A new type of assembled concrete beam–column joint based on a bolted connection was proposed, aiming to complete the post-earthquake node repair work by replacing precast beams and bolts. To study the seismic performance of the new beam–column joints, two full-scale components of the new joints were fabricated and subjected to low cyclic loading. The whole process from crack generation to component failure was investigated in detail, and seismic performance indicators such as the hysteresis curve, skeleton curve and stiffness degradation curve were compared and analyzed. Based on the experimental results, ABAQUS finite element software was applied to numerically simulate cast-in-place joints and test joints. Based on the failure mechanism of the new assembled beam–column semi-rigid joints, a stress analysis of semi-rigid joints was carried out. The research results show that the two new joints have good seismic performance and energy dissipation performance. Bolts and precast beams are the main stress components, and the repair of new joints can be completed by replacing bolts, which meets the seismic design concepts of “strong columns and weak beams” and “strong joints and weak components”. The larger the diameter of the bolts, the higher the load capacity and the lower the stiffness degradation rate. The finite element simulation results are high-accuracy and can well reflect the seismic performance of the components. It is found that cast-in-place joints are better in energy dissipation capacity than test joints, but the ultimate bearing capacity of test joints is better than that of cast-in-place joints. Based on the experimental stress characteristics of the nodal core zone, a mechanical analysis model of the nodal core zone of the new assembled concrete beam–column joints is proposed, and shear force calculation equations for the core zone of the new assembled concrete beam–column rigid joints and semi-rigid joints are derived.

Knowledge tracing (KT) is the core task of computer-aided education systems, and it aims at predicting whether a student can answer the next exercise (i.e., question) correctly based on his/her historical answer records. In recent years, deep neural network-based approaches have been widely developed in KT and achieved promising results. More recently, several researches further boost these KT models via exploiting plentiful relationships including exercise-skill relations (E-S), the exercise similarity (E-E) as well as skill similarity (S-S). However, these relationship information are frequently absent in many real-world educational applications, and it is a labor-intensive work for human experts to label it. Inspired by recent advances in natural language processing domain, we propose a novel pre-training approach, namely as SPAKT, and utilize self-supervised learning to pre-train exercise embedding representation without the need for expensive human-expert annotations in this paper. Contrary to existing pre-training methods that highly rely on manually labeling knowledge about the E-E, S-S, or E-S relationships, the core idea of the proposed SPAKT is to design three self-attention modules to model the E-S, E-E, and S-S relationships, respectively, and all of these three modules can be trained in the self-supervised setting. As a pre-training approach, our SPAKT can be effortlessly incorporated into existing deep neural network-based KT frameworks. We experimentally show that, even without using expensive annotations about the aforementioned three kinds of relationships, our model achieves competitive performance compared with state-of-the-arts. Our algorithm implementations have been made publicly available at https://github.com/Vinci-hp/pretrainKT .

The study of multiobjective optimization problems, such as resources, time limits, and cost based on network planning, can lead to an intelligent construction plan, which is very important for improving the comprehensive benefit of the project. In this article, a multiobjective static network planning optimization (i.e., sNP-RTC) model is proposed to integrate the heuristic local search algorithm and adaptive operation to improve the multiobjective particle swarm optimization (MOPSO) algorithm to solve the comprehensive “resource–duration–cost” optimization problem. The results show that this algorithm can achieve a unified and comprehensive Pareto frontier compared with the genetic MOPSO, providing a new method for solving resource–duration–cost optimization problems. Moreover, a dynamic “resource–time–cost” problem model for dynamic network planning (i.e., dNP-RTC) is presented and applied to improve the multiobjective subgroup algorithm to solve the model in real time after updating the network planning parameters. Compared with the static problem, the model increases the flexibility of problem solving, is more in-line with engineering practice, saves more cost in work scheduling, and has more application value for actual project scheduling.

A new assembled bolt-connected concrete beam–column joint is proposed, aimed at completing the repair of a post-earthquake node by replacing the bolts and precast beams. Low-cycle loading tests were performed on two new full-scale connections to investigate the effect of bolt strength on the seismic performance of the new connections. A finite element model was established based on the experimental node specimens and compared with the experimental results to verify the accuracy of the finite element simulation results. The seismic performance of the new joints under different axial ratios was studied using finite element software to determine the effect of the axial pressure ratio on the seismic performance of the new joints. Based on the research carried out, a new improved joint was designed, numerical models of the improved joint were established using finite element software, and the seismic performance of the improved joint was compared with the results of the experimental simulation to analyze the seismic performance of the improved joints. The results of the study showed that the bolts and precast concrete beams are the main load-bearing members in the period of service. The joint can be repaired by replacing the bolts and precast concrete beams under seismic action, which meets the new joint design concepts. The finite element simulation results are in good agreement with the experimental results. The larger the axial compression ratio, the earlier the failure stage of the concrete, and the faster the bearing capacity and ductility decrease. The larger the axial compression ratio, the higher the initial stiffness of the joints and the greater the rate of stiffness reduction. The bolt stress distribution of the modified and optimized joints is more satisfactory. This change in node form can improve the recovery efficiency of the joint to a certain extent.