Managing time shiftable devices Published on August 12, 2024 30 minute read Back to homepage
Heat pumps, induction cooking, electric vehicles (EVs) - our energy consumption is rapidly becoming more and more electric. And with good reason: electrification (switching to devices that use electrical energy) is an effective way of reducing carbon emissions when the used energy is produced in a renewable way, so it is one of many ways by which we can try to limit global warming and climate change. This shift doesn't come without technical difficulties however - let's dive in!
The average household uses around 10-25 kWh per day (depending on usage of air conditioning, electric heating, et cetera), with a peak power draw of up to 5 kW. Usually net energy production peaks due to solar panels occur around noon, and consumption peaks due to household and entertainment activities occur around 18:00-19:00. Fully charging an electric vehicle requires around 70 kWh, and usually draws power in the range of 3-11 kW - if this aligns with the usual peak power draw, this means a single household with a charging electric vehicle can match the peak power draw of about three households without electric vehicles!
Here are some examples of what the daily energy profile of a household with solar panels without electric vehicle look like (data from the European CoSSMic project, graph generated using Mermaid):
--- config: themeVariables: xyChart: plotColorPalette: "#000000, #00AAAA, #AA00AA, #AAAA00" --- xychart-beta x-axis "Time (hours)" ["0", "1", "2", "3", "4", "5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15", "16", "17", "18", "19", "20", "21", "22", "23", "24"] y-axis "Net power (kW)" -0.6 --> 1.75 line [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] line [0.550999999999476, 0.53899999999976, 0.524000000000342, 0.456000000000131, 1.17900000000009, 1.2510000000002, 0.565999999999804, 0.59900000000016, 0.0169999999993706, -0.361000000000331, -0.524999999999636, -0.295999999999822, -0.203999999999724, -0.335000000000036, 0.0799999999994725, -0.294000000000324, 0.0280000000007021, 0.734999999999673, 0.5, 1.0600000000004, 0.599999999999454, 0.955000000000837, 1.76599999999962, 0.802999999999884, 0.216000000000349] line [0.989999999999782,0.590000000000146,0.434000000000196,0.425000000000182,0.429999999999382,0.485000000000582,0.194999999999709,-0.0640000000003056,0.324000000000524,-0.179999999999836,-0.0399999999999636,-0.43100000000004,0.00599999999985812,-0.0850000000000364,-0.00999999999976353,-0.440000000000509,-0.320999999999913,0.792000000000371,1.26900000000023,0.920000000000073,1.2549999999992,1.45600000000013,0.739000000000488,0.59099999999944,0.690000000000509] line [0.690000000000509,0.804999999999382,0.454000000000633,0.453999999999724,0.46100000000024,0.25,0.255000000000109,0.244999999999891,0.800000000000182,1.08099999999968,0.518000000000029,0.460999999999785,-0.259999999999764,1.4399999999996,0.405000000000655,0.714999999999236,0.761000000000422,0.944000000000415,1.21599999999944,1.3739999999998,1.30100000000039,1.13400000000001,0.979999999999563,0.976000000000568,0.528999999999542]
Lots of devices within a neighborhood starting more or less simultaneously, as well as synchronised feeding back of energy from solar panels, can quickly become a problem; the demand for transmission of electricity can exceed the safe operating capacity of the grid, potentially causing components to overheat, ultimately causing grid failure. This phenomenon, called grid congestion, is becoming more prevalent the more our energy usage is electrified. Simply increasing capacity of the grid to support this electrification is often infeasible: a smart solution is required.
Smart grids
Energy providers normally adjust energy production to expected consumption, but what if we could get consumers to adjust their consumption behaviour according to production instead? Ideally their consumption would be spread more evenly across the day, and consumers would use their own generated energy as much as possible - stabilizing the net energy flow in the grid. Incentivizing this is known as demand-side management.
Demand-side management ideally happens automatically through devices that can be controlled to activate at a desirable time. For example, you could fill your dish washer or washing machine, and delegate activation to a controller that is aware of when your solar panels are giving you free energy. They could even be made aware of schedulable devices across households within the low-voltage network so devices can be scheduled when there is still excess power within the neighbourhood, or such that the total peak load on the transformer is minimized.
This requires the grid to be enhanced with communication and control capabilities, since coordination is required to prevent accidentally causing high peak loads by activating many devices at the same time. A grid equipped with such capabilities is commonly referred to as a smart grid, the controllable devices are commonly called distributed energy resources. But how would a controller for such a network actually work?
... continue reading