For most fuel rod codes, the time independent heat conduction equation, which is a steady heat conduction equation, is applied in fuel temperature calculation. However, it can affect the fuel temperature prediction in II condition, which the linear power has much change in some seconds. For improving the fuel temperature prediction in II condition, this paper gives a new numerical method, which combines classical thermal conduction integration method and the difference applied in time partial derivative. For guaranteeing the numerical method’s stability and convergence rate, the multi-dimension Newton-Raphson procedure are applied in fuel temperature calculation. This paper describes the theoretical deduction of the numerical method, and Halden fuel thermal conductivity model applied in fuel temperature calculation.
In order to verify new numerical method’s correctness, stability and convergence rate, the comparison between numerical solution and analytic solution is performed in 4 hypothetical conditions that the power transient duration is respectively 3s, 15s, 30s and 120s, the linear power changes from 15kW/m to 45 kW/m, and the fuel pellet surface temperature changes from 400 degree to 750 degree. And fuel density, specific heat and thermal conductivity are assumed as constants so that there exists analytic solution in this condition. The 4 hypothetical conditions have covered the worst II condition. According to the results in 4 hypothetical conditions, the fuel centerline temperature relative difference between numerical solution and theoretical solution is less than 0.6%, and the iterations are less than 5. So the numerical method possesses excellent correctness, stability and convergence, and this method has much potential in application in fuel rod code.