Learn decision trees with deep visual primitives

In this paper, we strive to propose a self-interpretable framework, termed PrimitiveTree, that incorporates deep visual primitives condensed from deep features with a conventional decision tree, bridging the gap between deep features extracted from deep neural networks (DNNs) and trees’ transparent decision-making processes. Specifically, we utilize a codebook, which embeds the continuous deep features into a finite discrete space (deep visual primitives) to distill the most common semantic information. The decision tree adopts the spatial location information and the mapped primitives to present the decision-making process of the deep features in a tree hierarchy. Moreover, the trained interpretable PrimitiveTree can inversely explain the constituents of the deep features, highlighting the most critical and semantic-rich image patches attributing to the final predictions of the given DNN. Extensive experiments and visualization results validate the effectiveness and interpretability of our method.     

A data-driven digital-twin prognostics method for proton exchange membrane fuel cell remaining useful life prediction

Prognostics and health management of proton exchange membrane fuel cell (PEMFC) systems have driven increasing research attention in recent years as the durability of PEMFC stack remains as a technical barrier for its large-scale commercialization. To monitor the health state during PEMFC operation, digital twin (DT), as a smart manufacturing technique, is applied in this paper to establish an ensemble remaining useful life prediction system. A data-driven DT is constructed to integrate the physical knowledge of the system and a deep transfer learning model based on stacked denoising autoencoder is used to update the DT with online measurement. A case study with experimental PEMFC degradation data is presented where the proposed data-driven DT prognostics method has applied and reached a high prediction accuracy. Furthermore, the predicted results are proved to be less affected even with limited measurement data.     

A practical development of engineering simulation-assisted educational AR environments

Engineering simulations have opened several gates for today’s chemical engineers. They are powerful tools to provide technical content as physics-based numerical solvers. Augmented reality (AR) and virtual reality (VR), on the other hand, are already underway to digitize environments in many fields. The combination of AR/VR environments and simulations in engineering education has been attracting widespread interest. Literature has demonstrated a massive amount of digital educational environments in several contexts as being complementary to conventional educational methods. Nevertheless, hosting technical content produced by engineering simulations with educational AR/VR is still challenging and requires expertise from multiple disciplines throughout the technical development. Present work provides a facile and agile methodology for low-cost hardware but content-wise rich AR software development. Inspired by the Covid-19 pandemic, a case study is developed to teach chemical-engineering concepts using a liquid-soap synthesis process. Accordingly, we assess and conclude the digital development process to guide inexperienced developers for the digitalization of teaching content. The present contribution serves as an example of the power of integrating AR/VR with traditional engineering simulations for educational purposes. The digital tool developed in this work is shared in the online version.     

Structural Aspects and Prediction of Calmodulin-Binding Proteins

Calmodulin (CaM) is an important intracellular protein that binds Ca²⁺ and functions as a critical second messenger involved in numerous biological activities through extensive interactions with proteins and peptides. CaM’s ability to adapt to binding targets with different structures is related to the flexible central helix separating the N- and C-terminal lobes, which allows for conformational changes between extended and collapsed forms of the protein. CaM-binding targets are most often identified using prediction algorithms that utilize sequence and structural data to predict regions of peptides and proteins that can interact with CaM. In this review, we provide an overview of different CaM-binding proteins, the motifs through which they interact with CaM, and shared properties that make them good binding partners for CaM. Additionally, we discuss the historical and current methods for predicting CaM binding, and the similarities and differences between these methods and their relative success at prediction. As new CaM-binding proteins are identified and classified, we will gain a broader understanding of the biological processes regulated through changes in Ca²⁺ concentration through interactions with CaM.     

Use of a smartphone-based Student Response System in large active-learning Chemical Engineering Thermodynamics classrooms

Smartphones are a promising tool as student response systems (SRS) for interactive teaching due to their widespread diffusion. Here, the main purpose is to assess the efficacy of smartphone-based SRS in large classroom settings of undergraduate Thermodynamics, as representative of engineering courses requiring high-level cognitive skills for problem solving. Four sets of multiple-choice questions were presented during the course. Overall, the results refer to 1055 students between control and SRS classes, each corresponding to a3 years period.One of the main results of this work is the strong linear correlation between the average questionnaire score and the final exam grade (R² = 0.91). A similar correlation, although with a lower value of R², is already found in the first questionnaire, thus showing the SRS high predictive power of class performance. The results of this study provide guidance for a quantitative use of smartphone-based SRS in teaching basic disciplines. The SRS monitoring capability allows early detection of struggling students, thus paving the way to personalized tutoring and improved student engagement in active learning practices. This approach is especially important in emergency situations, such as the SARS-Cov-2 pandemic, when distance learning is widely adopted, and remote interactive tools are highly needed.     

基于阶段注意力机制的电力线提取算法

电力系统维护是电力系统稳定运行的重要保障，应用智能算法的无人机电力巡检则为电力系统维护提供便捷。电力线提取是自主电力巡检以及保障飞行器低空飞行安全的关键技术，结合深度学习理论进行电力线提取是电力巡检的重要突破点。本文将深度学习方法用于电力线提取任务，结合电力线图像特点嵌入改进的图像输入策略和注意力模块，提出一种基于阶段注意力机制的电力线提取模型（SA-Unet）。本文提出的SA-Unet模型编码阶段采用阶段输入融合策略（Stage input fusion strategy， SIFS），充分利用图像的多尺度信息减少空间位置信息丢失。解码阶段通过嵌入阶段注意力模块（Stage attention module，SAM）聚焦电力线特征，从大量信息中快速筛选出高价值信息。实验结果表明，该方法在复杂背景的多场景中具有良好的性能。     

Identifying catalyst layer compositions of proton exchange membrane fuel cells through machine-learning-based approach

Membrane electrode assembly (MEA) is considered a key component of a proton exchange membrane fuel cell (PEMFC). However, developing a new MEA to meet desired properties, such as operation under low-humidity conditions without a humidifier, is a time- and cost-consuming process. This study employs a machine-learning-based approach using K-nearest neighbor (KNN) and neural networks (NN) in the MEA development process by identifying a suitable catalyst layer (CL) recipe in MEA. Minimum redundancy maximum relevance and principal component analysis were implemented to specify the most important predictor and reduce the data dimension. The number of predictors was found to play an essential role in the accuracy of the KNN and NN models although the predictors have self-correlations. The KNN model with a K of 7 was found to minimize the model loss with a loss of 11.9%. The NN model constructed by three corresponding hidden layers with nine, eight, and nine nodes can achieve the lowest error of 0.1293 for the Pt catalyst and 0.031 for PVA as a good additive blending in the CL of the MEA. However, even if the error is low, the prediction of PVA seems to be inaccurate, regardless of the model structure. Therefore, the KNN model is more appropriate for CL recipe prediction.     

Neural network-based flight control systems: Present and future

As the first review in this field, this paper presents an in-depth mathematical view of Intelligent Flight Control Systems (IFCSs), particularly those based on artificial neural networks. The rapid evolution of IFCSs in the last two decades in both the methodological and technical aspects necessitates a comprehensive view of them to better demonstrate the current stage and the crucial remaining steps towards developing a truly intelligent flight management unit. To this end, in this paper, we will provide a detailed mathematical view of Neural Network (NN)-based flight control systems and the challenging problems that still remain. The paper will cover both the model-based and model-free IFCSs. The model-based methods consist of the basic feedback error learning scheme, the pseudocontrol strategy, and the neural backstepping method. Besides, different approaches to analyze the closed-loop stability in IFCSs, their requirements, and their limitations will be discussed in detail. Various supplementary features, which can be integrated with a basic IFCS such as the fault-tolerance capability, the consideration of system constraints, and the combination of NNs with other robust and adaptive elements like disturbance observers, would be covered, as well. On the other hand, concerning model-free flight controllers, both the indirect and direct adaptive control systems including indirect adaptive control using NN-based system identification, the approximate dynamic programming using NN, and the reinforcement learning-based adaptive optimal control will be carefully addressed. Finally, by demonstrating a well-organized view of the current stage in the development of IFCSs, the challenging issues, which are critical to be addressed in the future, are thoroughly identified. As a result, this paper can be considered as a comprehensive road map for all researchers interested in the design and development of intelligent control systems, particularly in the field of aerospace applications.     

Integral barrier Lyapunov function-based adaptive fuzzy output feedback control for nonlinear delayed systems with time-varying full-state constraints

In this article, an adaptive fuzzy output feedback control method is presented for nonlinear time-delay systems with time-varying full state constraints and input saturation. To overcome the problem of time-varying constraints, the integral barrier Lyapunov functions (IBLFs) integrating with dynamic surface control (DSC) are applied for the first time to keep the state from violating constraints. The effects of unknown time delays can be removed by using designed Lyapunov-Krasovskii functions (LKFs). An auxiliary design system is introduced to solve the problem of input saturation. The unknown nonlinear functions are approximated by the fuzzy logic systems (FLS), and the unmeasured states are estimated by a designed fuzzy observer. The novel controller can guarantee that all signals remain semiglobally uniformly ultimately bounded and satisfactory tracking performance is achieved. Finally, two simulation examples illustrate the effectiveness of the presented control methods.     

基于启发式BP算法的软件缺陷预测模型

针对软件缺陷预测时缺陷数据集中存在的类别分布不平衡问题，结合上采样算法SMOTE与Edited Nearest Neighbor (ENN) 数据清洗策略，提出了一种基于启发式BP神经网络算法的软件缺陷预测模型。模型中采用上采样算法SMOTE增加少数类样本以改善项目中的数据不平衡状况，并针对采样后数据噪声问题进行ENN数据清洗，结合基于启发式学习的模拟退火算法改进四层BP神经网络后建立分类预测模型，在AEEEM数据库上使用交叉验证对提出的方案进行性能评估，结果表明所提出的算法能够有效提高模型在预测类不平衡数据时的分类准确度。