Integration of Wind Power in Medium Voltage Networks - Voltage Control and Losses

Since some years ago greenhouse gases and especially carbon dioxide have become one of the most widely debated issues. To reduce the emission of carbon dioxide and nuclear waste, renewable energy sources as for example wind power plants have become very successful. These generation units are normally small compared to common thermal power plants and they are often placed distributed and connected to the distribution network. Therefore they are also called distributed generation (DG) units. For the distribution network the connection of generation units is a challenge since distributed networks were planned and built and dimensioned for the connection of load. Customers are normally connected to the distribution network and therefore voltage quality is an important issue for distribution networks. Medium and low voltage distribution networks are quite passive today with none or only somewhat communication and voltage control. In many cases the only feasibility to control the network voltage is the on-load tap changer at the high voltage/medium voltage transformer. Voltage variations and varying network losses are two important aspects when connecting generation units to the distribution network. To reinforce an existing network by building new lines to cope with the voltage variations caused by DG units and in such a way increase the DG capacity in a distribution network is always a solution but it is an expensive one. In this thesis other solutions which allow an increase of the DG capacity without network reinforcement are considered. Three different methods for voltage control in medium voltage distribution networks are mentioned in this work: coordinated control of the on-load tap changer, reactive power consumption and active power curtailment. A voltage control algorithm to maintain the voltage within the limits at all network nodes has been developed within this work. The voltage control algorithm is able to perform three different control strategies: local on-load tap changer (OLTC) control, DG control and coordinated OLTC control. Local OLTC control is often used in medium voltage networks today. With DG control the voltage is controlled at the connection point of the DG units by the use of reactive power consumption and active power curtailment. In the case of coordinated OLTC control the voltage is controlled by a coordination of the on-load tap changer, reactive power consumption and active power curtailment. In addition to previous work done within this area voltage measurement values from already installed electricity meters are used as feedback for the OLTC control. The voltages obtained by the control algorithm and the network losses are simulated within this work. Simulations of the network voltages and the losses in the network were done on a generic network with three typical kinds of feeders. It is shown that the DG capacity in this generic network can be increased if the network becomes more active and the DG units are participating in the network voltage control as in the case of local DG control and coordinated OLTC control. Furthermore a more fair distribution of the curtailment has been tested but that increases the totally used curtailment. To exemplify on a real existing network, the network of Svalöv substation (Sweden) was analysed and transferred into the simulation tool. The Svalöv area has already today a lot of wind turbines connected and more are planned. The simulations from the generic network were repeated on this existing network and also here the DG capacity could be increased significantly by a more active voltage control in the network