Application of adaptive agents in decentralized energy management systems for the purpose of voltage stability in distribution grids

Nowadays, local utilizes face new challenges by operating their grids. This is mainly due to the increasing amount of distributed generation in the low and medium voltage level. If the injected power is not consummated locally but has to be transmitted over a long distance, voltages at busbars close to the source will rise which on the one hand is a potential threat to the equipment and on the other hand may lead to illegal system states and outages. Furthermore, measurement is sparsely spread in distribution grids. Thus, the grid state is not entirely observable and violations may not be detected. So local utilities trying to achieve a highly automated energy supply do not aim at a global governor but the grid has to be controlled locally using small adaptive systems which are able to make decisions autonomously. In this paper a new approach on the exploitation of reactive power reserves of inverter dominated distributed sources is introduced. Although X/R ratio differs from the transmission system reactive power offers a constricted but cost-effective possibility to control nodal voltages. So, first, a method to control single nodal voltages by using reactive power reserves of inverters is presented. After that, voltage stability is improved by intelligent agents which are able to work together.