Bridge structural monitoring and warning system application in Thailand — Experiences learned

Abstract

Engineering structures have been regarded as a one of the crucial foundations of societal and economic development of the nation. As significantly impact life quality and safety, monitoring and warning mechanisms are required. Structural members are considerably affected by several loadings imposed by normal operation, disaster and environment. Proper inspection and detection are thus crucial both during regular and unsafe events. An Enhanced Structural Health Monitoring System Using Stream Processing and Artificial Neural Network Techniques (SPANNeT) has been developed, tested and evaluated. SPANNeT applies wireless sensor network, real-time data stream processing and artificial neural network based upon the measured bending strains. Off-the-shelf sensor platform including strain gauge, data logger and base station were employed. Like other wireless sensor network (WSN) applications, several limitations based upon the resource constraint are recognized and additional restrictions are unforeseen. Several experiences learned from applying SPANNeT to an existing bridge structure in Bangkok, Thailand are described in this paper. Adaptations were conducted to the strain gauge to avoid inaccurate readings due to the non-uniform texture of the concrete. Furthermore, the distance between strain gauge and data logger was increased by using electrical cable to overcome device installation restriction and to tackle battery recharge problems. Even the distance between data logger and base station was decreased, completely reliable data delivery could not be achieved. At least 90% of successful data communication was observed. Such experiences can be considered as a preliminary guideline of system enhancement and deployment in the future.