基于高斯混合隐马尔科夫模型与人工神经网络的紧急换道行为预测方法

为了有效降低因驾驶员紧急换道行为而诱发的交通事故，提高道路交通事故链阻断效率，提出一种基于高斯混合隐马尔科夫模型（GMM-HMM）和人工神经网络（ANN）的紧急换道行为预测方法。首先利用GMM-HMM对车辆行驶状态以及驾驶行为连续观察序列进行换道意图辨识，采用ANN预测下一时段的驾驶行为，再预测换道过程中的横向加速度变化率，从而判断紧急换道的危险程度。驾驶员在环仿真实验及实车实验结果表明，该方法预测避险成功率达92.83%，实验避险成功率达90.32%。该方法能有效地对紧急换道行为进行提前警告与干预。     

氧化石墨烯改性碳纤维复合材料抗冲击行为研究

使用单层纳米氧化石墨烯（NGO）粒子对环氧树脂进行改性处理，采用真空辅助树脂传递模塑成型工艺制备了[±45/0/90]_2S铺层角度下的纯树脂及单层NGO改性碳纤维复合材料（CFRP）层合板。通过落锤冲击试验、超声C扫描检测、冲击后压缩试验等对纯树脂及单层NGO改性CFRP进行实验研究。结果表明，纯树脂及单层NGO改性CFRP在损伤阻抗及损伤容限实验中均存在拐点现象，且拐点出现在相同深度位置，其中纯树脂CFRP拐点位置为0.51 mm，单层NGO改性CFRP拐点位置为0.43 mm；相对于纯树脂CFRP，单层NGO改性CFRP可以显著提高复合材料的抗冲击性能及冲击后的压缩性能；通过对冲击后凹坑深度及凹坑面积进行数据模拟，可以用拟合公式实现对复合材料的损伤预测。     

A novel PEM fuel cell remaining useful life prediction method based on singular spectrum analysis and deep Gaussian processes

Accurate remaining useful life (RUL) prediction of proton exchange membrane fuel cells (PEMFCs) can assess the reliability of fuel cells to determine the occurrence of failures and mitigate their operational risk. However, is it quite challenging to design a high-precision prediction method because the implicit degradation details of PEMFCs are difficult to learn well from the measurement data with high-frequency noise. Recognizing this, a novel RUL prediction method based on singular spectrum analysis (SSA) and deep Gaussian process (DGP) is proposed in this paper. The SSA-based method is firstly employed to preprocess the measurement data, which can strengthen the effective information of PEMFC degradation data at the same time remove the noise and spikes that interfere with degradation prediction. As a deep structural model, DGP has strong feature learning ability which can represent the nonlinear details of degradation data and give more accurate prediction results. At the same time, it serves as a probabilistic model that can provide the confidence interval to enhance reliability of RUL prediction. The effectiveness of the proposed method is evaluated by experimental data of the PEMFCs under steady-state conditions, and the results show that the SSA-DGP method has higher accuracy and reliability than conventional methods.     

Performance degradation prediction of proton exchange membrane fuel cell using a hybrid prognostic approach

The proton exchange membrane fuel cell has been widely used for industrial systems; however, its performance gradually degrades during use. Therefore, the study on the performance degradation prediction of fuel cells is helpful to extend its lifespan. In this paper, a novel hybrid approach using a combination of model-based adaptive Kalman filter and data-driven NARX neural network is proposed to predict the degradation of fuel cells. The overall degradation trend (i.e., irreversible degradation process) is captured by an empirical aging model and adaptive Kalman filter. Meanwhile, the detail degradation information (i.e., reversible degradation process) is depicted by the NARX neural network. Moreover, the correlation analysis of the reversible voltage time series is carried out to obtain the number of delays of the NARX neural network based on the autocorrelation function and the partial autocorrelation function. Then, the total degradation prediction is the sum of the overall degradation prediction and the detail degradation prediction. Finally, the prognostic capability of the proposed method is verified by two aging datasets, and the results show the effectiveness and superiority of the proposed method which can provide accurate degradation forecasting and remaining useful life.     

An Atomistic Understanding of Allosteric Inhibition of Glutamate Racemase: a Dampening of Native Activation Dynamics

Glutamate racemases (GR) are members of the family of bacterial enzymes known as cofactor-independent racemases and epimerases and catalyze the stereoinversion of glutamate. D-amino acids are universally important for the proper construction of viable bacterial cell walls, and thus have been repeatedly validated as attractive targets for novel antimicrobial drug design. Significant aspects of the mechanism of this challenging stereoinversion remain unknown. The current study employs a combination of MD and QM/MM computational approaches to show that the GR from H. pylori must proceed via a pre-activation step, which is dependent on the enzyme's flexibility. This mechanism is starkly different from previously proposed mechanisms. These findings have immediate pharmaceutical relevance, as the H. pylori GR enzyme is a very attractive allosteric drug target. The results presented in this study offer a distinctly novel understanding of how AstraZeneca's lead series of inhibitors cripple the H. pylori GR's native motions, via prevention of this critical chemical pre-activation step. Our experimental studies, using SPR, fluorescence and NMR WaterLOGSY, show that H. pylori GR is not inhibited by the uncompetitive mechanism originally put forward by Lundqvist et al.. The current study supports a deep connection between native enzyme motions and chemical reactivity, which has strong relevance to the field of allosteric drug discovery.     

Three‐dimensional color prediction modeling of single‐ and double‐layered woven fabrics

In this research, the three‐dimensional structural and colorimetric modeling of three‐dimensional woven fabrics was conducted for accurate color predictions. One‐hundred forty single‐ and double‐layered woven samples in a wide range of colors were produced. With the consideration of their three‐dimensional structural parameters, three‐dimensional color prediction models, K/S‐, R‐, and L*a*b*‐based models, were developed through the optimization of previous two‐dimensional models which have been reported to be the three most accurate models for single‐layered woven structures. The accuracy of the new three‐dimensional models was evaluated by calculating the color differences ΔL*, ΔC*, Δh°, and ΔE_CMC(2:1) between the measured and the predicted colors of the samples, and then the error values were compared to those of the two‐dimensional models. As a result, there has been an overall improvement in color predictions of all models with a decrease in ΔE_CMC(2:1) from 10.30 to 5.25 units on average after the three‐dimensional modeling.     

Design of a Cost‐Reduced Flexible Plant for Supercritical Fluid‐Assisted Applications

A novel batch plant for supercritical CO₂ applications is proposed which is not equipped with expensive components, such as high‐pressure pumps, making it particularly suitable for bench‐scale use. For the first time, the use of a hanging scale is suggested to weigh the amount of CO₂ required for the experiment and the use of the thermodynamics to reach the working conditions. The rig is able to cover different applications, e.g., aerogel drying, impregnation, and extraction, showing high flexibility. An approximate cost analysis has been performed considering as a reference a 150‐mL vessel. It has been calculated that both the setup and running costs are considerably lower than the common batch and semicontinuous rigs.     

神经网络全局势函数在多相催化中的应用

当今的多相催化研究需要新的技术和方法从原子尺度上表征活性中心结构和反应中间体。本文作者课题组近期开发了理论模拟新技术来探索催化剂活性位点结构，即基于神经网络势函数的大规模原子模拟（LASP）软件中实现的全局神经网络势函数计算方法。本文介绍了该方法可以显著降低催化体系的计算代价，而维持与密度泛函理论同一级别的计算精度，从而解决多相催化中的许多复杂问题。本文对神经网络势函数方法的实现细节和目前已实现的应用场景进行了详细介绍。神经网络势函数可以用来预测材料晶体结构，理解高压氢化条件下TiO₂表面的结构演化和确定三元氧化物ZnCrO晶相中合成气制甲醇活性位点。最后文章分析了神经网络势函数的局限性和今后可能的三个研究方向，即材料性质预测、多元素体系神经网络势函数构造和化学反应拟合。     

A Novel Method of Heart Failure Prediction Based on DPCNNXGBOOST Model

The occurrence of perioperative heart failure will affect the quality of medical services and threaten the safety of patients. Existing methods depend on the judgment of doctors, the results are affected by many factors such as doctors’ knowledge and experience. The accuracy is difficult to guarantee and has a serious lag. In this paper, a mixture prediction model is proposed for perioperative adverse events of heart failure, which combined with the advantages of the Deep Pyramid Convolutional Neural Networks (DPCNN) and Extreme Gradient Boosting (XGBOOST). The DPCNN was used to automatically extract features from patient’s diagnostic texts, and the text features were integrated with the preoperative examination and intraoperative monitoring values of patients, then the XGBOOST algorithm was used to construct the prediction model of heart failure. An experimental comparison was conducted on the model based on the data of patients with heart failure in southwest hospital from 2014 to 2018. The results showed that the DPCNN-XGBOOST model improved the predictive sensitivity of the model by 3% and 31% compared with the text-based DPCNN Model and the numeric-based XGBOOST Model.     

MOOC Learner’s Final Grade Prediction Based on an Improved Random Forests Method

Massive Open Online Course (MOOC) has become a popular way of online learning used across the world by millions of people. Meanwhile, a vast amount of information has been collected from the MOOC learners and institutions. Based on the educational data, a lot of researches have been investigated for the prediction of the MOOC learner’s final grade. However, there are still two problems in this research field. The first problem is how to select the most proper features to improve the prediction accuracy, and the second problem is how to use or modify the data mining algorithms for a better analysis of the MOOC data. In order to solve these two problems, an improved random forests method is proposed in this paper. First, a hybrid indicator is defined to measure the importance of the features, and a rule is further established for the feature selection; then, a Clustering-Synthetic Minority Over-sampling Technique (SMOTE) is embedded into the traditional random forests algorithm to solve the class imbalance problem. In experiment part, we verify the performance of the proposed method by using the Canvas Network Person-Course (CNPC) dataset. Furthermore, four well-known prediction methods have been applied for comparison, where the superiority of our method has been proved.