Accuracy and Precision Evaluation of Unmanned Automatic Gas Detection Tube Measurement Device Using IoT

Abstract

The number of chemical accidents occurring nationwide when handling chemicals increased from 66 in 2022 to 116 in 2023. When a chemical accident occurs, response personnel must quickly detect the substances causing the accident and establish a plan for disaster prevention and resolution. However, due to the hazards and risks associated with accidents, ensuring the safety of response personnel while determining the origin of the accident is challenging. Generally, easy-to-use manual detector tubes are widely used during initial accident responses; however, using personal protective equipment, such as gloves and protective clothing, makes determining the origin difficult. An unmanned gas detection tube measuring device was patented in a previous study to address this issue. However, its use is limited because the accuracy of the device has not been evaluated. Therefore, in this study, we evaluated the accuracy and precision of the unmanned gas detector. Ammonia was selected as the target substance because it frequently causes accidents. To ensure the reliability of the unmanned gas-measuring device, a comparative analysis was conducted using the manual detector with a single gas-measuring device as a sensor. The manual detector showed a higher error compared to the standard error due to the delay in reading time and deviation in the individual reading. However, the measurement error of the unmanned gas detector was 19.8%, which was within the manual detector measurement error range; the accuracy was higher than that of the manual detector. Therefore, an unmanned gas detector is preferable over a manual detector.