On multiobjective volt-VAR optimization in power systems

The need for simultaneous optimization of reactive resources for the transmission and distribution system has long been recognized. If investment resources are limited, various combinations of solutions (voltage level, amount and type of reactive support, etc.) may impact a number of objectives (distribution losses, distribution feeder power factor, voltage profile for conservative voltage reduction, transmission losses, transmission capacity, voltage stability, etc.). While solutions have been proposed for subsets of the above problems, few algorithms have undertaken the simultaneous optimization of reactive resources in the transmission and distribution network. This paper attempts to address various issues that need to be solved: decomposition of the transmission model from the distribution model, design of an interface suitable for simultaneous optimization, and development of the methodology (multiobjective optimization based on building Pareto fronts with the help of custom-tailored genetic algorithms). Some of the issues discussed in this paper are illustrated on suitable examples and guidelines proposed for building a practical model that would incorporate all of the concerns with the modeling issues.