苏理云指导研究生熊浪、汪小意、潘纯权发表在《Applied Soft Computing》上的论文获得2024年川渝科技学术大会优秀论文二等奖！

       Xiong L.,  Su Liyun*, Wang X., Pan C., Dynamic Adaptive Graph Convolutional Transformer with Broad Learning System for Multi-Dimensional Chaotic Time Series Prediction, Applied Soft Computing, 2024. https://www.sciencedirect.com/science/article/pii/S1568494624002904

      按照川渝科技学术大会（以下简称“大会”）惯例，今年川渝两地将继续分别评选川渝科技学术大会优秀论文、年度川渝最具影响力十大学术活动、川渝一流学会和川渝一流科技期刊。经自愿申报、资格审查、专家评审等环节，并经市科协党组会审议，重庆市拟产生2024川渝科技学术大会优秀论文92篇，年度川渝最具影响力学术活动、年度川渝一流学会和年度川渝一流科技期刊各10个。现将相关名单予以公示。

http://www.cqast.cn/htm/1316451850609426432/web/content_1316451850609426432.htm

    论文摘要：

      Chaotic time series data is extensively applied in financial stocks, climate monitoring, and sea clutter, in which data fusion from various sources and multi-sensor information make accurate predictions of chaotic time series challenging under complex nonlinear conditions. Previous works focus on designing different model frameworks to capture the temporal dependence and extract richer nonlinear features to improve the accuracy of univariate chaotic time series prediction, which ignores the spatial dependence of multivariable. However, in this paper, we argue that spatial correlation among multiple variables is essential to improve the prediction accuracy of chaotic time series. To fill the gap, we innovatively propose a Dynamic Adaptive Graph Convolutional Transformer with a Broad Learning System (DAGCT-BLS), a GCN and Transformer-based model utilizing multivariate spatial dependence for multi-dimensional chaotic time series forecasting. In DAGCT-BLS, the multivariate chaotic time series are reconstructed into the phase space, and the reconstructed data are rapidly feature-extracted using a cascade network BLS with frozen weights to maximize the retention of chaotic properties and nonlinear relationships. Then, the Dynamic Adaptive Graph Convolutional Network (DAGCN) is proposed to capture the spatial correlation among the multiple variables. Finally, improved multi-head attention of the Transformer Encoder is used to capture the temporal dependence of the phase point sequence. Experiments of our proposed model on three datasets (Lorenz, Rossler, and Sea clutter) show that DAGCT-BLS can achieve the best prediction performance and have strong interpretability, and multivariate-based joint modeling of chaotic time series helps to improve the prediction performance.

    混沌时间序列数据广泛应用于金融股票、气候监测和海杂波等领域，在复杂的非线性条件下多种来源的数据融合和多传感器信息使得混沌时间序列的准确预测具有挑战性。以往的工作侧重于设计不同的模型框架来捕捉时间依赖性并提取更丰富的非线性特征，以提高单变量混沌时间序列预测的准确性，而忽略了多变量的空间依赖性。在本文中，充分利用多变量之间的空间相关性，创新性地提出了一种具有宽度学习系统的动态自适应图卷积变换器（DAGCT-BLS），这是一种基于图卷积神经网络（GCN）和变换器（Transformer）的融合模型，利用多元空间依赖性进行多维混沌时间序列预测。在DAGCT-BLS中，将多元混沌时间序列重建到相空间中，并使用具有冻结权重的级联宽度学习网络（BLS）快速提取重建数据的特征，以最大限度地保留混沌特性和非线性关系。然后，提出了动态自适应图卷积网络（DAGCN）来捕捉多个变量之间的空间相关性。最后，使用变压器编码器（Transformer）的改进多头注意力来捕捉相位点序列的时间依赖性。我们提出的模型在三个数据集（洛伦兹、罗斯勒和海杂波）上的实验表明，DAGCT-BLS可以实现最佳的预测性能，并具有很强的可解释性，基于多变量的混沌时间序列联合建模有助于提高预测性能。

 Applied Soft Computing：SCI 一区Top期刊.

     Applied Soft Computing is an international journal promoting an integrated view of soft computing to solve real life problems. Soft computing is a collection of methodologies, which aim to exploit tolerance for imprecision, uncertainty and partial truth to achieve tractability, robustness and low solution cost. The focus is to publish the highest quality research in application and convergence of the areas of Fuzzy Logic, Neural Networks, Evolutionary Computing, Rough Sets and other similar techniques to address real world complexities.

论文可以参见：

https://www.x-mol.com/groups/su_liyun/news/61604