Application of Ensemble Learning Using Weight Voting Protocol in the Prediction of Pile Bearing Capacity

Abstract

Accurate prediction of pile bearing capacity is an important part of foundation engineering. Notably, the determination of pile bearing capacity through an in situ load test is costly and time-consuming. Therefore, this study focused on developing a machine learning algorithm, namely, Ensemble Learning (EL), using weight voting protocol of three base machine learning algorithms, gradient boosting (GB), random forest (RF), and classic linear regression (LR), to predict the bearing capacity of the pile. Data includes 108 pile load tests under different conditions used for model training and testing. Performance evaluation indicators such as R-square (R2), root mean square error (RMSE), and MAE (mean absolute error) were used to evaluate the performance of models showing the efficiency of predicting pile bearing capacity with outstanding performance compared to other models. The results also showed that the EL model with a weight combination of w 1 = 0.482, w 2 = 0.338, and w 3 = 0.18 corresponding to the models GB, RF, and LR gave the best performance and achieved the best balance on all data sets. In addition, the global sensitivity analysis technique was used to detect the most important input features in determining the bearing capacity of the pile. This study provides an effective tool to predict pile load capacity with expert performance.

Model Test and Numerical Simulation of Water Conservancy Foundation Bearing Capacity

“…The determination of foundation bearing capacity is crucial in the design of foundation. Therefore, the determination of foundation bearing capacity of frozen soil under freezing conditions through tests is also critical for application in practical projects [ 25 – 34 ]. Through the test and analysis in this chapter, it is concluded that when the temperature gradually decreases, due to the cementation of ice in the soil, the bearing capacity of the foundation will significantly increase.…”

Ensemble learning based on relative accuracy approach and diversity teams

“…As many trainers can be added as much as a better performance [111]. Weight voting was used in many works [113]- [115], [119]- [124]. The relative accuracy 𝑎𝑐𝑐 𝑐 (𝑡) of a trainer 𝑡 ∈ 𝑇 with respect to a class 𝑐 ∈ 𝐶 can be stated as the times that 𝑡 classified an element 𝑥 ∈ 𝐷 correctly as an element from the class 𝑐 divided by the total number of elements in the class 𝑐 𝑎𝑐𝑐 𝑐 (𝑡) = |𝑝𝑟𝑒 𝑡 (𝑥∈𝐷)=𝑐|…”

An ensemble method based on weight voting method for improved prediction of slope stability

“…They concluded that stacking models provided more satisfactory performance compared with single model. Pham and Vu (2021) developed an ensemble earning algorithm based on weight voting method using three individual classifier including GB, RF, and classic LR for prediction of pile bearing capacity. They confirmed that the ensemble method provided the best performance for the evaluation of pile capacity in terms of evaluation indicators including R 2 , RMSE, and MAE, namely mean absolute error.…”