Development of Intelligent Electronic Nose for Livestock Industries

Abstract

This paper describes the development and application of our electronic nose (e-nose) machine and algorithms for monitoring of malodor emission. Malodor emission from livestock farms have been known as a significant source of odor nuisance resulting in increased complaints and environmental issues. E-nose technology has provided solutions to variety of these odorous problems with identification of the odor source and real-time monitoring of odor diffusion into the environment. This lab-developed electronic nose system consists of eight metal oxide semiconductor (MOX) gas sensors as sensing materials, including a humidity sensor and a temperature sensor. The gas sensors were specifically selected for the main volatile compounds emitted from the livestock farms. Principal Component Analysis (PCA) was used as classification algorithms combined with data analysis techniques, which enable the recognition of different odor samples. In this work, five commercial poultry farms were selected for odor monitoring using the in-lab developed e-nose machine. The field experiments were conducted at various sites in the farms such as chicken houses, manure storage, biogas facilities, downwind points and in nearby villages. Odor concentrations were mainly detected behind the poultry house fan (Max. 11 odor unit). In addition, the malodor sources have been investigated with the e-nose in each poultry farm. High concentrations of malodor were identified in the poultry farms with poor manure handling system and poor building management. In conclusion, the e-nose technology has the potential of providing monitoring capabilities for livestock and agricultural industries. The use of e-nose device is recommended to apply in the livestock farms to reduce neighbourhood complaints and risk on workers' health and safety.