Application of Multi-Sensor Data Fusion Method in the Ammonia Modification System

Abstract

This paper provides a method of multi-sensor data detection and fusion, aims to resolve the data measurement inaccurate problem in the process variable control system of the ammonia modification equipment. First, it establishes the sensor data measurement model and calculates mutual support level of the multiple sensors, using boundary value judgment method to produce the multiple-sensor mutual support level matrix, the total number of mutual support level for one sensor in the matrix is the weight value of this sensor, and finally the normalized weighted average data fusion (NWADF) method is used for data process.Experimental results show that the proposed method for multi-sensor data fusion can effectively be used to identify the abnormal data in multi-sensor measurement system and improve the accuracy and stability of the process variable data measurement.