Machine Learning as Surrogate to Building Performance Simulation: A Building Design Optimization Application

2018 ◽  
pp. 94-102
Author(s):  
Sokratis Papadopoulos ◽  
Wei Lee Woon ◽  
Elie Azar
Keyword(s):  
Machine Learning ◽  
Design Optimization ◽  
Building Design ◽  
Building Performance ◽  
Performance Simulation ◽  
Building Performance Simulation

scholarly journals Balancing daylight and overheating in low-energy design using CIBSE improved weather files

2019 ◽  
Vol 41 (2) ◽  
pp. 210-224 ◽  
Author(s):  
Eleonora Brembilla ◽  
Christina J Hopfe ◽  
John Mardaljevic ◽  
Anastasia Mylona ◽  
Eirini Mantesi
Keyword(s):  
Solar Radiation ◽  
Building Design ◽  
Low Energy ◽  
Solar Irradiation ◽  
Building Performance ◽  
Radiation Model ◽  
Performance Simulation ◽  
Building Performance Simulation ◽  
Irradiance Data ◽  
Solar Radiation Model

A new set of CIBSE weather files for building performance simulation was recently developed to address the need for better quality solar data. These are essential for most building performance simulation applications, particularly for daylighting studies and low-energy building design, which requires detailed irradiation data for passive solar design and overheating risk analysis. The reliability of weather data becomes paramount when building performance is pushed to its limits. Findings illustrate how principles of good window design can be applied to a case study building, built to the Passivhaus standard, and how its expected performance is affected by the quality of solar irradiation data. Analyses using test reference years were most affected by changes in the solar radiation model (up to 8.3% points), whereas for design summer years the maximum difference was 1.7% points. Adopting the new model caused overheating risk to be classified as more severe using test reference years than design summer years, prompting a discussion on the design summer year selection method. Irradiance data measured on-site were used as a benchmark to evaluate the new solar radiation model, which was found to significantly improve the accuracy of irradiance data within weather files and so the reliability of overheating assessments. Practical application: CIBSE weather files are widely used for compliance verification of building performance in the UK context. This paper tests how the introduction of a new solar radiation model in weather files will affect daylighting and overheating simulation results. Examples are given on how low-energy building design considerations driven by advanced simulation techniques can help reaching indoor visual and thermal comfort requirements.

scholarly journals [ANT]: A Machine Learning Approach for Building Performance Simulation: Methods and Development

2019 ◽  
Vol 3 (1) ◽  
pp. 205
Author(s):  
Mahmoud M. Abdelrahman ◽  
Ahmed Mohamed Yousef Toutou
Keyword(s):  
Machine Learning ◽  
High Performance ◽  
General Purpose ◽  
Ease Of Use ◽  
Building Performance ◽  
Performance Simulation ◽  
Building Performance Simulation ◽  
Wide Range ◽  
Performance Development ◽  
High Level

In this paper, we represent an approach for combining machine learning (ML) techniques with building performance simulation by introducing four methods in which ML could be effectively involved in this field i.e. Classification, Regression, Clustering and Model selection . Rhino-3d-Grasshopper SDK was used to develop a new plugin for involving machine learning in design process using Python programming language and making use of scikit-learn module, that is, a python module which provides a general purpose high level language to nonspecialist user by integration of wide range supervised and unsupervised learning algorithms with high performance, ease of use and well documented features. ANT plugin provides a method to make use of these modules inside Rhino\Grasshopper to be handy to designers. This tool is open source and is released under BSD simplified license. This approach represents promising results regarding making use of data in automating building performance development and could be widely applied. Future studies include providing parallel computation facility using PyOpenCL module as well as computer vision integration using scikit-image.

scholarly journals Using artificial neural network and WebGL to algorithmically optimize window wall ratios of high-rise office buildings

2021 ◽  
Author(s):  
Shenghuan Zhao
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Web Application ◽  
Building Design ◽  
Building Performance ◽  
Performance Simulation ◽  
Building Performance Simulation ◽  
Design Projects ◽  
Artificial Neural ◽  
Simulation Engines

Abstract By coupling parametric modeling, building performance simulation engines, and optimization algorithms, optimal design choices regarding predefined building performance objectives can be automatically obtained. This becomes an emerging research topic among scholars in the fields of architecture and built environment. However, it is not easy to apply this method to real building design projects, because of two main drawbacks: Building performance simulation is too time consuming, and the numerical visualization of final results is not intuitive for architects to make decisions. Therefore, this study tries to fill these two gaps by training an artificial neural network to replace simulation engines and developing a web application to speed up the 3D visualization of selected design choices. These two strategies are applied to optimize office towers’ window wall ratios in Hangzhou, China. Architects working on new design projects in that city can obtain the optimal group of window wall ratios for four facades in 2 s, faster than using simulation engines, which cost architects 2 weeks. Moreover, architects can also efficiently observe the appearance of design solutions with the web application. By improving its usability from these two aspects, this study significantly improves the applicability of algorithmic optimization for building design projects.

scholarly journals Assigning Robust Default Values in Building Performance Simulation Software for Improved Decision-Making in the Initial Stages of Building Design

2015 ◽  
Vol 2015 ◽  
pp. 1-11 ◽  
Author(s):  
Kyosuke Hiyama
Keyword(s):  
Design Patterns ◽  
Natural Ventilation ◽  
Building Design ◽  
Simulation Software ◽  
Building Performance ◽  
Performance Simulation ◽  
Existing Building ◽  
Building Performance Simulation ◽  
Convincing Argument ◽  
The Ideal

Applying data mining techniques on a database of BIM models could provide valuable insights in key design patterns implicitly present in these BIM models. The architectural designer would then be able to use previous data from existing building projects as default values in building performance simulation software for the early phases of building design. The author has proposed the method to minimize the magnitude of the variation in these default values in subsequent design stages. This approach maintains the accuracy of the simulation results in the initial stages of building design. In this study, a more convincing argument is presented to demonstrate the significance of the new method. The variation in the ideal default values for different building design conditions is assessed first. Next, the influence of each condition on these variations is investigated. The space depth is found to have a large impact on the ideal default value of the window to wall ratio. In addition, the presence or absence of lighting control and natural ventilation has a significant influence on the ideal default value. These effects can be used to identify the types of building conditions that should be considered to determine the ideal default values.

On the impact of local microclimate on building performance simulation. Part I: Prediction of building external conditions

2019 ◽  
Vol 12 (5) ◽  
pp. 735-746 ◽  
Author(s):  
Lucie Merlier ◽  
Loïc Frayssinet ◽  
Kévyn Johannes ◽  
Frédéric Kuznik
Keyword(s):  
Building Performance ◽  
Performance Simulation ◽  
Building Performance Simulation ◽  
External Conditions ◽  
The Impact

Assessment of solar radiation data quality in typical meteorological years and its influence on the building performance simulation

2021 ◽  
pp. 111251
Author(s):  
Facundo Bre ◽  
Rayner Maurício e Silva Machado ◽  
Linda K. Lawrie ◽  
Drury B. Crawley ◽  
Roberto Lamberts
Keyword(s):  
Solar Radiation ◽  
Data Quality ◽  
Building Performance ◽  
Performance Simulation ◽  
Building Performance Simulation ◽  
Radiation Data
Sign in / Sign up

Export Citation Format

Share Document