Chapter 5. Distribution Network Representation in the Presence of Demand Response

In order to support and stimulate the Demand Response (DR) in distribution networks, updated and new monitoring and control functionalities are required together with a revision of the Distribution System Operator's (DSO's) control centre architecture. As a consequence, a considerable amount of information is made available at the different control levels that needs to be properly handled. However, not all the information has to be considered in details; a relevant simplification can be achieved by adopting a distribution network representation in which only the information that is actually relevant is treated. To exploit the benefits accruing from DR, the relevant information concerns the DR supplier location, both for commercial and technical purposes. From the technical point of view, the question arises on how detailed such a locational information should be in order to allow the DSO the monitoring and control of the system. Each consumer, or small group of consumers, is connected to a different node of the network; however, a network model detailing each and every single node of the distribution system would be not convenient, in particular if the power injections of consumers in those nodes presents a similar impact on the network behaviour. In order to ensure a correct and secure operation of the network, for the DSO it could be enough to identify in advance the nodes/lines with likely uncorrect operation (overload, or undervoltage/overvoltage), and to consider and represent as one equivalent consumer all those consumers whose demand has a similar impact on the considered constraints. In this view, a useful concept is the one of Load Area (LA); basically a LA can be seen as a group of prosumers whose power injection has a similar impact on the distribution grid operating conditions. By the LA concept, all the relevant data for the monitoring and control of distribution systems can be recognized, and a compact yet accurate representation of the relationships between such data can be derived as a possible reduced modeling of the LA. In this chapter a possible application of LAs to distribution network is considered. With the adoption of LAs, the information available for monitoring and control functionalities can be significantly reduced, and only the relationships among limited and well recognizable sets of data need to be considered for representing a given distribution system. From that, a relevant reduction results in the amount of information to be treated to describe the network. In the following, the steps needed for both the LA identification and representation are recalled. Three networks of varying size are taken as examples on which the LA concept and representation are demonstrated.