A real-time optimization control method is proposed to enhance engine thrust response and enlarge its maximum thrust during emergent flight. This real-time optimization control is model based, and the on-board engine predictive model is devised by a multi-input multi-output recursive reduced least squares support vector regression method. Two emergency engine control modes during engine emergent acceleration process, the overthrust mode and the faster response mode, are redesigned by specifying relative objective function with engine operation constraints. Namely, the overthrust mode is for getting much bigger maximum thrust, and the faster response mode is for obtaining faster engine response. In the initial engine acceleration stage, the two modes have the same objective function for fastest thrust response within necessary operation constraints. After that, the objective function varies to maximum thrust in the overthrust mode, whereas the faster response mode keeps the objective for fastest response unchanged. For solving the online optimization problem, a feasible sequential quadratic programming algorithm is utilized here. The simulation results demonstrate that, first, engine response ability can be improved in faster response mode. Second, not only more thrust was provided over the maximum value but also fast response ability was achieved in the overthrust mode. Moreover, compared to conventional operation nowadays, argumentation of compressor guided vane angle control and broadening operation restrictions were deeply studied and verified in the new schemes.