An Data Correction Method for Hydrological Monitoring Based on Improved BP Neural Network
For hydrological monitoring, the missing and distorted sensor data may directly affect the reliability of the acquired information. To address such problems, an information fusion algorithm for sensor data correction based on the spatio-temporal correlation of hydrological monitoring information is proposed in this paper. A monitoring station unit whose core device is FPGA (Field Programmable Gate Array) is employed as hardware platform and fusion of the data collected by the monitoring station unit is performed using an improved BP (Back Propagation) neural network. This work uses the horizontal and vertical correlation of flow velocity distribution to correct flow velocity. The simulation experimental results show that this algorithm can be used for the correction of both random and gross error of sensor data.