Electricity flow is generally invisible, silent, and not something that most humans want to touch, so understanding how charge moves around can be fairly unintuitive at first. There are plenty of analogies to help understand its behavior, such as imagining a circuit as a pipe of water, with pressure standing in for voltage and flow standing in for current. But you can flip this idea in reverse and use electric circuits to model other complex phenomena instead. [Oxx], for example, is using circuit theory to model his home’s heating systems.
To build his model, he’s using LTSpice, a free circuit simulation program. Using voltage to model temperature and current to model heat flow, he’s set up a model for his home to compare the behavior of a heat pump and a propane furnace. A switch model already in LTSpice with built-in hysteresis takes the place of the thermostat. Using temperature data for a single day in January [Oxx] can see how each of his two heating systems might behave, and the model for the heat pump is incredibly close to how the heat pump behaved in real life.
The model includes all kinds of data about the system, including the coefficient of performance of the heat pump and its backup electric resistive heater, and the model is fairly accurate at predicting behavior. Of course, it takes a good bit of work to set up the parameters for all of the components since our homes and heating systems won’t be included in LTSpice by default, but it does show how powerful an electric circuit analog can be when building models of other systems. If you’ve never used this program before, we’ve featured a few guides to getting started that you can take a look at.
Thanks to [Jarvis] for the tip!
If your only tool for analyzing dynamical systems is SPICE, every control system’s dynamics looks like a circuit.
Well done, and an interesting hack. 😁
Using analogs (identifying non-EE systems which follow the same form of differential equations as circuits w/ Rs and Cs) is a pretty old technique and can help give insight especially when the system is quite similar to well-studied electrical systems (e.g. 2nd order resonant systems).
That said, as your system starts to include non-linear things or the controller you want to try is complicated, it gets harder to make things fit into what SPICE can easily do.
That’s when general purpose tools come in handy, such as the ODES part of scikit (python) or the DiffEq packages in SciML (Julia). These will have a significant learning curve (for the language if you don’t know it and for the particular solver package) but will let you go farther.
Hmmmmm. A dynamic simulator for my control system. . . . This will require some investigation. Nice tip!
Analog circuit analysis techniques can be applied to thermal systems.
The following equivalences are accurate to about 10%..
ELECTRICAL THERMAL
1 amp 1 watt
1 farad 1 gram aluminum
1 volt 1 degree C
1 second 1 second
1 ohm 1 degreeC/watt
So, with appropriate unit scaling, Spice can be used to model simple
thermal systems.
I’m curious about a few things regarding your model. How did you go about calibrating the coefficient of performance for the heat pump and the backup electric resistive heater?
The COP curve used was fabricated from a review of old heat pump data. I could have used the actual house data and computed more accurate numbers but my haste to try it out got in the way. As far as the actual resistive heat energy draw…an energy watt meter was giving time and usage for the heat pump. When it kicked in…big jump in watts being used.