Department of Architecture

Office : Design Building Room 654

Phone : (02)2771-2171 Ext. 2915

Email : huangch@ntut.edu.tw

Education : Dr.-Ing. Karlsruhe Institute of Technology

Expertise : Eco-city/urban heat island,Urban ecology and urban design, Sustainable low-carbon buildings, Sustainable community/building planning,

roject subject :

Building energy saving and net-zero emissions, improving urban thermal environment

Project goals :

• Simulation of urban regional heat dissipation efficiency of mesoscale wet cooling towers
• Simulation of heat flow and air flow field in double-skin facade (DSF) wall
• Research on urban canopy heat island and boundary layer heat island
• Low carbon energy saving building

Research methods and applications :

1. Application of Cradle CFD scSTREAM

Simulation of regional urban heat dissipation efficiency of mesoscale by wet cooling towers

Simulation of regional urban heat dissipation efficiency of mesoscale by wet cooling towers.

The results show that the tower can reduce the temperature up to 1.5°C within 100 meters around the tower, showing its possibility of improving the urban heat island effect.

Simulation of heat flow and air flow field in double-skin facade (DSF).

Simulation of heat flow and air flow field in double-skin facade (DSF).

The results show the potential to reduce temperatures at pedestrian heights by 1°C-3°C.

 

2. Experimental measurement of urban thermal characteristics of vertical structures on the ground surface

The texture of the ground surface has a specific impact on its vertical structure. Environmental physical parameters such as heat capacity, latent heat, wind field, heat flux, and heat enthalpy were measured in the field for observation and analysis.

A four-axis UAV equipped with sensors is used to explore the vertical structure of the heat island. The results show that the texture of the ground surface has a specific impact on the vertical structure, and the form of urban planning does have a huge impact on the thermal environment.

 

Applying big data, machine learning, structural equations, Python algorithms

 Statistical discussion of the average floating height and standard deviation of the boundary layer and canopy, and structural equation model analysis to explore what are the most important factors affecting urban thermal environment in different urban types. Deep learning explores the shape of the entire vertical structure of heat on the earth's surface.

Using Python programming language, NumPy mathematical function library and Pandas data operations to organize and classify data, Statistical correlation between heat islands and meteorological big data - air pressure, temperature, humidity, wind speed, wind direction, cloud cover, etc. would be analyzed.

Using deep learning to process a large amount of data, automatically process data, and use variables: temperature, humidity, enthalpy, altitude, wind speed, and cloud cover, patterns would be derived from deep learning by extracting seasons, time periods, and comfort levels. Through multiple iterations, gradient descent is used for parameter optimization to bring the loss function of the predicted value closest to the true value.

Research results :

1. Impact of asphalt concrete pavement on urban thermal environment

Asphalt concrete contributes to the increasingly serious urban heat island effect. Changing the thermal properties of materials is a key method to reduce urban heat islands. By changing factors such as thermal conductivity, specific heat, reflectivity, and emissivity, it promotes surface cooling.

 

2. 2. Urban Heat dissipation strategies beyond urban boundary layers UHI

• The thermal physical parameters of asphalt pavement were obtained experimentally.
• Measured radiant cooling materials improve asphalt to radiate heat through the urban boundary layer, reducing heat accumulation in the city.
• To Analyze and predict the effect evaluation of the cooling pavement through simulation calculations.

Heat dissipation strategies beyond urban boundary layers UHI

Solar radiation causes the shell walls to accumulate heat sources, causing the heat energy between the walls to be dissipated into the indoor environment as heat, indirectly affecting the indoor environmental temperature of the building.

The shell wall receives most of the heat source and transmits it to the room. In addition to absorbing the total amount of solar radiation, it also takes into account the direct reflection of the wall itself, the long-wave radiation released after absorption by the wall, and the heat convection generated by the air and the exterior wall.

3. Long-wave cooling effect of radiant cooling coating on building exterior walls

 

4. Urban heat dissipation simulation beyond the urban boundary layer UHI

• Using machine learning to analyze sounding balloon meteorological big data, we explore the vertical structure of urban heat islands.
• Through the inversion calculation of python in the previous step, we found that high and low cloud cover seems to have little effect on the boundary layer UHI fluctuation.
• Wind speed can effectively reduce the height of the urban boundary layer. High wind speed will directly blow away the ground heat, so the boundary layer is lower. The difference between high and low wind speeds affecting the boundary layer is about 400m.
• The boundary layer is highly generalized under the influence of humidity and appears to be less regular.
• In winter, the floating difference between the two urban types widens.
• The results show that wind speed can effectively reduce the height of the urban boundary layer ~
  The average boundary layer height at low wind speed is about 450-500m.
  The average boundary layer height of high wind speed is about 100-150m.
• The high induction of the influence of humidity on the boundary layer shows the irregularity of the pattern of humidity.
• High and low cloud cover seems to have little effect on boundary layer fluctuations.

Machine learning analyzes meteorological big data and explores the vertical structure of urban heat islands