Energy3D is a free, open-source tool we are developing from scratch to empower students to design, make, and test energy-efficient model houses.

Today we had some students design and make their own houses. One student succeeded in designing a model house after her own real house. The first two screenshots show her model under the sun in different months.

Like many architecture design tools, students can “walk” into their own design and imagine “living in the house” virtually. The other two screenshots show two close-ups: one from the outside and the other from the inside. The fifth image is a physical house made of foam board and assembled, based on this design.

If they are satisfied with their designs, students can “print” out their houses, cut out all the pieces, and assemble them.

We feel that computer-aided design tools such as Energy3D would be a big help to students when they are undertaking complicated engineering design challenges such as making a house. 3D reasoning is usually difficult for most students. A What-You-See-Is-What-You-Get (WYSIWYG) CAD tool can help them think through.

You may be wondering why we want to develop this tool. Many students complain that their science and engineering projects in schools are not challenging enough to be interesting. Many teachers do strive to make their student projects more attractive. However, they lack appropriate educational tools to do so. Energy3D is an attempt to provide teachers and students with cutting-edge tools that can teach and learn modern full-cycle engineering processes–from design to manufacturing to test– through an interesting project about energy-efficient houses. We hope this tool would intrigue, inspire, and prepare students for STEM careers.

PS on 7/28: This is another building designed by the same student, which shows an intersecting roof. These examples show that Energy3D could be used to design quite a variety of architecture. It turns out that roof is the most difficult part to design using tools such as Google’s SketchUp. We are trying to simplify that part by figuring out algorithms that would enable easy editing of roofs. Our work focuses on two directions. First, an algorithm is needed to automatically generate a roof of a given type based on the boundary walls the user has laid. Second, the topological transformations between different types of roofs need to be identified so that we can build the user interface for adjusting the roof easily.