Anomaly Detection in Particulate Matter Sensor Using Hypothesis Pruning Generative Adversarial Network

World Health Organization (WHO) provides the guideline for managing the Particulate Matter (PM) level because when the PM level is higher, it threats the human health. For managing PM level, the procedure for measuring PM value is needed firstly. We use Tapered Element Oscillating Microbalance (TEOM)-based PM measuring sensors because it shows higher cost-effectiveness than Beta Attenuation Monitor (BAM)-based sensor. However, TEOM-based sensor has higher probability of malfunctioning than BAM-based sensor. In this paper, we call the overall malfunction as an anomaly, and we aim to detect anomalies for the maintenance of PM measuring sensors. We propose a novel architecture for solving the above aim that named as Hypothesis Pruning Generative Adversarial Network (HP-GAN). We experimentally compare the several anomaly detection architectures to certify ours performing better.

Anomaly Detection in Particulate Matter Sensor Using Hypothesis Pruning Generative Adversarial Network

World Health Organization (WHO) provides the guideline for managing the Particulate Matter (PM) level because when the PM level is higher, it threats the human health. For managing PM level, the procedure for measuring PM value is needed firstly. The Beta Attenuation Monitor (BAM)-based PM sensor can be used for measuring PM value precisely. However, BAM-based sensor occurs not only high cost for maintaining but also cause of lower spatial resolution for monitoring PM level. We use Tapered Element Oscillating Microbalance (TEOM)-based sensors, which needs lower cost than BAM-based sensor, as a way to increase spatial resolution for monitoring PM level. The disadvantage of TEOM-based sensor is higher probability of malfunctioning than BAM-based sensor. In this paper, we aim to detect malfunctions for the maintenance of these cost-effective sensors. In this paper, we call many kinds of malfunctions from sensor as anomaly, and our purpose is detecting anomalies in PM sensor. We propose a novel architecture named with Hypothesis Pruning Generative Adversarial Network (HP-GAN) for anomaly detection. We present the performance comparison with other anomaly detection models with experiments. The results show that proposed architecture, HP-GAN, achieves cutting-edge performance at anomaly detection.

Field Comparison of TSI DustTrak versus TEOM in Two Poultry Houses

The TSI DustTrak monitor has been used for particulate matter (PM) monitoring at various animal facilities. The instrument determines PM concentrations based on the principle of light scattering. Several assumptions (e.g., particle size, refractive index, and density) are imposed during the calibration process; however, they may not apply to PM emanating from agricultural settings. In this study, PM10 monitoring was conducted at a broiler house and a layer breeding house with four collocated instruments: three DustTrak monitors and one tapered element oscillating microbalance (TEOM). Being a federal equivalent method (FEM) for PM10 monitoring, TEOM was selected here as a transfer standard for assessing the field performance of DustTrak. Results revealed a good linearity between DustTrak and TEOM PM10 readings (R2 =0.92 and 0.85 in the broiler and layer breeding houses, respectively). However, DustTrak significantly underestimated PM10 concentrations in both houses. To correct for the monitoring bias by DustTrak, an average correction factor was derived from correlation analysis that characterized the ratio of DustTrak’s PM10 response to TEOM’s. The factor was calculated as 0.267 for the broiler house and 0.244 for the layer breeding house. Mie scattering simulation was performed to further verify the derived correction factors. A factor of 0.204 was estimated from the simulation, and it accorded well with experimental results. A dependence of the correction factor on PM10 concentration was noted in both poultry houses, indicating the feasibility of developing a concentration-dependent correction factor for future monitoring efforts. Such a relationship could also be explained by Mie scattering. This study is expected to facilitate a better understanding of the limitations and perspectives of the TSI DustTrak and other light scattering PM monitors for agricultural air quality research. Keywords: DustTrak, Mass concentration, Mie scattering, PM10, Poultry, TEOM.

Base adsorption mechanism over zeolite catalysts at different Al contents probed by the tapered element oscillating microbalance (TEOM)

The TEOM is able to quantify with high accuracy the extent of physisorption and chemisorption of base probe molecules over zeolite surfaces at different Al contents.