Uncertainty Evaluation For Virtual Instrument Measurements Based On Monte Carlo Method

According to the composing and the work principle of a virtual instrument measurement system, this paper analyzed the uncertainty sources of a compositive transducer and a given data acquisition board ,which contribute to the uncertainty of the final virtual instrument measure result. On the basis of the elementary presentation of Monte Carlo computing method, a firenew uncertainty evaluation theory for virtual instrument measurements was proposed. Its essential principles were discussed as follows: According to probability incident simulation technique (Monte Carlo method), lots of given distribution random variables are generated so as to simulate random errors of the virtual instrument measurement chain , then simulative random errors and sampling datum are synchronously accumulated in a virtual instrument data processing module, a series of pseudo measurement results are acquired, meanwhile, the statistical standard deviation of the results is regarded as a token of the virtual instrument measurement ultima result uncertainty. In allusion to different data acquisition driving modes of a virtual instrument measurement system,the author compiled a random sampling errors simulation module of the PCI—6024 data acquisition board. The module enrich uncertainty evaluation function of the virtual instrument, it can resolve the uncertainty evaluation problem of a virtual instrument effectively. Then the author compiled a virtual instrument measurement uncertainty roboticized evaluation software which based on Monte Carlo method, it can implement roboticized evaluation of measurement uncertainty in a general virtual instrument system. In the experiment of a virtual instrument signal effective value measure,the author adopted the random sampling errors simulation module of the PCI—6024 data acquisition board proposed, to implement simulation evaluation of the signal effective value measure.The uncertainty evaluation results based on Monte Carlo computing method were coincident with the ones of the GUM theoretic method, it demonstrated the validity of the uncertainty evaluation theory based on Monte Carlo method in virtual instrument measurements.