Entropy production in quasi-isentropic compression (QIC) is critically important for understanding the properties of materials under extreme conditions. However, the origin and accurate quantification of entropy in this situation remain long-standing challenges. In this work, a framework is established for the quantification of entropy production and partition, and their relation to microstructural change in QIC. Cu₅₀Zr₅₀ is taken as a model material, and its compression is simulated by molecular dynamics. On the basis of atomistic simulation-informed physical properties and free energy, the thermodynamic path is recovered, and the entropy production and its relation to microstructural change are successfully quantified by the proposed framework. Contrary to intuition, entropy production during QIC of metallic glasses is relatively insensitive to the strain rate γ̇ when γ̇ ranges from 7.5 × 10⁸ to 2 × 10⁹/s, which are values reachable in QIC experiments, with a magnitude of the order of 10^-2k_B/atom per GPa. However, when γ̇ is extremely high (>2×109/s), a notable increase in entropy production rate with γ̇ is observed. The Taylor–Quinney factor is found to vary with strain but not with strain rate in the simulated regime. It is demonstrated that entropy production is dominated by the configurational part, compared with the vibrational part. In the rate-insensitive regime, the increase in configurational entropy exhibits a linear relation to the Shannon-entropic quantification of microstructural change, and a stretched exponential relation to the Taylor–Quinney factor. The quantification of entropy is expected to provide thermodynamic insights into the fundamental relation between microstructure evolution and plastic dissipation.

核事故的发生具有不可预测性和破坏性，为应急车辆制定合理的疏散计划将危险区域的人员撤离至安置点，可以有效减少人员所受到的伤害。针对核事故下应急车辆路径规划问题，以累积辐射剂量为评价指标，提出了一种基于混合蚁群算法（Hybrid ant colony algorithm，HACO）的车辆路径规划方法。首先，利用模糊网络建立了时间窗内疏散路径平均通行时间期望模型，同时结合累积辐射剂量计算模型，建立了能够随时间变化的动态累积辐射剂量计算模型。然后在蚁群算法迭代过程中引入模拟退火算法，并且在邻域搜索中引入A*算法启发式思想，提高了算法全局寻优能力。为进一步提高算法的局部搜索能力，引入帕累托排序方式，在蚁群算法信息素更新方式中加入距离对信息素增量的影响。仿真结果表明：HACO算法相较于蚁群算法平均收敛值提高了31%，稳定性提高了30%，能够为核事故下疏散路径规划预案的制定提供技术支持。

该文提出了一种基于手机的电磁波实验教学设计，将手机变身为微波功率计，利用Wi-Fi路由器作为发射信号源，开展电磁波传播特性、电磁波极化特性和电磁屏蔽效能的实验学习。该实验设备普及、操作简便且实验时间和地点不受限制，能够满足互联网时代学生的多样化学习需求；另外，该实验设计结合生活实践，鼓励学生自由发挥，对提升学习兴趣、培养创新思维有积极作用。