Thermal Insight would be a standalone app that optimizes indoor environments by predicting thermal comfort. With simulations from trained models to help enhance occupant comfort and energy efficiency, evaluating metrics like PMV, PPD, and MRT. This tool would help improve well-being and productivity while achieving efficiency goals.
Abstract
- Objective: Improve thermal comfort in residential buildings in Istanbul using AI-driven analysis and historical data.
- Scope: Analyze thermal comfort elements such as temperature, humidity, and building design in residential buildings within Istanbul using simulation process.
Methodology
- Data Collection:
- Climate Data: Historical temperature, humidity, and seasonal variation data from Istanbul.
- Building Data: Design and thermal performance data of residential buildings.
- Simulation Tools: Grasshopper, Ladybug, and Honeybee for simulations.
- AI Models: Implementation of Pix2Pix for thermal comfort prediction.
Use Case
- Components:
- MRT Model: Utilize the Mean Radiant Temperature (MRT) model for thermal comfort analysis.
- Flask Web Application Script: Develop a web application using Flask to interface with the model.
- PIX2PIX Model: Use the Pix2Pix model in Keras for thermal comfort prediction and visualization.
Data Analysis
- Simulation Data:
- Generated Unit Types: Distribution of different types of units generated.
- Floor Plan Elements: Proportions of different elements (windows, walls, doors) in floor plans.
- Thermal Analysis Results: Categorization into hot and cold thermal analysis.
- AI Predictions: Accuracy and efficiency of Pix2Pix model predictions.
Results and Discussion
- Findings: Summary of key findings from simulations and AI predictions.
- Implications: Discuss the implications of findings for improving thermal comfort.
- Recommendations: Suggestions for building design improvements and policy changes.
Outlook
- Expansion: Plans for expanding the project to other cities.
- Continued Research: Areas for further research and development.
AIA Studio Seminars
Project Team:
- Manolya Nielsen
- Hetal Bharwanu
- Agustin Salas
Thermal Insight: Optimizing Indoor Analysis is a project of Iaac, Institute for Advanced Architecture of Catalonia developed at Master in Advanced Computation for Architecture & Design in 2023-24 by Students: Manolya Nielsen, Hetal Bharwanu, Agustin Salas, under the supervision of Faculty: Angelos Chronis, Vasiliki Fragkia, Stanislava Fedorova.
