论文标题
部分可观测时空混沌系统的无模型预测
Improving Capstone Research Projects: Using Computational Thinking to Provide Choice and Structured Active Learning
论文作者
论文摘要
储层计算是预测湍流的有力工具，其简单的架构具有处理大型系统的计算效率。然而，其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间，然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用，该系统模拟了湍流的若干特征。我们表明，使用径向基函数作为非线性投影器，即使只有部分观测并且不知道控制方程，也能稳健地捕捉复杂的系统非线性。最后，我们表明，当测量稀疏、不完整且带有噪声，甚至控制方程变得不准确时，我们的网络仍然可以产生相当准确的预测，从而为实际湍流系统的无模型预测铺平了道路。
This work presents a structured systematic process for undergraduate capstone research projects embodying computational thinking (CT) practices. Students learn to conduct research with a decision support system utilizing CT. The system is demonstrated through a case study of a capstone research project course. The course is a 3rd year single semester capstone in an aviation program. CT was integrated over a decade, through 21 semesters of coordinating and delivering the course. The CT practices evolved and were utilized for more aspects over time. The CT system facilitated a significant reduction in staff workload by eliminating the need for direct one-on-one supervision and enabling the streamlining of marking. This resulted in fairer marking by eliminating supervisor bias. Student feedback shows a high degree of satisfaction, with comments highlighting choice and learning.
