Multi-sensor poly information fusion technology and its application

Radio frequency capacitance sensor is usually used in the intelligent inspection of the water-content in oil product. This method depends on the great difference between the water dielectric constant and oil. The dielectric constant of water is so much bigger than that of the oil that the radio frequency showed by them is much different. In the process of the inspection, we found that the output of the sensor is determined by one parameter and furthermore the output date of the whole system is also changed when the temperature of environment is changed. That is to say the system has certain cross sensitivity about the environment and it can also affect the function of the whole system and measurement precision. In order to improve the measurement precision and maintain the temperature stability of the system we use poly information fusion technology to make a further inspection about the temperature and the moisture voltage. After the fusion process, the sensor can put out the information in any form of temperature, water-content, temperature voltage or water-content voltage, of which you want to choose depends on what fitting equation you want to take. There are a lot of methods in the bisensor fusion. We used two of them: curved surface fitting and neural network. We examined the sensor by lubricant oil and the result showed that the sensor's zero temperature coefficient, sensitivity temperature coefficient, temperature stability and measurement precision is much more accurate than those before the fusion technology is not used. The use of information fusion technology in the water-content inspect system of oil product can greatly improve the distinguish ability of and the speed to get a highly accurate measurement result.