A new constitutive model of a magneto-rheological fluid actuator using an extreme learning machine method
Penulis/Author
Irfan Bahiuddin, S.T., M.Phil., Ph.D. (1); Saiful Amri Mazlan (2); Mohd Ibrahim Shapiai (3); Seung Bok Choi (4); Fitrian Imaduddin (5); Mohd Azizi Abdul Rahman (6); Mohd Hatta Mohammed Ariff (7)
Tanggal/Date
2018
Kata Kunci/Keyword
Abstrak/Abstract
tIn this work, a new constitutive model of a magneto-rheological fluid (MRF) actuator is proposed using anextreme learning machine (ELM) technique to enhance the prediction accuracy of the field-dependentactuating force. After briefly reviewing existing rheological constitutive models of MRF actuator, ELMalgorithm is formulated using a single-hidden layer feed-forward neural network. In this formulation,both the magnetic field and measured shear rates are used as inputs variables, while the shear stresspredicted from the ELM training is used as an output variable. Subsequently, in order to validate theeffectiveness of the proposed model, the target defined as the error between the prediction and measureddata is set. Then, the fitness of the training and prediction performances is evaluated using a normalizedroot mean square error (NRMSE) method. It is shown that the shear stress estimation based on the ELMmodel using sinusoidal activation function is more accurate than conventional rheological constitutivemodels such as Herschel-Bulkley model. It is also demonstrated that the proposed model is capable ofpredicting the field-dependent yield stress which is defined as an actuating force of the MRF actuatorwithout causing significant errors.
