Volt var regulation algorithm using distributed transformer low voltage analogues

Voltages on electricity distribution networks are managed by automated control systems. Without management or control, the voltage will fluctuate throughout the day with changes in consumer loads, network conditions and changes in solar power output. The control algorithm under study in this paper currently uses only local measurements to guide decision making. This project aims to model and assess a control system algorithm which incorporates distributed voltage measurements. Electricity distribution networks are traditionally a top down arrangement with power flowing to customers in a one-way flow. This made it possible, with good network planning, to manage voltage using only localised measurements. Increased complexity of networks, and advances in communication system technology allow an opportunity to use voltage measurements distributed throughout the network to fine tune the decision making of automated voltage management systems. MATLAB has been used to analyse historical data and create a suitable model of the network for testing. The existing volt var algorithm was replicated in MATLAB and the performance of the modelling was tested against historical data. The algorithm was then modified to incorporate the measurements from the downstream distribution transformer analogue readings. The outcomes of the research are to identify the potential benefits of modifying the information sources available to the volt var calculations being made by the substation control system. This is still being investigated. The results of the modelling of the modified volt var algorithm suggest that there is a benefit to modifying the VVR algorithm to incorporate the voltage data sources that are available. The modified algorithm demonstrates a reduced Tap Change frequency and improved voltage tolerances. Currently the substation automation system is not aware of changes in network topography beyond the substation fence. Further work includes developing a function within the network management system to keep the substation automation system in step with changes to the network supplied from the substation. The network model used in this study is constructed with a number of approximations to simplify the large number of calculations being made. Further work to define and quantify the material benefit of making changes to the existing algorithm will require a more complex and accurate model