Modeling Approach
For this assignment I chose to do a solar analysis in kwh/ft^2/day (for the day of May 20, 2022) and a panel directness value of different alternatives to views of San Francisco.
For 2 Units
Step 1 – Create two custom nodes to measure new evaluation metrics for the building form test cases created in Module 5
For both of my new nodes, I changed the BuildingForm.SelectWallSurfaces node that was provided in the examples to take in an input of dynamo geometry surfaces instead of grabbing Revit elements. This change allowed me to then select the wall surfaces of my dynamo generated design to develop my new evaluation metrics of Solar Analysis and Directness Value:
Solar Analysis Custom Node Walkthrough:
This custom node takes in weather data from San Francisco, CA from sunrise to sunset on May 20th, 2022 to calculate the total cumulative solar insolation value in wh/m^2 for each alternative. I then further changed the value to kwh/ft^2 to reflect the amount of solar energy that hits the building (if the building were fully passive):
Directness Value Computation Custom Node Walkthrough:
This node calculates the total directness of each alternative to two building forms (simulating directness to buildings in the Bay). I created two towers in Revit and then inputted these forms to be added into the total directness from my building form to the two buildings:
Step 2 – Integrate the custom nodes into Module 5 testing node logic and report the new evaluation values for each test case
My original form was an octagon shaped tower that can be twisted in the middle and top profiles. I originally used a radius of 100 ft and a top height of 500ft:
I made a copy of my testing function from Module 5 and incorporated the new evaluation nodes into it. In Module 5 I created an octagon building form using dynamo that changed the radius of three profiles and the top height of the volume created to test different alternatives. For this module, I incorporated the solar analysis and directness values as well.
I created a list of values from the new testing node for each test case that included the input parameter values I tested, the gross floor area, gross surface area, and gross volume, the two new evaluation metrics, and an extra calculation of the total wall area. Though I did not use the wall area calculation for this assignment, I plan to use it in the future for calculation metrics. As far as the metrics I tested, I am using the radius generated from using dynamo geometry like I did in Module 5 and a top height value that changes the total height of the building.
Input parameter values tested: 70, 75, 80, 85 ft radius
Top values tested: 400, 500, 600 ft
For 3 Units
Step 3 – Design a Single-Objective Optimization Scheme for Using the Computed Evaluators to Compare and Rank the Building Form Alternatives
For my main design strategy, I weighted the solar insulation two times more than views. This is because I want to conduct more energy analysis into my building forms in the future. While directness to other buildings is good to have in terms of views and distance, I think energy analysis deserves a higher weight because it can help design better strategies for lowering the energy needed to power a building. Especially in the Bay Area, places are more likely to have heating loads and use passive ventilation as a form of cooling. Therefore, I made building directness weighted to a value of 1 and solar analysis weighted to a value of 2. I used a custom node to normalize the values. Here is an overview of the total custom node:
The custom node outputs are as shown:
Step 4 – Implement Single-Objective Optimization Scheme
I created a list of the maximum and minimum values for each test case to determine how to normalize and weigh my two main evaluations of solar analysis and directness value equally:
By using the custom node provided in the example, I edited the node to reflect the two values I was interested in assessing. The explanation of how I weighed the values is in Part 3:
I found that the 80 ft radius, 600 ft tall tower had the best directness to the two towers (by a couple small decimal places) and also had a high value for solar insulation. The 80 ft radius, 500 ft tall tower and the 85 ft radius, 500 ft tower had the same total weighted score so they were the 2nd and 3rd best alternatives based on the tests.
The best score is shown in gold and the two following scores are highlighted in silver:
For 4 Units
Step 5 – Create a Simple Revit or Rhino Model Illustrating the Recommended Design
To visualize the best alternative, I used the node I created in Module 5 to generate a visualization of the best alternative that takes the input of the radius and top height for the alternative with the best score. I then grabbed the wall surfaces and panelized them to be able to place adaptable components onto the external walls.
I used the building wall node that I adapted for a dynamo form again so I would be able to use my dynamo generated wall surfaces instead of grabbing them from a Revit form:
Step 6 – Provide Visual Feedback By Adjusting the Appearance of the Panels
I adapted the panels based on their directness to the sun. By using the Rect_Panel with Resizable Opening: glazing adaptive component I changed the panels depending on their outward normal values to the sun which then changes their color depending on how much solar exposure each panel receives and increases Frame Width A and B if the panels have more sun exposure:
Revit Visualization with solar settings and shadows:
Revit Form using adaptive panel forms:
Lessons Learned:
1) I would like to further expand my weighting system so that it does not just depend on the two values. A more sophisticated weighting system would allow me to better compare alternatives in more detail so that two values do not end up with the same weight. Some of the metrics that I would like to further explore are the cost per floor and total cash flows that would be possible assuming 20% affordable housing for a building and/or mixed use leasing spaces.
2) There are so many metrics that can go into design and it is hard to just choose a couple to consider. Moving forward I would like to focus on energy efficiency and calculating building loads because I am also working on an office building project for the class CEE 226E. So far we have mostly worked on energy modeling in terms of building loads and values, but I think I would be able to grasps concepts in both that class and this class if I can visualize the building loads and costs per building floor.
3) Custom nodes are interesting to incorporate into designs. At first, I was intimidated by the ability to develop custom nodes, but I feel more confident now that I was able to adapt nodes to give the outputs that I wanted and have the proper inputs as needed.