The performance of Atkinson gasoline engine was simulated and analyzed based on the 1D engine model at working condition of the part load of 3 000 r?min-1. The results show that with the increase of EGR rate or LIVC, the engine fuel consumption is increased in the case that the throttle position, airfuel ratio, and ignition timing keep unchanged. In order to improve the Atkinson engine fuel economy，the united simulation model of MATLAB and GTpower was set up in the wide open throttle working condition. The values of EGR rate, LIVC, ignition timing, and airfuel ratio were iteratively optimized by using a redesigned genetic algorithm. The fuzzy clustering method was adopted to analyze all the previous generations for obtaining the best fuel consumption and the corresponding control parameters at different load points.