An accurate estimation of pile response to loading is a challenging task due to the complexity of the soil–pile interactions and uncertainties in the soil properties. Conventional methods of predicting pile load–settlement relationship either oversimplify the problem or require the parameters that are difficult to determine in the laboratory. In this study, a high-order neural network (HON) is developed to simulate the pile load–settlement curve using properties of the pile and SPT data along the depth of pile embedment as inputs. The results indicated a significant improvement in the quality of HON predictions over that of BPN, RBF and GRNN models. Based on the comparisons with the predictions of elastic and hyperbolic models, the proposed HON model provides better predictions than existing theoretical models.
|Number of pages||9|
|Journal||Engineering Applications of Artificial Intellegence|
|Publication status||Published - Aug 2011|
- Artificial neural network
- Higher-order neural network
- Soil-pile interaction
- Load-settlement behaviour