In many cogeneration systems, one or more boilers are used in hot standby to meet the plant demand of steam in case of failure or upset in the cogeneration unit. Such boilers need to quickly respond to sudden and large steam load changes. However, fast changes in the firing rate cause transient changes in both the drum-boiler steam pressure and drum level, in addition to the potential of developing of thermal stresses in the walls of steam risers. A genetic algorithm (GA) based optimization scheme is proposed for tuning the conventional boiler control loops to maximize the ability of the boiler to respond to large steam demand while keeping the fluctuations in pressure, drum level, and feed rate within acceptable operation limits. A nonlinear model for an actual boiler is first built, validated, and then, it is used to demonstrate the performance of the boiler with the proposed control loop optimization.