Browsing by Author "Layth, Yousif"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
Item UTILIZING ARTIFICIAL NEURAL NETWORKS TO PREDICT HUMAN BODY EXERGY CONSUMPTION(2022-03-10) Layth, Yousif; TURHAN, Cihan; LOTFISADIGH, BahramThe American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE) defines thermal comfort as "the state of mind that conveys happiness with the thermal environment”. Energy and Matter can scatter as a system and move toward equilibrium with their surrounding environment, and this is referred to as exergy in thermodynamics. Predicted Mean Vote (PMV)/Percentage of Predicted Dissatisfied (PPD) model and adaptive thermal comfort approach are the two most widely used methods for assessing thermal comfort. However, it is also possible to apply the exergy notion to the human body system as an index of thermal comfort. The relationship between a person's exergy balance and their level of thermal comfort is that effectively dissipating heat and water from the body is essential to human well-being. For this reason, the lowest human body exergy consumption rate mostly gives the optimum thermal comfort level. In this thesis, an Artificial Intelligence-based work was conducted in a room of engineering faculty of the Atilim University, Ankara, Turkey, with an occupant being inside the room to obtain the best condition for his exergy and thermal comfort. Human body exergy consumption is extracted via a computer programme and environmental parameters are measured by objective sensors. Then, an Artificial Neural Network (ANN) model is developed in Python environment. A back propagation and sigmoid function is used in the neural network technique. A total of 133 data are included in the ANN model, with 75% (99 datasets) being used for training and the remaining for testing. A Mean Absolute Percentage Error (MAPE) of 1.98 and an accurate prediction rate (R2 ) of 0.91 are found under the provided conditions, indicating a good coordination between the artificial neural network model outputs and the human body exergy data. Simplicity, speed of analysis, and learning from restricted data sets are all features of an ANN model over human body exergy simulation. This thesis presents a novel concept that uses an ANN model to determine how much exergy rate people consume (HBExC). This is because artificial neural networks (ANNs) are the most commonly used artificial intelligence technique in the field of buildings and thermal comfort fields. After all, they can handle nonlinear variables' interactions rapidly and correctly, especially, exergy concept which has complex nonlinear relationships between its variables.