Abstract

Most of measurement sensors used for maintenance are continuously exposed to various environmental factors such as transportation and rainfall, so the possibility of breakage increases gradually. The maintenance measurement sensor of domestic subway tunnel shows an average of 14.2% to 14.8% of loss rate after about 5 to 6 yearsfrom installation, and it shows a sensor loss rate of about 13.9% in case of foreign countries. As a result, it can be seen that an average of 15% of maintenance measurement sensors at home and abroad cannot send measuring values after 5~6 years. In order to continuously collect accurate data, measurement data must be recovered by performing repair or replacement of the sensor, but some lost measurement sensors are buried after installation. So, there are many difficulties in repairing sensors, including cost and time. Therefore, in this paper, we propose lost measurement sensor data estimation technology based on data trend analysis using adjacent sensors.