Electricity Networks & Unfounded Assumptions
The problems concerning electricity network covered by a recent Euractiv article (https://www.euractiv.com/section/energy-environment/news/growing-demand-puts-eu-electricity-grid-under-pressure) were predicted by research undertaken in 2019 by PWR and Challoch.
Electrification is seen as the key route to EU decarbonisation for both space heating (using mostly heat pumps (HPs) and transport (using electric vehicles EVs). Although it is recognised that action needs to be taken to reinforce electricity networks, only cursory consideration, until now, has been given to the impact of the two new loads on local electricity networks.
PWR and Challoch spent 2 man years of effort in 2019, quantifying the impact of HPs and EV charging and the installation of roof top solar (with and without batteries) on electrical networks in suburban areas. The question answered was: what happens to a low voltage (LV) network powered by a single medium-voltage/low voltage transformer (typical of 99% of EU networks) when new loads and generation are added.
Most suburban electrical networks were designed on the assumption that gas or oil would be used for heating. Exceptions include France (some networks are designed to support electrical heating) and cities/urban areas where district heating systems are sometimes deployed (Northern & Eastern EU Member States and Germany). The metric used to design suburban networks is called “after diversity maximum demand” (ADMD). The number is usually in the range 1 to 1.5kW per household. Typically 200 to 250 householdsare connected to an MV/LV transformer with a rating of 500kW. The maximum transformer load, usually on a winters night, will be in the range 200kW through to 375kW. The transformer and associated cables will be sized to reflect this load. Electrical power demand per house per year is around 3.5MWh (reflected in Eurostat data).
A house with a floor space of 120m2 with loft insulation and double glazing consumes 15 to 18MWh/year of gas (or oil) for heating and hot water demand. If the house changes the heating system to an HP with a coefficient of performance of 3 then the electrical consumption will be the range 5MWh to 6MWh per year. Improvements to the thermal efficiency of the house (e.g. cavity wall insulation) can reduce overall heat demand to 9MWh/year. HP electricity consumption will thus be 3MWh/year. Adding an HP to existing household loads will at least double the ADMD from 1 to 1.5 to 2 to 2.5kW. Adding EV charging further compounds the problem.
A heat pump penetration of around 30% on a given LV network will exceed the capacity of the transformer. However, before this occurs, a related problem will occur, voltage drop due to the load exceeding the carrying capacity of the LV cable supplying the houses connected to it. As noted, LV cables are sized for an ADMD of 1 to 1.5kW. Connecting more load to a given cable will lead to a fall in the voltage supplied to households particularly those sitting at the end of the cable. EU distribution network operators (DNOs) who own and control electricity networks are contractually obliged to provide power within specified limits, typically +/-10% of 230 volts. If it falls outside of those limits the DNO has fix the problem.
Most electrical networks in suburban areas feature LV cables that are laid directly in the pavement or road at a depth of roughly 0.5 metres. Sitting with them are: telephone cables, gas pipes, water and sewerage pipes, cables to power street lighting, cable TV and Medium Voltage electricity cables. Replacing the LV cable with a larger one is perfectly feasible by digging up by hand the old cable and putting in the new one which then needs to be connected to each house. The EU suburban landscape and associated networks was built over a period of 50 years or more. Given this reality, how long would it take to make EU LV networks serving suburbs fit for an all-electric future? Decades?
There is an alternative to the wholesale reinforcement of LV distribution networks: deploy micro-CHP systems to households in parallel with the deployment of HPs. Micro-CHP systems such as fuel-cells provide space heat, domestic hot water and electrical power. They can provide network support exactly where it is needed, on the LV network, and at a time when it is needed, i.e. a winter’s night when HPs are running at full capacity.
PWR and Challoch modelling suggests that a mix of HPs and micro-CHPs would greatly reduce the need for massive LV network replacement whilst at the same time reducing the impact of EV charging (which ideally would take place in the period 2200 to 0600hrs). Findings from the research were presented in early 2020 to various EU DNOs who agreed with both the problems identified and the possible solutions.
One objection to micro-CHP is that at the moment such systems use natural gas. Thus if there was a 50/50 mix of HPs and micro-CHPs there would only be partial decarbonisation of heating. However, green hydrogen could replace natural gas. Furthermore, our research identified another problem: roof-top PV in the summer causes voltage rise on LV networks even with relatively low PV penetrations of 20 to 25%. The solution to this is to use electrolysers connected to the LV network to convert excess electricity into hydrogen and inject it into the gas network. This ultimately leads to an energy system in which gas and electricity networks are operated as complementary to each other. That this is barely under consideration demonstrates the paucity of strategic thinking within the EU and its various institutions.
