Distributed price-directive decomposition applications in power systems operations

A number of important problems in the area of power systems operations and planning can be modeled as optimization problems that are amenable to solution by decomposition techniques. A major drawback on large problems, however, is that computer run times are often long, thereby limiting the utility of this approach for real-time operations. Using the constrained economic dispatch model as a case study, this paper demonstrates that distributed decomposition on a hypercube multiprocessor can be extremely effective in controlling the growth in run times as problem size is increased.< >