APPLYING BIOGEOGRAPHY-BASED MULTI-LAYER PERCEPTRON NEURAL NETWORK TO PREDICT CALIFORNIA BEARING CAPACITY VALUE OF STABILIZED POND ASH WITH LIME AND LIME SLUDGE

In this study, a hybrid biogeography-based multi-layer perceptron neural network (BBO-MLP) with different number of hidden layers (one up to three) was developed for predicting the California bearing capacity (CBR) value of pond ash stabilized with lime and lime sludge. To this aim, model had five variables named maximum dry density, optimum moisture content, lime percentage, lime sludge percentage and curing period as inputs, and CBR as output variable. Regarding BBO-MLP models, BBO-MLP1 has the best results, which its R2 stood at 0.9977, RMSE at 0.7397, MAE at 0.476, and PI at 0.0104. In all three developed models, the estimated CBR values specify acceptable agreement with experimental results, which represents the workability of proposed models for predicting the CBR values with high accuracy. Comparison of three developed models supply that BBO-MLP1 outperform others. Therefore, BBO-MLP1 could be recognized as proposed model.