Aslay, Semi Emrah

The aim of this study is to investigate building energy efficiency by integrating hybrid modelling approaches and interpretable models into architectural design processes and engineering solutions while considering environmental impacts. At the same time, it is aimed to achieve the most efficient building design possible. Separate analyses were carried out using a total of 6,913 data from Rutland (185) and Salford (6718) cities. The data were grouped into carbon emission information, architectural information, lighting information, personal heating information, and main heating system information to form a dataset. Light Gradient Boosting Machine (LightGBM) was preferred as the base model and Particle Swarm Optimisation (PSO) method was applied for hyperparameter optimisation. The hybrid model created in this way is called PSO-LightGBM. The optimization process was carried out using software in both R Studio and Python environments, utilizing seven different hyperparameters. Apart from the hybrid model used as a method, 2 different SHAP analyses, neural network based, and tree based, were performed to clearly explain the parameter relationships. The PSO-LightGBM hybrid model provided more successful predictions compared to the basic LightGBM model. While R2 values improved between 0.82 and 0.90 in the Rutland dataset, this value increased from 0.8687 to 0.8901 for test data and from 0.8538 to 0.9091 for training data in the Salford dataset. R2 values show an improvement of 7% in the Rutland dataset and maximum 6% in the Salford dataset. When the reduction in error rates is evaluated, it is found that the greatest improvement is in the Mean Squared Error (MSE) metric. MSE decreased by 17% in the Rutland dataset and by 4% in the Salford dataset. According to the SHAP Analysis results, CO2 emissions have the largest impact on energy consumption, while primary fuel types, number heated rooms and individual heating systems are other important parameters. While the tree-based SHAP model is more sensitive to physical parameters, the neural network model is more sensitive to indirect relationships. In both analyses, the communal heating system type has the lowest impact. In order to improve building energy efficiency, high efficiency individual boiler systems should be preferred, architectural approaches that optimise the number of heated rooms and smart heating solutions should be used, and central heating systems should be modernised. The results highlight the effectiveness of hybrid modelling approaches with SHAP analyses based on different baselines to ensure the integration of environmental impacts, architectural design processes and engineering solutions in terms of building energy efficiency. Furthermore, the findings contribute to the importance of interdisciplinary work in buildings to improve energy efficiency. Future studies can focus on the development of building energy performance prediction models that can be explained by integrating various learning algorithms and different optimisation techniques, using regions with high cooling demand, large datasets and different building types.

Köse Özturan, Merve; Kurnuç Seyhan, Aslıhan

Reducing heat transfer from the building will reduce the energy cost required to heat the building. Significant energy savings of 25 % to 50 % can be achieved when the appropriate thickness of thermal insulation is applied to the building walls. In this study, calculations were made to determine the optimum insulation thickness of some thermal insulation materials used to reduce the heat transfer from buildings in Erzincan, located in Türkiye's fourth climate zone. The optimum thickness of the thermal insulation to be applied to the external walls was determined using the heating degree day method, taking into account the effect of solar radiation according to the four main directions. Optimum insulation thicknesses and payback periods were calculated for combinations of two building materials (pumice and bricks), four fuels (fuel oil, LPG, natural gas, and coal), and three insulation materials (XPS, EPS, and Rock wool). The numbers of heating degree days were calculated as 2229 in the north, 1425 in the east and west, and 1118 in the south direction for the four main wall directions. The lowest optimum insulation thickness for these heating degree day values, which minimizes the total heating cost, was obtained for pumice/XPS/natural gas combination. The optimum insulation thicknesses for this combination were calculated as 0.037 m in the north, 0.025 m in the east and west, and 0.02 m in the south direction. The calculated payback periods for all combinations range from 0.9 to 4 years and the payback period of Pumice/XPS/Natural Gas combination is 4 years.

Dogan, Ahmet; Kayaci, Nurullah; Demir, Hakan; Kemal Sevindir, Mustafa

Hydronic radiant systems, which provide a thermal comfort environment, come to the fore in different building applications due to their energy saving potential and being integrated with renewable energy systems. The aim of this study is to present a full-scale experiment to compare the thermal comfort performances of radiant ceiling, wall and wall + ceiling systems coupled ground source heat pump and direct expansion fan coil system implemented in Yildiz Technical University Science and Technology Application and Research Center, Istanbul, Turkey. Air velocity, horizontal and vertical air temperature differences, mean radiant temperature and humidity values, which are four physical parameters that can affect the thermal satisfaction of office workers for 9 different experimental conditions, are measured for fan coil and all radiant systems. The results indicate that vertical temperature differences between ankle and head level vary from 0.9 °C to 1.4 °C for both fan coil system and all radiant system configurations in different location of the test room. Acceptable overall comfort in indoor environments Class A only showed stable performance with an average of 0.12 PMV and 5.3 % PPD when ceiling + wall radiant panels were actively cooled. The result shows that the difference between room air temperature and room operating temperature varies between 0.45 °C and 0.6 °C when cooling loads are met by radiant systems. This may not make it possible to use indoor air temperature instead of indoor operating temperature in the control of radiant systems. Moreover, the thermal environment of the radiant system are more uniform than fan coil system due to low air temperature stratification, air velocity and small differences between mean radiant temperature and indoor air temperature.

Şahin, Mustafa; Oʇuz, Yüksel; Büyüktümtürk, Fuat

The illuminance of a lighting system decreases over time both due to pollution on fixtures, lamps, and wall surfaces, and decreases and failures in the luminous flux of lighting components. Such decreases are not recognizable, as they occur slowly and continuously. Numerous studies have concluded that eyesight weakens due to incorrect lighting or decreases in luminous flux of lighting components over time. However, a sufficient and stable level of illuminance is obligatory for the establishment of healthy working environments. Following the installation of a lighting system, measurement of the illuminance distribution in that environment can become a tiring and time-consuming process. In this study, an artificial neural network (ANN) model is established as an alternative to these measurements and is trained to estimate the illuminance distribution in interior spaces. Thanks to an installed ANN model, real-time estimation of illuminance distribution in an environment was realized. With these estimated values, the mean illuminance in the environment was calculated, in turn allowing the depreciation multiplier of the system to be obtained. As a result, the percentage of energy in the system that was not transformed into light was determined, allowing either improvements or fundamental changes to be made to the system, thus preventing energy waste and light loss. Furthermore, ergonomic comfort conditions of the illuminated environment were kept at the desired level.