Objectives  Experimental fluid dynamics(EFD)and computational fluid dynamics(CFD) are the two main methods for studying fluid mechanics problems, and they both have uncertainty analysis problems. The uncertainty of ship model resistance testing in circulating water channels is studied, and the uncertainty caused by the precision limit, balance calibration and data acquisition process are analyzed respectively.

  Methods  It is difficult to solve the sensitivity coefficient of uncertainty using the EFD method because of the specific inflow unevenness and turbulence intensity of circulating water channels. According to the actual situation of circulating water channels at Shanghai Jiao Tong University, a related mathematical model is established, and the influence of inflow unevenness and turbulence intensity on total resistance is obtained via CFD simulation. The corresponding uncertainty component is obtained by applying the uncertainty theory, and the uncertainty of the CFD calculation is also specially analyzed.

  Results  The results show that in the uncertainty analysis of CFD, the corrected and uncorrected calculation results are effectively confirmed, which indicates the feasibility of using CFD simulation to carry out research. The relative uncertainty of ship model resistance testing in a circulating water tank is 1.91%. Turbulence intensity has the greatest influence on resistance and plays a leading role; it should be considered in order to reduce turbulence intensity as much as possible when designing circulating water channels. The second most influential aspect is the uncertainty component caused by unevenness. The uncertainty components caused by the precision limit, balance calibration and data acquisition process are not significant.

  Conclusions  Using the CFD method to analyze certain components of EFD uncertainty is feasible and has certain reference significance.