Industrial Big Data Research Articles

Research on Privacy Protection Based on Joint Learning in Power Industry Big Data Analysis

In the era of big data, protecting the privacy of smart grid data is critical in ensuring the integrity and confidentiality of that data. Utilizing large amounts of energy data to gain insight into electricity consumers’ consumption patterns helps develop power supply strategies. This article presents a Big Data-assisted Joint learning process (BDA-JLP), taking data security issues posed by big data in the electric power industry into consideration for privacy protection using K-anonymity and L-diversity as a foundation. A blockchain with JLP electric utility investigation is being conducted, part of the existing trading model split into phases. To begin, an attribute is chosen to categorize the input database. The comparable class number K and sensitive attribute value category L are limited by the number of original predecessors in the source data table, simplifying the calculation. A mathematical equation is then developed to determine the distance between first cousins multiplied by their combined weight. Linear and clustering with binary K are used to categorize data tables. Cluster and generalize initial data sets, considering how the attribute values’ internal range changes. The asymmetric encryption method uses two distinct keys for encryption and decryption ensuring that the blockchain system is completely secure. Simulated data show that the BDA-JLP mechanism proposed here has a privacy ratio of 98.3 percent, scalability of 97.0%, improved data management and data protection ratio of 98.2 percent, customer satisfaction ratio of 98.4 percent, and a low energy consumption ratio of 23.9% when compared to other methods currently available.

Application of Industrial Big Data Cloud Control Platform Based on Fusion Transmission Sensor

In order to solve the problem that the traditional operation and maintenance system of photovoltaic operation and power station is mainly based on statistical analysis and multisite and multilevel deployment mode, which wastes software and hardware resources, is not easy to expand and has extremely low efficiency, which cannot meet the urgent needs of users to reduce costs and increase efficiency. This research proposes a methodology for integrating big data, cloud computing, and photovoltaic operation and maintenance. This method constructs the business model, data model, application model, and technical model of the photovoltaic power station operation and maintenance cloud platform. The results obtained are as follows: the system provides diagnostic services for the application system of a 30 MWp photovoltaic power station in a certain place, and a total of 87 defects are found, the defect elimination rate is 88.51%, and the monthly power generation of the power station is increased by 122,529 kWh; the use effect of the system in this research after it goes online. The evaluation is 93.04 points, ranging from very satisfactory (A) to satisfactory (B) and biased towards A, indicating that the use effect is good. It is proved that the successful research and promotion of the system in this research will be of great significance to improve the intelligent operation and maintenance level of photovoltaic power plants and improve the operation and maintenance efficiency of photovoltaic power plants.

Intelligent Health Monitoring of Machine Tools Using a Bayesian Multibranch Neural Network

Online health monitoring of machine tools is an essential technique for tool life extension, manufacturing productivity improvement and product quality improvement. In the era of industrial big data, numerous deep learning (DL)-based methods have been proposed to achieve these goals. However, in complex and dynamic manufacturing processes, practical concerns such as uncertainty quantification and anti-noise capabilities of the DL-based methods are rarely considered. Thus, in this study, a novel multi-branch Bayesian Neural Network (BNN) is developed for the reliable and robust online health monitoring of Computer Numerical Control (CNC) machine tools. With the proposed model, the heterogeneous fault information extracted from multiple sensors can be simultaneously integrated in a deep convolutional neural network (DCNN)-multiple layer perceptron (MLP)-based multi-branch neural network to enhance the health monitoring accuracy and robustness. Furthermore, the proposed multi-branch neural network is extended into a BNN to improve its uncertainty quantification capabilities. The proposed method is evaluated on the tool wear tests of three cutting tools. Tool wear estimation results indicate that the proposed method outperforms comparative methods and achieves the best prediction accuracy and robustness on all three health monitoring tasks investigated in this study. We also found that the proposed method can accurately classify tool wear stages and reach up to 95% mean classification accuracy, which is the best among comparative methods. Also, measures, such as coverage probability of estimation interval (EICP) and normalized mean estimation interval width (NMEIW), are used to assess the capability of quantifying the confidence intervals (CIs) of the tool wear estimations. Results show that the proposed method achieves superior CIs quantification performance with the average EICP and NMEIW values of 95.77% and 0.27 on all three health monitoring tasks.

A Bibliometric Analysis of the Reliability Assessment Technology Based on Accelerated Degradation

Background: In the face of the development trend of high-end manufacturing servitization, the reliability standard of manufacturing products gradually increases. Objective: In order to accurately predict the product life cycle, the accelerated degradation evaluation technology could be applied to significantly shorten the experiment duration. As the technologies of intelligent manufacturing and industrial big data develop, the theory of accelerated degradation evolves as well. Methods: Based on the scientific knowledge mapping, co-author network and co-existence network, 22283 pertinent articles since the year 2010 have been collected to conduct a bibliometric study. Results: The results show that the accelerated degradation reliability assessment spans over many research fields, and achieves great development in the mathematical modeling and experiment verification. Conclusion: To further the study, more efforts are expected in the areas such as building effective evaluation systems and enhancing the credibility of the assessment outcomes, as more advanced sensory data and wireless communication technologies become available.

Research on the Construction of University Campus Economic Management System Based on the Concept of Big Data

At present, in order to adapt to the development of the era of big data, more and more industries use the concept of big data, and the economic and financial management system is no exception. In the context of the concept of big data, the campus economy has also ushered in new opportunities and challenges. Combined with the basic knowledge of big data, this paper makes a systematic research on industrial economic information analysis, which will contribute to the application practice of industrial big data. Using the value of big data will promote revolutionary changes in the work of the school’s economic management department. Taking this as the starting point, this paper systematically analyzes the analysis methods of campus economic information under the background of big data and its role in macro decision-making.

Research on the Construction of Application-Oriented Undergraduate Data Science and Big Data Technology Courses

In order to conduct research and analysis on the construction of application-oriented undergraduate data science and big data technology courses, the professional development characteristics of universities and enterprises should be taken into consideration, the development trend of the big data industry should be scrutinized, and professional application-oriented talents should be cultivated in line with job requirements. This paper expounds the demand for capacity-building professional development in application-oriented undergraduate data science and big data technology courses, conducts research and analysis on the current situation of professional development, and puts forward strategies in hope to provide reference for capacity-building professional development.

Research on the Reform of Higher Education Training Mode under the Background of Big Data and Internet of Things

With the rapid development of information processing, storage, and transmission technologies, the demand for talents from all walks of life in the society has increased, which has put forward higher requirements for the diversity of Chinese higher education. In China’s higher education, undergraduate education is in a fundamental position. China’s undergraduate education has undergone several reforms in talent training, but there are still many problems in the innovative talent training model. With the gradual improvement of the Internet of Things infrastructure, various applications characterized by big data are deeply embedded in people’s daily life. In the era of big data, the amount of data is larger and the structure is more complex. It penetrates into all fields of human life and shows unprecedented vitality and vigor. Especially in the field of higher education, the era of big data provides more information resources for education and learning and also brings new opportunities for the cultivation of applied talents in local undergraduate colleges and universities. This paper deeply studies the training mode of applied talents in local undergraduate colleges and universities under the background of big data and the Internet of Things. Based on the data analysis results of big data application technicians and industry talents, professional ability-course structure data, this paper completes the design of the corresponding talent training mode analysis system. Under the application of the system designed in this paper, it is possible to realize the analysis of the differences in the knowledge structure requirements of talents in various positions and the needs of talents’ ability, as well as the optimization and reform of the talent training curriculum system. The design of this paper can play an important guiding role in talent training in the environment of big data and Internet of Things.

Using meta-learning for automated algorithms selection and configuration: an experimental framework for industrial big data

Advanced analytics are fundamental to transform large manufacturing data into resourceful knowledge for various purposes. In its very nature, such “industrial big data” can relay its usefulness to reach further utilitarian applications. In this context, Machine Learning (ML) is among the major predictive modeling approaches that can enable manufacturing researchers and practitioners to improve the product quality and achieve resource efficiency by exploiting large amounts of data (which is collected during manufacturing process). However, disposing ML algorithms is a challenging task for manufacturing industrial actors due to the prior specification of one or more algorithms hyperparameters (HPs) and their values. Moreover, manufacturing industrial actors often lack the technical expertise to apply advanced analytics. Consequently, it necessitates frequent consultations with data scientists; but such collaborations tends to cost the delays, which can generate the risks such as human-resource bottlenecks. As the complexity of these tasks increases, so does the demand for support solutions. In response, the field of automated ML (AutoML) is a data mining-based formalism that aims to reduce human effort and speedup the development cycle through automation. In this regard, existing approaches include evolutionary algorithms, Bayesian optimization, and reinforcement learning. These approaches mainly focus on providing the user assistance by automating the partial or entire data analysis process, but they provide very limited details concerning their impact on the analysis. The major goal of these conventional approaches has been generally focused on the performance factors, while the other important and even crucial aspects such as computational complexity are rather omitted. Therefore, in this paper, we present a novel meta-learning based approach to automate ML predictive models built over the industrial big data. The approach is leveraged with development of, AMLBID, an Automated ML tool for Big Industrial Data analyses. It attempts to support the manufacturing engineers and researchers who presumably have meager skills to carry out the advanced analytics. The empirical results show that AMLBID surpasses the state-of-the-art approaches and could retrieve the usefulness of large manufacturing data to prosper the research in manufacturing domain and improve the use of predictive models instead of precluding their outcomes.

Physical Metallurgy Guided Industrial Big Data Analysis System with Data Classification and Property Prediction

Various computational analysis systems based on machine learning (ML) methods have been established for the analysis of steel industrial data. However, limited by the extensibility of one regression strategy, it is difficult to obtain a generic property prediction model for multiple types of steels. To solve this problem, this study proposes a novel industrial big data analysis system that combines ML classification and regression models with key physical metallurgy (PM) variables. First, the database is obtained from an industrial production line and carefully preprocessed. Then, multiple types of steels are categorized into five classes using a K‐nearest neighbor (KNN) algorithm, and suitable ML algorithms are selected for each category to maximize the performance. Considering the role of PM variables in improving the model accuracy, some relevant parameters (the Ac1 temperature, Ac3 temperature, and flow stress) are introduced to guide the further optimization of the ML process. The proposed industrial analysis system has more accurate prediction and higher flexibility than the model that directly uses the original dataset. With a rational combination of different regression strategies, the present results clearly demonstrate that the extensibility of the proposed property prediction model is significantly improved for industrial big data.

Hybrid Parallel Stochastic Configuration Networks for Industrial Data Analytics

As a class of randomized learner model, stochastic configuration networks (SCNs) have been successfully applied in a few data analytics tasks. Given the industrial big data modeling tasks, however, the original SCNs potentially lead to excessive training time. To this end, this article extends SCNs to a hybrid parallel version, termed hybrid parallel SCNs (HPSCNs). In the hybrid parallel learning algorithm, two SCNs are synchronously constructed. The difference between the two of them is that one employs a point-incremental algorithm, and another one adopts a block-incremental algorithm. Moreover, a data parallel method is established based on a dynamic block strategy to accelerate the establishment of candidate node pool for each SCN. Additionally, this article proposes an adaptive hyperparameter adjustment method, which allows the hyperparameters in the supervisory mechanism to be automatically adjusted along with the learning process. Comparative experiments are first conducted through four large-scale benchmark datasets, followed by the fully discussion on the algorithm parameters. Finally, a practical industrial application case is made, where a grinding particle size soft sensor is developed based on HPSCN, showing the effectiveness of the proposed algorithm.

Research trends of supply chain management practice before and during pandemic: A bibliometric analysis

The field of supply chain management has been undergoing numerous changes before and during the pandemic. Emerging areas have caught the attention of researchers across the globe. The purpose of this study is to provide a bibliometric review and analysis of the research trends on supply chain management practice (SCMP) in the pre-pandemic and mid-pandemic periods. We collect and filter data from the Scopus database and conduct bibliometric analysis. Before and during pandemic periods, 293 and 211 papers were selected, respectively, from 9,246 publications. The researcher utilizes VOS viewer software to analyze the findings by carrying out citation analysis and keyword analysis. The study reveals seven current research streams of high interest within SCMP: Big Data, Sustainable Supply Chain Management, Circular Economy, Industry 4.0, and Green Supply Chain. Supply chain management and sustainability are the most frequently occurring keywords. The most influential countries in pre-pandemic and mid-pandemic periods are USA and UK, respectively, with 59 and 37 publications. The Journal of Cleaner production was the most significant source of publications. This study is among the few pieces of research revealing current SCMP research trends during the pandemic. By considering the current SCMP concepts, business and policy decision-makers can adopt the modern concepts of the field to identify new opportunities by expanding capabilities in big data analytics, sustainable development, and industry 4.0 technologies.

An ABGE-aided manufacturing knowledge graph construction approach for heterogeneous IIoT data integration

The Industrial Internet of Things (IIoT) provides a foundation for the development of emerging digital servitization paradigm in smart manufacturing. The deep integration of massive heterogeneous IIOT data plays a critical role in realising manufacturing digital servitization. However, there is a knowledge gap between different manufacturing fields, which brings a challenge for efficient integration and leverage of industrial big data. For this purpose, a Framework of Manufacturing Knowledge Graph (FMKG) is proposed, which is used to extracts industry knowledge triples from multi-source heterogeneous data to integrate domain knowledge. Also, an attention-based graph embedding model (ABGE) is proposed to discover and complement the implicit missing relationships in the knowledge graph to obtain a complete industrial knowledge graph. The effectiveness of the ABGE model has been verified on several knowledge graph data sets. And an aerospace enterprise production process was taken as an example to establish a product quality knowledge graph, which proved the feasibility of the proposed method.

Carbon Emission Calculation and Influencing Factor Analysis Based on Industrial Big Data in the "Double Carbon" Era.

The arrival of the “double carbon” era indicates that industrial carbon reduction with high energy consumption and high carbon emission is imperative. From the perspective of carbon emission driving factors, industrial carbon emission is decomposed into five influencing factors: energy intensity, energy structure, industrial structure, economic efficiency, and employee scale. Taking the data of 41 subindustries of industrial industry in Liaoning Province from 2010 to 2019 as the research sample, the carbon emission is calculated. The LMDI model is used to analyze and point out the difference in the driving contribution of carbon emissions of each subindustry. The results show that the total carbon emission of Liaoning province gradually decreases, decreases for the first time in 2014, and gradually turns from flat to upward. Economic efficiency is the only and most important reason to drive the increase of industrial carbon emissions in Liaoning Province, and energy efficiency is the primary factor to curb carbon emissions. High carbon industries play a significant role in promoting the increase of carbon emissions, while the medium and low carbon industries have a better effect on restraining carbon emissions. It provides reference and theoretical basis for the government to adjust the industrial structure, control industrial overcapacity, and realize the “double carbon” goal as soon as possible. It is of great significance for the country to optimize energy layout, ensure energy security, and implement the new energy strategy.

A Stochastic Learning Algorithm for Machine Fault Diagnosis

Industrial big data bring a large number of high-dimensional sample datasets. Although a deep learning network can well mine the internal nonlinear structure of the dataset, the construction of the deep learning model requires a lot of computing time and hardware facilities. At the same time, there are some nonlinear problems such as noise and fluctuation in industrial data, which make the deep architecture extremely complex and the recognition accuracy of the diagnosis model difficult to guarantee. To solve this problem, a new method, named stochastic learning algorithm (SL), is proposed in this paper for dimension reduction. The proposed method consists of three steps: firstly, to increase the computational efficiency of the model, the dimension of the high-dimensional data is reduced by establishing a random matrix; secondly, for enhancing the clustering influence of the sample, the input data are enhanced by feature processing; thirdly, to make the clustering effects more pronounced, the noise and interference of the data need to be processed, and the singularity value denoising method is used to denoise training data and test data. To further prove the superiority of the SL method, we conducted two sets of experiments on the wind turbine gearbox and the benchmark dataset. It can be seen from the experimental results that the SL method not only improves the classification accuracy but also reduces the computational burden.

Towards big industrial data mining through explainable automated machine learning

Industrial systems resources are capable of producing large amount of data. These data are often in heterogeneous formats and distributed, yet they provide means to mine the information which can allow the deployment of intelligent management tools for production activities. For this purpose, it is necessary to be able to implement knowledge extraction and prediction processes using Artificial Intelligence (AI) models, but the selection and configuration of intended AI models tend to be increasingly complex for a non-expert user. In this paper, we present an approach and a software platform that may allow industrial actors, who are usually not familiar with AI, to select and configure algorithms optimally adapted to their needs. Hence, the approach is essentially based on automated machine learning. The resulting platform effectively enables a better choice among the combination of AI algorithms and hyper-parameters configurations. It also makes it possible to provide features of explainability of the resulting algorithms and models, thus increasing the acceptability of these models in practicing community of the users. The proposed approach has been applied in the field of predictive maintenance. Current tests are based on the analysis of more than 360 databases from the subjected field.

System construction for comprehensive industrial ecosystem oriented networked collaborative manufacturing platform (NCMP) based on three chains

In industrial manufacturing, with the deep integration and development of advanced digital technologies such as industrial big data, industrial Internet and industrial artificial intelligence with manufacturing industry, the manufacturing model is gradually developing and evolving from the direction of networking, platform, integration, collaboration and ecology. Therefore, for industrial manufacturing process, the research of construction of comprehensive industrial ecosystem oriented Networked Collaborative Manufacturing Platform (NCMP) system is great industrial application value by integrating network characteristics, platform elements, integrated mechanism, collaborative model and ecological format. In this paper, the system model of NCMP is constructed from four perspectives, including network perspective, collaborative manufacturing perspective, platform perspective, and industrial ecosystem perspective. Then, a system framework of NCMP based on three chains (manufacturing chain, value chain, and industrial chain) is proposed. Three collaborative subsystems of NCMP (vertical collaboration subsystem from multi perspective, horizontal collaboration subsystem from multi process, end to end collaboration subsystem from multi operator) are constructed. At last, a comprehensive system construction for NCMP in automobile industry is giving. The research results show that the most important elements to construct NCMP are purchasing department collaboration, demand department collaboration, design collaboration, data integration between enterprises and factories, etc.

Rethinking the Value of Just-in-Time Learning in the Era of Industrial Big Data

Just-in-time learning (JITL) has become a widely used industrial process modeling tool. With the advent of the industrial big data era, rich data information has brought new opportunities to JITL. Specifically, the completeness of data samples in the era of big data provides an important premise and support for the JITL method, prompting us to rethink the application value of JITL in the context of industrial big data. At the same time, the huge amount of data causes certain difficulties in data searching, which is a key issue for JITL. In this article, a parallel computing strategy is adopted to divide the entire computational searching task into several subtasks, and assign the subtasks to parallel computing nodes to complete parallel searching. In this way, not only the full advantage of big data information is effectively utilized, but also the search capability and efficiency under big data are improved. In addition, in order to improve the real-time nature of JITL, a model library management (MLM) strategy is adopted, and the query similar samples are used to operate with existing similar models. And by selectively adding new data, the database management (DBM) strategy is also developed, which not only alleviates the problem of information redundancy, but also reduces the search pressure caused by the increasingly large database. Obviously, MLM and DBM are particularly important as JITL auxiliary tools under industrial big data. Combining parallel computing, a parallel JITL (P-JITL) framework is proposed. As an example, the variational Bayesian factor regression model is transformed into the parallel Bayesian-JITL method for big process data modeling, which is further extended to a nonlinear form. To evaluate the feasibility and efficiency of the developed methods, a real industrial case is demonstrated.

Financial Technology, Big Data Enterprise Financing Constraints and Big Data Industry Development: Empirical Analysis Based on Mediating Effect and Threshold Effect

Based on theoretical analysis, we select the relevant data of 30 provinces (autonomous regions and municipalities) in China from 2013 to 2019, and empirically test the impact of financial technology on the development of big data industry and its mechanism using dynamic panel data model, mediating effect test method and threshold effect model. The benchmark regression results show that the regression coefficient of financial technology to big data industry is significantly positive at the significance level of 10%, indicating that the financial technology can directly promote the development of big data industry. The regression coefficient of the dynamic lag term of big data industry is negative, but not significant, indicating that the dynamic lag effect of big data industry is not obvious. The mediating effect test results show that the financial technology can indirectly promote the development of big data industry by alleviating the big data enterprise financing constraints. The big data enterprise financing constraints have a partial mediating effect, and the mediating effect account for 27.63% of the total effect. In addition, the threshold effect test results show that the direct effect of financial technology on big data industry is significantly enhanced when the development level of financial technology is higher than 5.8790, that is, there is a positive threshold effect of financial technology directly promoting the development of big data industry. However, the indirect effect of financial technology on big data industry is relatively weak when the development level of financial technology is higher than 5.4328, that is, financial technology indirectly promotes the development of big data industry by alleviating the big data enterprise financing constraints, which has a negative threshold effect.

Industrial artificial intelligence based energy management system: Integrated framework for electricity load forecasting and fault prediction

Forecasting accuracy electricity load can help industrial enterprises optimise production scheduling based on peak and off-peak electricity prices. The electricity load forecasting results can be provided to an electricity system to improve electricity generation efficiency and minimize energy consumption by developing electricity generation plans in advance and by avoiding over or under the generation of electricity. However, because of the different informatization levels in different industries, few reliable intelligent electricity management systems are applied on the power supply side. Based on industrial big data and machine learning algorithms, this study proposes an integrated model to forecast short-term electricity load. The hybrid model based on the hybrid mode decomposition algorithms is proposed to decompose the total electricity load signal. To improve the generalisation ability of the forecasting model, a dynamic forecasting model is proposed based on the improved hybrid intelligent algorithm to forecast the short-term electricity load. The results show that the accuracy of the proposed dynamic integrated electricity load forecasting model is as high as 99%. The integrated framework could forecast abnormal electricity consumption in time and provide reliable evidence for production process scheduling.

Big Data Precision Marketing Approach under IoT Cloud Platform Information Mining

In this article, an in-depth study and analysis of the precision marketing approach are carried out by building an IoT cloud platform and then using the technology of big data information mining. The cloud platform uses the MySQL database combined with the MongoDB database to store the cloud platform data to ensure the correct storage of data as well as to improve the access speed of data. The storage method of IoT temporal data is optimized, and the way of storing data in time slots is used to improve the efficiency of reading large amounts of data. For the scalability of the IoT data storage system, a MongoDB database clustering scheme is designed to ensure the scalability of data storage and disaster recovery capability. The relevant theories of big data marketing are reviewed and analyzed; secondly, based on the relevant theories, combined with the author's work experience and relevant information, a comprehensive analysis and research on the current situation of big data marketing are conducted, focusing on its macro-, micro-, and industry environment. The service model combines the types of user needs, encapsulates the resources obtained by the alliance through data mining for service products, and publishes and delivers them in the form of data products. From the perspective of the development of the telecommunications industry, in terms of technology, the telecommunications industry has seen the development trend of mobile replacing fixed networks and triple play. The development of emerging technologies represented by the Internet of Things and cloud computing has also led to technological changes in the telecommunications industry. Operators are facing new development opportunities and challenges. It also divides the service mode into self-service and consulting service mode according to the different degrees of users' cognition and understanding of the service, as well as proposes standardized data mining service guarantee from two aspects: after-sales service and operation supervision. A customized data mining service is a kind of data mining service for users' personalized needs. And the intelligent data mining service guarantee is proposed from two aspects of multicase experience integration and group intelligence. In the empirical research part, the big data alliance in Big Data Industry Alliance, which provides data mining service as the main business, is selected as the research object, and the data mining service model of the big data alliance proposed in this article is applied to the actual alliance to verify the scientific and rationality of the data mining service model and improve the data mining service model management system.