Monitoring Of Manufacturing Processes Research Articles

Deep Convolutional Neural Networks as a Rapid Screening Tool for Complex Additively Manufactured Structures

Additively manufactured metamaterials such as lattices offer unique physical properties such as high specific strengths and stiffnesses. However, additively manufactured parts, including lattices, exhibit a higher variability in their mechanical properties than wrought materials, placing more stringent demands on inspection, part quality verification, and product qualification. Previous research on anomaly detection has primarily focused on using in-situ monitoring of the additive manufacturing process or post-process (ex-situ) x-ray computed tomography. In this work, we show that convolutional neural networks (CNN), a machine learning algorithm, can directly predict the energy required to compressively deform gyroid and octet truss metamaterials using only optical images. Using the tiled nature of engineered lattices, the relatively small data set (43 to 48 lattices) can be augmented by systematically subdividing the original image into many smaller sub-images. During testing of the CNN, the prediction from these sub-images can be combined using an ensemble-like technique to predict the deformation work of the entire lattice. This approach provides a fast and inexpensive screening tool for predicting properties of 3D printed lattices. Importantly, this artificial intelligence strategy goes beyond ‘inspection’, since it accurately estimates product performance metrics, not just the existence of defects.

Recent advances in material development and process monitoring in polymer additive manufacturing

AbstractThis article provides an overview of the recent progress made in the field of polymer based additive manufacturing. The research works discussed in this article was presented at the Polymers in Additive Manufacturing session of the Next‐Gen Manufacturing Topical Conference at the 2019 American Institute of Chemical Engineers Annual Meeting. It is worth noting that the developments reviewed in this paper are by no means exhaustive; however, we focus only on the research presented in the aforementioned session for the sake of conciseness. The major focus of this article is on the utilization of macromolecular systems in the emerging field of additive manufacturing. In particular, the article highlights both experimental and computational efforts that utilize polymers in additive manufacturing. Current research trends addressing all methods of polymer‐based additive manufacturing, including material extrusion, stereolithography, and direct ink writing, have been analyzed. Finally, prospective avenues to achieve more fundamental understanding and hence further insights into improving additive manufacturing of polymers have also been presented.

Analysis of chemometric models applied to Raman spectroscopy for monitoring key metabolites of cell culture.

The Food and Drug Administration (FDA) initiative of Process Analytical Technology (PAT) encourages the monitoring of biopharmaceutical manufacturing processes by innovative solutions. Raman spectroscopy and the chemometric modeling tool partial least squares (PLS) have been applied to this aim for monitoring cell culture process variables. This study compares the chemometric modeling methods of Support Vector Machine radial (SVMr), Random Forests (RF), and Cubist to the commonly used linear PLS model for predicting cell culture components-glucose, lactate, and ammonia. This research is performed to assess whether the use of PLS as standard practice is justified for chemometric modeling of Raman spectroscopy and cell culture data. Model development data from five small-scale bioreactors (2 × 1 L and 3 × 5 L) using two Chinese hamster ovary (CHO) cell lines were used to predict against a manufacturing scale bioreactor (2,000 L). Analysis demonstrated that Cubist predictive models were better for average performance over PLS, SVMr, and RF for glucose, lactate, and ammonia. The root mean square error of prediction (RMSEP) of Cubist modeling was acceptable for the process concentration ranges of glucose (1.437 mM), lactate (2.0 mM), and ammonia (0.819 mM). Interpretation of variable importance (VI) results theorizes the potential advantages of Cubist modeling in avoiding interference of Raman spectral peaks. Predictors/Raman wavenumbers (cm-1 ) of interest for individual variables are X1139-X1141 for glucose, X846-X849 for lactate, and X2941-X2943 for ammonia. These results demonstrate that other beneficial chemometric models are available for use in monitoring cell culture with Raman spectroscopy.

Real-time in situ monitoring of manufacturing process and CFRP quality by relative resistance change measurement

In situ monitoring of resin flow, impregnation of carbon fiber fabrics, and curing during composite manufacturing are very important for determining the quality of composite parts. In conventional methods, sensors, such as optical fibers and strain gages, are bonded to or embedded in the composites for measuring the changes in mechanical and chemical properties. Although they can detect resin curing behavior and impregnation of carbon fibers, they may adversely affect the manufacturing process or structural integrity of the composites. In this study, carbon fiber itself was used as a sensor that minimizes the degradation of mechanical properties and increases the efficiency of monitoring the manufacturing process. The change in the electrical resistance of carbon fiber fabrics was monitored during the various manufacturing processes when the resin flowed through the carbon fiber fabric and curing progressed. The effectiveness of this monitoring method was confirmed, and it is expected to be applicable in monitoring the quality of the finished composite parts.

Stamping Process Modelling in an Internet of Production

Sharing data between companies throughout the supply chain is expected to be beneficial for product quality as well as for the economical savings in the manufacturing industry. To utilize the available data in the vision of an Internet of Production (IoP) a precise condition monitoring of manufacturing and production processes that facilitates the quantification of influences throughout the supply chain is inevitable. In this paper, we consider stamping processes in the context of an Internet of Production and the preliminaries for analytical models that utilize the ever-increasing available data. Three research objectives to cope with the amount of data and for a methodology to monitor, analyze and evaluate the influence of available data onto stamping processes have been identified: (i) State detection based on cyclic sensor signals, (ii) mapping of in- and output parameter variations onto process states, and (iii) models for edge and in-network computing approaches. After discussing state-of-the-art approaches to monitor stamping processes and the introduction of the fineblanking process as an exemplary stamping process, a research roadmap for an IoP enabling modeling framework is presented.

Intelligent data-driven monitoring of high dimensional multistage manufacturing processes

Recent advances in cyber-physical systems and the Internet of things (IoT) have enabled the possible development of smart production systems. However, the complexity of such a system has posed significant challenges for traditional quality engineering methods, especially in monitoring and diagnosis of system performance. The traditional practices for monitoring or controlling multistage systems either treat each stage as an individual entity or model all stages as a whole. The formal approach mainly focuses on the most critical stages while ignores information from the other stages. In contrast, the latter approach attempts to build one model to account for all stages. In a complex production system, this latter approach is impractical, if not impossible. This research provides a control strategy by proposing an intelligent process monitoring system for high dimensional multistage processes using predictive models built from historical data. A repository dataset is used to demonstrate the implementation of the proposed framework.

Intelligent data-driven monitoring of high dimensional multistage manufacturing processes

Intelligent data-driven monitoring of high dimensional multistage manufacturing processes

Smart Connected Sensors, Industrial Big Data, and Real-Time Process Monitoring in Cyber-Physical System-based Manufacturing

Smart Connected Sensors, Industrial Big Data, and Real-Time Process Monitoring in Cyber-Physical System-based Manufacturing

Feasibility Investigation for Online Elemental Monitoring of Iron and Steel Manufacturing Processes using Laser-Induced Breakdown Spectroscopy

The metallurgical industries are very important for social development. In order to improve the metallurgical techniques and quality of products, the real-time analysis and monitoring of iron and steel manufacturing processes are very significant. Laser-induced breakdown spectroscopy (LIBS) has been studied and applied for the contents measurement of iron and steel. In this paper, the remote open-path LIBS measurement was studied under different sample temperature, lens to target distance (LTD), sample angle conditions to clarify its online measurement features. The 3D profile measurement system of parallel laser beam fringes projection was also developed to measure the sample profile at different sample temperature. The measurement results demonstrated the robustness of remote open-path LIBS system and 3D profile measurement system. However, the correction is necessary to enhance the detection ability of LIBS online measurement. In order to improve the precision and accuracy of real-time elemental measurement, an innovative co-axial laser beam measurement system combining LIBS and 3D profile techniques is proposed to automatically adjust the focus unit and measure the sample components. The further study of this promising method will be developed for online application of iron and steel manufacturing processes.

Continuous manufacturing process monitoring of pharmaceutical solid dosage form: A case study

Continuous Manufacturing (CM) of pharmaceutical drug products is a rather new approach within the pharmaceutical industry. In the presented paper, a GMP continuous wet granulation line used for clinical production of solid dosage forms was investigated with a thorough monitoring strategy regarding process performance and robustness. The line was composed of the subsequent continuous unit operations feeding – twin-screw wet-granulation – fluid-bed drying – sieving and tableting; the formulation of a new pharmaceutical entity in development was selected for this study. In detail, a Design of Experiments (DoE) was used to evaluate the impact of the three main factors (amount of water, filling rate, and shear force in twin-screw granulator) on the tablet quality. The process was monitored via in-process control (IPC) tests (e.g. weight, hardness, disintegration, and loss-on-drying), Process Analytical Technologies (PAT), and through the analysis of the process parameters (multivariate process control). The tested formulation was very robust to the large process variation of the DoE: all IPC results were in specification, the PAT probes provided stable results for the content uniformity and no critical variations can be detected in the process parameters. An adequate monitoring strategy was presented and the robustness of the process with one formulation has been demonstrated. In summary, this continuous process in combination with smart formulation development allows the robust production of constant quality tablets. The synergy between PAT, process data science and IPC creates an adequate monitoring framework of the continuous manufacturing line.

Potential use of laser-induced breakdown spectroscopy combined laser cleaning for inspection of particle defect components on silicon wafer

The contamination control of silicon wafer surface is more and more strict. Many investigations have been done to inspect defects on silicon wafer. However, rare studies have been reported on defect component inspection, which is also critical to trace the source of defects and monitor manufacturing processes in time. In order to inspect the components of contaminated particles on silicon wafer, especially with a high-speed, in-line mode and negligible damage, a dual nanosecond pulse laser system with both wavelengths at 532 nm is designed, in which one laser pumps the particles away from the wafer surface with negligible damage, the other laser breaks down the particles in the air above the wafer surface to obtain the emission lines of the contaminated particles by a spectroscopy with intensified charge coupled device. The sensitivity of the dual pulse laser system is evaluated. The particle dynamic process after pump is analyzed. The results in this work provide a potential on-line method for the semiconductor industry to trace the sources of defects during the manufacture process.

Combined Approach for Determining Diuron in Sugarcane and Soil: Ultrasound‐Assisted Extraction, Carbon Nanotube‐Mediated Purification, and Gas Chromatography–Electron Capture Detection

Diuron is a urea herbicide that is frequently detected in surface water, groundwater, and marine waters. However, there are few methods or guidelines reported on ensuring the quality of sugarcane and soil. In this study, a method was developed for detecting diuron to ensure the quality and safety of food and sugar. Mass spectrometry was used to identify 3,4-dichloroaniline as a marker for the thermal decomposition of diuron, and thus, as a representative component for quantitative diuron analysis. This approach can be used to rapidly detect trace amounts of diuron. In addition, ultrasound-assisted extraction (UAE) and carbon nanotube column purification were used in conjunction with gas chromatography-electron capture detection to detect diuron. The method was then evaluated for its accuracy, detection limit, and viability. The effects of extraction solvent, ultrasound time, and ultrasound power on the extraction efficiency of the analyte from sugarcane and soil were also investigated. The efficiency and optimum conditions of UAE were examined through single-factor experiments and Box-Behnken design (BBD). The optimal extraction conditions were identified as follows: acetonitrile as the extraction solvent, extraction temperature of 27°C, extraction time of 3.4min, and ultrasound power of 70 W. Under these conditions, high linearity was achieved for diuron concentrations of 0.01 to 5.0 mg/L, and the purification correlation coefficient was consistently greater than 0.998. Hence, gas chromatography, combined with UAE and BBD, offers superior efficiency extraction, which is sufficiently accurate and precise for pesticide residue analysis. PRACTICAL APPLICATION: We developed an accurate and cost-effective method for detecting diuron (a commonly used herbicide) in soil and sugar samples. We performed experiments to determine the optimum detection conditions for our method. This method can be used for online monitoring of sugar manufacturing processes to ensure food safety and quality.

Manufacturing Process Monitoring in Terms of Energy Management Improving

Seeking out new technologies and deploying them for competitive advantage is a key priority for any company. In this context, Internet of Things (IoT) promise to change the way we live and work. IoT could help us in overtaking the top global challenges, as energy management, agriculture and food security, environment and natural resources security, etc., but to achieve this vision, "things" need to sense their environment and share this information among them as well as with us to offer intelligent decision-making. Nowadays, IoT become even more interesting due to the existence of Low Power Wide Area (LPWA) networks. LPWA is a key component of a wider IoT technology wave. In this paper, a LoRaWAN based solution is presented, for manufacturing process monitoring, in order to improve energy management in plants. The authors present a minimally invasive device, with which many product manufacturing data can be obtained for analysis and further improvements. This paper is an extended variant of [14].

M3D aerosol jet printed buckypaper multifunctional sensors for composite structural health monitoring

Abstract Sensors placed along various stages of manufacturing processes can identify errors and ensure the quality of a product. Specially, sensors are becoming essential for more complicated composite fabrication due to coupling together heat transport, curing kinetics, resin rheology and flow as well as pressure effects. In this research, embedded composite sensor fabrication, maskless mesoscale materials deposition system has been used to print electrically conductive silver patterns on the carbon nanotube (CNT) sheets or buckypaper (BP) to measure its resistance change, and thus used as embedded sensors in composite laminates for both manufacturing process and structure health monitoring. Three different samples of nanotube sheets were used in this work and compared. The samples exhibited slightly different behaviors in terms of resistive response during the resin infusion and curing process with vacuum bagging, yet also demonstrated similar patterns, which indicated their sensing capabilities. Upon the resin flow, the sensor resistance dropped ∼20–30% from the initial value, then increased up to 120–145% compared to its minimum value at the final curing stage. After curing process, embedded sensors could monitor the impact or structural deformation of a composite part.

In Situ Additive Manufacturing Process Monitoring With an Acoustic Technique: Clustering Performance Evaluation Using K-Means Algorithm

Additive manufacturing (AM) is based on layer-by-layer addition of materials. It gives design flexibility and potential to decrease costs and manufacturing lead time. Because the AM process involves incremental deposition of materials, it provides unique opportunities to investigate the material quality as it is deposited. Development of in situ monitoring methodologies is a vital part of the assessment of process performance and understanding of defects formation. In situ process monitoring provides the capability for early detection of process faults and defects. Due to the sensitivity of AM processes to different factors such as laser and material properties, any changes in aspects of the process can potentially have an impact on the part quality. As a result, in-process monitoring of AM is crucial to assure the quality, integrity, and safety of AM parts. There are various sensors and techniques that have been used for in situ process monitoring. In this work, acoustic signatures were used for in situ monitoring of the metal direct energy deposition (DED) AM process operating under different process conditions. Correlations were demonstrated between metrics and various process conditions. Demonstrated correlation between the acoustic signatures and the manufacturing process conditions shows the capability of acoustic technique for in situ monitoring of the additive manufacturing process. To identify the different process conditions, a new approach of K-means statistical clustering algorithm is used for the classification of different process conditions, and quantitative evaluation of the classification performance in terms of cohesion and isolation of the clusters. The identified acoustic signatures, quantitative clustering approach, and the achieved classification efficiency demonstrate potential for use in in situ acoustic monitoring and quality control for the additive manufacturing process.

An efficient transient temperature monitoring of fused filament fabrication process with physics-based compressive sensing

Sensors play an important role in manufacturing processes. Different types of sensors have been used in process monitoring to ensure the quality of products. As a result, the cost of quality control is rising. Processing a large amount of sensor data for real-time process monitoring is also challenging. Recently, a Physics-Based Compressive Sensing (PBCS) approach was proposed to reduce the number of sensors and the amount of data collection associated with manufacturing process monitoring. PBCS significantly improves the compression ratio from traditional compressed sensing by incorporating the knowledge of physical phenomena in specific applications. In this article the PBCS approach is demonstrated with the dynamic process of fused filament fabrication where the constantly changing temperature field needs to be continuously monitored. A transient thermal model for PBCS is formulated. Based on the model, three-dimensional thermal distributions in manufacturing processes can be efficiently monitored by reconstructing distributions from sparse samplings in both spatial and temporal domains. The systematic error from reconstruction can also be predicted and compensated based on a Gaussian process uncertainty quantification approach.

Failure Monitoring and Recovery System During Manufacturing Process

Originally designed for industrial use, numerical control (CNC) machines have been increasingly present incommercial and residential environments, especially due to the popularization of low-cost 3D printers and machines with a similar control system, such as small laser cutting machines. Most of such machines are controlled by G-codes, however, open, simple and generic tools for remote control, monitoring and open management of G-codes sent to them are rare. In the context of Industry 4.0, including remote operation and monitoring of such machines, their use has become fundamental. This article introduces an open and generic CNC device control software platform that manages and controls the machine via remote web, from anywhere, with any device (computer or smartphone). The system involves failure detection during the manufacturing process and remote monitoring of sensors and enables the pausing or continuation of manufacturing or restarting processes from the point of interruption. The article addresses the systems characteristics, its implementation and the experimental results from its use with two 3D printers.

CEP ONLINE: A WEB-ORIENTED EXPERT SYSTEM FOR STATISTICAL PROCESS CONTROL

ABSTRACT In this paper, a new software for Statistical Process Control (SPC) is proposed. The system, the so-called CEP Online, was developed based on statistical computing resources of well-known free softwares, such as HTML, PHP, R and MySQL under an online server with operating system Linux Ubuntu. The main uni and multivariate SPC tools are available for monitoring and evaluation of manufacturing and non-manufacturing production processes over time. Some advantages of the new software are: (i) low operational cost, since it is cloud-based, only needing a computer connected to the Internet; (ii) easy to use with great interaction with the user; (iii) it does not require investment in any specific hardware or software; (iv) real time reports generation on process condition monitoring and process capability. Thus, the CEP Online offers for SPC practitioners fast, efficient and accurate SPC procedures. Therefore, CEP Online becomes an important resource for those who have no access to non-free softwares, such as SAS, SPSS, Minitab and STATISTICA. To the best of our knowledge, the CEP Online is unique with respect to its characteristics.

Автоматизация измерения неравномерности магнитной проводимости зубцовой зоны электрических микромашин

A technique for measuring the permeance nonuniformity in rotating transformers based on analyzing the harmonic composition of their permeance is presented. For estimating the permeance nonuniformity, it is proposed to use its influence on the spatial change of the magnetic flux vector position. The suggested technique can be used for diagnosing the operation of not only rotating transformers and selsyns, but also other AC electrical machines. The use of the proposed permeance nonuniforrmity measurement method allows the measurement to be made within a shorter period of time and yields a real permeance variation law. Two schematic designs of experimental benches for implementing the developed technique are proposed, and the bench equipment structure and bench operation principle are described. By using the experimental bench equipment it is possible to obtain extensive data characterizing spectral nonuniformity of the permeance of prototypes and to estimate the accuracy characteristics of rotating transformers and selsyns. A technological bench made according to a simpler design makes it possible to carry out prompt checks of permeance nonuniformity during manufacture of rotating transformers and selsyns, and in periodically performing their calibration checks. The experimentally obtained oscillograms are presented. The external appearance of the obtained graph allows a judgment to be made about the level of rotational pulsations caused by the rotating transformer magnetic circuit anisotropy and the level of pulsations determined by the tooth zone permeance nonuniformity. The use of the proposed technique makes it possible to simplify the permeance nonuniformity measurement process, to visualize it, and to automate technological monitoring of the electrical machinery manufacturing process.

An Application of 3D Model Reconstruction and Augmented Reality for Real-Time Monitoring of Additive Manufacturing

This paper presents a novel scan-based method for the real-time monitoring of additive manufacturing processes. Most traditional scanning techniques used for generating 3D models capture the only outer shape of the object after completion of the printing process. The method proposed in this paper differs as it relies on a layer-by-layer scanning of the 3D object directly during the printing process. This strategy has been successfully implemented with a fused filament 3D printer (PRUSA i3 MK3). Furthermore, in order to offer an increased interaction between the obtained 3D model and the user, a virtual environment has been developed for the augmented reality glasses HoloLens. The novelty of this method lies in the layer-by-layer 3D model reconstruction of both the outer shape and the inner layers of the printed part. It enables the user, directly during the printing process, to view and detect potential defects, not only at the surface but also in the inner layers of the printed object. Therefore, it can provide detailed information about the build quality and can be used as the basis of a decision-making tool.