The connection of Distributed Generators (DGs) to a distribution network causes technical concerns for Distribution Network Operators (DNOs) which include power flow management, loss increase and voltage management problems. An Active Network Management System can provide monitoring and control of the distribution network as well as providing the infrastructure and technology for full integration of DGs into the distribution network. The Optimal Power Flow (OPF) method is a valuable tool in providing optimal control solutions for active network management system applications. The research presented here has concentrated on the development of a multi-objective OPF to provide power flow management, voltage control solutions and network optimisation strategies. The OPF has been shown to provide accurate solutions for variety of network topologies. It is possible to apply time-series of load and generation data to the OPF in a loop, generating optimal network solutions to maintain the network within thermal and voltage limits. The OPF incorporates not only the DG real power output maximisation, but also network loss minimisation as well as minimising the dispatch of DG reactive power. This investigation uses a direct Interior Point (IP) method as the solution methodology which is speed efficient and converges in polynomial time. Each objective function has been assigned a weighting factor, making it possible to favour one objective function and ignore the others. Contributions to enhance the performance of the IP OPF algorithm include a new generic barrier parameter formulation and a new swing bus formulation to model energy export/import in the main optimisation routine. A Terminal Voltage Regulator Mode (TVRM) and Power Factor Regulation Mode (PFRM) for DG were incorporated in the main optimisation routine. The main motivation is to compare these two decentralised DG control methods in terms of the achieving the maximum DG real power generation. The DG operation methods of TVRM and PFRM are compared with the optimisation results obtained from centralised dispatch in terms of the DG capacity achieved as it produces the optimum overall network solution. A suitable value of the droop and local voltage regulator dead-bands were determined for particular DGs. Furthermore, the effect of these decentralised DG control methods on distribution network losses are considered in a measure to assess the financial implications from a DNO's perspective.