C I R E D 17 th International Conference on Electricity Distribution Barcelona, 12-15 May 2003 UEP_Padilha_A1 Session 4 Paper No 82 - 1 - VOLTAGE REGULATION IN DISTRIBUTION NETWORKS WITH DISPERSED GENERATORS Antonio PADILHA; Iara F. E. D. DENIS; Rade M. CIRIC Universidade Estadual Paulista - Brazil padilha@dee.feis.unesp.br ABSTRACT Various investigations have showed that dispersed generators integrated into utilities’ distribution networks could affect the host distribution networks in number of ways. This paper reports an investigation of the voltage regulation possibilities in the distribution networks with dispersed generators. In addition, the impact of different voltage regulation actions on loss allocation in the distribution networks with dispersed generators is also reported. Results obtained from several case studies using IEEE 34 nodes test network are presented and discussed. Key words: voltage regulation; dispersed generations; power flow; loss allocation. INTRODUCTION Several investigations have showed that dispersed generators (DGs) integrated into utilities’ distribution networks (DNs) could affect the host DNs in number of ways [1-9]. Previous experience has shown that the integration of DGs into DNs could create safety and technical problems. They may contribute to fault currents, cause voltage flickers, interfere with the process of voltage control, increase losses, etc. Since the DNs with DGs are not passive, all issues about planning, building, maintaining and operation of the DNs become very interesting and need a re-investigation. Actually, overall model of the distribution system should be renewed, since the impact of DGs on the DNs expansion planning and operation is significant. The Distribution Management System (DMS) functions like load (state) estimation, power flow calculation, network reconfiguration, supply restoration, fault analysis, relay setting, Q-V regulation etc. are significantly affected by the DGs in the DN. It means that the DMS functions should be re-considered and probably modified in order to respect the presence of DGs in the DNs. For example, the presence of DGs in the DN will improve the quality of state estimation in the DN, since the voltages in the DG’s nodes are known. Power flow calculation is of course affected by the DGs, as well as, the network reconfiguration in order to minimize the power losses. Supply restoration after the fault of feeder or supply transformer, should respect the presence of the DGs in the DN as well, since alternate variants of power supply could be quite different comparing to the passive DNs. Similarly the DGs change the voltages and reactive power flow in the DNs and consequently Q-V regulation should be re- considered, etc. This paper reports the voltage regulation issues when DGs are integrated into DNs. The impact of the DGs on the voltage regulation, reactive power flow and loss allocation performances of the distribution systems in steady state is also investigated. Results obtained from several case studies using IEEE 34 nodes test network are presented and discussed. VOLTAGE REGULATION The basic tool for analyzing the voltage regulation performances of DNs with DGs is an efficient power flow program. The DG may operate in one of the following modes: 1) In parallel operation with the feeder where DG is designated to supply a large load with fixed real and reactive power output; 2) To output power at a specified power factor; 3) To output power at a specified terminal voltage. Considering power flow computations, the DG node in the first two cases can be represented as a PQ node. It requires just a little modification in the power flow algorithm, actually the current is injected into the bus. In the third case where the source controls the voltage magnitude at the corresponding node, the node is refereed to as a PV node. If the computed reactive power generation is out of the reactive generation limits, the reactive power generation is set to that limit and the unit acts as a PQ node. Some dispersed storage units may also act as a constant current but for purposes of the power flow the PQ model is adequate. In last decade, different procedures for handling PV nodes have been proposed [10- 14]. Special single–phase and three-phase power flow methods have been developed for radial and weakly meshed network analysis. Experience showed that very good results in handling PV nodes in large-scale DNs are obtained using the backward/forward procedure, i.e. branch-oriented methods. These methods may be classified as follows: current summation methods, power summation methods and admittance summation methods [14]. In the proposed methodology for determining the impact of DGs on the voltage regulation performance, the efficient and robust compensation method proposed in [11] is applied. In this method, PV node sensitivity matrix is used to eliminate voltage magnitude mismatch for all PV nodes. The problem of compensating PV node voltage magnitude is transferred to the determining reactive current injection for each PV node, so that the voltage magnitude of this node is equal to the scheduled value. Since relation between reactive current and voltage magnitude of the DG is nonlinear, desired reactive current of the DG is determined iteratively. Connection of the DGs in the DNs can result in voltages that may be out the statutory limit [15-17]. Generally there are two cases of DG that can operate in the DNs: induction generator (wind-turbine) or synchronous generator (gas, diesel turbine etc.). In case of wind-turbine, DG’s voltage can be greater than substation high voltage/medium voltage HV/MV), namely voltage rise at the DG node can occur. In case of synchronous generator substation voltage is usually maintain greater than DG node voltage. Voltage rise - mismatch between generator voltage and substation voltage, ∆V r in case of induction DG is given by equation (1) [15]. The voltage