ANN and ANFIS Approaches to Calculate the Heating and Cooling Degree Day Values: The Case of Provinces in Turkey

• Published in: Arabian journal for science and engineering (2011) 2019-09, Vol.44 (9), p.7581-7597
• Main Authors: Işık, Erdem, İnallı, Mustafa, Celik, Erkan
• Format: Article
• Language: English
• Subjects: Adaptive systems; Air conditioners; Air conditioning; Algorithms; Artificial neural networks; Back propagation; Building design; Cooling; Cost analysis; Demand analysis; Engineering; Fuzzy systems; Heating; Humanities and Social Sciences; Inference; Literature reviews; Mathematical analysis; Meteorological services; multidisciplinary; Research Article - Mechanical Engineering; Science
• ISSN: 2193-567X; 1319-8025; 2191-4281
• DOI: 10.1007/s13369-019-03852-4

Energy analysis plays a vital role in developing an optimum and cost-effective design for the heating, ventilating and air conditioning system of a building. The values of heating degree days (HDD) and cooling degree days (CDD) are used to analyze the demand for energy for buildings. In this study, the HDD and CDD values for Turkey were calculated using estimated temperature values. An artificial neural network (ANN) and an adaptive network-based fuzzy inference system (ANFIS) were applied for the estimation of temperature values. The Levenberg–Marquardt training algorithm was used for the feedforward backpropagation of the ANN, whereas the Sugeno-type fuzzy inference system was used for the ANFIS. Meteorological data from the last 11 years were used for the proposed approaches, and these were applied to different stations ( n  = 50) of the Turkish State Meteorological Service. In relation to the insulated building design, base temperatures have been considered as 16 °C and 18 °C for the HDD and 22 °C and 24 °C for the CDD based on a literature review. Two novel maps were prepared for the CDD and HDD. The calculated degree days and the maps are consistent with those from the Turkish State Meteorological Service.