A method for establishing a bearing residual life prediction model for process enhancement equipment based on rotor imbalance response analysis

A rotating packed bed is a typical chemical process enhancement equipment that can strengthen micromixing and mass transfer.During the operation of the rotating packed bed,the nonreactants and products irregularly adhere to the wire mesh packing in the rotor,thus resulting in an imbalance in the vibration of the rotor,which may cause serious damage to the bearing and material leakage.This study proposes a model prediction for estimating the bearing residual life of a rotating packed bed based on rotor imbalance response analysis.This method is used to determine the influence of the mass on the imbalance in the vibration of the rotor on bearing damage.The major influence on rotor vibration was found to be exerted by the imbalanced mass and its distribution radius,as revealed by the results of orthogonal experiments.Through implementing finite element analysis,the imbalance response curve for the rotating packed bed rotor was obtained,and a correlation among rotor imbalance mass,distribution radius of imbalance mass,and bearing residue life was established via data fitting.The predicted value of the bearing life can be used as the reference basis for an early safety warning of a rotating packed bed to effectively avoid accidents.

Clustered Federated Learning with Weighted Model Aggregation for Imbalanced Data

As a promising edge learning framework in future 6G networks,federated learning(FL)faces a number of technical challenges due to the heterogeneous network environment and diversified user behaviors.Data imbalance is one of these challenges that can significantly degrade the learning efficiency.To deal with data imbalance issue,this work proposes a new learning framework,called clustered federated learning with weighted model aggregation(weighted CFL).Compared with traditional FL,our weighted CFL adaptively clusters the participating edge devices based on the cosine similarity of their local gradients at each training iteration,and then performs weighted per-cluster model aggregation.Therein,the similarity threshold for clustering is adaptive over iterations in response to the time-varying divergence of local gradients.Moreover,the weights for per-cluster model aggregation are adjusted according to the data balance feature so as to speed up the convergence rate.Experimental results show that the proposed weighted CFL achieves a faster model convergence rate and greater learning accuracy than benchmark methods under the imbalanced data scenario.

Attenuate Class Imbalance Problem for Pneumonia Diagnosis Using Ensemble Parallel Stacked Pre-Trained Models

Pneumonia is an acute lung infection that has caused many fatalitiesglobally. Radiologists often employ chest X-rays to identify pneumoniasince they are presently the most effective imaging method for this purpose.Computer-aided diagnosis of pneumonia using deep learning techniques iswidely used due to its effectiveness and performance. In the proposed method,the Synthetic Minority Oversampling Technique (SMOTE) approach is usedto eliminate the class imbalance in the X-ray dataset. To compensate forthe paucity of accessible data, pre-trained transfer learning is used, and anensemble Convolutional Neural Network (CNN) model is developed. Theensemble model consists of all possible combinations of the MobileNetv2,Visual Geometry Group (VGG16), and DenseNet169 models. MobileNetV2and DenseNet169 performed well in the Single classifier model, with anaccuracy of 94%, while the ensemble model (MobileNetV2+DenseNet169)achieved an accuracy of 96.9%. Using the data synchronous parallel modelin Distributed Tensorflow, the training process accelerated performance by98.6% and outperformed other conventional approaches.

Relationship between effort-reward imbalance, job satisfaction, and intention to leave the profession among the medical staff of Qom University of Medical Sciences

Objective: The effort–reward imbalance(ERI) model claims that work that has the characteristics of high effort and low reward has a mutual defect between input and output and this imbalance may result in sustained and long-lasting results. This study aimed to investigate the relationship between the intention to leave the profession(ILP) and ERI and job satisfaction among the medical staff in Qom Province.Methods: A descriptive-correlative study was conducted on 202 medical staff in Qom Province based on random sampling in 2018. Demographics checklist, standard ILP, job satisfaction, and Siegrist’s ERI questionnaires were used for data collection. The chisquared test, independent t-test, and one-way Analyses of Variance(ANOVA) were used to analyze data.Results: The mean age of employees was 32.04 ± 7.9 years, and 165(87.1%) of the employees were women. The results showed that the medical staff was willing to leave their profession at a moderate level(40.7 ± 10.3). There was no significant relationship between demographics and ILP. Nevertheless, a significant and inverse relationship was observed between ERI(r:0.318, P < 0.01) and ILP(r: 0.197, P < 0.01). Leave the profession(LP) had a negative correlation with the dimensions of job descriptive index(JDI) such as job, manager, coworker and wage score(P < 0.01, r:-0.147, r:-0.262, r:-0.292, r:-0.271, r:-0.396).Conclusions: According to the results, managers need to make sure that their working staff is rewarded as they deserve. According to the results, managers need to ensure that the reward factor is observed for the staff, while an ERI imbalance may contribute to ILP of the staff. On the other hand, it leads to job satisfaction.

Exploring effort-reward imbalance and professional quality of life among health workers in Cape Town,South Africa:a mixed-methods study

Background:In the context of a growing appreciation for the wellbeing of the health workforce as the foundation of high-quality,sustainable health systems,this paper presents findings from two complementary studies to explore occupational stress and professional quality of life among health workers that were conducted in preparation for a task-shifting intervention to improve antenatal mental health services in Cape Town.Methods:This mixed-methods,cross-sectional study was conducted in public sector Midwife Obstetric Units and associated Non-Profit Organisations in Cape Town.Semi-structured interviews and a quantitative survey were conducted among facility-and community-based professional and lay health workers.The survey included demographic as well as effort-reward imbalance(ERI)and professional quality of life(PROQOL)questionnaires to examine overall levels of work-related psychosocial stress and professional quality of life,as well as differences between lay and professional health workers.Qualitative data was analysed using a thematic content analysis approach.Quantitative data was analysed using STATA 12.Results:Findings from 37 qualitative interviews highlighted the difficult working conditions and often limited reward and support structures experienced by health workers.Corroborating these findings,our quantitative survey of 165 professional and lay health workers revealed that most health workers experienced a mismatch between efforts spent and rewards gained at work(61.1%of professional and 70.2%of lay health workers;p=0.302).There were few statistically significant differences in ERI and PROQOL scores between professional and lay health workers.Although Compassion Satisfaction was high for all health worker groups,lay health workers also showed elevated levels of burnout and compassion fatigue,with community-based health workers particularly affected.Conclusions:Findings of this study add to the existing evidence base on adverse working conditions faced by South African public-sector health workers that should be taken into consideration as national and local governments seek to‘re-engineer’South Africa’s Primary Health Care system.Furthermore,they also highlight the importance of taking into consideration the wellbeing of health workers themselves to develop interventions that can sustainably foster resilient and high-quality health systems.

Primary logistics planning of oil products under the imbalance of supply and demand

This paper intends to complete the primary logistics planning of oil products under the imbalance of supply and demand. An integrated mathematical programming model is developed to simultaneously find the balance between supply and demand, and optimize the logistics scheme. The model takes minimum logistics cost and resource adjustment cost as the objective function, and takes supply and demand capacity, transportation capacity, mass balance, and resource adjustment rules as constraints.Three adjustment rules are considered in the model, including resource adjustment within oil suppliers,within oil consumers, and between oil consumers. The model is tested on a large-scale primary logistics of a state-owned petroleum enterprise, involving 37 affiliated refineries, 31 procurement departments,286 market depots and dedicated consumers. After the unified optimization, the supply and demand imbalance is eased by 97% and the total cost is saved by 7%, which proves the effectiveness and applicability of the proposed model.

mLysPTMpred: Multiple Lysine PTM Site Prediction Using Combination of SVM with Resolving Data Imbalance Issue

Post-translational modification (PTM) increases the functional diversity of proteins by introducing new functional groups to the side chain of amino acid of a protein. Among all amino acid residues, the side chain of lysine (K) can undergo many types of PTM, called K-PTM, such as “acetylation”, “crotonylation”, “methylation” and “succinylation” and also responsible for occurring multiple PTM in the same lysine of a protein which leads to the requirement of multi-label PTM site identification. However, most of the existing computational methods have been established to predict various single-label PTM sites and a very few have been developed to solve multi-label issue which needs further improvement. Here, we have developed a computational tool termed mLysPTMpred to predict multi-label lysine PTM sites by 1) incorporating the sequence-coupled information into the general pseudo amino acid composition, 2) balancing the effect of skewed training dataset by Different Error Cost method, and 3) constructing a multi-label predictor using a combination of support vector machine (SVM). This predictor achieved 83.73% accuracy in predicting the multi-label PTM site of K-PTM types. Moreover, all the experimental results along with accuracy outperformed than the existing predictor iPTM-mLys. A user-friendly web server of mLysPTMpred is available at http://research.ru.ac.bd/mLysPTMpred/.

Integrating deep learning and logging data analytics for lithofacies classification and 3D modeling of tight sandstone reservoirs

The lithofacies classification is essential for oil and gas reservoir exploration and development.The traditional method of lithofacies classification is based on"core calibration logging"and the experience of geologists.This approach has strong subjectivity,low efficiency,and high uncertainty.This uncertainty may be one of the key factors affecting the results of 3 D modeling of tight sandstone reservoirs.In recent years,deep learning,which is a cutting-edge artificial intelligence technology,has attracted attention from various fields.However,the study of deep-learning techniques in the field of lithofacies classification has not been sufficient.Therefore,this paper proposes a novel hybrid deep-learning model based on the efficient data feature-extraction ability of convolutional neural networks(CNN)and the excellent ability to describe time-dependent features of long short-term memory networks(LSTM)to conduct lithological facies-classification experiments.The results of a series of experiments show that the hybrid CNN-LSTM model had an average accuracy of 87.3%and the best classification effect compared to the CNN,LSTM or the three commonly used machine learning models(Support vector machine,random forest,and gradient boosting decision tree).In addition,the borderline synthetic minority oversampling technique(BSMOTE)is introduced to address the class-imbalance issue of raw data.The results show that processed data balance can significantly improve the accuracy of lithofacies classification.Beside that,based on the fine lithofacies constraints,the sequential indicator simulation method is used to establish a three-dimensional lithofacies model,which completes the fine description of the spatial distribution of tight sandstone reservoirs in the study area.According to this comprehensive analysis,the proposed CNN-LSTM model,which eliminates class imbalance,can be effectively applied to lithofacies classification,and is expected to improve the reality of the geological model for the tight sandstone reservoirs.

The Model of Dependence of the Temperature of the Surface Layer of Atmosphere from the Earth's Albedo and Thermal Inertia of the Hydrosphere

The assumption of M. Milankovich about the constancy of the Earth's albedo during the interglacial period was replaced with the alternative one. The model was developed where anomalies of the average annual temperature of the surface atmosphere were related with interannual changes in the planetary albedo and the thermal inertia of the hydrosphere. The surface temperature changes due to albedo actual and model changes were calculated. Possible external causes of albedo changes were considered.

基于Stacking集成学习的热轧带钢凸度诊断模型

在热连轧生产过程中,凸度是重要的质量指标,过程数据的非平衡性限制了数据驱动模型的预测效果,为提高模型的预测精度,提出一种融合SMOTE和Stacking集成算法的热轧带钢凸度诊断模型。首先,采用SMOTE过采样方法处理凸度相关数据集,降低数据非平衡分布导致的影响;然后,构建以轻量级梯度提升机(LightGBM)、支持向量机(SVM)、K近邻(KNN)和随机森林(RF)为基学习器,逻辑回归(LR)为元学习器的Stacking集成模型,最后,使用某2160 mm热轧带钢实际生产数据进行模型验证。研究结果表明,诊断模型的准确率、少数类召回率、平衡F分数、几何平均值和ROC曲线下面积分别为0.9580、0.9595、0.9573、0.9589和0.9579,与XGBoost、LightGBM、KNN、SVM和随机森林模型对比,预测效果最优,证明了Stacking集成算法能够有效增强诊断模型的泛化能力,具有优良的诊断性能。

面向类不均衡数据的多任务博弈概率分类向量机

在工程实际中获取的故障样本往往会呈现不均衡特点,同时传统的分类模型也会存在局限性。针对这些问题,基于稀疏贝叶斯理论、模糊隶属度等理论,提出了一种多任务博弈概率分类向量机(MGPCVM)分类方法。首先,在MGPCVM的目标函数中,设计了博弈因子,将不同类样本质心间的博弈信息赋予每个样本特定的样本质心敏感值,以解决传统分类器对不平衡数据集分类表现较差的问题;然后,在贝叶斯框架理论下,采用截断高斯先验分布的方法,使样本参数的正负与对应的标签信息相一致,且使样本质心敏感值产生了稀疏估计;最后,将MGPCVM方法应用于两种不同实验平台采集的滚动轴承实验数据处理,进行了故障诊断有效性验证。研究结果表明:在不同的不平衡比(IR)下,MGPCVM方法的准确率均保持在95%以上,相对于支持向量机(SVM)、概率分类向量机(PCVM)等方法提升了4%~8%;与典型向量式分类方法相比,MGPCVM方法可以在不平衡数据条件下表现出优越的分类性能,适用于实际工况中数据失衡的分类问题。

高等教育治理现代化的张力失衡及其调适路径——基于公共价值战略三角模型的分析框架

高等教育治理现代化是一个涉及价值、主体、结构等多元治理要素的动态发展过程,对其主要特征的全面把握是新时代推进高等教育治理现代化的前提和基础。借助公共价值战略三角模型,从使命管理、政治管理、运营管理三个维度构建分析框架,阐释我国高等教育治理现代化的主要特征,剖析高等教育治理现代化进程中出现的价值张力失衡、权力张力失衡和结构张力失衡问题。要有效解决我国高等教育治理现代化的张力失衡问题,需要以价值耦合凝聚高等教育“公共性”价值共识,以利益共融打造共建共享的高等教育利益共同体,以结构优化构建纵向连通、横向协作的高等教育治理机制,促进高等教育公共价值最大化的实现。

基于模型融合和生成网络的有效阵位智能决策方法

军事智能技术是当前最具活力的前沿领域和未来无人装备发展的必然趋势。针对无人平台在复杂环境下自主决策可靠性和实时性的双重需求和现有基于规则推演的作战仿真技术在动态性和灵活性方面的不足,采用原理分析与实验验证的研究方法,在某型无人平台射击实验数据集的基础上,围绕攻击决策的有效阵位识别环节,将其转换为机器学习领域类别不平衡的二分类问题,综合采用相关性分析、特征工程、模型融合技术构建高实时性和灵活性的有效阵位智能决策模型,并提出基于ICGAN-Stacking不平衡分类架构对少数类样本进行定向扩充,实现数据增强和模型性能提升。实验结果表明:所提方法召回率提升了4.1%、精确度提升了0.4%、F1值提升了1.5%、AUC值达到90.9%,能够满足无人平台执行作战任务实时性和可靠性需求。

基于去噪扩散概率模型不平衡样本增强的暂态稳定评估

准确有效的电力系统暂态稳定评估对电力系统安全稳定运行具有重要意义。目前,基于深度学习的暂态稳定评估方法面临着时序特征空间表征困难、样本类别严重不平衡等问题,影响到评估结果的可信度。为了弥补现有研究的不足,提出了一种基于去噪扩散概率模型(DDPM)不平衡样本增强的电力系统暂态稳定评估方法。首先,构建改进HSV颜色模型对高维数据进行二维图像化处理,从而直观表征高维数据,便于后续训练;然后,基于DDPM算法对不平衡失稳样本空间进行表征学习,规模化生成概率同分布的增强样本,进而解决类别不平衡问题;最后,提出梯度加权类激活映射卷积神经网络以构建暂态稳定评估模型,提升模型的可信度与可解释性。IEEE 39节点系统测试的仿真结果表明,所构建的模型相较于其他方法具备更高的稳定性判别精度,且对失稳样本的识别率显著提高,验证了所提方法的有效性。

婚姻市场性别结构对女性婚姻流动距离的影响及异质性分析

在关于婚姻流动距离的研究中,个体对于经济社会资源等方面的偏好因素的影响已经被广泛认同。然而地区性别结构因素作为地区婚姻市场的一种重要的宏观因素,关于其对于婚姻流动距离的影响还值得深入探讨。基于婚姻搜寻理论,利用多层模型证实了地区婚姻市场性别结构对中国女性婚姻流出距离的影响。研究发现:(1)总的来看,未婚人口性别比高的地区,女性婚姻流出距离相对更短,当地女性更可能在较近的地区结婚。(2)性别结构的影响在不同收入水平、城乡之间和不同受教育程度的女性之间存在异质性,户籍地为低收入地区、乡村地区或受教育程度较低的女性婚姻流动距离受到婚姻市场性别结构的影响较大。

中国耕地补充与撂荒的空间关系识别及其失衡归因

研究目的:探究耕地补充与耕地撂荒的空间匹配关系,为耕地保护和开发提供科学依据。研究方法:遥感监测、空间自相关分析、空间失衡指数、地理探测器和增强回归树模型。研究结果:(1)近20年中国耕地年均补充263.22×10^(4) hm^(2),年均补充率为1.39%,集中分布于西北、东北和西南地区。耕地年均撂荒226.68×10^(4) hm^(2),年均撂荒率为1.19%,主要分布于黄河中游和西南地区。(2)耕地补充和耕地撂荒之间以轻度失衡和不失衡为主,但极度失衡和重度失衡的单元数量逐年增加。至2020年,中度、重度和极度失衡的单元占16.87%,主要分布在西北、黄河中游、西南和南部沿海地区。(3)高—高集聚和低—低集聚是耕地补充和撂荒失衡的主要空间集聚模式。其中,高—高集聚区集中在西北、黄河中游、西南和南部沿海地区,低—低集聚区集中在东北、长江中游、西南和北部沿海地区。(4)坡度和年平均降水量为耕地补充与撂荒空间失衡主要的自然资源驱动因子,非农从业人数、农业机械总动力是主要的社会经济驱动因子。研究结论:耕地补充和撂荒之间空间失衡具有复杂性且受当地自然条件和经济社会因素的交互影响,未来应基于不同的情景和目标综合确定最优的利用模式,为高质量耕地保护提供科学依据。

基于MSSMOTE-CNN模型的空调冷水机组故障诊断

针对冷水机组运行过程中数据类别不平衡问题,提出基于马氏距离进行“三角”区域插值的MSSMOTE方法对故障数据进行扩充,将得到的数据输入CNN模型进行训练,实现对冷水机组中7种故障的诊断。在不同扩充比例下和同一种数据类型下分别进行仿真测试,结果显示:在扩充比例为4时,MSSMOTE-CNN模型对于正常样本测试的准确率和F 1-score分别达到0.961和0.971,能够较准确识别出冷水机组的故障类型。

槐耳清膏治疗大鼠肉芽肿性小叶性乳腺炎模型的作用研究

[目的]探讨槐耳清膏通过降低炎症水平治疗肉芽肿性小叶性乳腺炎(granulomatous lobular mastitis,GLM)大鼠的作用。[方法]将雌性SD大鼠随机分为正常组、模型组、槐耳组[槐耳清膏4.5 g/(kg·d)]、激素组[醋酸泼尼松龙0.0018 g/(kg·d)],利用GLM病变组织与弗氏完全佐剂混合后的组织匀浆诱导GLM大鼠模型。给药14 d后,对各组大鼠进行乳腺大体观察,以大鼠乳腺炎症指数评分表进行炎症指数评分,苏木精-伊红染色后进行乳腺组织病理学观察,并通过酶联免疫吸附测定(enzyme-linked immunosorbent assay,ELISA)法检测大鼠血清炎症细胞因子的表达水平,采用16S rRNA扩增子测序技术测定各组大鼠肠道菌群的多样性、丰度以及肠道菌群相对丰度。[结果]与正常组比较,模型组大鼠乳房皮色红,出现明显肿块,且乳腺炎症指数显著上升(P<0.05);病理学改变包括形成以乳腺小叶为中心的肉芽肿,伴大量淋巴细胞、浆细胞等炎症细胞浸润,证实成功诱导GLM大鼠模型。与模型组比较,槐耳组及激素组大鼠乳房外观红肿改善,肿块变软,炎症指数均下降(P<0.05,P<0.05),乳腺炎症程度明显改善,炎症细胞减少,血清炎症因子白细胞介素-6(interleukin-6,IL-6)、白细胞介素-17A(interleukin-17A,IL-17A)、肿瘤坏死因子-α(tumor necrosis factor-α,TNF-α)水平显著降低(P<0.05,P<0.05,P<0.05),乳腺组织淋巴细胞、浆细胞等炎症细胞浸润减少。模型大鼠肠道菌群发生明显变化,差异具有统计学意义(P<0.05)。模型组大鼠肠道菌群中乳杆菌目、放线菌门、双歧杆菌目、红蝽菌纲丰度下调,柄杆菌目和脱硫菌门丰度上调。槐耳清膏干预后,模型大鼠肠道菌群中梭菌纲丰度下调,拟杆菌门、红螺菌目丰度上调。[结论]槐耳清膏能够缓解GLM模型大鼠炎症症状,有效减轻GLM模型大鼠乳腺组织炎症细胞浸润,抑制上皮细胞增生及炎性肉芽肿的形成,降低GLM模型大鼠血清炎症细胞因子水平,同时调节大鼠肠道菌群,其作用机制可能跟增强大鼠自身免疫力、调节肠道菌群,进而有效缓解炎症反应有关。

不平衡电网电压下虚拟同步发电机模型预测控制

虚拟同步发电机(Virtual Synchronous Generator,VSG)控制算法,可为电网提供惯性并参与电压及频率调节,但在电网电压出现不平衡时会导致并网电流不平衡及有功、无功功率波动的问题,严重影响了并网电能质量。针对此问题,提出了一种不平衡电网下VSG模型预测控制策略,通过基于快速电压矢量选择的模型预测控制策略来控制所重构的正序及负序电流分量,达到有功功率恒定、无功功率恒定及电流平衡的控制目标。此外还引入负序电流调节系数,进而实现这三个目标的协调控制。

苏浙两省人口城镇化和土地城镇化失衡问题研究:不同发展模式视角

本文从不同发展模式视角,探讨了苏浙两省的城镇化失衡问题,得出以下主要结论:一是江苏省城镇化的发展与“苏南模式”的阶段性演变密切相关,第一阶段是人口城镇化与土地城镇化双抑制阶段,第二阶段受股份制改革及吸引外资的影响,是对人口城镇化和土地城镇化双推动阶段,第三阶段是政府引导下人口城镇化和土地城镇化协调发展阶段;二是“温州模式”下的浙江省人口城镇化与土地城镇化协调发展没有明显的阶段性特征,总体上“商业为主+民营经济”的经营方式对土地城镇化贡献不大,反而会通过增加农民收入等方式制约人口城镇化的发展。本文在以上结论基础上,提出了相应的对策,以期为矫正城镇化失衡问题提供帮助。