Prediction of Concrete Compressive Strength Based on ISSA-BPNN-AdaBoost DOI Open Access
Ping Li,

Zichen Zhang,

Jiming Gu

и другие.

Materials, Год журнала: 2024, Номер 17(23), С. 5727 - 5727

Опубликована: Ноя. 22, 2024

Strength testing of concrete mainly relies on physical experiments, which are not only time-consuming but also costly. To solve this problem, machine learning has proven to be a promising technological tool in strength prediction. In order improve the accuracy model predicting compressive concrete, paper chooses optimize base learner ensemble model. The position update formula search phase sparrow algorithm (SSA) is improved, and piecewise chaotic mapping adaptive t-distribution variation added, enhances diversity population improves algorithm's global convergence abilities. Subsequently, effectiveness improvement strategy was demonstrated by comparing improved (ISSA) with some commonly used intelligent optimization algorithms 10 test functions. A back propagation neural network (BPNN) optimized ISSA as learner, boosting (AdaBoost) train integrate multiple learners, thus establishing an based (ISSA-BPNN-AdaBoost) prediction Then comparison experiments were conducted other models single two datasets. experimental results show that ISSA-BPNN-AdaBoost exhibits excellent both datasets can accurately perform strength, demonstrating superiority strength.

Язык: Английский

A multi-mechanism balanced advanced learning sparrow search algorithm for UAV path planning DOI
Chao Yang, Hong Yang, Donglin Zhu

и другие.

Cluster Computing, Год журнала: 2024, Номер 27(5), С. 6623 - 6666

Опубликована: Март 5, 2024

Язык: Английский

Процитировано

4
An enhanced sparrow search swarm optimizer via multi-strategies for high-dimensional optimization problems DOI
Shuang Liang, Minghao Yin, Geng Sun

и другие.

Swarm and Evolutionary Computation, Год журнала: 2024, Номер 88, С. 101603 - 101603

Опубликована: Май 18, 2024

Язык: Английский

Процитировано

3
DYNet: A Printed Book Detection Model Using Dual Kernel Neural Networks DOI Creative Commons
Lubin Wang, Xiaolan Xie, Peng Huang

и другие.

Sensors, Год журнала: 2023, Номер 23(24), С. 9880 - 9880

Опубликована: Дек. 17, 2023

Target detection has always been a hotspot in image processing/computer vision research, and small-target is frequently encountered problem the field of target detection. With continuous innovation technology, people hope that small targets can reach real-time accuracy large-target In this paper, model based on dual-core convolutional neural networks (CNN) proposed, which mainly used for intelligent books production line printed books. The composed two modules, including region prediction module suspicious search module. uses CNN to predict blocks large context. different from above find tiny predicted blocks. Comparative testing four book samples using shows better compared other models.

Язык: Английский

Процитировано

2
Prediction of Concrete Compressive Strength Based on ISSA-BPNN-AdaBoost DOI Open Access
Ping Li,

Zichen Zhang,

Jiming Gu

и другие.

Materials, Год журнала: 2024, Номер 17(23), С. 5727 - 5727

Опубликована: Ноя. 22, 2024

Strength testing of concrete mainly relies on physical experiments, which are not only time-consuming but also costly. To solve this problem, machine learning has proven to be a promising technological tool in strength prediction. In order improve the accuracy model predicting compressive concrete, paper chooses optimize base learner ensemble model. The position update formula search phase sparrow algorithm (SSA) is improved, and piecewise chaotic mapping adaptive t-distribution variation added, enhances diversity population improves algorithm's global convergence abilities. Subsequently, effectiveness improvement strategy was demonstrated by comparing improved (ISSA) with some commonly used intelligent optimization algorithms 10 test functions. A back propagation neural network (BPNN) optimized ISSA as learner, boosting (AdaBoost) train integrate multiple learners, thus establishing an based (ISSA-BPNN-AdaBoost) prediction Then comparison experiments were conducted other models single two datasets. experimental results show that ISSA-BPNN-AdaBoost exhibits excellent both datasets can accurately perform strength, demonstrating superiority strength.

Язык: Английский

Процитировано

0