State of the art in applications of machine learning in steelmaking process modeling

With the development of automation and informatization in the steelmaking industry,the human brain gradually fails to cope with an increasing amount of data generated during the steelmaking process.Machine learning technology provides a new method other than production experience and metallurgical principles in dealing with large amounts of data.The application of machine learning in the steelmaking process has become a research hotspot in recent years.This paper provides an overview of the applications of machine learning in the steelmaking process modeling involving hot metal pretreatment,primary steelmaking,secondary refining,and some other aspects.The three most frequently used machine learning algorithms in steelmaking process modeling are the artificial neural network,support vector machine,and case-based reasoning,demonstrating proportions of 56%,14%,and 10%,respectively.Collected data in the steelmaking plants are frequently faulty.Thus,data processing,especially data cleaning,is crucially important to the performance of machine learning models.The detection of variable importance can be used to optimize the process parameters and guide production.Machine learning is used in hot metal pretreatment modeling mainly for endpoint S content prediction.The predictions of the endpoints of element compositions and the process parameters are widely investigated in primary steelmaking.Machine learning is used in secondary refining modeling mainly for ladle furnaces,Ruhrstahl–Heraeus,vacuum degassing,argon oxygen decarburization,and vacuum oxygen decarburization processes.Further development of machine learning in the steelmaking process modeling can be realized through additional efforts in the construction of the data platform,the industrial transformation of the research achievements to the practical steelmaking process,and the improvement of the universality of the machine learning models.

Augmented Industrial Data-Driven Modeling Under the Curse of Dimensionality

The curse of dimensionality refers to the problem o increased sparsity and computational complexity when dealing with high-dimensional data.In recent years,the types and vari ables of industrial data have increased significantly,making data driven models more challenging to develop.To address this prob lem,data augmentation technology has been introduced as an effective tool to solve the sparsity problem of high-dimensiona industrial data.This paper systematically explores and discusses the necessity,feasibility,and effectiveness of augmented indus trial data-driven modeling in the context of the curse of dimen sionality and virtual big data.Then,the process of data augmen tation modeling is analyzed,and the concept of data boosting augmentation is proposed.The data boosting augmentation involves designing the reliability weight and actual-virtual weigh functions,and developing a double weighted partial least squares model to optimize the three stages of data generation,data fusion and modeling.This approach significantly improves the inter pretability,effectiveness,and practicality of data augmentation in the industrial modeling.Finally,the proposed method is verified using practical examples of fault diagnosis systems and virtua measurement systems in the industry.The results demonstrate the effectiveness of the proposed approach in improving the accu racy and robustness of data-driven models,making them more suitable for real-world industrial applications.

Research on Welding Quality Traceability Model of Offshore Platform Block Construction Process

Quality traceability plays an essential role in assembling and welding offshore platform blocks.The improvement of the welding quality traceability system is conducive to improving the durability of the offshore platform and the process level of the offshore industry.Currently,qualitymanagement remains in the era of primary information,and there is a lack of effective tracking and recording of welding quality data.When welding defects are encountered,it is difficult to rapidly and accurately determine the root cause of the problem from various complexities and scattered quality data.In this paper,a composite welding quality traceability model for offshore platform block construction process is proposed,it contains the quality early-warning method based on long short-term memory and quality data backtracking query optimization algorithm.By fulfilling the training of the early-warning model and the implementation of the query optimization algorithm,the quality traceability model has the ability to assist enterprises in realizing the rapid identification and positioning of quality problems.Furthermore,the model and the quality traceability algorithm are checked by cases in actual working conditions.Verification analyses suggest that the proposed early-warningmodel for welding quality and the algorithmfor optimizing backtracking requests are effective and can be applied to the actual construction process.

Data Processing Model of Coalmine Gas Early-Warning System

The data processing mode is vital to the performance of an entire coalmine gas early-warning system, especially in real-time performance. Our objective was to present the structural features of coalmine gas data, so that the data could be processed at different priority levels in C language. Two different data processing models, one with prior- ity and the other without priority, were built based on queuing theory. Their theoretical formulas were determined via a M/M/1 model in order to calculate average occupation time of each measuring point in an early-warning program. We validated the model with the gas early-warning system of the Huaibei Coalmine Group Corp. The results indicate that the average occupation time for gas data processing by using the queuing system model with priority is nearly 1/30 of that of the model without priority..

Establishment of Person and Enterprise Process Model in Product Data Management System

Before the implementation of product data management (PDM) system, person model and enterprise process model (EPM) must be firstly established. For the convenience of project management, all the related users must be allocated to the “Person User Role Group” net. Based on the person model and the direction of information flow, the EPM is established subsequently. The EPM consists of several release levels, in which the access controls are defined. The EPM procedure shows the blueprint of the workflow process structure. The establishment of person model and EPM in an enterprise has been instanced at the end of this paper.

Fault monitoring based on mutual information feature engineering modeling in chemical process

A large amount of information is frequently encountered when characterizing the sample model in chemical process.A fault diagnosis method based on dynamic modeling of feature engineering is proposed to effectively remove the nonlinear correlation redundancy of chemical process in this paper.From the whole process point of view,the method makes use of the characteristic of mutual information to select the optimal variable subset.It extracts the correlation among variables in the whitening process without limiting to only linear correlations.Further,PCA(Principal Component Analysis)dimension reduction is used to extract feature subset before fault diagnosis.The application results of the TE(Tennessee Eastman)simulation process show that the dynamic modeling process of MIFE(Mutual Information Feature Engineering)can accurately extract the nonlinear correlation relationship among process variables and can effectively reduce the dimension of feature detection in process monitoring.

PERFORMANCE EVALUATION METHOD FOR BUSINESS PROCESS OF MACHINERY MANUFACTURER BASED ON DEA/AHP HYBRID MODEL

A set of indices for performance evaluation for business processes with multiple inputs and multiple outputs is proposed, which are found in machinery manufacturers. Based on the traditional methods of data envelopment analysis （DEA） and analytical hierarchical process （AHP）, a hybrid model called DEA/AHP model is proposed to deal with the evaluation of business process performance. With the proposed method, the DEA is firstly used to develop a pairwise comparison matrix, and then the AHP is applied to evaluate the performance of business process using the pairwise comparison matrix. The significant advantage of this hybrid model is the use of objective data instead of subjective human judgment for performance evaluation. In the case study, a project of business process reengineering （BPR） with a hydraulic machinery manufacturer is used to demonstrate the effectiveness of the DEA/AHP model.

XML-based Data Processing in Network Supported Collaborative Design

In the course of network supported collaborative design, the data processing plays a very vital role. Much effort has been spent in this area, and many kinds of approaches have been proposed. Based on the correlative materials, this paper presents extensible markup language （XML） based strategy for several important problems of data processing in network supported collaborative design, such as the representation of standard for the exchange of product model data （STEP） with XML in the product information expression and the management of XML documents using relational database. The paper gives a detailed exposition on how to clarify the mapping between XML structure and the relationship database structure and how XML-QL queries can be translated into structured query language （SQL） queries. Finally, the structure of data processing system based on XML is presented.

On the Efficiency of a CFD-Based Full Convolution Neural Network for the Post-Processing of Field Data

The present study aims to improve the efficiency of typical procedures used for post-processing flow field data by applying a neural-network technology.Assuming a problem of aircraft design as the workhorse,a regression calculation model for processing the flow data of a FCN-VGG19 aircraft is elaborated based on VGGNet(Visual Geometry Group Net)and FCN(Fully Convolutional Network)techniques.As shown by the results,the model displays a strong fitting ability,and there is almost no over-fitting in training.Moreover,the model has good accuracy and convergence.For different input data and different grids,the model basically achieves convergence,showing good performances.It is shown that the proposed simulation regression model based on FCN has great potential in typical problems of computational fluid dynamics(CFD)and related data processing.

Sedimentary Microfacies and Porosity Modeling of Deep-Water Sandy Debris Flows by Combining Sedimentary Patterns with Seismic Data： An Example from Unit I of Gas Field A, South China Sea

Sandy debris flow deposits are present in Unit I during Miocene of Gas Field A in the Baiyun Depression of the South China Sea. The paucity of well data and the great variability of the sedimentary microfacies make it difficult to identify and predict the distribution patterns of the main gas reservoir, and have seriously hindered further exploration and development of the gas field. Therefore, making full use of the available seismic data is extremely important for predicting the spatial distribution of sedimentary microfacies when constructing three-dimensional reservoir models. A suitable reservoir modeling strategy or workflow controlled by sedimentary microfacies and seismic data has been developed. Five types of seismic attributes were selected to correlate with the sand percentage, and the root mean square （RMS） amplitude performed the best. The relation between the RMS amplitude and the sand percentage was used to construct a reservoir sand distribution map. Three types of main sedimentary microfacies were identified： debris channels, fan lobes, and natural levees. Using constraints from the sedimentary microfacies boundaries, a sedimentary microfacies model was constructed using the sequential indicator and assigned value simulation methods. Finally, reservoir models of physical properties for sandy debris flow deposits controlled by sedimentary microfacies and seismic inversion data were established. Property cutoff values were adopted because the sedimentary microfacies and the reservoir properties from well-logging interpretation are intrinsically different. Selection of appropriate reservoir property cutoffs is a key step in reservoir modeling when using simulation methods based on sedimentary microfacies control. When the abnormal data are truncated and the reservoir properties probability distribution fits a normal distribution, microfacies-controlled reservoir property models are more reliable than those obtained from the sequence Gauss simulation method. The cutoffs for effective porosity of the debris channel, fan lobe, and natural levee facies were 0.2, 0.09, and 0.12, respectively; the corresponding average effective porosities were 0.24, 0.13, and 0.15. The proposed modeling method makes full use of seismic attributes and seismic inversion data, and also makes the property data of single-well depositional microfacies more conformable to a normal distribution with geological significance. Thus, the method allows use of more reliable input data when we construct a model of a sandy debris flow.

Detection of Alzheimer’s disease onset using MRI and PET neuroimaging:longitudinal data analysis and machine learning

The scientists are dedicated to studying the detection of Alzheimer’s disease onset to find a cure, or at the very least, medication that can slow the progression of the disease. This article explores the effectiveness of longitudinal data analysis, artificial intelligence, and machine learning approaches based on magnetic resonance imaging and positron emission tomography neuroimaging modalities for progression estimation and the detection of Alzheimer’s disease onset. The significance of feature extraction in highly complex neuroimaging data, identification of vulnerable brain regions, and the determination of the threshold values for plaques, tangles, and neurodegeneration of these regions will extensively be evaluated. Developing automated methods to improve the aforementioned research areas would enable specialists to determine the progression of the disease and find the link between the biomarkers and more accurate detection of Alzheimer’s disease onset.

Application of Model-Based Data Transmission Techniques to Gravitational Model Data

The transmission of scientific data over long distances is required to enable interplanetary science expeditions. Current approaches include transmitting all collected data or transmitting low resolution data to enable ground controller review and selection of data for transmission. Model-based data transmission (MBDT) seeks to increase the amount of knowledge conveyed per unit of data transmitted by comparing high-resolution data collected in situ to a pre-existing (or potentially co-transmitted) model. This paper describes the application of MBDT to gravitational data and characterizes its utility and performance. This is performed by applying the MBDT technique to a selection of gravitational data previously collected for the Earth and comparing the transmission requirements to the level required for raw data transmis-sion and non-application-aware compression. Levels of transmission reduction up to 31.8% (without the use maximum-error-thresholding) and up to 97.17% (with the use of maximum-error-thresholding) resulted. These levels significantly exceed what is possible with non-application-aware compression.

Multimode Process Monitoring Based on the Density-Based Support Vector Data Description

Complex industry processes often need multiple operation modes to meet the change of production conditions. In the same mode,there are discrete samples belonging to this mode. Therefore,it is important to consider the samples which are sparse in the mode.To solve this issue,a new approach called density-based support vector data description( DBSVDD) is proposed. In this article,an algorithm using Gaussian mixture model( GMM) with the DBSVDD technique is proposed for process monitoring. The GMM method is used to obtain the center of each mode and determine the number of the modes. Considering the complexity of the data distribution and discrete samples in monitoring process,the DBSVDD is utilized for process monitoring. Finally,the validity and effectiveness of the DBSVDD method are illustrated through the Tennessee Eastman( TE) process.

OSS Effort Expense Optimization Based on Wiener Process Model and GA

Various open source software are managed by using several bug tracking systems. In particular, the open source software extends to the cloud service and edge computing. Recently, OSF Edge Computing Group is launched by OpenStack. There are big data behind the internet services such as cloud and edge computing. Then, it is important to consider the impact of big data in order to assess the reliability of open source software. Various optimal software release problems have been proposed by specific researchers. In the typical optimal software release problems, the cost parameters are defined as the known parameter. However, it is difficult to decide the cost parameter because of the uncertainty. The purpose of our research is to estimate the effort parameters included in our models. In this paper, we propose an estimation method of effort parameter by using the genetic algorithm. Then, we show the estimation method in section 3. Moreover, we analyze actual data to show numerical examples for the estimation method of effort parameter. As the research results, we found that the OSS managers would be able to comprehend the human resources required before the OSS project in advance by using our method.

面向云边端协同的多模态数据建模技术及其应用

云边端协同架构中数据类型多样,各级存储资源与计算资源存在差异,给数据管理带来新的挑战.现有数据模型或者数据模型的简单叠加,都难以同时满足云边端中多模态数据管理和协同管理需求.因此,研究面向云边端协同的多模态数据建模技术成为重要问题.其核心在于,如何高效地从云边端三层架构中得到满足应用所需的查询结果.从云边端三层数据的数据类型出发,提出了面向云边端协同的多模态数据建模技术,给出了基于元组的多模态数据模型定义,设计了6种基类,解决多模态数据统一表征困难的问题;提出了云边端协同查询的基本数据操作体系,以满足云边端业务场景的查询需求;给出了多模态数据模型的完整性约束,为查询优化奠定了理论基础.最后,给出了面向云边端协同多模态数据模型的示范应用,并从数据存储时间、存储空间和查询时间这3个方面对所提出的数据模型存储方法进行了验证.实验结果表明,所提方案能够有效地表示云边端协同架构中的多模态数据.

基于特征匹配的逆向工程模型重建方法

逆向工程(RE)也称反求工程,以实物、样件、软件(包括图样、程序、技术文件等)或影像(图片、照片等)为对象,应用产品设计方法学、系统工程学、计算机辅助技术方法进行系统分析和研究,探索掌握其关键技术,进而开发出同类或更先进产品。从数据采集、数据处理和模型构建3个阶段给出了RE中的模型重建过程。首先基于特征的点云匹配方法进行数据采集,获取物体表面的点云数据,然后采取降噪、平滑等方式进行预处理,最后通过曲面拟合、曲面特征提取和数据拼接完成实物到模型的建模。同时给出了各阶段的通用方法和实际应用的注意事项。RE可快速准确地确定实物的结构特点、内部构造等信息,并为后续设计提供参考。

基于Transformer的司法文书命名实体识别方法

命名实体识别是自然语言处理领域的关键任务之一,是实现下游任务的基础。目前针对司法领域的相关研究相对较少,司法系统的信息化和智能化转型仍有许多问题亟需解决。相比其他领域的文本,司法文书存在专业性强、语料资源少等局限,导致现有的司法文书识别结果较低。因此,从以下3方面开展研究:首先,提出了一种多标签层级迭代的文本标注方式,可以对原始司法文书文本进行自动化标注,同时有效地提升司法文书命名实体识别任务的实体识别效果;其次,提出了一种交融式的Transformer神经网络模型,对汉字固有属性的深层特征进行了充分利用,用于对司法文书进行命名实体识别;最后,对所提出的标注方法和模型与其他神经网络模型进行了对比实验。所提出的文本标注方式可以较为准确地实现司法文书的标注任务;同时,所提出的模型在通用数据集中相对于对照模型有较大的提高,并在司法领域数据集中取得了良好的效果。

数模联动的多特征工件加工能耗预测方法研究

在实际切削加工过程中材料去除率是不断变化的,现有将其视为恒量的能耗建模方法难以实现能耗准确预测。为了提高切削过程能耗预测精度,提出了一种基于材料去除率的数模联动加工能耗预测方法。首先,基于切削过程刀具与工件的接触关系分析了切入、完全切入和切出阶段材料去除率变化规律,并对相应的加工能耗特性进行了分析;其次,提出了数据驱动的刀具切入,切出阶段加工能耗预测方法,以及模型驱动的完全切入阶段加工能耗预测方法,实现加工过程能耗准确预测;最后,利用实验案例验证了所提模型及方法的有效性,为今后研究能耗预测精度奠定了基础。

基于赛教融合的智能交通课程实践教学研究

结合全国大学生智能汽车竞赛百度智慧交通创意组竞赛项目,设计智能交通课程实践教学内容,从竞赛内容凝练与课程知识点结合的教学案例和实验项目,实现赛教资源融合和赛教一体化实践平台共享。设计的实验教学项目融合了图像处理、数据分析、人工智能、智能车硬件平台等内容,包括数据采集、数据分析、数据集构建,深度学习模型建立与智能车平台部署等实践环节。学生从编写简单程序开始,不断提高,到构建复杂的深度学习模型实现智能车自动巡航控制实验。将竞赛任务具有挑战性和对抗性的特点融入课程教学,吸引学生兴趣,增强学生求知欲,提升学生创新能力和工程实践能力。

基于OPC UA的纤维缠绕机信息模型开发和应用

为实现对复合材料纤维缠绕机生产过程的监测,以缠绕机及其辅助设备为研究对象,针对实际加工设备的多样性造成的加工数据采集的异构性,基于OPC UA建立缠绕机的信息模型架构。根据OPC基金会发布的OPC UA规范和相关的行业标准,建立纤维缠绕机系统的信息模型和实例化。基于开源项目open62541,导入编译后的缠绕机的信息模型代码,自定义开发出符合使用规范的OPC UA服务器,然后利用UA客户端实现对服务器的连接,客户端能够通过服务器查询系统的加工数据,解决加工数据的异构性问题以达到监测缠绕机生产加工过程的目的;最后验证了建立的纤维缠绕机信息模型能够正确导入到开源项目open62541中,实现了加工数据监测。