Research On Measurement System Accuracy Loss Mechanism & Model

"Science is based on measurement", the level of measurement technology is the important evaluation standard for technology development of a country. Even if the influences of environment factors are not considered, the accuracy measuring system, especially with mechanical structure, will decrease in the process of actual use as time went go. So improving measurement accuracy has drawn attention to many developed countries.After deep study in measurement system & dynamic character, the research on accuracy loss mechanism of measurement system has been put forward to find the reason for accuracy loss and to build the accuracy loss function from the fields of mechanics and electrics. Support vector machine and BP neural network are used to establish prediction model of time series and to ensure the future development trend of equipment. According to the accuracy loss function, the future accuracy can be predicted, and based on the prediction result, the system effective life can be determination.In order to prove effectively the theory of the accuracy loss mechanism & model of the dynamic measurement system, a simple angle displacement measuring system is selected to make an accelerating experiment study of accuracy loss. After getting the magnitude error of angle displacement measuring system, the accuracy loss law of the system is obtained. The two modeling method support vector machine and BP neural network are building the prediction model of the system of accuracy loss. Using this model can achieve real time forecast of accuracy loss about measurement system.