CSDL Bài trích Báo - Tạp chí
Khoa Xây Dựng
1 Training artificial neural network regression based on the generalized delta rule : a case study in modeling the compressive strength of concrete = Huấn luyện mạng nơ-ron thần kinh nhân tạo dùng cho phân tích hồi quy dựa trên quy tắc delta khái quát: ứng / Hoang Nhat Duc, Nguyen Quoc Lam, Pham Quang Nhat // .- 2023 .- Số 02(57) .- P. 3-9 .- 690
This paper presents the algorithms for training an artificial neural network (ANN) for regression analysis; the algorithm is based on the generalized delta rule. The training method of a simple neuron model and an ANN model are presented and generalized. The models are then programed in Visual C# .NET and applied to predict the compressive strength of concrete mixes. Three datasets, collected from the literature, are used to demonstrate the applications of the models.
2 XGBoost regression for estimating bearing capacity of concrete piles = Sử dụng hồi quy XGBoost để đánh giá sức chịu tải của cọc bê tông / Tran Thu Hien, Hoang Nhat Duc // .- 2023 .- Số 03(58) .- P. 3-11 .- 690
This paper uses XGBoost to predict bearing capacity of concrete piles. The proposed model is trained and tested against a dataset of 472 samples collected from static load tests in Vietnam. The results indicate that the default XGBoost model consistently outperforms the Deep Neural Network (DNN) regression. XGBoost is a suitable tool for engineers to predict pile bearing capacity.
3 Training deep neural network for regression analysis with the generalized delta rule : a case study in modeling the shear strength of soil = Huấn luyện mạng nơ-ron thần kinh nhân tạo sâu dùng cho phân tích hồi quy dựa trên quy tắc delta khái quát: ứng dụn / Hoang Nhat Duc, Tran Xuan Linh, Nguyen Quoc Lam // .- 2023 .- Số 03 (58) - Tháng 6 .- P. 23-30 .- 690
This article presents the method for training a deep artificial neural network (DANN) for regression analysis; this method is based on the generalized delta rule. To illustrate the rule, a DANN with two hidden layers is used. The model’s construction is described in the form of mathematical equations. Subsequently, a DANN program is written in Visual C# .NET. This program is tested with the task of estimating the shear strength of soil samples.
4 Linear regression models for predicting the compressive strength of rice husk ash-blended concrete = Ứng dụng các mô hình hồi quy tuyến tính cho việc dự báo cường độ chịu nén của bê tông có chứa tro trấu / Hoang Nhat Duc, Nguyen Quoc Lam // .- 2023 .- Số 5(60) .- P. 39-43 .- 690
This paper constructs linear regression models for estimating the compressive strength (CS) of rice husk ash (RHA)- blended concrete. The conventional multiple linear regression model and multivariate power function-based model are employed. Experimental results show that the performance of the latter is better than that of the former. The multivariate power function-based regression model can achieve good prediction results with a mean absolute percentage error of 14%.
5 Applications of Google OR-Tools in solving construction management linear optimization problems = Ứng dụng công cụ Google OR-Tools trong giải các bài toán tối ưu hóa tuyến tính trong quản lý dự án xây dựng / Hoang Nhat Duc, Nguyen Quoc Lam // .- 2021 .- Số 06(49) .- P. 3-7 .- 692
This study demonstrates the application of Google OR-Tools as a powerful, easy-to-use, and open source software suite in tackling linear programs. Herein, this open source tool has been integrated with Visual C# programming and the optimization models are formulated in Microsoft Visual Studio. Two optimization case studies are used in this study to illustrate the implementation of the Google OR-Tools.
6 Experimental study on influence of rice husk ash on mortar compressive strength at different temperatures = Nghiên cứu thực nghiệm ảnh hưởng của tro trấu tới cường độ của vữa ở các nhiệt độ khác nhau / Tran Thu Hien // .- 2022 .- Số 03(52) .- P. 51-59 .- 624
An experimental study on compressive strength of mortar mixes containing different amounts of rice husk ash (5, 10, 20%) has been investigated. The impact of different elevated temperatures (150, 300, 450, 750oC) on mortar strength is also examined. The experiments found that RHA can be used in its untreated form up to 10% as cementitious replacement without detriment of mortar compressive strength.
7 Artificial neural network with adaptive moment estimation training approaches for prediction of punching shear capacity of steel fibre reinforced concrete slabs = Sử dụng mạng nơ-ron thần kinh nhân tạo với phương pháp huấn luyện ước tính mô men tự thích n / Hoang Nhat Duc // .- 2022 .- Số 03(52) .- P. 18-22 .- 690
stimating punching shear capacity (PSC) of steel fibre reinforced concrete slabs (SFRCS) is a crucial task in structural design. This study investigates the performances of artificial neural networks trained by the adaptive moment estimation (Adam) method in dealing with the task of interest. To alleviate overfitting problem, decoupled weight decay (AdamW) and L2 regularization (AdamL2) are used.
8 Influence of rice husk ash on mortar compressive strength at different temperatures : machine learning based modelling = Ảnh hưởng của tro trấu tới cường độ của vữa ở các nhiệt độ khác nhau : mô hình hóa bằng máy học / Tran Thu Hien, Hoang Nhat Duc // .- 2022 .- Số 05(54) .- P. 27-36 .- 624
This research is dedicated to the extension of the body of knowledge by proposing the application of three machine learning models for predicting compressive strength of mortar mixtures containing RHA. The Artificial Neural Network (ANN), Least Squares Support Vector Regression (LS-SVR) and Multivariate Adaptive Regression Splines (MARS) models were selected.
9 Image processing-based automatic gradation of stone aggregates = Tự động hóa việc xác định cấp phối hạt của cốt liệu đá sử dụng kỹ thuật xử lý ảnh / Hoang Nhat Duc, Nguyen Quoc Lam, Pham Quang Nhat // .- 2022 .- Số 05(54) .- P. 37-42 .- 624
This paper aims to employ image processing methods to develop a simple tool for automatic gradation of stone aggregates. The methods of image thresholding, Gaussian filtering, median filtering, morphological closing, and contour analysis are employed.
10 A review of metaheuristic optimized machine learning regression with applications in construction engineering = Khảo sát các mô hình học máy được tối ưu hóa bởi các thuật toán tìm kiếm với ứng dụng cho phân tích hồi quy trong ngành xây dựng / Hoang Nhat Duc, Nguyen Quoc Lama, Tran Van Duc // .- 2022 .- Số 05(54) .- P. 43-49 .- 624
This article aims at reviewing state-of-the-art research works involving the use of metaheuristic optimized machine learning regression models. Recent research articles published in the time period of 2019-2021 are surveyed.