Research On Optimization Of Calibration Intervals And Calibration Conclusion Risk Assessment For Measuring Instruments

With the increase in the using of measuring instruments,the uncertainty will continue to increase.In order to ensure that the measuring instrument measuring characteristics in the maximum allowable error range,setting up a reasonable calibration interval is necessary in the daily measurement and testing work.In addition,we can not guarantee the calibration result hundred percent to be accurate because of the error in standard instrument.It also has certain risk.Therefore,it is of great significance to study the prediction method of calibration interval to determine the calibration interval of the instrument scientifically and rationally and reduce the risk of misjudgment of calibration conclusionsCalibration interval determination and calibration conclusion risk assessment are the two important issues in the calibration process.This paper focuses on these two aspects of the study.(1)At present,the prediction of the calibration interval of measuring instruments is mainly reflected in the calibration data processing.Most of the traditional methods can only predict the performance degradation of the instrument,the accuracy is not high.Based on the characteristics of the calibration history data,this paper proposes a gray neural network prediction model to predict the calibration interval.The system error and the random error of measurement instrument predicted by the gray GM(1,1)model and the BP neural network.Finally constitute a combined forecasting model,and the maximum information is retained.The experimental results show that the gray neural network prediction model has good precision and prediction effect.(2)Because of the uncertainty of the standard instrument,there is a risk of misjudgment of the calibration conclusion.This risk is often overlooked.In reducing the risk of misjudgment is also limited to improve the measurement uncertainty ratio.To solve this problem,the risk of single calibration conclusion is analyzed in this paper,and verifies the risk assessment of DMM calibration conclusion through experiment.Based on the single calibration conclusion risk,the risk of the overall calibration is obtained with the change of influencing factors through the statistical analysis of the calibration process made by Monte Carlo simulation.A new compression factor is used in the overall calibration conclusion risk assessment to balance the risk of false accept and the risk of false reject,which provides a new method to reduce the risk of misjudgment.