Abstract: To acquire the accurate tunnel environment information and comprehensively evaluate the tunnel air quality, a series of highway tunnel environment perception system (TEPS) based on the Internet of Things and cloud computing is designed and developed. Highprecision sensors measuring pollutant concentration, temperature and pressure, visibility, and wind speed and direction are arranged and integrated in a tunnel. An STM32 microprocessor is adopted for realtime monitoring of environmental data, and a 4G mobile communication network is used for remote data transmission, realizing the cloud storage and realtime display of the tunnel environment monitoring data. Compared with various special detectors, the mean absolute percentage errors of CO and NO2 concentrations, visibility, and wind speed of the proposed system are 3.36% and 1.73%, 5.43%, and 2.06%, respectively. Results demonstrate that the TEPS is highly integrated, can be easily deployed, and provide stable operation. These performance characteristics overcome the shortcomings of portable monitoring devices, such as single collection content and insufficient data accuracy, and those of the traditional fixed tunnel monitoring systems, such as complex structure and scattered layout. By operating the TEPS online for a long time, suitable data can be measured to support the exploration of the variations in air quality in the tunnel caused by traffic and dynamical adjustment of the ventilation systems.

Key words: highway tunnel, environment perception, air quality; ventilation control, Internet of Things, cloud computing