Research on reinforced corrosion monitoring system based on Multi-sensor data fusion and wireless sensor network

This paper applies multi-sensor data fusion technology and wireless sensor network (WSN) to monitor and predict steel corrosion parameters in real-time. To overcome the difficulties and low accuracy in identifying rebar corrosion, this study selected five parameters for data fusion, including chloride ion concentration, pH value, rebar corrosion potential, and internal temperature and humidity of concrete. A three-level data fusion structure is designed with corresponding fusion algorithms chosen for each level. The primary fusion is completed through data cleaning and median average filtering methods, followed by using adaptive weighting algorithms to fuse sensor data of the same type to obtain parameter characteristics of the region. Finally, an improved PSO-BP neural network fuses the data from the previous level of fusion to achieve prediction of steel corrosion. Experimental results show that the steel corrosion monitoring system based on multi-sensor data fusion technology and WSN has higher reliability and accuracy compared to traditional corrosion monitoring methods.