Abstract: A system for simulating and optimizing a powerhouse, designed to furnish electrical power to a process plant, incorporates a plurality of units for simulating operation of the individual components of the powerhouse, and for calculating the cost of operation of the system as a whole, including the cost of fuel for the powerhouse components and the cost of purchased power needed to satisfy the power demand of the process plant beyond the capacity of the powerhouse. An adaptive search routine varies all of the operative parameters of the powerhouse, on a random basis, and repeats the simulation, in order to identify a combination of parameters which represents a lower cost solution. The range of random variation is reduced, as lower cost solutions are not found within a given number of repetitions, in order to identify a precise convergence on the optimum solution.