Gas Fired Combi System by Artificial Neural Network Fuel Quantity Estimation Example of Isparta

Mehmet Erhan Şahin

Makale detayları

Gas Fired Combi System by Artificial Neural Network Fuel Quantity Estimation Example of Isparta

Yazarlar:

Mehmet Erhan Şahin

Yayıncı:

Isparta University of Applied Sciences

Ana Başlıklar:

Elektrik Mühendisliği

Dergi:

Uluslararası Teknolojik Bilimler Dergisi Vol.15 Issue 1, pp.11-18

Belge Tipi:

Makale

Yayın Tarihi:

11/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001290584

pISSN:

1309-1220

Anahtar kelimeler:

Özet:

 The use of natural gas, which is a gas fuel, is increasing in the world because it is an environmentally friendly clean energy source In Turkey, natural gas discharge protection is maintained during the day. However, a large amount of foreign dependency is purchased. Natural gas has to be stocked depending on its use in Turkey.Due to this advantage, the annual consumption estimate of the natural gas we consume is also extremely important. Especially with the increase in natural gas costs in recent years, it is necessary to make important calculations about how much annual expenditure will be made. Today, energy costs can be reduced by administrative energy use. In this study, the temperature values of the rooms, the outside air and the fuel amount of the system were measured and recorded with a measurement system developed in a house using natural gas fired combi boiler system for a 10-year-old house with an east-north facing sheathing of 120 m2 in the province of Isparta. 181 data for the months of October, November, December, January, February, and March were collected. 141 of these data were used for training, 10 for validation and 30 for testing. The created artificial neural network model is trained with training data. In a standard winter season, daily natural gas amounts consumed according to room temperatures and outside air temperature, with the help of artificial neural network ANN model, the annual natural gas consumption amount was estimated at 98 with 0.118 MSE error.

 The use of natural gas, which is a gas fuel, is increasing in the world because it is an environmentally friendly clean energy source In Turkey, natural gas discharge protection is maintained during the day. However, a large amount of foreign dependency is purchased. Natural gas has to be stocked depending on its use in Turkey.Due to this advantage, the annual consumption estimate of the natural gas we consume is also extremely important. Especially with the increase in natural gas costs in recent years, it is necessary to make important calculations about how much annual expenditure will be made. Today, energy costs can be reduced by administrative energy use. In this study, the temperature values of the rooms, the outside air and the fuel amount of the system were measured and recorded with a measurement system developed in a house using natural gas fired combi boiler system for a 10-year-old house with an east-north facing sheathing of 120 m2 in the province of Isparta. 181 data for the months of October, November, December, January, February, and March were collected. 141 of these data were used for training, 10 for validation and 30 for testing. The created artificial neural network model is trained with training data. In a standard winter season, daily natural gas amounts consumed according to room temperatures and outside air temperature, with the help of artificial neural network ANN model, the annual natural gas consumption amount was estimated at 98 with 0.118 MSE error.

 The use of natural gas, which is a gas fuel, is increasing in the world because it is an environmentally friendly clean energy source In Turkey, natural gas discharge protection is maintained during the day. However, a large amount of foreign dependency is purchased. Natural gas has to be stocked depending on its use in Turkey.Due to this advantage, the annual consumption estimate of the natural gas we consume is also extremely important. Especially with the increase in natural gas costs in recent years, it is necessary to make important calculations about how much annual expenditure will be made. Today, energy costs can be reduced by administrative energy use. In this study, the temperature values of the rooms, the outside air and the fuel amount of the system were measured and recorded with a measurement system developed in a house using natural gas fired combi boiler system for a 10-year-old house with an east-north facing sheathing of 120 m2 in the province of Isparta. 181 data for the months of October, November, December, January, February, and March were collected. 141 of these data were used for training, 10 for validation and 30 for testing. The created artificial neural network model is trained with training data. In a standard winter season, daily natural gas amounts consumed according to room temperatures and outside air temperature, with the help of artificial neural network ANN model, the annual natural gas consumption amount was estimated at 98 with 0.118 MSE error.

 Bir gaz yakıt olan doğalgazın çevreye duyarlı temiz enerji kaynağı olmasından dolayı dünyadaki kullanımı giderek artmaktadır. Türkiyede doğalgaz çıkarımı noktasında çalışmalar gün geçtikçe yapılmaktadır. Ancak büyük miktarda dışa bağımlı olarak satın alınmaktadır. Doğalgaz Türkiyede kullanımına bağlı olarak stok yapılmak durumundadır. Bu avantajından dolayı tüketmiş olduğumuz doğalgazın yıllık tüketim tahmini de son derece önem arz etmektedir. Özellikle son yıllarda doğalgaz maliyetlerinin artmasıyla birlikte ne kadarlık bir yıllık harcamanın yapılacağı konusunda önemli hesaplamalar yapılması gerekmektedir. Günümüzde, yönetimsel enerji kullanımları sayesinde enerji maliyetlerinde düşüşler sağlanabilmektedir. Bu çalışmada, Isparta ilinde bulunan 120 m2 büyüklüğünde doğu kuzey cepheli mantolaması bulunan 10 yıllık bir konut için doğalgaz yakıtlı kombi sistemi kullanan bir evde geliştiren bir ölçüm sistemi ile odaların, dış havanın sıcaklık değerleri ve sistemin yakıt miktarı ölçülerek kayıt altına alınmıştır. Ekim, Kasım, Aralık, Ocak, Şubat, Mart aylarına ait 181 adet veri ele alınmıştır. Bu verilerin 141 tanesi eğitim, 10 tanesi doğrulama ve 30 tanesi test için kullanılmıştır. Oluşturulan yapay sinir ağı modeli eğitim verileri ile eğitilmiştir. Standart bir kış sezonunda oda sıcaklıklarına ve dış hava sıcaklığına göre tüketilen günlük doğalgaz miktarları, yapay sinir ağı YSA modeli yardımıyla eğitilmiş 0.118 MSE hata ile doğalgaz yıllık tüketim miktarı 98 oranında tahmin edilmiştir.

Makale detayları

Gas Fired Combi System by Artificial Neural Network Fuel Quantity Estimation Example of Isparta

Yazarlar:

Mehmet Erhan Şahin

Yayıncı:

Isparta University of Applied Sciences

Ana Başlıklar:

Elektrik Mühendisliği

Dergi:

Uluslararası Teknolojik Bilimler Dergisi Vol.15 Issue 1, pp.11-18

Belge Tipi:

Makale

Yayın Tarihi:

11/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001290584

pISSN:

1309-1220

Artifical Neural Networks

Energy Management

Microcontroler

AutomationYapay Sinir Ağları

Enerji Yönetimi

Mikrodenetleyici

Otomasyon

Sayı

Makale detayları

Yazarlar: Mehmet Erhan Şahin

Yayıncı: Isparta University of Applied Sciences

Ana Başlıklar: Elektrik Mühendisliği

Dergi: Uluslararası Teknolojik Bilimler Dergisi Vol.15 Issue 1, pp.11-18

Belge Tipi: Makale

Yayın Tarihi: 11/2023

Belge Dili: Türkçe

Document ID (Org No.): 2001290584

pISSN: 1309-1220

Sayı

Yazarlar:

Mehmet Erhan Şahin

Yayıncı:

Isparta University of Applied Sciences

Ana Başlıklar:

Elektrik Mühendisliği

Dergi:

Uluslararası Teknolojik Bilimler Dergisi Vol.15 Issue 1, pp.11-18

Belge Tipi:

Makale

Yayın Tarihi:

11/2023

Belge Dili:

Türkçe

Document ID (Org No.):

2001290584

pISSN:

1309-1220