- For 2 or More Units: Create Two New Evaluator Nodes
I designed the node to create the calculation of total construction price and labor cost and DeadLoad. In this way, we can directly obtain the new project cost after adjusting different parameters. DeadLoad is one of the examples I use to calculate building load forces. We can automatically calculate new loads without having to repeat calculations for each new design. Below are the sources of the data I used：
“A study found that transparent BIPV panels installed on office building windows can reduce energy costs by up to 33.5 percent.”
- For 3 or More Units: Develop a Single-Objective Optimization Scheme
First, combine the two custom nodes in stage1 into a single node. What should be noted here is the output composition of the list. The problem I have here is my use of list.map, which can convert the function into the required list for subsequent use. When Rescale Column Values and Compute Combined Score, resultsdataRow is not connected so that both subsequent functions can use the data. That is to say, I can get the numerical value directly after trying to connect, but I cannot directly get the maximum and minimum corresponding data later.
According to the influencing factors and their proportions that I set that I think are important, the larger the data obtained, the more in line with my design expectations. I think the best choice is 400-height and 30-rotation because 30 degree’s rotation and 40 degree’s rotation have same optimization weight values and 30 degree’s rotation need less construction efforts.
Sometimes there is too much intuitive data that prevents us from judging which design is the best. Through weighted design, we can design corresponding weights according to the aspects we are concerned about, so as to directly obtain a numerical value for us to directly judge.
- For 4 Units: Visualize the Recommended Alternative
What should be noted here is that we need to choose the appropriate panelized family type, otherwise it will not be used on buildings with distorted designs. Secondly, we need to remove nulls in the list, otherwise subsequent adjustments will not be possible. Adding more watch nodes will help identify existing problems
Small design changes in architectural form can cause disparities in unpredictable outcomes. The calculation of the Dead Load of a building is important for structural design, so we can choose the one that is more important. The overall design cost will also affect the choice. For example, a design with a small gap will lead to a large price gap, so we will choose the cheaper one.
Use node to calculate and analyze the average and relative levels of solar energy to guide us to choose a design that is more in line with expectations. For example, we can choose a design with a higher solar energy reception rate to utilize more solar energy.
I pay more attention to the result of designing GFA, because more area can be used to be more efficient and maintain sustainable, so I designed his weight to 2.GSA and GV are equally important to me, because these are what we The scope of green and sustainable applications can be measured. The 400-height and 30-rotation design met my needs: more interior space and a suitable area for sustainable use. While other designs had similar results, this set of designs was less difficult to construct.
There are indeed subtle differences that cannot be seen in a single assessment. The gap between the total construction price of the design result and the Dead Load is often not huge, which makes it impossible for us to immediately determine which design is the optimal solution. Optimization Scheme can better reflect the differences caused by different designs than a single evaluation.