Data-driven and numerical approaches to predict thermal comfort in traditional courtyards

Highlights

• Thermal comfort in courtyards is studied by means of Envi-met software and genetic programming.

• A new formulae is developed using Genetic Programming to predict thermal comfort in courtyards.

• The Envi-met simulations are validated using field measurements.

• A set of practical charts for the prediction of thermal comfort in historical courtyards is developed.

• Error band charts are used to assess the reliability of the GP developed models.

Abstract

This paper studies the climactic performance of the 10 traditional courtyards located in warm-dry climates of Kashan and cold climates of Ardabil based on shading and sunlit coverage. The modelling process comprises two sections: first, a number of numerical simulations are run using Envi-met software to detail the shading and sunlit percentage, PET and PMV in the samples of interest. These numerical models are validated on the basis of the results made available by field observations. Such validation revealed an excellent agreement between the numerical solution and the benchmarking data. Afterwards, GP is used to evolve some equations for predicting PET and PMV using the data points derived from the numerical simulations.

The results suggest that regarding the thermal indices (PET and PMV), there is a high correlation between the shadow and sunlit effects and thermal comfort in Kashan's houses in comparison with Ardabil houses. However, in tropical regions (Kashan), summer shading and winter sunlit have a greater effect on thermal comfort and temperature adjustment than cold regions. Moreover, the statistical criterion, as well as reliability analysis and contour plots show that the GP developed formulas can be exploited in predicting the PET and PMV based on the shading percentage.

Introduction

Today, compatibility with climate and ecological sustainability are among the key issues in many fields of science and technology. Issues like thermal comfort, energy-saving and controlling the energy exchange between humans, the artificial and natural environments are an integral part of the urban design studies. Therefore, due to the increasing population and urban density as well as increasing the earth heat, the attention is drawn to the effect of design on outdoor thermal comfort. In order to gain access to sustainable designs in populated places, it has been tried to improve the quality of open spaces and provide comfort. As an open or semi-open space, the traditional courtyards are of interest in contemporary architecture. They play an important element in absorbing radiant energy, reducing energy consumption and providing the residents with thermal comfort. The problems associated with thermal comfort can cause significant unforeseen expenditures to courtyards’ functionality. Therefore, study the effects of contributing parameters on thermal comfort in courtyards is fundamental.

Thermal comfort is a mental condition that expresses the satisfaction of the thermal conditions of the environment. The thermal comfort will be possible only when the person is in thermal equilibrium (no heat is stored in the body). The type of user activity, type of coverage, as well as individual preferences are effective in human perception of thermal comfort [1]. In the exchange of human body heat with the surrounding environment, the clothes are an effective factor and reduce the body's contact with the environment [2]. The type of coverage is an insulator chosen by humans to adapt themselves to the surrounding environment. Also, the amount of energy resulting from body metabolism is a function of human activity that directly affects the thermal comfort of the body. Typically, the total energy resulting from muscle movement is converted into heat energy. The human body does not have separate receptors for feeling the ambient temperature; that’s why the human perception of the environment condition is dependent on the air temperature, relative humidity, wind, and sunlight make [3]. Regarding the above-mentioned factors, it is essential to conduct climatic studies and identify the environmental features affected by climate, as well as to use its results in optimizing the space structures in the designs especially in tropical areas.

It is clear that one of the design goals in tropical areas is to prevent sunlit and create shadows on the surfaces and walls of the building, while in cold areas, avoiding the shadows on the surfaces is very necessary. The proportions, directions, materials and form of the traditional courtyard play a major role in determining the amount of radiation absorbed in the various levels of the yard and providing the visual and thermal comfort for the residents.

Famous Indicators such as the physiological equivalent thermometer (PET) and the predicted mean vote (PMV) indices have been suggested as the most important criterion in human thermal comfort. The PMV index is a 7-degree thermal energy distribution scale which ranges from −3.5 (cold) to +3.5 (hot). Zero on this scale indicates a neutral heat sensation. The PMV index actually estimates the average score of a large group of people (in terms of thermal comfort). The PET is a well-known physiological temperature indicator derived from the human body energy balance equation, which is shown in Celsius degrees. Table 1 summarized the ranges of PET and PMV indexes with the equivalent thermal comfort.

Attempts have been made to provide a warm new outdoor comfort since the 1970s. These studies first focused on the indicators of the thermal comfort in the closed space and its remaking with respect to outdoor conditions. In addition to climatic conditions, thermal comfort is affected by the surrounding environment, surface cover, evaporation and transpiration, shading, or the factors created by natural and man-made factors. The latter factors help us to measure the performance of designed spaces and utilize the climatic elements affecting thermal comfort. Shabbier [4] showed that the summer temperature in the streets in where the height to width ratio equals to one is four times larger than that in comparison with the streets with the height to width ratio equals to three.

Soflaei et al. [5] studied and analyzed the shadow of ten houses in Kerman. They introduced the optimal traditional courtyard in terms of dimensions and angle of deviation to the north by considering the shadow based on the thermal comfort temperature. Mahmoudi et al. [6] provided the effective thermal design criteria to enhance the design of open spaces with a climate approach by comparing the physiological equivalent temperature and environmental physical factors. Taghvaie et al. [7] in their research on Shiraz gardens concluded that Jahan Nama garden was 11 °C cooler in summer and in winter warmer than the local climate. In Delghsha garden, it is cooler 2 to 7 °C in the summer and warmer than the local climate in the winter. Despite the fact that there are many studies on energy issues and thermal comfort indicators, the effect of shadow and radiation on thermal comfort indicators has not been studied adequately.

One of the most successful samples of climatic responsive architecture is traditional courtyard houses located in the hot climate of Iran, which have been designed with careful attention to climatic requirements through various passive design strategies like Showdan, Khishkhan, Shabestan, Hozkhaneh, courtyard, windcatcher, and air-vent of dome-roof to provide indoor thermal comfort for residents [5]. According to Fathy [8], one of the areas to look for energy-efficient architecture might be to go back and reinvestigate traditional architecture such as courtyards houses. He believed that in the context of sustainability where ecological issues are of prime importance, the courtyard can still be perceived as an important design element that functions both as a social space as well as something that reduces the carbon footprint of the building. A courtyard housing typology is a set of building types of familial and historic dwelling that the living spaces is arranged around a central courtyard [5] . It is one of the world's oldest housing typology that can be pursued back to 5,000 years ago, when were constructed in the Middle East and China [8], as a protective mechanism both from harsh weather and unfriendly neighbours [9].

Zamani et al. [10] investigated that design of courtyard in residential buildings is an efficient strategy to obtain sustainability and energy consumption. They also mentioned that details of courtyards in houses can reduce energy consumption and better microclimate of buildings. In this study, most significant of courtyard design (such as opening, geometry, and orientation) were assessed by evaluating the energy performance, indoor temperature and natural ventilation in a different climate. Manioglua and Orala [11], explained that the geometry of the courtyard form (shape) affects considerably the shadows produced on the building envelope, and consequently the received solar radiation and the cooling and heating loads in hot-dry climate. They defined the courtyard’s shape based on the shape factor (W/L); the ratio of courtyard width (W) to courtyard length (L). They also examined the variation of the obtained heating and cooling loads as a result of changing the building form with the proportion of the courtyard. Heidari [12] proposed a design guideline for courtyards in the desert climate of Yazd based on the air movement and thermal comfort results of case studies. He concluded that humidity in the studied courtyards could be improved through landscaping, and a pond could significantly affect thermal comfort. The depth-to-width ratio of a courtyard is an important factor of airflow pattern. Based on Saeed [13] investigation, the shapes of courtyards nowadays have taken more dynamic forms other than the rectilinear shapes of the traditional footprint, in response to several factors such as a designer’s creativity, as well the contextual needs such as site restriction, and specific functions. This has resulted in the creation of new and modern shapes typified as U, L, T or Y. Biabani Moghadam Baboli et al. [14] worked on design variants and the adaptive role of the traditional courtyard houses in the moderate climate of Iran. They presented thirty-four cases of courtyard houses in the moderate climate of Babul in terms of its adaptive potential to suit the users’ comfort requirements. Their findings illustrated that the plan of traditional houses was designed to provide natural cross-ventilation, with proper utilization of openings in both the northern and southern side of the building. Rajapaksha et al. [15] investigated the potential of a courtyard for passive cooling in a single-story high mass building in a warm humid climate. Their results revealed that there is a significant correlation between wall surface temperatures and indoor air temperatures. Mouslia and Semprinia [16] analyzed the influence of the thermal performance of building structures and natural ventilation on the indoor thermal comfort for traditional houses and courtyard located in Damascus old city. Their results showed the importance of high thermal mass structure, high-performance materials, openings (areas and positions), the size and shaded area, the presence of trees and water fountain, which have effectively effect on natural ventilation to provide indoor thermal comfort for residents. Hassan [17] investigated the potential of a ventilated courtyard for passive cooling in a small building in a hot desert climate in New Aswan City, Egypt. The results show that the courtyard orientation and the courtyard geometry are among the most significant factors, which affect the thermal performance of the courtyard building model. Ahmad et al. [18] studied a traditional courtyard house within a six centuries old indigenous urban cluster and compared it to a modern detached house within a new urban development under summer and winter climates of Ghadames, Libya. They showed the thermal comfort superiority of an indigenous courtyard house over a modern pavilion-type house. Soflaei et al. [19] have focused on the thermal performance of traditional courtyards in the hot-arid climate of Iran. They conducted several field survey research, to analyze six important courtyard design variants including the orientation, extension, rotation angle, dimensions, and proportions of enclosed and open spaces, as well as physical bodies (opaque walls), transparent surfaces (openings), and natural elements (water and soil) in 29 valuable traditional houses from 10 ancient desert cities of Iran, where located in four mesoclimates of BWhs, BWks, BShs, and BSks. Results of this comprehensive quantitative study, show that Iranian traditional central courtyards were designed based on a careful attention to orientation and geometrical properties regarding the physical and natural parameters to act as an effective microclimate modifier.

The studies of thermal comfort and energy in various buildings are categorized into three groups: experimental or field evaluations, numerical simulations and data-driven approaches. All the mentioned approaches are employed in the present paper. First, using field measurements of thermal comfort, a dataset is provided to benchmark the numerical and data-driven models. Then, the collected dataset is used to validate the numerical simulations. Finally, the verified results of numerical simulations are used to evolve the soft computing models using GP.

Being user-friendly, producing robust results and owing easy-to-use interface, Envi-met software is very convenient to utilize in analyzing the thermal comfort of various buildings and courtyards. Hence, many researchers have utilized it to simulate various issues especially thermal comfort [20], [21], [22], [23]. Moracinio et al. [24] studied the effectiveness of trees in improving thermal comfort using the Envi-met® model and concluded that the tree configuration, including leaf surface index, tree height, trunk height affect the level of the thermal comfort. Tudert and Mayer [25] studied the influence of the geometric shape of streets and vegetation on thermal comfort in urban roads using the Envi-met® model. Ma et al. [26] investigated outdoor thermal comfort using a field questionnaire survey and Envi-met simulations. Their result suggested that the most pleasant times for walking are early morning (8:00 a.m. to 11:00 a.m.) and after 18:00p.m. Ghaffarianhoseini et al. [27] studied the recognition of the thermally uncomfortable area using field observations and numerical simulations via Envi-met and IES-VE softwares. They offered some suggestions to better the design quality of outdoor spaces regarding thermal comfort. Accordingly, in this paper, version 4 of the Envi-met® model is implemented to evaluate the PET and PMV in various courtyards and do numerical simulations.

Recently, data-driven methods such as genetic programming (GP) have been successfully exploited for the prediction of thermal comfort in the various buildings, such as optimization of HVAC system energy consumption in a building using artificial neural network and multi-objective genetic algorithm [28], the analysis of thermal comfort in presence of large radiant surfaces [29], [30], [31] and visual comfort to reduce the energy consumption [32], thermal sensation prediction using soft computing approaches [33], the simulation of visitors’ thermal comfort at urban areas in hot and arid climate by NN-ARX soft computing method [34], adaptation of ANFIS model to assess thermal comfort of an urban square in moderate and dry climate [35]. Considering the above-mentioned studies, GP as a robust and promising data-driven approach can predict thermal comfort. To the best knowledge of the authors, there is a gap in exploiting GP to predict thermal comfort criterion (e.g. PET and PMV). Moreover, GP can produce transparent and accurate formulae in the form of mathematical equations [36]. Using this capability, in this paper some accurate and physically-sound formulae are presented for the prediction of PET and PMV.

The majority of previous researches have emphasized the environmental role of the courtyard to provide indoor thermal comfort for occupants. However, there is no study yet available that investigates the effects of shadow and sunlit on the thermal comfort in outdoor (open) spaces such as courtyards, to the best knowledge of the authors. Further, the studies of thermal comfort in various places are numerical or experimental, and so there is not an accurate formula to predict thermal comfort on the basis of its contributing parameters. Therefore, the main aims and novelties of this paper are: (1) to study the effects of the shading in the summer and sunlit in the winter on the thermal comfort in courtyards; (2) to present some mathematical equations and charts to predict the thermal comfort criterion based on the measurement dataset.

The paper is structured as follows: Section 2 gives a detailed overview of methodology and materials. This section contains the introduction to climate data, the properties of case studies, detailed information on genetic programming and data points derived from the numerical simulations. The evaluation of numerical simulation results and the development of GP models are presented in Section 3, which is followed by the summary and conclusion of Section 4. Finally, Section 5 closes the paper.