Drag Force Estimation of a Truck Trailer Model Using Artificial Neural Network
Küçük Resim Yok
Tarih
2016
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Prediction of the drag forces acting on a truck trailer with/without spoiler is carried out by using artificial neural network (ANN). ANN model data set include the experiments of spoiler positions which have zero level to trailer front corner, -2 mm, -4.5 mm, -9 mm, +4.5 mm and +9 mm and truck trailer without spoiler. The experiments were carried out in the wind tunnel in the range of the free stream velocity between 4.6 m/s and 19.3 m/s, corresponding the Re number range, 1.0×105 -5.0×105. Mean absolute percentage error (MAPE) for training, validation and testing is 2.24%, 3.75% and 4.58% in the prediction of the drag forces, respectively. Prediction performance of the developed ANN model has a very good accuracy. According to the drag coefficients results, Reynolds number independence for truck trailer model is obtained at Reynolds number between 1.97×105 and 4.89×105. For spoiler position cases, while minimum drag coefficient acting on truck trailer with spoiler is seen at - 2mm offset, maximum drag coefficient is seen at -9 mm offset
Prediction of the drag forces acting on a truck trailer with/without spoiler is carried out by using artificial neural network (ANN). ANN model data set include the experiments of spoiler positions which have zero level to trailer front corner, -2 mm, -4.5 mm, -9 mm, +4.5 mm and +9 mm and truck trailer without spoiler. The experiments were carried out in the wind tunnel in the range of the free stream velocity between 4.6 m/s and 19.3 m/s, corresponding the Re number range, 1.0×105 -5.0×105. Mean absolute percentage error (MAPE) for training, validation and testing is 2.24%, 3.75% and 4.58% in the prediction of the drag forces, respectively. Prediction performance of the developed ANN model has a very good accuracy. According to the drag coefficients results, Reynolds number independence for truck trailer model is obtained at Reynolds number between 1.97×105 and 4.89×105. For spoiler position cases, while minimum drag coefficient acting on truck trailer with spoiler is seen at - 2mm offset, maximum drag coefficient is seen at -9 mm offset
Prediction of the drag forces acting on a truck trailer with/without spoiler is carried out by using artificial neural network (ANN). ANN model data set include the experiments of spoiler positions which have zero level to trailer front corner, -2 mm, -4.5 mm, -9 mm, +4.5 mm and +9 mm and truck trailer without spoiler. The experiments were carried out in the wind tunnel in the range of the free stream velocity between 4.6 m/s and 19.3 m/s, corresponding the Re number range, 1.0×105 -5.0×105. Mean absolute percentage error (MAPE) for training, validation and testing is 2.24%, 3.75% and 4.58% in the prediction of the drag forces, respectively. Prediction performance of the developed ANN model has a very good accuracy. According to the drag coefficients results, Reynolds number independence for truck trailer model is obtained at Reynolds number between 1.97×105 and 4.89×105. For spoiler position cases, while minimum drag coefficient acting on truck trailer with spoiler is seen at - 2mm offset, maximum drag coefficient is seen at -9 mm offset
Açıklama
Anahtar Kelimeler
Mühendislik, Makine, Mühendislik, Ortak Disiplinler
Kaynak
International Journal of Automotive Engineering and Technologies
WoS Q Değeri
Scopus Q Değeri
Cilt
5
Sayı
4
