Nonlinear system identification of large-scale smart pavement systems
| Autor: | Sankha Bhowmick, Bao-Liang Chen, Rajib B. Mallick, Yeesock Kim |
|---|---|
| Rok vydání: | 2013 |
| Předmět: |
Adaptive neuro fuzzy inference system
geography geography.geographical_feature_category Nonlinear system identification Artificial neural network Automatic control Water flow Computer science General Engineering System identification Process (computing) Inlet Fuzzy logic Computer Science Applications Artificial Intelligence Simulation |
| Zdroj: | Expert Systems with Applications. 40:3551-3560 |
| ISSN: | 0957-4174 |
| DOI: | 10.1016/j.eswa.2012.12.062 |
| Popis: | This paper proposes a novel model for predicting complex behavior of smart pavements under a variety of environmental conditions. The mathematical model is developed through an adaptive neuro fuzzy inference system (ANFIS). To evaluate the effectiveness of the ANFIS model, the temperature fluctuations at different locations in smart pavement systems equipped with pipe network systems under solar radiations is investigated. To develop the smart pavement ANFIS model, various sets of input and output field experimental data are collected from large-scale experimental test beds. The solar radiation and the inlet water flow are used as input signals for training complex behavior of the smart pavement ANFIS model, while the temperature fluctuation of the smart pavement system is used for the output signal. The trained model is validated using 20 different data sets that are not used for the training process. It is demonstrated from the simulation that the ANFIS identification approach is effective in modeling complex behavior of the pavement-fluid system under a variety of environmental conditions. Comparison with high fidelity data proves the viability of the proposed approach in pavement health monitoring setting, as well as automatic control systems. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect