Dự báo khả năng chịu tải nén lệch tâm của cột thép nhồi bê tông bằng mô hình học máy được tối ưu hóa bằng thuật toán Jellyfish Search=Predicting the eccentric axial load capacity of concrete filled steel tube columns using a machine learning model optimized by Jellyfish Search algorithm

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Ks Trần Hữu Thắng, Hữu Anh Tuấn Nguyễn, Ths Lê Thị Thùy Linh, Ts Trương Đình Nhật

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Tạp chí Xây dựng, 2024

Mô tả vật lý: tr.76

Bộ sưu tập: Metadata

ID: 483990

Thêm vào giỏ

Concrete-filled steel tubes (CFST) are becoming increasingly popular in civil and transportation projects due to their significant advantages over traditional reinforced concrete structures. This study aims to develop a predictive model using the Jellyfish Search (JS) optimization algorithm to automatically optimize machine learning parameters for predicting the eccentric compressive strength of CFST. A dataset of 499 samples, comprising 11 input variables and the target variable ofeccentric compressive capacity, was used to develop individual machine learning models (ANN, SVR, CART, LR) and ensemble models (Voting, Bagging, Stacking). After building and comparing these models, the most accurate one was selected for combination with the JS optimization algorithm to achieve the highest-performing predictive model. The results are highly promising, with R = 0.9949, MAE = 46.8157 kN, RMSE = 7.2097 kN, MAPE = 7.67%, and SI = 0.00 (Rank = 1),indicating that this model holds great potential for CFST design and structural analysis. Keywords: Concrete filled steel columns
eccentric axial load
machine learning model
optimization algorithm.Cột thép nhồi bê tông (Concrete-filled steel tube, CFST) ngày càng được ưa chuộng trong các công trình dân dụng và giao thông hiện nay bởi các ưu thế vượt trội của nó so với kết cấu bê tông cốt thép thông thường. Nghiên cứu này tập trung vào xây dựng mô hình dự báo dựa trên thuật toán tối ưu hóa Jellyfish Search (JS) để tự động tìm kiếm các thông số của mô hình học máy nhằm dự báo độ nén lệch tâm của CFST. Một bộ dữ liệu gồm 499 mẫu với 11 biến đầu vào và đầu ra là khả năng chịu nén lệch tâm của cột được sử dụng để xây dựng các mô hình học máy đơn ANN, SVR, CART, LR và mô hình hỗn hợp Voting, Bagging, Stacking. Sau khi xây dựng và so sánh mô hình trên, mô hình có độ chính xác nhất được chọn để kết hợp với thuật toán tối ưu hóa JS nhằm tạo ra mô hình dự báo có hiệu suất cao nhất. Kết quả thu được rất khả quan với với R= 0.9949, MAE= 46.8157 kN, RMSE= 7.2097 kN và MAPE= 7.67%, SI=0.00 (Rank=1) chứng tỏ đây là một mô hình đầy hứa hẹn để sử dụng trong thiết kế và phân tích kết cấu CFST. Từ khóa: Cột ống thép nhồi bê tông
tải trọng lệch tâm
mô hình học máy
tối ưu hóa.

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR