This paper reports an investigation on the tensile resistance of fully-penetrated butt-welded connections of (ultra-)high-strength steels considering the strength softening of the heat-affected zone (HAZ) and the strengthening due to the transverse constraint. Firstly, a parametric study is carried out to analyse the tensile behaviour of fully-penetrated butt-welded S700 connections using validated finite element (FE) models. Based on the FE results, the effect of the investigated parameters on the ultimate resistance of the connections is quantified. Consequently, a Macro prediction model is proposed, neglecting the detail of HAZ inhomogeneous microstructures but considering five design parameters, which are the width and thickness of the specimen, the width of HAZ, the softening ratio, and the matching ratio. Finally, the proposed Macro model is validated against the available experimental data of fully-penetrated butt-welded (ultra-)high-strength steel connections in literature. The results show that the Macro model is proven successful in accurately quantifying the effect of HAZ, concerning the strength softening and the transverse constraint strengthening, on the tensile resistance of fully-penetrated butt-welded (ultra-)high-strength steel connections. It is also found that 157 and 96 out of 185 experiments are predicted within a 5% deviation range using the Macro model and prEN 1993-1-8:2021, respectively, proving that the Macro model is more consistent with the experimental data compared to the prediction of prEN 1993-1-8:2021.

The problem of learning-based control for robots has been extensively studied, whereas the security issue under malicious adversaries has not been paid much attention to. Malicious adversaries can invade intelligent devices and communication networks used in robots, causing incidents, achieving illegal objectives, and even injuring people. This article first investigates the problems of optimal false data injection attack scheduling and countermeasure design for car-like robots in the framework of deep reinforcement learning. Using a state-of-the-art deep reinforcement learning approach, an optimal false data injection attack scheme is proposed to deteriorate the tracking performance of a robot, guaranteeing the tradeoff between the attack efficiency and the limited attack energy. Then, an optimal tracking control strategy is learned to mitigate attacks and recover the tracking performance. More importantly, a theoretical stability guarantee of a robot using the learning-based secure control scheme is achieved. Both simulated and real-world experiments are conducted to show the effectiveness of the proposed schemes.