On-line Process Monitoring Research Articles

Online monitoring of batch processes combining subspace design of latent variables with support vector data description

Online monitoring of batch processes combining subspace design of latent variables with support vector data description

Plasma optical emission spectroscopy based on feedforward neural network

Optical emission spectroscopy (OES) has been widely applied to plasma etching, material processing, development of plasma equipment and technology, as well as plasma propulsion. The collisional-radiative model used in OES is affected by the deviation of fundamental data such as collision cross sections, thus leading to the error in diagnostic results. In this work, a novel method is developed based on feedforward neural network for OES. By comparing the error characteristics of the new method with those of the traditional least-square diagnostic method, it is found that the neural network diagnosis method can reduce the transmission of basic data deviation to the diagnosis results by identifying the characteristics of the spectral vector. This is confirmed by the experimental results. Finally, the mechanism of the neural network algorithm against fundamental data deviation is analyzed. This method also has a good application prospect in plasma parameter online monitoring, imaging monitoring and mass data processing.

Monitoring and Diagnosis of Multistage Manufacturing Processes Using Hierarchical Bayesian Networks

In recent years, manufacturing systems have given rise to manufacturing big data due to the rapid developments in sensor and information technology and that has fueled data-driven research techniques towards addressing the issues in multistage quality control and diagnosis. In this paper, a unified framework with dual Hierarchical Bayesian Networks (HBNs) has been presented for simultaneous online process monitoring and fault diagnosis of a multistage manufacturing system. To achieve this, a novel AMDS (Absolute Mean Deviation of States) control chart has been developed for monitoring the unobserved inputs. The AMDS control chart is built on the AMDS statistic, which is calculated using the inferred states distribution generated utilizing the HBNs of the unobserved inputs. Discrete event simulation results of the two-stage process demonstrate that the methodology can successfully detect process changes and diagnose the root causes of the change. In addition, it can also identify the time at which the fault has occurred and the type (mean shift or variance change) and nature (step faults or slow drifts) of the change. The robustness of the proposed methodology is extensively tested against multiple randomly generated non-linear quadratic process models for two-stage systems.

Self-Detecting the Measurement Error of Electronic Voltage Transformer Based on Principal Component Analysis-Wavelet Packet Decomposition

The poor long-term stability of measurement error is the main problem in the operation of an electronic voltage transformer (EVT). Prevalent methods of error detection involve calibrating it regularly with a standard voltage transformer, but this is not conducive to the timely detection of the degradation in the error of the EVT. In light of this, this paper proposes a method to self-detect error based on principal component-wavelet packet decomposition. A standard of comparison for error detection is first established based on the characteristics of three-phase symmetrical operation. Principal component analysis is then applied to the measurement data of the three-phase EVTs, and their error is mapped to the online process monitoring of squared prediction error (SPE) statistics. Furthermore, the SPE statistics are decomposed by a three-level wavelet packet to eliminate the influence of the time-varying characteristics of node imbalance on the results of detection. The results of experiments show that the proposed method can detect the measurement error with an accuracy of 0.2 class without a standard.

Use of a Carbon Dioxide Measurement System to Control the Process of Obtaining Breathing Air for Hyperbaric Oxygen Conditions

Abstract Maintaining a stable carbon dioxide content in the process of producing breathing air is important both for the safety of divers performing underwater work and for avoiding financial losses resulting from poor product quality. This paper deals with the implementation of safety measurement systems for online control of the breathing air production process. On the basis of a qualified control system, the capability of the rationalised process was assessed, identifying both its current status and its potential for improvement in terms of eliminating defects caused by excessive carbon dioxide content. For process reasons, the effectiveness of online process monitoring was evaluated against the previously used periodic sample control by means of laboratory methods of instrumental analysis. The analysis was conducted at KTPP AMW1 for the compressed air supply system of the DGKN - 1202 complex.

Reconfiguration modulation strategy analysis for proactive adaptability of the monitored inverter to on-line monitoring system of IGBT modules

To realize on-line monitoring for insulated-gate bipolar transistor (IGBT) modules in the three-level neutral point clamped (NPC) inverters, a reconfiguration modulation strategy is proposed. The saturated voltage drop (VCE) is one of the most common diagnostic signs of IGBTs’ healthy status due to the advantages of monitoring method such as simple theory, mature implementation, and accurate measurement, and the method to extract VCE in the monitoring system has been researched in the previous study. On this basis, considering realizing on-line monitoring process, essential supported hardware and proactive on-line monitoring environment are needed to guarantee the safe and friendly on-line monitoring implementation. Analysis has been made to show that the on-line monitoring system is able to work regularly by just making use of the inherent O state of a three-level NPC inverter in some circumstances. But to guarantee the monitoring process and the safety of the monitored NPC inverter, the novel reconfiguration strategy is a kind of necessary and essential means to create an on-line monitoring environment, and it is proposed and researched in this paper. Through theoretical derivation, the optimization of reconfiguration strategy is obtained. The reconfiguration strategy proposed in this research has been achieved and its validity is proved based on experimental results.

Application of Near-Infrared Spectroscopy Analysis Technology to Total Nucleosides Quality Control in the Fermented Cordyceps Powder Production Process.

Product quality control is a prerequisite for ensuring safety, effectiveness, and stability. However, because of the different strain species and fermentation processes, there was a significant difference in quality. As a result, they should be clearly distinguished in clinical use. Among them, the fermentation process is critical to achieving consistent product quality. This study aims to introduce near-infrared spectroscopy analysis technology into the production process of fermented Cordyceps powder, including strain culture, strain passage, strain fermentation, strain filtration, strain drying, strain pulverizing, and strain mixing. First, high performance liquid chromatography (HPLC) was used to measure the total nucleosides content in the production process of 30 batches of fermented Cordyceps powder, including uracil, uridine, adenine, guanosine, adenosine, and the process stability and interbatch consistency were analyzed with traditional Chinese medicine (TCM) fingerprinting, followed by the near-infrared spectroscopy (NIRS) combined with partial least squares regression (PLSR) to establish a quantitative analysis model of total nucleosides for online process monitoring of fermented Cordyceps powder preparation products. The model parameters indicate that the established model with good robustness and high measurement precision. It further clarifies that the model can be used for online process monitoring of fermented Cordyceps powder preparation products.

Improved conversion of residual MSW biomass waste to sugars using online process monitoring and integrated contamination control

Enzymatic saccharification of lignocellulosic biomass to soluble sugars is vulnerable to microbial contamination. Identification of early stage indicators of contamination could allow for more rapid intervention and improved sugar retention. Here we validate the use of dissolved oxygen (DO) and pH as early-stage indicators, and used to evaluate the effectiveness of several antimicrobial agents (sodium azide, benzisothiazolinone, hydrogen peroxide and tetracycline). It was found that benzisothiazolinone (BIT) and sodium azide performed most favourably (8.93 and 9.89 g glucose/L, respectively), controlling contamination for the entire 24 hour time course assessed. We then describe the development of a novel integrated control system with online monitoring of DO/pH and automatic administration of discrete doses of antimicrobial agent to the reaction once environmental DO dropped below a pre-defined threshold. Further, we show that FTIR can be used for continuous real-time glucose quantification during enzyme hydrolysis as an alternative to traditional offline measurement methods.

Monitoring Volatility Change for Time Series Based on Support Vector Regression.

This paper considers monitoring an anomaly from sequentially observed time series with heteroscedastic conditional volatilities based on the cumulative sum (CUSUM) method combined with support vector regression (SVR). The proposed online monitoring process is designed to detect a significant change in volatility of financial time series. The tuning parameters are optimally chosen using particle swarm optimization (PSO). We conduct Monte Carlo simulation experiments to illustrate the validity of the proposed method. A real data analysis with the S&P 500 index, Korea Composite Stock Price Index (KOSPI), and the stock price of Microsoft Corporation is presented to demonstrate the versatility of our model.

Development of chemometric models using Vis-NIR and Raman spectral data fusion for assessment of infant formula storage temperature and time

This study evaluated the potential of Vis-NIR and Raman spectral data fusion combined with PLS and SVM chemometric models developed using a large dataset (n = 1700) of commercial infant formula (IF) samples to (i) discriminate between different IF storage temperature (20, 37 °C) and (ii) predict IF storage time (0–12 months). Three interval-based PLS variable selection methods (forward interval PLS (FiPLS), backward interval PLS (BiPLS) and synergy interval PLS (SiPLS)) and SVM-recursive feature elimination (SVM-RFE) methods were compared for model development. The best IF storage temperature discrimination model was developed using SVM classification (SVMC) and Vis-NIR spectra (400–2498 nm) (AccuracyCV = 99.82%, AccuracyP = 100%). SVM regression (SVMR) models developed using medium level data fusion (features selected by SVM-RFE) had the lowest root mean square error (RMSE) values for IF samples stored at either temperature, 20 °C or 37 °C (RMSECV = 0.7–0.8, RMSEP = 0.6–0.9). Industrial relevanceSpectroscopic technologies, including Vis-NIR and Raman spectroscopy have been widely applied for process analysis and increasingly for on-line process monitoring in areas of chemicals, food processing, agriculture and pharmaceuticals, etc. Due to their rapid measurement and minimal or no sample preparation, they are highly suitable for in-line process monitoring. This study demonstrates that Vis-NIR and Raman process analytical tools either individually or combined may be employed for quality assessment and process control of IF manufacture.

Acid number, viscosity and end-point detection in a multiphase high temperature polymerisation process using an online miniaturised MEMS Fabry-Pérot interferometer

Recent advances in the latest generation of MEMS (micro-electro-mechanical system) Fabry-Pérot interferometers (FPI) for near infrared (NIR) wavelengths has led to the development of ultra-fast and low cost NIR sensors with potential to be used by the process industry. One of these miniaturised sensors operating from 1350 to 1650 nm, was integrated into a software platform to monitor a multiphase solid-gas-liquid process, for the production of saturated polyester resins. Twelve batches were run in a 2 L reactor mimicking industrial conditions (24 h process, with temperatures ranging from 220 to 240 °C), using an immersion NIR transmission probe. Because of the multiphase nature of the reaction, strong interference produced by process disturbances such as temperature variations and the presence of solid particles and bubbles in the online spectra required robust pre-processing algorithms and a good long-term stability of the probe. These allowed partial least squares (PLS) regression models to be built for the key analytical parameters acid number and viscosity. In parallel, spectra were also used to build an end-point detection model based on principal component analysis (PCA) for multivariate statistical process control (MSPC). The novel MEMS-FPI sensor combined with robust chemometric analysis proved to be a suitable and affordable alternative for online process monitoring, contributing to sustainability in the process industry.

Acoustic emission and machine learning for in situ monitoring of a gold–copper ore weakening by electric pulse

The excessive energy consumption from the mining industry are currently receiving international attention. A promising method able to enhance significantly the comminution process efficiency worldwide is by using electric pulse fragmentation treatment. However, to insure a minimum energy consumption in real scale operation, an online process monitoring is of utmost importance. This work presents an in situ and real-time monitoring method by combining acoustic emission sensor and advanced machine learning algorithms. The proposed method was developed on a gold-copper ore in well-controlled single stone experiments and in semi-continuous process, reproducing a real industrial environment. In single stone experiment, the pulse energies was varied from 200 to 750 J leading to three weakening behaviours; no discharge, surface discharge and fragmentation. Acoustic signals for these categories have been decomposed with wavelet packets, and sub-band energies have been chosen as features. Then, only the most informative features were selected via standard linear principal component analysis. Finally, the classification was performed via a traditional support vector machine. In the semi-continuous experiments, an unsupervised learning method was used for classification task based on Laplacian support vector machine. Results for single stone tests showed accuracy above 90% for the three categories. For semi-continuous tests, we demonstrated that the unsupervised classification can be applied efficiently to estimate the amount of weakening of the passed through ore. We are very confident that the proposed method can be easily industrialised to monitor in situ and in real-time the electric discharge process within a comminution operation.

Screening of brominated pyrolysis products of tetrabromobisphenol a by integrating controllable heating device with ambient mass spectrometry

Pyrolysis, an effective recycle strategy for resins or plastic containing tetrabromobisphenol A (TBBPA), often lead to formation of more toxic decomposition chemicals. Online collection coupled with off-line analysis for qualitative and quantitative analysis of the residue often induce information loss of these pyrolysis products. In this study, a temperature controllable thermal decomposition device coupled with an ambient mass spectrometry was directly employed for on-line monitoring of the pyrolysis process of TBBPA. Thermal products through debromination, methylation, methoxylation and β-scission have all been detected. Based on the mass spectra of TBBPA decomposition products, the number of bromine atoms and the m/z values, about 8 brominated decomposition products/intermediates have been identified. Temperature played the critical role for the pyrolysis processes. It was found that almost no TBBPA degraded at low temperature (100−150 °C), intensified decomposition happened at about 300 °C and small molecules formed in large quantities at > 450 °C.

A predictive Bayesian approach to EWMA and CUSUM charts for time-between-events monitoring

ABSTRACT This article introduces Bayesian predictive monitoring of time-between-events using Cumulative Sum (CUSUM) and Exponentially Weighted Moving Average (EWMA) control charts with predictive control limits. It is shown that the proposed methodology not only overcomes the requirement of a large Phase-I data set to establish control limits, but also feasible for online process monitoring. In addition to Bayesian memory-type charts with dynamic control limits, a comparison of the frequentist sequential charts, designed by using the unbiased and biased estimator of the process parameter, is also given in this article. For the performance evaluation of the predictive TBE chart in the presence of practitioner-to-practitioner variability, we use the average of the in-control average run length (AARL) and the standard deviation of the in-control run length (SDARL).

Heating behaviour of plasma sprayed TiOx/Cr2O3 coatings for injection moulding

Mould tempering in injection moulding influences part quality as well as process efficiency. Currently established systems are either slow or require an additional process step. A thermally sprayed heating coating on the tool cavity surface enables temperature control in close proximity of the polymer melt.In this study, TiOx/Cr2O3 was used as heating coating covered with a protective coating of Al2O3. The coatings were applied by atmospheric plasma spraying using a three cathode spray gun. The heating behaviour on the surface as well as the phase stability and the thermal durability of the coating system have been investigated. Furthermore, resistance measurements of the TiOx/Cr2O3 coating at different temperatures have been conducted as a negative temperature coefficient in semiconductive materials may enable an additional utilization for precise online process monitoring.

Online monitoring and control of a cyber-physical manufacturing process under uncertainty.

Recent technological advancements in computing, sensing and communication have led to the development of cyber-physical manufacturing processes, where a computing subsystem monitors the manufacturing process performance in real-time by analyzing sensor data and implements the necessary control to improve the product quality. This paper develops a predictive control framework where control actions are implemented after predicting the state of the manufacturing process or product quality at a future time using process models. In a cyber-physical manufacturing process, the product quality predictions may be affected by uncertainty sources from the computing subsystem (resource and communication uncertainty), manufacturing process (input uncertainty, process variability and modeling errors), and sensors (measurement uncertainty). In addition, due to the continuous interactions between the computing subsystem and the manufacturing process, these uncertainty sources may aggregate and compound over time. In some cases, some process parameters needed for model predictions may not be precisely known and may need to be derived from real time sensor data. This paper develops a dynamic Bayesian network approach, which enables the aggregation of multiple uncertainty sources, parameter estimation and robust prediction for online control. As the number of process parameters increase, their estimation using sensor data in real-time can be computationally expensive. To facilitate real-time analysis, variance-based global sensitivity analysis is used for dimension reduction. The proposed methodology of online monitoring and control under uncertainty, and dimension reduction, are illustrated for a cyber-physical turning process.

Quality Monitoring for Micro Resistance Spot Welding with Class-Imbalanced Data Based on Anomaly Detection

Micro resistance spot welding (MRSW) is an important technology widely used in electronics manufacturing for micro component joining. For the joining of micro enameled wire, quality control is heavily dependent on manual inspection till now. In this paper, a quality monitoring approach based on isolation forest (iForest) is proposed to identify abnormal welds and normal welds. Electrode voltage and welding current of over 110,000 spot welds were collected from a production line. The dynamic resistance and heat input were calculated for all welds and used for feature extraction. A class imbalance problem existed in the collected dataset because abnormal welds were far fewer than normal welds. The anomaly detection model based on iForest was established for the imbalanced data classification after comparison with other methods such as one-class (support vector machine) SVM and local outlier factor. Test results show that the similarity of dynamic resistance profile and heat input compared with the previous ten welds are valid features for detecting a part of the abnormal welds. The iForest model is effective for distinguishing incomplete fusion welds from normal welds with high efficiency. It can assist in the on-line quality monitoring of enameled wire welding process in production.

Soft-Sensing Estimation of Optical Density for PHA Production Using Multilayer Perceptron Neural Network

Biomass concentration is an important indicator of production rate in polyhydroxyalkanoates (PHA) fermentation process. In current practice, measurement of biomass concentration is done off-line by laboratory analysis that is unsuitable for online process monitoring and control. Soft-sensor is often used as an alternative that provides an estimate of hard to measure parameters from easy to measure process data. However, most of these studies use simulated data or data generated from mathematical model that was developed without full consideration of process and measurement uncertainty. In this study, a soft-sensor is developed from real production data for PHA fermentation in pilot-scale bioreactor with the appropriate data pre-processing techniques applied to process data that was obtained from this system. Multilayer perceptron (MLP) neural network is used to estimate biomass concentration using secondary process parameters such as dissolved oxygen (DO), temperature, pH and agitation speed as inputs. Different models are developed based on different batches of production data and various network architecture in order to study the appropriate integration of process data and network topology that gives the best model accuracy. Results indicate that the biomass soft-sensor developed using MLP-ANN provides a better estimate of biomass in comparison to radial basis function (RBF) neural network and support vector regression (SVR) methods. The developed soft-sensor can be further used in monitoring and control of production output.

Advanced Process Monitoring in Additive Manufacturing

AbstractLaser‐based additive manufacturing processes like laser metal deposition (LMD) and laser powder bed fusion (LPBF) are becoming more and more popular in industry. In demanding applications, however, every part has to be tested non‐destructively in a subsequent step. Online process monitoring can provide detailed documentation for each part and has the potential to reduce the post‐process inspection costs. BIAS has developed an advanced monitoring approach for in‐process temperature field measurement using a coaxial two‐channel pyrometric camera.

Online Process Monitoring Based on Vibration-Surface Quality Map for Robotic Grinding

In robotic manufacturing, the machining vibration can easily affect the quality of the processed surfaces due to poor stiffness and variable stiffness of the robot arm. It is important to monitor the processing quality. Therefore, an online process monitor method for robotic grinding is proposed in this article. Since the surface quality is difficult to be measured directly online, a vibration-surface quality map is proposed to show the mapping relationship between the vibration signals and the surface images. Based on this map, the grinding states are classified to lack, normal, and overgrinding, which formed the training set data. Then a convolutional neural network classifier is trained to detect the grinding states during robotic machining only with the vibration signals. To establish the vibration-surface quality map, the symmetrized dot pattern and sparse autoencoder methods are utilized to extract the features from the raw vibrations and images data, and a gray-level co-occurrence matrix index is further used to build the bridge between them. The experimental results in this article have proved the effectiveness and potential of our method for online robotic grinding quality monitoring.