A comparison between support vector regression and hybrid approach of adaptive Neuro-fuzzy inference system optimized by imperialist competitive algorithm for the prediction of energy loss of run-off-road vehicles

Hojjat Jafari

Abstract


The present paper is aimed at modeling the energy loss of run-off-road vehicles using artificial intelligence techniques. Support vector regression (SVR), adaptive Neuro-fuzzy inference system (ANFIS) and adaptive neuro-fuzzy inference system optimized by imperialist competitive algorithm (ANFIS-ICA) methodologies were developed to obtain the best model in terms of several statistical criteria including average absolute relative error (AARE), mean square error (MSE), mean relative error (MRE) and coefficient of determination (R2). A total of 50 data points were derived from experiments at five levels of wheel load (i.e. 0, 0.5, 1, 1.5, and 2 kN), five levels of tire inflation pressure (i.e. 70, 100, 140, 175, and 210 kPa) with two replicates in a capacious soil bin facility utilizing a single wheel-tester. The obtained results revealed that SVR outperformed ANFIS and ANFIS-ICA approaches at the lowest predicting error. These findings shed light on some of the underlying principles of SVR, ANFIS-ICA and ANFIS in prediction of energy loss of run-off-road vehicles. 


Full Text:

PDF

References


Bekkar M G. Off road locomotion. Research and Development in Terramechanics. The University of Michigan Press, Ann Arbor, Michigan (1960)

Taghavifar H., Mardani A. A knowledge-based Mamdani fuzzy logic prediction of the motion resistance coefficient in a soil bin facility for clay loam soil. Neural Computing and Applications. 2013, 23(1): 293-302.

Berneti S M. A Hybrid Approach Based on the Combination of Adaptive Neuro-Fuzzy Inference System and Imperialist Competitive Algorithm: Oil Flow Rate of the Wells Prediction Case Study. International Journal of Computational Intelligence Systems. 2013, 6(2), 198-208.

Pousinho H M I, Mendes V M F, Catalão J P S. Short-term electricity prices forecasting in a competitive market by a hybrid PSO–ANFIS approach. International Journal of Electrical Power & Energy Systems. 2012, 39(1): 29-35.

Jiang H M, Kwong C K, Ip WH, Wong TC. Modeling customer satisfaction for new product development using a PSO-based ANFIS approach. Applied Soft Computing. 2012, 12(2): 726-734.

Pousinho H M I, Mendes V M F, Catalao J P S. A hybrid PSO–ANFIS approach for short-term wind power prediction in Portugal. Energy Conversion and Management. 2011, 52:397–402.

Ho W H, Tsai J T, Lin B T, Chou J H. Adaptive network-based fuzzy inference system for prediction of surface roughness in end milling process using hybrid Taguchi-genetic learning algorithm. Expert Systems with Applications. 2009, 36: 3216–3222.

Cheng-Yuan T, Yi-Leh W, Chien-Chin P, Fundamental matrix estimation by multiobjective genetic algorithm with Taguchi's method. Applied Soft Computing. 2012, 12(1): 553-558.

Atashpaz-Gargari E, Lucas C. Imperialist Competitive Algorithm: An algorithm for optimization inspired by imperialistic competition. IEEE Congress on Evolutionary Computation. 2007, 4661–4667.

Cortes C, Vapnik V. Support-vector networks, Machine Learning, September 1995, 20 (3): 273-297.

Al-Anazi A.F., Gates I.D., Support vector regression for porosity prediction in a heterogeneous reservoir: A comparative study, Computers & Geosciences 36 (2010) 1494–1503

Mardani A, Shahidi K, Rahmani A, Mashoofi B, Karimmaslak H. Studies on a long soil bin for soil-tool interactions. Cercetări Agronomice în Moldova. 2010, 142(2):5-10.

Takagi T, Sugeno M. Fuzzy Identification of System and Its Applications to Modelling and Control. IEEE Transactions on Systems, Man, and Cybernetics. 1985, 15(1): 116-132.

Jang J S R. ANFIS: Adaptive-Network-Based Fuzzy Inference System, IEEE Transactions on Systems, Man, and Cybernetics. 1993, 23(3): 665-685.

Burges C J C. A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery. 1998, 2 (2):121–167.

Vapnik V. The Nature of Statistical Learning Theory. Springer New York, 1995 MSE: 2.81884


Refbacks

  • There are currently no refbacks.