Modelling Future Prosumer Nodes and their Impact on Network Planning - a Case Study

Abstract

This paper presents a pair of reasonably likely scenarios for low voltage (LV) electricity consumption and distributed generation and investigates how these scenarios will impact a realistic approximation of an existing section of LV network generated by a distribution network-planning algorithm using falsified but realistic existing load data. Some simplifying assumptions are made due to the difficulty of modelling the consumption of individual customers, but general methodology is set up and illustrated that will suit a more fine-grained analysis as data becomes available. The scenario loading uses simulated consumption time series data for two heating scenarios: EV consumption using uncontrolled and sensible charging scenarios, and photovoltaic rooftop generation time series data from earlier research at Aalto University. The results clearly illustrate that smart EV charging can alleviate the need for network reinforcements but the network upgrade costs imposed by the most pessimistic uncontrolled EV charging are also quantified.