A blast furnace fault monitoring algorithm with low false alarm rate:Ensemble of greedy dynamic principal component analysis-Gaussian mixture model

The large blast furnace is essential equipment in the process of iron and steel manufacturing. Due to the complex operation process and frequent fluctuations of variables, conventional monitoring methods often bring false alarms. To address the above problem, an ensemble of greedy dynamic principal component analysis-Gaussian mixture model(EGDPCA-GMM) is proposed in this paper. First, PCA-GMM is introduced to deal with the collinearity and the non-Gaussian distribution of blast furnace data.Second, in order to explain the dynamics of data, the greedy algorithm is used to determine the extended variables and their corresponding time lags, so as to avoid introducing unnecessary noise. Then the bagging ensemble is adopted to cooperate with greedy extension to eliminate the randomness brought by the greedy algorithm and further reduce the false alarm rate(FAR) of monitoring results. Finally, the algorithm is applied to the blast furnace of a large iron and steel group in South China to verify performance.Compared with the basic algorithms, the proposed method achieves lowest FAR, while keeping missed alarm rate(MAR) remain stable.

Minimax entropy-based co-training for fault diagnosis of blast furnace

Due to the problems of few fault samples and large data fluctuations in the blast furnace(BF)ironmaking process,some transfer learning-based fault diagnosis methods are proposed.The vast majority of such methods perform distribution adaptation by reducing the distance between data distributions and applying a classifier to generate pseudo-labels for self-training.However,since the training data is dominated by labeled source domain data,such classifiers tend to be weak classifiers in the target domain.In addition,the features generated after domain adaptation are likely to be at the decision boundary,resulting in a loss of classification performance.Hence,we propose a novel method called minimax entropy-based co-training(MMEC)that adversarially optimizes a transferable fault diagnosis model for the BF.The structure of MMEC includes a dual-view feature extractor,followed by two classifiers that compute the feature's cosine similarity to representative vector of each class.Knowledge transfer is achieved by alternately increasing and decreasing the entropy of unlabeled target samples with the classifier and the feature extractor,respectively.Transfer BF fault diagnosis experiments show that our method improves accuracy by about 5%over state-of-the-art methods.

Finite Element Simulation of Thermal Field in Double Wire Welding

In this paper, the thermal field of double wire welding is simulated by using ANSYS software. Simulation results were shown that the total heat input (E) is the most significant parameters to change the value of t8/5;By the mean of rationally controlling the proportion of the front arc heat input (E1) in the total heat input (E) and appropriately selecting double wire spacing (L), It is effective means to get the double wire welding thermal cycle. By the way of simulation, it is possible to manage the thermal input in the double welding wires and to control the temperature field and cooling rate that are fundamental for the final joint quality, it is great importance guidance to optimize the double wire welding process parameters.

紫杉醇联合阿帕替尼对比单药紫杉醇二线治疗胃癌的真实世界前瞻性研究

目的:观察紫杉醇联合阿帕替尼对比单药紫杉醇二线治疗胃癌的疗效及安全性.方法:选取2017年3月至2018年3月就诊于内蒙古医科大学赤峰临床医学院经一线(氟尿嘧啶类联合铂类)治疗后进展的HER-2阴性的晚期胃癌患者60例,分为单药组和联合组,每组30例.单药组给予单药紫杉醇化疗,联合组给予紫杉醇联合阿帕替尼治疗,观察的主要终点为两组患者的无进展生存期(progression-free survival,PFS),次要终点为疾病控制率(disease control rate,DCR)、客观缓解率(objective response rate,ORR)、安全性.结果:联合组患者的PFS优于单药组患者,两组相比差异有统计学意义(P<0.001).联合组DCR、ORR均高于单药组,但差异均无统计学意义(均P>0.05);联合组高血压发生率高于单药组(P<0.05),其余不良反应发生率差异均无统计学意义(均P>0.05).结论:紫杉醇联合甲磺酸阿帕替尼对比单药紫杉醇二线治疗胃癌,联合组PFS、DCR、ORR均优于单药组,虽然联合组DCR、ORR较单药组相比差异无统计学意义,但联合组组患者的PFS较单组延长,差异有统计学意义(P<0.05),且联合组不良反应可耐受,安全性良好.

Intelligent Ironmaking Optimization Service on a Cloud Computing Platform by Digital Twin

The shortage of computation methods and storage devices has largely limited the development of multiobjective optimization in industrial processes.To improve the operational levels of the process industries,we propose a multi-objective optimization framework based on cloud services and a cloud distribution system.Real-time data from manufacturing procedures are first temporarily stored in a local database,and then transferred to the relational database in the cloud.Next,a distribution system with elastic compute power is set up for the optimization framework.Finally,a multi-objective optimization model based on deep learning and an evolutionary algorithm is proposed to optimize several conflicting goals of the blast furnace ironmaking process.With the application of this optimization service in a cloud factory,iron production was found to increase by 83.91 t∙d^(-1),the coke ratio decreased 13.50 kg∙t^(-1),and the silicon content decreased by an average of 0.047%.

Dust Distribution Study at the Blast Furnace Top Based on k-Sε-u_(p)Model

The dust distribution law acting at the top of a blast fumace(BF)is of great significance for understanding gas flow distribution and mitigating the negative influence of dust particles on the accuracy and service life of detection equipment.The harsh environment inside a BF makes it difficult to describe the dust disthibution.This paper adresses this problem by proposing a dust distribution k-Sε-u_(p)model based on interphase(gas-powder)coupling.The proposed model is coupled with a k-Sεmodel(which describes gas flow movement)and a u_(p)model(which depicts dust movement).First,the kinetic energy equation and turbulent dissipation rate equation in the k-Sεmodel are established based on the modeling theory and single Green-function two scale direct interaction approximation(SGF-TSDIA)theory.Second,a dust particle mnovement u_(p)model is built based on a force analysis of the dust and Newton's laws of motion.Finally,a coupling factor that descibes the interphase interaction is proposed,and the k-Sε-u_(p)model,with clear physical meaning.ligorous mathematical logic,and adequate generality,is dleveloped.Siumulation results and o-site verification show that the k-Sε-u_(p)model not only has high precision,but also reveals the aggregate distribution features of the dust,which are helpful in optimizing the installation position of the detection equipment and imnproving its accuracy and service life.

Itinerant ferromagnetism entrenched by the anisotropy of spin-orbit coupling in a dipolar Fermi gas

We investigate the itinerant ferromagnetism in a dipolar Fermi atomic system with the anisotropic spin–orbit coupling(SOC),which is traditionally explored with isotropic contact interaction.We first study the ferromagnetism transition boundaries and the properties of the ground states through the density and spin-flip distribution in momentum space,and we find that both the anisotropy and the magnitude of the SOC play an important role in this process.We propose a helpful scheme and a quantum control method which can be applied to conquering the difficulties of previous experimental observation of itinerant ferromagnetism.Our further study reveals that exotic Fermi surfaces and an abnormal phase region can exist in this system by controlling the anisotropy of SOC,which can provide constructive suggestions for the research and the application of a dipolar Fermi gas.Furthermore,we also calculate the ferromagnetism transition temperature and novel distributions in momentum space at finite temperature beyond the ground states from the perspective of experiment.

Data-driven soft sensors in blast furnace ironmaking:a survey

The blast furnace is a highly energy-intensive,highly polluting,and extremely complex reactor in the ironmaking process.Soft sensors are a key technology for predicting molten iron quality indices reflecting blast furnace energy consumption and operation stability,and play an important role in saving energy,reducing emissions,improving product quality,and producing economic benefits.With the advancement of the Internet of Things,big data,and artificial intelligence,data-driven soft sensors in blast furnace ironmaking processes have attracted increasing attention from researchers,but there has been no systematic review of the data-driven soft sensors in the blast furnace ironmaking process.This review covers the state-of-the-art studies of data-driven soft sensors technologies in the blast furnace ironmaking process.Specifically,wefirst conduct a comprehensive overview of various data-driven soft sensor modeling methods(multiscale methods,adaptive methods,deep learning,etc.)used in blast furnace ironmaking.Second,the important applications of data-driven soft sensors in blast furnace ironmaking(silicon content,molten iron temperature,gas utilization rate,etc.)are classified.Finally,the potential challenges and future development trends of data-driven soft sensors in blast furnace ironmaking applications are discussed,including digital twin,multi-source data fusion,and carbon peaking and carbon neutrality.