To analyse the rapid changes occurring in residential energy supply chain, specifically the residential electricity grid, a simulator has been developed. As the electricity grid and load profile of a neighbourhood depends on many factors the entire energy grid of a neighbourhood can be modelled and analysed on various levels. This includes, but is not limited to, gas imported from the gas grid and heat being distributed over, e.g., a district heating system. The simulator allows us to quickly test and validate models and control methodologies for the various part of residential energy supply chains.

Model of a house

The corner stones of the simulation models are the households which together combine into the considered residential grids. To ensure the capability of describing most (future) scenarios in a residential energy supply chain, the house is modelled using various generic and specific appliances, all connected to a single household connection. These (future) appliances allow the alteration of their load profile by virtue of offering flexibility in when and/or how much energy they consume. Various classes with some examples are given below.

  • Uncontrollable electric devices; lighting, TV, desktop PC.
  • Controllable electric devices; smart washing machines, smart dishwashers.
  • Heat consuming devices; the central heating system, shower.
  • Local electricity generation; rooftop mounted PV.
  • Local heat generation; heat pump, micro-CHP (co-generates electricity).
  • Storage; electric vehicle, battery, heat vessel.

Together these devices are capable of significantly altering the households electricity consumption profile when controlled in an intelligent manner. In many cases, combinations offer even more flexibility, e.g., local heat generation combined with storage. In each house a controller is present that can manage the flexibility of the various appliances present. An example of a modelled household is given in Figures 1 and 2. Each device is equipped with a controller to test various control strategies. Note that if the controller is set to simply pass on the device’s load profile, the scenario without control is simulated. To ensure the feasibility of the model, devices are interconnected using energy stream. These streams ensure that that energy is always preserved, e.g., that energy can only be consumed in a house if it is produced there or imported from the grid. Devices are connected to controllers using data links. These data links are also used to interconnected controllers to specificy which controllers can communicate directly with eachother.

Model of house appliances
Figure 1: Various elements indicating the appliances in a house and their controllers.

Model of house connections
Figure 2: The various connections made in and between houses.

Model of a neighbourhood

A neighbourhood is modelled by interconnecting groups of houses with power lines. The power flow across these lines is calculated during the simulations, ensuring that the scenarios and control strategies can be tested for feasibility with respect to the physical grid. Also, these models can be used to calculate, e.g., the losses across the lines in a neighbourhood. Note that this is in contrast to the energy stream used inside the house which can effectively be considered as lines with zero resistance. Furthermore, it is possible to interconnect the controllers of various houses on different levels, e.g., feeder or transformer to model the effect of control strategies on each of these levels. An example of a model for a neighbourhood in the town of Lochem is given in Figure 3. Such models allow us to quickly check the impact of control strategies on residential energy supply chain that exist now and how well these systems cope with changes expected in the future.

Model of Lochem
Figure 3: An overview of the model of a neighbourhood in Lochem, both with (bottom)
and without (top) an overlaid map.

Model output

Even with a moderate amount of appliances and houses in a model, the number of different parameters of interest in the model is huge. Thus it is important to be able to have quick access to the relevant data after simulations have finished. To this end we developed a sophisticated user interface, where relevant data can be extracted and quickly compared to previous results. A few examples of the user interface can be seen in Figure 4. This particular simulation involves 16 futuristic houses with PV and a relatively large CHP with heat buffer. The case study investigates the potential for such a group of houses to work disconnected from the electricity grid.

Simulation results
Figure 3: An overview of a futuristic scenario of 16 houses with the electricity
requirements for a winter (top) and summer (bottom) week. The white line is
without control and the green line is with control.