Project 2 - Algorithms and Scripting for Parametric Design

 

Practice:  British Museum Great Court

The project's goal was to improve the roof of the British Museum's Great Court by refining the initial parametric model from Project 1. This involved using advanced modeling techniques to focus on different parametric modeling aspects. This phase was a continuation of Project 1, which used the Kangaroo physics engine to simulate dynamic effects, and aimed to make important corrections to the first version of the project. 

To achieve this, tools like Rhino/Grasshopper for parametric modeling, genetic algorithms for optimizing design elements such as cost, daylight factors, and structural performance, and AI technologies for predictive analytics and visualizing the final model were used. Scripting within Grasshopper automated cost calculations and material optimization, while the Ladybug and Honeybee plugins provided environmental simulations to enhance daylight.

    

Cost and Material Specifications

The project utilized Grasshopper's scripting capabilities to establish material thickness and specification parameters. One crucial aspect was setting the unit prices for each material used, which facilitated the automated calculation of the overall cost. Accurate cost assessments and material takeoffs were achieved by specifying the precise areas of glass panels and deducting the metal framework. This approach ensured that the project remained within the budgetary constraints.

Simplification of Mesh Geometry

To facilitate the practice of different complex optimization methods, the project incorporated a simplified mesh rectangle instead of the original complex pattern. This adjustment allowed for more straightforward application of optimization algorithms, reducing computational overhead and simplifying the analysis without compromising the depth of the optimization processes.

Daylight Analysis

To evaluate the daylight autonomy factor, we used Ladybug and Honeybee plugins. Based on the input material specifications, we adjusted the Skylight Ratio component. Additionally, we imported weather data for College Station to calculate the natural light distribution within the structure.

Structural Analysis

During the structural analysis phase, Karamba3D was utilized to establish and define the necessary load supports and material cross-sections for evaluating the structural integrity of the design. This setup allowed displacement analysis under different load conditions, providing valuable insights into the structural stability of the design and the effectiveness of the chosen materials and their configurations.

Single Objective Optimization

Galapagos was used for single objective optimization, focusing on individual parameters such as cost reduction, structural integrity, or daylight enhancement. This method allowed for targeted improvements, making it possible to isolate and refine specific aspects of the design one at a time, thereby achieving optimal solutions for distinct challenges.

Multi-Objective Optimization with Wallacei

Wallacei was utilized for multi-objective optimization, simultaneously addressing three key objectives: Minimizing cost efficiency, Minimizing daylight autonomy, and minimizing structural displacement. The U and V indices, along with material dimensions, were inputted as parameters into the genetic algorithm for a thorough evaluation of potential design configurations.

AI-Enhanced Rendering of 3D Models

A tool powered by artificial intelligence (https://mnml.ai) was incorporated into the project to render the 3D model, which improved the visualization process.