It is difficult to control the complex object with lagging uncertainty and nonlinearity effectively. To solve this kind of control problem, this paper presents a self-correction fuzzy controller with multiple weighted factors based on genetic algorithm. According to information achieved on line, it finds the global optimum weighted factors with a high speed by the improved genetic algorithm so that to amend and perfect the control rules. It also has done some simulation experiments in the tobacco-redrying control process. The simulation results demonstrate that this kind of control method can achieve good performance.