This paper aims to solve the problems in the trajectory optimization during the cruise phase that the edge weights are closely related to the selection of the preceding path， and the calculation cost of the evaluation function is too heavy under different objectives like the least fuel consumption， the shortest flight time， and the lowest cost. First， a process is built， involving the GRIB2-format grid meteorological data interpolation and calculation of aircraft performance parameters. Second， the efficiency is augmented by the improvement of the A* algorithm evaluation function coefficient and the step size of performance calculation. Results for typical flight tasks under different optimization objectives are compared and the effects of factors such as cost index， mass of aircraft and flight level are discussed. The A* algorithm that incorporates the flight performance calculation into the edge weights， requires more calculation time than the traditional shortest path algorithm. However， compared with the Dijkstra’s algorithm， it can find the optimal trajectory more quickly under the same performance-based optimization objective.