Method for detecting soil total nitrogen contents based on pyrolysis and artificial olfaction

Soil nitrogen is an essential nutrient element for crop growth and development, important indicator of soil fertility characteristics. This study proposed a method based on pyrolysis artificial olfaction to quickly accurately determine the total (STN) content. A muffle furnace was used pyrolyze samples, ten different types oxide semiconductor gas sensors were construct sensor array detect samples’ gas. The response curves tested at temperatures 200°C, 300°C, 400°C, 500°C times 1 min, 3 5 10 min obtain optimal state samples. temperature time min. area, maximum value, average differential coefficient, variance gradient 8th-second transient value curve extracted olfactory feature space 121×10×7 (121 numbers, seven eigenvalues). Back-propagation neural network algorithm (BPNN), partial least squares regression (PLSR), combined with back-propagation (PLSR-BPNN) establish prediction model STN Moreover, coefficient determination (R2), root mean square error (RMSE), ratio performance deviation (RPD) as indicators results. test results showed that R2 PLSR, BPNN, PLSR-BPNN models 0.89033, 0.81185, 0.92186, RMSE values 0.24297, 0.37370, 0.21781, RPD 2.9964, 1.9482, 3.3426, respectively. established by has highest smallest RMSE, can achieve accurate content, therefore rated “excellent”. detection in this achieves low-cost, rapid, provides new reference measurement STN. Keywords: pyrolysis, system, nitrogen, array, methods DOI: 10.25165/j.ijabe.20221503.7086 Citation: Li M W, Zhu Q H, Liu Xia X M, Huang D Y. Method detecting contents olfaction. Int J Agric & Biol Eng, 2022; 15(3): 167–176.