Accuracy Comparison of ARX and ANFIS Model of PM Brake Lining Wear Behavior Yildiz Technical University

Format:

Conference/Event

Author/Creator:

Isin, Isin, author.

Contributor:

Istif, Ilyas

Uzunsoy, Deniz

Uzunsoy, Erdem

Conference Name:

SAE 2013 World Congress & Exhibition (2013-04-16 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 2013

Summary:

The brake friction materials in an automotive brake system play important role in the overall braking performance of a vehicle. A previous study by the same authors was focused on wear testing for a 1040 steel disc interacting with Powder metallurgy (PM) copper-based brake lining material with and without MoS₂ additive at constant applied load and sliding velocity. In this paper, a non-Linear Autoregressive model (ARX) Model structure with sigmoid network having one hidden layer and nonlinear ANFIS (Adaptive Neuro-Fuzzy Inference System) model structure was used to find the best possible wear prediction results and both approaches have been applied to simulate wear behavior of the brake lining material. Preliminary results showed that ARX provides closer results to the experiments than the ANFIS model. As a result, nonlinear ARX modeling can be used as an effective tool in the prediction of brake lining material properties instead of time-consuming experimental processes

Notes:

Vendor supplied data

Publisher Number:

2013-01-1216

Access Restriction:

Restricted for use by site license