The use of protective barriers is one of the most common approaches to protect buildings and their occupants against blast and vehicle impacts. They increase the stand-off distance between the explosive source and the building. However, the protection capabilities of barriers with openings have not yet been thoroughly studied. The present paper discusses the shock wave attenuation effect of protective barriers made of steel posts with a hollow cross section. In the experiments, the steel posts are located at a distance of 5 m to the building to be protected. Prior to the experiments, numerical models were developed to predict the blast loads numerically. Overpressure-time history measurements (side-on and reflected) were made at various distances in front of and behind the barrier. The experimental data were used to, for example, validate the numerical models. The experimental and numerical results showed that barriers can reduce the blast loads relative to the scenario in which no barriers were present. Considerable reductions in peak side-on and reflected overpressure and impulse were observed behind the barriers. Furthermore, after validation, parametric studies are carried out to investigate the influence of further parameters on the overpressure reduction behind the barriers, that is, the number of posts or the spacing between posts, the cross-sectional shapes of posts, and the arrangement of posts (single-layer or multilayer, aligned or staggered). These studies showed that a barrier without openings is not always necessary to offer the desired protection because barriers with openings can also show satisfactory results. Hence, the necessary amount of material (steel in this case) and, thus, the construction cost can be considerably reduced.
Power transformers play an important role in power system. In recent years, with the rapid development of UHV power grid, transformer explosions occurred occasionally, often followed by fires, which caused serious economic loss and casualty. For disaster mitigation of transformer explosion, a balance design concept for power transformer against arc fault explosion was proposed, which includes vulnerability analysis of transformer structure, optimization design, and pressure-relief design scheme. For realizing the balance design, nonlinear dynamic finite element analysis of an 800kV converter transformer considering possible arc failure occurrence and fluid-structure interaction was carried out based on LS-DYNA. The transformer model included tank, ascending flange stiffening rib and winding. The method used to achieve gas production successfully simulates high pressure gas loading. Propagation of pressure wave, pressure distribution on transformer structure, and dynamic response characteristic of the transformer were observed in different typical arc fault occurrence. Weak points of the structure were identified and corresponding design measures about stiffening rib and stress concentration location are proposed to reach optimal design. The design and arrangement of the pressure-relief device for typical arc fault occurrence were suggested. A pressure relief device was set on ascending flange which closed to arc fault. Its quick trigger changed the response characteristic of structure and decreases the peak stress. The efficacy of the proposed design concept was validated through numerical analysis. The research findings from this paper provide reference for improving explosion resilience design of transformer.
Truss layout optimization under complex constraints has been a hot and challenging problem for decades that aims to find the optimal node locations, connection topology between nodes, and cross-sectional areas of connecting bars. Monte Carlo Tree Search (MCTS) is a reinforcement learning search technique that is competent to solve decision-making problems. Inspired by the success of AlphaGo using MCTS, the truss layout problem is formulated as a Markov Decision Process (MDP) model, and a 2-stage MCTS-based algorithm, AlphaTruss, is proposed for generating optimal truss layout considering topology, geometry, and bar size. In this MDP model, three sequential action sets of adding nodes, adding bars, and selecting sectional areas greatly expand the solution space and the reward function gives feedback to actions according to both geometric stability and structural simulation. To find the optimal sequential actions, AlphaTruss solves the MDP model and gives the best decision in each design step by searching and learning through MCTS. Compared with existing results from the literature, AlphaTruss exhibits better performance in finding the truss layout with the minimum weight under stress, displacement, and buckling constraints, which verifies the validity and efficiency of the established algorithm.
Abstract This paper develops a new empirical formula for the prediction of the triple point path in irregular shock reflection cases. Numerical simulations using a two-dimensional axisymmetric multi-material arbitrary Lagrangian–Eulerian formulation are employed to obtain the data of fluid density. Using the data of fluid density and nodal coordinates, the gradients of fluid density are determined and then used to generate numerical schlieren images. Based on these images, the triple point paths are derived and compared with the models of the Unified Facilities Criteria (UFC) and Natural Resources Defense Council (NRDC) as well as two models from the open literature. It is found that the numerically derived triple point paths are in good agreement with those predicted by a recently published model in the open literature for the typical ground range of shock wave propagation of up to 6 m. Considering the whole distance range, it is found that the agreement of different models of the triple point path with the numerical ones depends on the considered blast scenario, i.e., the scaled charge height. For small-scaled charge heights, the model of the UFC and the recently published model in the open literature are in better agreement with the numerical results than the other two models, whereas the NRDC model has the best agreement with the numerical results for large-scaled charge heights. Based on the numerical results, a new empirical formula is proposed for the prediction of the triple point path, which is valid for a wide range of the scaled charge heights from 0.5 to 3.5 m/kg 1/3 and scaled ground distances up to 15 m/kg 1/3 .
Truss layout design aims to find the optimal layout, considering node locations, connection topology between nodes, and cross-sectional areas of connecting bars. The design process of trusses can be represented as a reinforcement learning problem by formulating the optimization task into a Markov Decision Process (MDP). The optimization variables such as node positions need to be transformed into discrete actions in this MDP; however, the common method is to uniformly discretize the design domain by generating a set of candidate actions, which brings dimension explosion problems in spatial truss design. In this paper, a reinforcement learning algorithm is proposed to deal with continuous action spaces in truss layout design problems by using kernel regression. It is a nonparametric regression way to sample the continuous action space and generalize the information about action value between sampled actions and unexplored parts of the action space. As the number of searches increases, the algorithm can gradually increase the candidate action set by appending actions of high confidence value from the continuous action space. The value correlation between actions is mapped by the Gaussian function and Euclidean distance. In this sampling strategy, a modified Confidence Upper Bound formula is proposed to evaluate the heuristics of sampled actions, including both 2D and 3D cases. The proposed algorithm was tested in various layout design problems of planar and spatial trusses. The results indicate that the proposed algorithm has a good performance in finding the truss layout with minimum weight. This implies the validity and efficiency of the established algorithm.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)