Performance Analysis of MAU-9 Electronic-Nose MOS Sensor Array Components and ANN Classification Methods for Discrimination of Herb and Fruit Essential Oils

Abstract

The recent development of MAU-9 electronic sensory methods, based on artificial olfaction detection of volatile emissions using an experimental metal oxide semiconductor (MOS)-type electronic-nose (e-nose) device, have provided novel means for the effective discovery of adulterated and counterfeit essential oil-based plant products sold in worldwide commercial markets. These new methods have the potential of facilitating enforcement of regulatory quality assurance (QA) for authentication of plant product genuineness and quality through rapid evaluation by volatile (aroma) emissions. The MAU-9 e-nose system was further evaluated using performance-analysis methods to determine ways for improving on overall system operation and effectiveness in discriminating and classifying volatile essential oils derived from fruit and herbal edible plants. Individual MOS-sensor components in the e-nose sensor array were performance tested for their effectiveness in contributing to discriminations of volatile organic compounds (VOCs) analyzed in headspace from purified essential oils using artificial neural network (ANN) classification. Two additional statistical data-analysis methods, including principal regression (PR) and partial least squares (PLS), were also compared. All statistical methods tested effectively classified essential oils with high accuracy. Aroma classification with PLS method using 2 optimal MOS sensors yielded much higher accuracy than using all nine sensors. The accuracy of 2-group and 6-group classifications of essentials oils by ANN was 100% and 98.9%, respectively.

Application of MOS Gas Sensors Coupled with Chemometrics Methods to Predict the Amount of Sugar and Carbohydrates in Potatoes

“…For the six-group classification, based on the type of extract, the polynomial and RBF functions had a classification accuracy of 98.9 in the C-SVM method, while in the Nu-SVM method the classification accuracy of linear functions and RBF was 100% in learning and 98.9% in validation. Similar results have been reported for other crops, such as grape leaves [ 44 ], fruit juices [ 45 ], essential oils [ 4 , 16 ], coffee bean [ 46 ], corn [ 47 ], and cucumbers [ 48 ].…”

“…The detection system consists of sensors placed inside the measurement chamber. Commercial sensors, such as metal oxide sensors (MOSs), are the most common ones available for the detection of volatile organic compounds; nine metal oxide semiconductor sensors (MOSs) have been tested [ 4 , 16 ]. The names of the sensors and the main applications are MQ3 (alcohol), MQ4 (urban gases and methane), MQ8 (hydrogen), MQ9 (CO and combustible gas), MQ135 (steam ammonia, benzene, and sulfide), MQ136 (sulfur dioxide), TGS813 (CH4, C3H8, and C4H10), TGS822 (steam organic solvents), and TGS2620 (alcohol and steam organic solvents).…”

Impact of Coffee Bean Roasting on the Content of Pyridines Determined by Analysis of Volatile Organic Compounds

“…Organic volatile compounds can be defined as a family of carbon-containing chemicals exhibiting high vapor pressure at ambient temperature. They have biological, chemical, or physical emission sources [ 15 , 16 , 17 , 18 , 19 , 20 ]. In coffee roasting, a combination of these processes is the emission source [ 21 , 22 ].…”

Development of an Online Detection Setup for Dissolved Gas in Transformer Insulating Oil

“…These methods have certain limitations, especially in multicomponent gas detection. With the development of VOCs detection technology, some scholars use various types of sensors for multi-component gas detection [13] in many fields, such as purified essential oil detection in commercial plant products [14], alcoholic beverages and perfume quality control [15], bakery food product classification [16], and coffee processing [17]. However, in the field of DGA real-time detection, it is necessary to use the high anti-interference, fast response time and high-performance detection method that does not consume the test samples [18].…”