Grain Moisture Sensor Data Fusion Based on Improved Radial Basis Function Neural Network

Difficulty was known to get satisfactory measurement effect on precision in capacitive grain’s moisture measurement due to many influencing factors, such as temperature, species, compaction and so on. The data confusion method of Radial Basis Function (RBF) nerve network is adopted. With improved orthogonal optimal method, the RBF nerve network’s weight factors can be obtained. This method can avoid artificially selected the number of hidden units, which can cause low learn precision or over learn. Tests showed that the improved RBF network algorithm reduces the network structure, greatly enhances the learning speed of calculation. By using of the improved RBF nerve network, the precision for wheat’s moisture measurement has been improved. Keyword: Radial Basis Function Never Network, moisture measurement, k-means clustering, hidden layer neuron.