A Study on Model of Multisensor Information Fusion and its Application

Abstract

Multisensor information fusion seeks to combine data from multiple sensors to measure the variables that may not be possible from a single sensor alone, reducing signals uncertainty and improving the accuracy performance of the measuring. The two main parts in multisensor information fusion system are the fusion model and fusion algorithm. In this paper, a radial basis function (RBF) neural network model with a variable bias is adopted in multisensor information fusion system, and an improved learning algorithm is proposed. This information fusion model is used in boiler drum water level measurement. By using this fusion model the drum level measurement precision is improved, and the influence of the "ghost water level" to the drum level measurement can be eliminated. The simulation results illustrate that the drum level measurement with the multisensor information fusion is more accurate and reliable than the traditional method, and the algorithm of information fusion is effective