A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Load validity discrimination for telescopic boom of rock drilling jumbo
Abstract The telescopic boom is an essential component in the rock drilling jumbo. To avoid the influence of the invalid load schemes on optimization design, it is urgent to find a method that can efficiently discriminate the validity of load schemes. Based on the combination of Classroom-like Generative Adversarial Network (CLGAN) and Conditional Generative Adversarial Network (CGAN), Conditional Classroom-like Generative Adversarial Network (CCLGAN) is proposed in this paper to discriminate the validity of load schemes. And comparative experiments are designed to compare the effects of different training strategies, loss functions, and discriminators with different structures on the discrimination ability of discriminators. The comparison results show that the improved training strategy, Wasserstein loss, and the discriminator with 2 output nodes have better performance. The optimal CCLGAN is trained to discriminate the load schemes, and the discriminator in the optimal CCLGAN including 5 different types of generators has the best discrimination performance.
Highlights Load validity discrimination is regarded as a classification problem. The discriminator in proposed CCLGAN is used for load validity discrimination. Influence of different factors on the performance of discriminator are analyzed. Optimal CCLGAN is investigated to further improve the accuracy of discriminator.
Load validity discrimination for telescopic boom of rock drilling jumbo
Abstract The telescopic boom is an essential component in the rock drilling jumbo. To avoid the influence of the invalid load schemes on optimization design, it is urgent to find a method that can efficiently discriminate the validity of load schemes. Based on the combination of Classroom-like Generative Adversarial Network (CLGAN) and Conditional Generative Adversarial Network (CGAN), Conditional Classroom-like Generative Adversarial Network (CCLGAN) is proposed in this paper to discriminate the validity of load schemes. And comparative experiments are designed to compare the effects of different training strategies, loss functions, and discriminators with different structures on the discrimination ability of discriminators. The comparison results show that the improved training strategy, Wasserstein loss, and the discriminator with 2 output nodes have better performance. The optimal CCLGAN is trained to discriminate the load schemes, and the discriminator in the optimal CCLGAN including 5 different types of generators has the best discrimination performance.
Highlights Load validity discrimination is regarded as a classification problem. The discriminator in proposed CCLGAN is used for load validity discrimination. Influence of different factors on the performance of discriminator are analyzed. Optimal CCLGAN is investigated to further improve the accuracy of discriminator.
Load validity discrimination for telescopic boom of rock drilling jumbo
Lin, Lin (author) / Lv, Yancheng (author) / Guo, Hao (author) / Liu, Jie (author) / Tong, Changsheng (author) / Zu, Lizheng (author) / He, Wenhui (author)
2022-06-14
Article (Journal)
Electronic Resource
English