Nhảy đến nội dung

Development of neuro-fuzzy and neuro-bee predictive models for prediction of the safety factor of eco-protection slopes


Maryam Safa, Puteri Azura Sari, Mahdi Shariati*, Meldi Suhatril, Nguyen Thoi Trung, Karzan Wakil, Majid Khorami

Source title: 
Physica A: Statistical Mechanics and its Applications, 550: 124046, 2020 (ISI)
Academic year of acceptance: 

This study is aimed to investigate the surface eco-protection techniques for cohesive soil slopes along the selected Guthrie Corridor Expressway (GCE) stretch by way of analyzing a new set of intelligence techniques namely neuro-bee, artificial neural network (ANN) and neuro-fuzzy. Soil erosion and mass movement which induce landslides have become one of the disasters faced in Selangor, Malaysia causing enormous loss affecting human lives, destruction of property and the environment. Establishing and maintaining slope stability using mechanical structures are costly. Hence, biotechnical slope protection offers an alternative which is not only cost effective but also aesthetically pleasing. To reach the aim of the current study, a field investigations and numerical studies were conducted and a suitable database was prepared and established. By preparing factor of safety (FOS) as a single output parameter and a combination of the most important parameters on that, the desired models have been designed based on training and test patterns. In order to evaluate and compare the prediction precision of the developed models, a series of statistical indices, such as root mean squared error (RMSE), determination coefficient (R-square) and variance account for (VAF) are calculated. Many intelligence models with the most effective parameters on the mentioned models were developed to predict FOS. Based on the simulation results and the measured indices, it was found that the proposed neuro-fuzzy model with the lowest system error and highest R-square performs better as compared to other proposed ANN and neuro-bee models. Therefore, the neuro-fuzzy can provide a new applicable model to effectively predict the FOS of the slopes due to the fact that it is able to combine the advantages of the ANN and fuzzy inference system to indicate a high prediction capacity in solving problem of slope stability.