landfill boundaries and near vicinity, comparison of e-nose data with meteorological measurements | [20] |
| waste disposal, landfill areas | 6 to 8 tin oxide sensors | multilinear regression | correlation (R2 = 0.88) response of TGS822 sensor with odour concentration in the range of 0 to 1500 ouE/m3 | [21] |
| waste incineration plant | 6QMB; 400 mL/min absorption, 40 mL/min desorption; | PCA | correlation of particular sensor response with odour concentration in the range of 0 to 500 ouE/m3, detection of charcoal filter conditions | [22] |
| composting plant | 6 MOS; 200 mL/min | PCA | Correlation of e-nose response with odour concentration in the range of 0 to 1500 ouE/m3, classification of air contamination from compost hall | [23] |
| composting plant | EOS835 25; 6MOS; 3 min meas./12 min recovery | PCA | 96.4% classification accuracy of quantitative recognition test of odour source, high correlation (R = 0.95) e-nose response with odour concentration in the range of 20 to 50 ouE/m3 | [24] |
| composting plant | EOS3; 6MOS; EOS9; 6MOS; 3 min meas./9 min recovery | PCA, Fourier | 72% classification accuracy of quantitative recognition test of odour source, high correlation (R = 0.89) e-nose response with odour concentration in the range of 0 to 100 ouE/m3, response of e-nose compliant with human perception of nuisance | [25] |
| poultry farm | 12 sensors (MOS, hybrid, tin dioxide, tungsten oxide) | ANN | accurate odour strength prediction with e-nose (R = 0.93) | [26] |
| poultry shed | 24 MOS, 500 mL/min | PCA, PLS | high correlation of e-nose with odour concentration (R = 0.94) in the range of 0 to 4000 ouE/m3 | [27] |
| piggery building | Aromascan A32S; 32CP | PCA, PLS, ANN (Matlab) | correlation of e-nose with odour concentration PCA (R2 = 0.44), PLS R2 = 0.79, ANN R2 = 0.62 with, estimation of biofilter efficiency | [28] |
| swine manure storage | Aromascan A32S; 32 CP | ANN 12 hid. neur. (NeuroShell2) | low correlation (R2 = 0.46) e-nose with odour concentration in the range of 0 to 5000 ouE/m3 | [29] |
| piggery effluent ponds | Aromascan A32S; 32 CP | PCA, ANN (20 hid. neur., Matlab) | high correlation of e-nose with odour concentration (R = 0.98) in the range of 0 to 90 ouE/m3 | [30] |
| cattle slurry | Odourmapper; 20 CP (polyindol); 100 mL/min; Aromascan A32S; 32 CP (polypyrrole); 100 mL/min | PCA (Genstat) | correlation of average e-nose response with odour concentration in the range of 0 to 1000 ouE/m3 | [31] |
| livestock waste | 20× CP polypyrrolee | PCA | discrimination of different odours | [32] |
| rendering plant | 6QMB + 6MOS + O2 + H2O + CO + CO2; 3 s sample pulse, 60 s post-sampling measurement | PCA, PLS | sufficient correlation of e-nose with odour concentration in the range of 1000 to 30000 ouE/m3; biofilter controls | [33] |
| buildings material | KAMINA; 38 oxide gas sensor on chip | LDA, PLS | discrimination of different materials, correlation of e-nose response with perceived smell intensity in the range of 0 to 16 π (π is comparative unit determined on basis acetone vapours) | [34] |
PCA—principal component analysis, CP—conducting polymer, MOS—metal oxide semiconductors, PLS—partial least squares, ANN—artificial neural network, QMB—quartz crystal microbalance, TON—threshold odour number, CCA—canonical correlation analysis, DFA—discriminant function analysis, LDA—linear discriminant analysis, ouE/m3—European odour unit per cubic meter.