Contains a python script for preparing, searching and post-processing aerodynamic database for buildings. The database holds pressure time-series measured at points(taps) distributed over the building surface. The aerodynamic data set can be:
- Experimetal measurment
- CFD simulation (i.e., LES)
- Full-scale feild measurment
The aerodynamicDatabaseLib.py
is a python library to save and retrive any database entry. The core class of the library is the WindLoadData
class which contains all the atributes of a database entry. The library also holds HighRiseData
and LowRiseData
that implement fuctions for manipulating wind load data for high-rise and low-rise buildings, respectively.
The database is saved into JSON file format. Each database entry is stored using two associated files. The first file ends with "_info.json" and contains main atributes of the database entry, which include the building geometry, wind characterstics, data type, unit system, etc. The HighRiseData
class has a function implemented to write this file for high rise building. Typical "_info.json" file for aerodynamic database looks like this:
{
windSpeed: 10.6924
width: 0.2
depth: 0.1
height: 0.3
heightToWidth: 1.5
widthToDepth: 2
duration: 32
timeUnit: "sec"
lengthUnit: "m"
samplingRate: 1000
windDirection: 0
exposureType: "Open"
roughnessLength: 0.03
dataType: "CFD"
fileName: "HR_CFD_1.50_2.00_0.00_0.030"
}
The other file ends with just ".json" and holds all the information "_info.json" file contains and additional data about measurment locations and the actual pressure time-series. This file is usually big and the "_info.json" is used instead for searching purpose.
Each database entry is named based on the following convention. The name contains building type (high-rise or low-rise), data type (experiment, CFD or field), height to width ratio, width to depth ratio, wind direction, aerodynamic roughness length (z0). For example HR_CFD_4_1.5_90_0.03
represents a high-rise building aerodynamic database extracted from CFD simulation with height/width and width/depth ratio of 4 and 1.5 for 90 degree wind direction in a terrain having 0.03 m aerodynamic roughness height.
For searching a particular entry in the database, generally the following information is required:
- Building type (high-rise or low-rise)
- Height to width ratio
- Width to depth ratio
- Wind direction
- Exposure condition i.e. aerodynamic roughness length (z0)
The current capability of the aerodynamic database is very limitted. Future work is needed to enhance its capability in the following key directions:
- Populate the database with more data (both experimetal and CFD). Performing CFD simulations variying different parameters listed above seem a feasible direction.
- Implement database interpolation technique for cases that can not be found in the database. This requires developing interpolation scheme based on devreduced order modeling or machine learning techniques.