Programmable thermostats allow Energy3D to study the effects of human behavior on home energy use

Figure 1

Energy3D Version 5.1 has incorporated a new feature: programmable thermostats (Figure 1). This allows users to add a programmable thermostat to each building in the design. Such a thermostat allows building researchers to model the occupants' schedules and thermal comfort preferences, making it possible to include a human dimension into building simulation.

Figure 2

Figure 2 shows the hourly heating cost of a house on January 1st when the thermostat temperature is set to be 18, 19, 20, 21, and 22 °C. Obviously, the higher temperature you set, the higher the cost will be. Notice, however, that the hourly heating cost might be close to zero around noon because the solar energy alone might be enough to provide the heating (it depends on the size of the house, the size of the windows, and the locations of the windows).

Figure3

The thermostat has a graphical user interface that can be used to set the temperature of a building by hour, week, and month (Figure 3). This flexible user interface can adjust the temperature of a thermostat in four different ways:

• Drag "All" button up or down to increase or decrease the temperatures in all the hours of all the days in a week of the selected month. 
• Drag the hour button up or down to increase or decrease the temperatures in the selected hour of all the days in a week of the selected month.
• Drag a gray temperature button on the right up or down to increase or decrease the temperatures in all the hours of the selected day;
• Drag any temperature button up or down to increase or decrease the temperature in the selected hour of the selected day.

In the future, we plan to extend this feature so that Energy3D can model "smart" thermostats such as the Nest Learning Thermostat, which can adapt to the occupants' specific needs and schedules.

Visualizing thermal equilibration: IR imaging vs. Energy2D simulation

Figure 1

A classic experiment to show thermal equilibration is to put a small Petri dish filled with some hot or cold water into a larger one filled with tap water around room temperature, as illustrated in Figure 1. Then stick one thermometer in the inner dish and another in the outer dish and take their readings over time.

With a low-cost IR camera like the FLIR C2 camera or FLIR ONE camera, this experiment becomes much more visual (Figure 2). As an IR camera provides a full-field view of the experiment in real time, you get much richer information about the process than a graph of two converging curves from the temperature data read from the two thermometers.

Figure 2

The complete equilibration process typically takes 10-30 minutes, depending on the initial temperature difference between the water in the two dishes and the amount of water in the inner dish. A larger temperature difference or a larger amount of water in the inner dish will require more time to reach the thermal equilibrium.

Another way to quickly show this process is to use our Energy2D software to create a computer simulation (Figure 3). Such a simulation provides a visualization that resembles the IR imaging result. The advantage is that it runs very fast -- only 10 seconds or so are needed to reach the thermal equilibrium. This allows you to test various conditions rapidly, e.g., changing the initial temperature of the water in the inner dish or the outer dish or changing the diameters of the dishes.

Figure 3

Both real-world experiments and computer simulations have their own pros and cons. Exactly which one to use depends on your situation. As a scientist, I believe nothing beats real-world experiments in supporting authentic science learning and we should always favor them whenever possible. However, conducting real-world experiments requires a lot of time and resources, which makes it impractical to implement throughout a course. Computer simulations provide an alternative solution that allows students to get a sense of real-world experiments without entailing the time and cost. But the downside is that a computer simulation, most of the time, is an overly simplified scientific model that does not have the many layers of complexity and the many types of interactions that we experience in reality. In a real-world experiment, there are always unexpected factors and details that need to be attended to. It is these unexpected factors and details that create genuinely profound and exciting teachable moments. This important nature of science is severely missing in computer simulations, even with a sophisticated computational fluid dynamics tool such as Energy2D.

Here is my balancing of this trade-off equation: It is essential for students to learn simplified scientific models before they can explore complex real-world situations. The models will give students the frameworks needed to make sense of real-world observation. A fair strategy is to use simulations to teach simplified models and then make some time for students to conduct experiments in the real world and learn how to integrate and apply their knowledge about the models to solve real problems.

A side note: You may be wondering how well the Energy2D result agrees with the IR result on a quantitative basis. This is kind of an important question -- If the simulation is not a good approximation of the real-world process, it is not a good simulation and one may challenge its usefulness, even for learning purposes. Figure 4 shows a comparison of a test run. As you can see, the while the result predicted by Energy2D agrees in trend with the results observed through IR imaging, there are some details in the real data that may be caused by either human errors in taking the data or thermal fluctuations in the room. What is more, after the thermal equilibrium was reached, the water in both dishes continued to cool down to room temperature and then below due to evaporative cooling. The cooling to room temperature was modeled in the Energy2D simulation through a thermal coupling to the environment but evaporative cooling was not.

Figure 4

Harvard study uses Energy2D to simulate origins of life

How do prebiotic oligomer synthesis reactions, crucial to the origins of life, occur in natural geological conditions? These reactions could happen through certain thermal cycles driven by natural nuclear reactors in the Precambrian period. Origins of Life and Evolution of Biospheres, the Journal of the International Astrobiology Society,  recently published a research paper by Dr. Zachary R. Adam from the Department of Earth and Planetary Sciences, Harvard University, titled "Temperature oscillations near natural nuclear reactor cores and the potential for prebiotic oligomer synthesis." The paper reported a theoretical study based on our Energy2D software.

The critical finding of the paper is that Energy2D models indicate that "water-moderated, convectively-cooled natural fission reactors in porous host rocks create temperature oscillations that resemble those employed in polymerase chain reaction (PCR) devices to artificially amplify oligonucleotides." The simulations show that "this temperature profile is characterized by short-duration pulses up to 70–100 °C, followed by a sustained period of temperatures in the range of 30–70 °C, and finally a period of relaxation to ambient temperatures until the cycle is restarted by a fresh influx of pore water."

This study concludes that "for a given reactor configuration, temperature maxima and the time required to relax to ambient temperatures depend most strongly on the aggregate effect of host rock permeability in decreasing the thermal expansion and increasing the viscosity and evaporation temperature of the pore fluids. Once formed, fission-fueled reactors can sustain multi-kilowatt-level power production for 105–106 years, ensuring microenvironmental longevity and chemical output."

This is the FOURTH published scientific research paper in a year -- as far as I know -- that involves the applications of Energy2D. This is a strong endorsement of its scientific validity from the science community. The software has impacted science, engineering, and education as I envisioned when starting its development six years ago. In the new year, the development will continue in my spare time. A recent update is that the latest version of Energy2D has provided both Windows and Mac installers that do not require Java to run, potentially making it easier for people to install and update the program.

An infrared investigation on a Stirling engine

Figure 1

The year 2016 marks the 200th anniversary of an important invention of Robert Stirling -- the Stirling engine. So I thought I should start this year's blogging with a commemoration article about this truly ingenious invention.

A Stirling engine is a closed-cycle heat engine that operates by cyclic compression and expansion of air or other gas by a temperature difference across the engine. A Stirling engine is able to convert thermal energy into mechanical work.

You can buy an awesome toy Stirling engine from Amazon (perhaps next Christmas's gift for some inquisitive minds). If you put it on top of a cup of hot water, this amazing machine will just run until the hot water cools down to the room temperature.

Figure 2

Curious about whether the Stirling circle would actually accelerate the cooling process, I filled hot water into two identical mugs and covered one of them with the Stirling engine. Then I started the engine and observed what happened to the temperature through an IR camera. It turned out that the mug covered by the engine maintained a temperature about 10 °C higher than the open mug in about 30 minutes of observation time. If you have a chance to do this experiment, you probably would be surprised. The flying wheel of the Stirling engine seems to be announcing that it is working very hard by displaying fast spinning and making a lot of noise. But all that energy, visual and audible as it is, is no match to the thermal energy lost through evaporation of water from the open hot mug (Figure 1).

How about comparing the Stirling engine with heat transfer? I found a metal box that has approximately the same size and same thickness with our Stirling engine. I refilled the hot water to the two mugs and covered one with the metal box and the other with the Stirling engine. Then I started the engine and tracked their temperatures through the IR camera. It turned out that the rates of heat loss from the two mugs were about the same in about 30 minutes of observation. What this really means is that the energy that drove the engine was actually very small compared with the thermal energy that is lost to the environment through heat transfer (Figure 2).

This is understandable because the speed of the flying wheel is only a small fraction of the average speed of molecules (which is about the speed of sound or higher). This investigation also suggests that the Stirling engine is very efficient. Had we insulated the mug, it would have run for hours.