Abstract
In recent times, the issue of the gap between electricity supply and demand has been addressed using distributed generation (DG) technology. When DG technology is properly placed within a transmission and distribution network, there is always improvement in power loss reduction, voltage profile and system reliability. Proper placement can be achieved by evaluating the optimal values for voltage deviation, real power loss and bus voltage. Thus, this study presents a multi-objective DG optimisation model that minimizes of voltage deviation and real power loss, while maximising voltage stability factor. The model was formulated as a multi-objective problem and solved using an improved Strength Pareto Evolutionary Algorithm (SPEA-2) technique of optimisation to DG problem. The Nigeria-31 bus, 330 KV, transmission network was considered as test case. The results obtained were validated with the standard IEEE 30-bus system. In addition, this study presented scenarios where 1, 2 and 3 DGs were placed into these test cases. Based on the SPEA-2 implementation, the optimisation run time for the Nigerian network and the IEEE network were 2229.55 and 2039.42 secs, respectively. The optimal bus location of the three DGs (whose capacities are: 8.7793MW/6.1272MVAr, 8.1806MW/4.7778MVAr and 7.9567MW/4.6281MVAr respectively) for the IEEE 30 bus were 5,7 and 26 buses, respectively, while for the Nigerian-31 bus were 14, 15 and 17 (whose capacities are: 22.9693MW/15.2956MVAr, 27.3711MW/21.7274MVAr and 30.9910MW/15.4069MVAr respectively) respectively. For the placement of these DGs, the power loss reduction in the IEEE-30 bus is 16.17MW, 15.28MW, 14.07MW respectively and 64.21MVAr, 61.19MVAr, 56.13MVAr respectively. While for the Nigeria-31 bus system; the reduction in power loss is: 34.66MW, 33.99MW, 33.82MW and 411.85MVAr, 403.45MVAr, 400.54MVAr respectively. The results obtained showed that the total power losses reduced as DGs are sited at the optimal locations for the two test cases when compared with cases when DGs were not considered.

Key words: Distributed generators, power loss, voltage profile, optimal placement, evolutionary algorithm