A methodology for designing a machine to generate electricity using the forces of gravity and buoyancy is provided which generates an output sufficient to sustain the machine's operation and provide a remainder amount of electricity for commercial purposes. The machine has two independent electricity generating units. Output work, U.sub.o, for each generating unit is based on the kinetic energy of a buoyant shuttle falling under the influence of gravity, and each unit's input requirement, U.sub.i is based on the work required to manipulate a volume of water during shuttle transit through water tanks of the machine. The methodology is based on a pre-selected output power P.sub.o which is used to establish machine component configurations. The shuttle's kinetic energy is then compared to U.sub.i to evaluate the machine's operational efficiency.

A system and method for predicting power output of a wind farm are disclosed. The method includes determining first and second parameter values of a power curve for a plurality of wind turbines. A second relationship is determined between the densities associated with the wind turbines and the values of the first parameter. A third relationship is determined between the densities associated with the wind turbines and the values of the second parameter. A value of the first parameter for a specified wind farm density is determined based on the second relationship. A value of the second parameter for the specified wind farm density is determined based on the third relationship. An indication of a power output for the specified wind farm density is generated by applying the determined values of the first and second parameters to the power curve.