Learning from emergent behaviour in networks to design non-peaky low-voltage electricity networks

It is a well-established fact that peaks in electricity demand in low voltage distribution networks increase operational costs and carbon emissions, as well as endangering network stability . Hence the reduction of peaks is of strategic importance to distributed network operators.
The most common contributor to high peak demand in the residential sector is simultaneously occurring electricity consumption factors.
In the literature, Demand Side Management (DSM) represents a broad class of electricity demand scheduling techniques at network level.
However, these primarily operate on flattening the peaks either at a single building or global network level.
In this paper we explore, for the first time, the emergent properties of two different peak-coordination behaviours - taking into account predictability of behaviour and interdependence between the network components.
The results suggest that smart controllers, coordinating and shifting the timings of electricity demand within neighbourhoods of buildings, could significantly improve network operation.
This would be particularly valuable for developing countries where, due to population and economic growth, electricity demand is predicted to rapidly increase posing a threat to resilience of low-voltage networks.