Loss allocation in radial distribution networks with various distributed generation and load models

Abstract This paper proposes a new method for loss allocation in radial distribution networks (DNs) considering different models of distributed generation (DG) and load in context of a deregulated environment. In the proposed method, a direct relation between real/reactive power flow in a branch and its losses has been developed without taking any assumption and approximation. Suitable expressions/relations for network power flow have been developed employing power summation algorithm. The developed expressions do not contain any cross-terms. For allocating the losses among network participants, the proposed method uses a circuit based branch oriented approach. Using only power flow results, this method employs a backward sweep network reduction technique to allocate the network losses to load/DG at various nodes. This method does not require additional step of normalization to collect the exact amount of total network losses. In the present study, different types of DG, e.g. DG injecting only real power, DG injecting only reactive power, DG injecting real power and absorbing reactive power, and DG injecting both real and reactive power are considered to allocate losses. In addition to this, various load models based on impact of voltage variation on real/reactive power consumption are also considered. To test the proposed method, modified 9-node and 33-node radial DNs have been considered. In order to show the effectiveness of the proposed method, its numerical results have been compared with those by other methods available in the literature.