Indexing Odour by using electronic nose

Abstract

Odour produced from landfill had become a serious problem in Malaysia as 70% of the complaints received by environmental regulators were about the odour. At present, Malaysia is still developing its odour standard permitted level as for haze and water. Currently, the accepted standard to provide odour concentration is olfactometry which relies on human panels. This method is labour-intensive, expensive and the equipment is bulky, also human panels are fatigue and unable to produced consistence results. Many studies had shown the application of the QCM sensors for odour detection. Among those are detection of organic vapour in the environment detection of odour emission from composting facility, etc. The aim of this work is to see if e-Nose could be used to measure odour concentration. The nose system comprises of an array of mass sensitive sensors (quartz crystal microbalance, QCM) as well as data acquisition and a pattern recognition tool namely principal components analysis (PCA). The performance of the sensor was further investigated using supervised pattern recognition techniques such as discriminate function analysis (DFA). This later data processing technique gives better classification of morning and evening air quality.