Multi-sensor Monitoring System Research Articles

Multivariate data analysis on multisensor measurement for inline process monitoring of adenovirus production in HEK293 cells.

In the era of Biopharma 4.0, process digitalization fundamentally requires accurate and timely monitoring of critical process parameters (CPPs) and quality attributes. Bioreactor systems are equipped with a variety of sensors to ensure process robustness and product quality. However, during the biphasic production of viral vectors or replication-competent viruses for gene and cell therapies and vaccination, current monitoring techniques relying on a single working sensor can be affected by the physiological state change of the cells due to infection/transduction/transfection step required to initiate production. To address this limitation, a multisensor (MS) monitoring system, which includes dual-wavelength fluorescence spectroscopy, dielectric signals, and a set of CPPs, such as oxygen uptake rate and pH control outputs, was employed to monitor the upstream process of adenovirus production in HEK293 cells in bioreactor. This system successfully identified characteristic responses to infection by comparing variations in these signals, and the correlation between signals and target critical variables was analyzed mechanistically and statistically. The predictive performance of several target CPPs using different multivariate data analysis (MVDA) methods on data from a single sensor/source or fused from multiple sensors were compared. An MS regression model can accurately predict viable cell density with a relative root mean squared error (rRMSE) as low as 8.3% regardless of the changes occurring over the infection phase. This is a significant improvement over the 12% rRMSE achieved with models based on a single source. The MS models also provide the best predictions for glucose, glutamine, lactate, and ammonium. These results demonstrate the potential of using MVDA on MS systems as a real-time monitoring approach for biphasic bioproduction processes. Yet, models based solely on the multiplicity and timing of infection outperformed both single-sensor and MS models, emphasizing the need for a deeper mechanistic understanding in virus production prediction.

Online Detection of Laser Welding Penetration Depth Based on Multi-Sensor Features.

The accurate online detection of laser welding penetration depth has been a critical problem to which the industry has paid the most attention. Aiming at the laser welding process of TC4 titanium alloy, a multi-sensor monitoring system that obtained the keyhole/molten pool images and laser-induced plasma spectrum was built. The influences of laser power on the keyhole/molten pool morphologies and plasma thermo-mechanical characteristics were investigated. The results showed that there were significant correlations among the variations of the keyhole-molten pool, plasma spectrum, and penetration depth. The image features and spectral features were extracted by image processing and dimension-reduction methods, respectively. Moreover, several penetration depth prediction models based on single-sensor features and multi-sensor features were established. The mean square error of the neural network model built by multi-sensor features was 0.0162, which was smaller than that of the model built by single-sensor features. The established high-precision model provided a theoretical basis for real-time feedback control of the penetration depth in the laser welding process.

Modified Fusing-and-Filling Generative Adversarial Network–based few-shot image generation for GMAW defect detection using multi-sensor monitoring system

Modified Fusing-and-Filling Generative Adversarial Network–based few-shot image generation for GMAW defect detection using multi-sensor monitoring system

Hierarchical Deep Reinforcement Learning for Self-Powered Monitoring and Communication Integrated System in High-Speed Railway Networks

To align with the vision of future intelligent high-speed railway (HSR) networks, integrating sensor monitoring and remote communication are challenging for ensuring the lightweight of train equipment, high-quality transmissions, and dynamic interaction between monitoring and communication. In this paper, we propose a self-powered multisensor monitoring and communication integrated system in HSR. A low-power backscatter communication working framework of the self-powered monitoring system is designed in the monitoring network model, and a finite Gaussian mixture model (GMM) clustering method is used to analyze the communication cell coverage area in the communication network model. Aiming to minimize the total task completion time, we formulate a data monitoring and remote communication problem with the energy transfer constraint, data collection constraint, and transmission data rate constraint. As for the non-convex minimum time optimization problem, we develop a novel option-based hierarchical deep reinforcement learning (OHDRL) method to deal with the complex continuous variation characteristics of the monitoring and communication integrated HSR system. The system learns to select options at a high level, and the action is executed according to the policy of the selected option at a low level. This approach enables us to handle stochastic HSR environments, closed-loop policies, and goals in a temporal abstraction way. Numerical results reveal that the proposed algorithm for the integrated monitoring and communication HSR achieves a significantly higher reward and more stable learning performance than other algorithms in the literature.

Coaxial and synchronous monitoring of molten pool height, area, and temperature in laser metal deposition

Laser Metal Deposition is an additive manufacturing technology that enables to realize of one-of-a-kind large components with a free-form geometry. However, the advantages of laser metal deposition from the economic and sustainability points of view can be attenuated by the presence of defects in the realized components. The process should be real-time monitored to immediately identify the onset of defects or process drifts that may undermine the outcome of the build, also enabling the possibility to immediately act against them with feedback control. Many of the proposed solutions in monitoring the Laser Metal Deposition process exploit the use of only one sensor or more but decoupled ones, which does not provide a comprehensive overview of the process. In this work, a coaxial multi-sensor monitoring system is proposed and exploited to capture the variation of three different process signatures. A custom laser triangulator, a near-infrared camera and a ratio pyrometer for simultaneous monitoring of the molten pool height, molten pool area, and molten pool temperature, respectively, are coaxially integrated. Thanks to the possibility to extract concurrent (in spatial and time domains) signals directly from the workpiece area and in real-time, a comprehensive status of the deposition can be deduced. The influence of the process parameters on the sensed outputs is investigated, and the relationships between the process signatures can be researched. While the molten pool area and its temperature are mostly affected by the laser power, the scan speed is more impactful on the molten pool height. A good linear correlation between the molten pool temperature and the area is assessed, while no correlation with the molten pool height is found, making it almost independently manipulatable.

A bacteriorhodopsin multisensor system for qualitative and quantitative monitoring of methanol, ethanol, propanol, and butanol under extreme conditions.

One of the most serious problems in waste biodegradation and biofuel production is the lack of adequate systems for monitoring reaction media. It has been demonstrated that the bacteriorhodopsin of Halobacterium salinarum is capable of generating photoelectric signals that can be modulated as a function of a chemical environment containing ethanol, methanol, propanol or butanol. The chemical modification of retinal (proton substitution with a fluorine atom at the 10, 12, or 14 position) and genetic modification of protein (aspartic acid 96 substituted with asparagine) may enhance the responses of bacteriorhodopsin systems. The responses of single elements to alcohols form characteristic response patterns. These patterns constitute the basis for the construction of the biosensor, a bacteriorhodopsin multisensor system equipped with artificial neural network methodology for monitoring these alcohols under extreme environmental conditions such as high or low pH and high temperature. It is, to the author's knowledge, the first time that the application of a constructed biosensor for monitoring thermophilic (55 °C) production of ethanol during paper and pulp wastewater degradation and thermophilic (55 °C) methanol digestion in methanol-rich wastewater from pulp and paper factories has been presented.

Towards Online Health Monitoring of Robotic Arm

Robot arms exhibit complex dynamic behaviours as their joints move at different angular speeds, acceleration, torques, and rotation at various angles. These operations differ from that of rotating machines, which often move at fixed continuous speeds. Since the majority of health monitoring strategies have been designed for the latter, it is a challenge to develop a reliable and intelligent health monitoring system for robots that addresses the non-stationary nature of their signals; often requiring synergy of instrumentation, analytical and information technologies with knowledge and experience in design, operation and maintenance. This article presents preliminary findings regarding the extraction of crucial components from data obtained from an industrial robot arm, with the ultimate goal of designing a multi-sensor measurement system for online health monitoring. This approach serves as an alternative to the conventional method that relies on vibration signal analysis for detecting anomalies and predicting remaining useful life (RUL), when integrated with machine learning techniques. The primary aim of the proposed system is the online identification of operational anomalies, deterioration, or damage that may adversely affect the arm's reliability and safety. To achieve this goal, the measurement system can employ wavelet analysis and decision trees to accurately track each joint of the robot arm during operation.

Analysis of temperature characteristics for overwintering bee colonies based on long-term monitoring data

Honey Bees are very important to ecological environment and human society. However, the collapse of bee colonies in winter has frequently occurred in recent years, so condition monitoring of bee colonies is vital to maintain bee colonies health. Considering that the temperature has the strongest ability of recognizing the status of bee colonies, this paper uses a multi-sensor system to investigate the long-term air temperature inside beehives for overwintering bee colonies. Firstly, different from other studies in which it is usually a single temperature observation or a small number of samples and a short observation time, the multi-sensor monitoring system is adopted to carry out a continuous experiment for more than 30 Italian bee colonies. The experiment time covers two complete overwintering periods. Subsequently, considering the impact of different time scales on the monitoring of bee colony status, the long-term (several months) data analysis is based on the daily average data, while the short-term (typical weeks) data analysis is performed on the hourly data. Finally, the correlation between air temperature and other monitoring parameters (entrance counts, humidity, sound) was computed. The results demonstrate that the air temperature change inside the beehive has an evident regularity during the overwintering period, such as the fixed temperature threshold for the outdoor activities of bees, periodic fluctuation by the hour and the smallest temperature difference between inside and outside. These findings will enable the assessment of the health state of overwintering bee colonies by use of temperature-derived indicators.

On multi-sensor monitoring of fiber laser fusion cutting

Laser cutting is a well-established industrial process for sheet metal applications. However, cutting thick plates is still accompanied by problems because of the characteristic limited process parameter window. Since cutting by means of fiber lasers has become dominant, tailored solutions are required in such systems for industrial applications. The development of a robust real-time monitoring system, which adapts the process parameters according to a specific quality requirement, implies a significant step forward towards automated laser cutting and increases the process robustness and performance. In this work, a coaxial multi-sensor monitoring system is tested for fiber laser cutting of stainless steel thick plates. A high-speed camera and a photodiode sensor have been selected for this investigation. Experiments at different cutting speeds, representing primary cut quality cases, have been conducted and various features of the obtained process zone signals have been examined. Finally, the feasibility of industrial application of the developed setup for high-power fiber laser cutting is discussed, followed by several implementation recommendations.

Temperature Prediction of Photovoltaic Panels Based on Support Vector Machine with Pigeon-Inspired Optimization

The output stability of the photovoltaic (PV) system is directly affected by temperature change of PV panels. In this paper, a novel temperature prediction method of PV panels with support vector machine (SVM) is proposed, which can solve the temperature prediction problem in a complex environment. In order to optimize parameters of SVM, a Pigeon-Inspired Optimization (PIO) method is given. Meanwhile, the delay factor (DF) is added to improve the PIO algorithm for avoiding the problem of local optimum. Moreover, a multisensor monitoring system of PV is established, and the collected data of temperature are used to train and verify the accuracy of the model. Finally, the proposed method is evaluated using synthetic and actual data sets. Simulation results show that the DFPIO-SVM can obtain better predictive performance.

Generation of Vortex Optical Beams Based on Chiral Fiber-Optic Periodic Structures.

In this paper, we consider the process of fiber vortex modes generation using chiral periodic structures that include both chiral optical fibers and chiral (vortex) fiber Bragg gratings (ChFBGs). A generalized theoretical model of the ChFBG is developed including an arbitrary function of apodization and chirping, which provides a way to calculate gratings that generate vortex modes with a given state for the required frequency band and reflection coefficient. In addition, a matrix method for describing the ChFBG is proposed, based on the mathematical apparatus of the coupled modes theory and scattering matrices. Simulation modeling of the fiber structures considered is carried out. Chiral optical fibers maintaining optical vortex propagation are also described. It is also proposed to use chiral fiber-optic periodic structures as sensors of physical fields (temperature, strain, etc.), which can be applied to address multi-sensor monitoring systems due to a unique address parameter—the orbital angular momentum of optical radiation.

Open source arc analyzer: Multi-sensor monitoring of wire arc additive manufacturing

Low-cost high-resolution metal 3-D printing remains elusive for the scientific community. Low-cost gas metal arc wire (GMAW)-based 3-D printing enables wire arc additive manufacturing (WAAM) for near net shape applications, but has limited resolution due to the complexities of the arcing process. To begin to monitor and thus control these complexities, the initial designs of the open source GMAW 3-D printer have evolved to include current and voltage monitoring. Building on this prior work, in this study, the design, fabrication and use of the open source arc analyzer is described. The arc analyzer is a multi-sensor monitoring system for quantifying the processing during WAAM, which includes voltage, current, sound, light intensity, radio frequency, and temperature data outputs. The open source arc analyzer is tested here on aluminum WAAM by varying wire feed rate and measuring the resultant changes in the sensor data. Visual inspection and microstructural analysis of the printed samples looking for the presence of porosity are used as the physical indicators of quality. The value of the sensors was assessed and the most impactful sensors were found to be the light and radio frequency sensors, which showed arc extinction events and a characteristic “good weld” peak frequency.

Design of a Multi-sensor Monitoring System for Additive Manufacturing Process

Additive manufacturing (AM) technology is becoming increasingly feasible in industrial applications. Although AM systems continue to improve, the lack of repeatability and stability of these technologies remains an obstacle to industrial breakthroughs. Thus, AM process monitoring equipment has to be set up to supervise product quality and detect defects. This study first reviews the various categories of defects, relevant AM process signatures, and optical monitoring methods of powder bed fusion processes proposed so far. Then, according to the detection requirements, an optical process monitoring system based on multiple sensors is proposed, which is equipped with a white light imaging channel, an infrared imaging channel, and a polarization imaging channel. The aberrations can be optimized using Zemax software by appropriately selecting the optimization function operands. Thus, the design requirements for the multi-sensor monitoring system for the AM process can be achieved.

Artificial Neural Network–Based Multisensor Monitoring System for Collision Damage Assessment of Lithium‐Ion Battery Cells

The sales of electric vehicles (EVs) have seen a significant upward trend in recent years. The occurrence of severe collision accidents has raised awareness concerning safety issues of lithium‐ion batteries (LIBs) which are the core components of EV energy storage systems. When collision accidents of vehicles occur, in most cases, the battery cells will immediately burn and even explode, whereas in some cases, despite not directly showing visual damages, the battery cells may lead to delayed catastrophic failures. Herein, the assessment of the battery cells collision damage based on sensor signal data is focused on by identifying potentially unsafe cells. A campaign of collision experimental tests and a series of electrical performance tests are conducted on single‐cell specimens. A cell damage characterization procedure is proposed and implemented on the battery cells after the collision tests. The impact force and z‐axis acceleration signal features and respective damage classes are input to an artificial neural network (ANN) pattern classifier to train a model to assess the battery cell collision damage.

Design of a Wearable Wireless Multi-Sensor Monitoring System and Application for Activity Recognition Using Deep Learning

Human activity monitoring and recognition systems assist experts in evaluating various health problems including obesity, cardiac diseases and, sports injury detection. However, these systems have two challenging points; monitoring activities for outdoor applications and extracting relevant features using hand-crafted techniques from multi-dimensional and large datasets. To address these challenges, we have focused on new dataset generation for activity recognition, a novel design of a sensor-based wireless activity monitoring system, and its application to deep learning neural networks. The designed monitoring system consists of one master and four slave devices, and can collect and record acceleration and gyroscope information. The slave devices were attached on arm, chest, thigh, and shank areas of the human body. Activity data were collected and recorded from sixty healthy people for thirteen activity types including drink from cup and cleaning table. These activities were divided into three activity categories as basic, complex, and all, which is the combination of basic and complex activities. Obtained datasets were fed into deep learning neural networks namely convolutional neural network (CNN), long-short term memory (LSTM) neural networks, and convolutional LSTM (ConvLSTM) neural networks. The performance of each neural network for each category type was separately examined. The results show that ConvLSTM outperforms CNN and LSTM as far as activity recognition is concerned.

Online quantitative evaluation of expulsion in resistance spot welding

Weld expulsion is a universal problem during resistance spot welding (RSW) process and has potential damage to welding products and surrounding facilities. This study presents a new approach to the online evaluation of expulsion intensity using the amount of molten metal ejected during the expulsion. Through the newly developed multi-sensor monitoring and high-speed camera system, instantaneous features of process signals involving dynamic resistance, electrode force, and electrode displacement are simultaneously monitored and comparatively studied. The experimental results show that the amount of sudden decrease in the force and displacement signals is proportional to the weight of the expelled metal, but no apparent relationship is observed between the sudden resistance drop and the expulsion severity. Based upon further analysis, the abrupt change of electrode force is proven an optimal feature to precisely assess the magnitude of expulsion for a servo gun system. After calibration, the error of expulsion estimation could be limited to ±16.8 mg (±2σ) at a 95% confidence level, and the accuracy is robust to varying workpieces ranging from low carbon steel to high strength steel. This study may pave the way to online control the expulsion to the minimum level.

Implantable multi-sensor system for hemodynamic controlling

Abstract A miniaturized implantable multi-sensor system for cardiovascular monitoring of physiological parameters is presented. High accuracy pressure measurements within the vessel can be performed by a capacitive pressure sensor. Additional information about the patient, e. g., sudden movement, inclination or increased temperature can be obtained by additional sensor components such as an acceleration sensor and a temperature sensor unit. This information facilitates compensation of interferences for more accurate pressure measurements. A multi-functional ASIC enables, amongst others, sensor signal processing, power management and telemetric communication with extracorporeal electronics. Sensor chips, the multi-functional ASIC and passive components are assembled on a LTCC circuit board in which an antenna coil is integrated for telemetric energy and data transmission at a frequency of 13.56 MHz. In order to support further miniaturization, the implant shall be encapsulated with a stack of very thin and hermetic ceramics applied by ALD instead of using bulky metal housings. Further encapsulation with polymers, which can be functionalised with appropriate biomolecules, is necessary for a proper shape, a biocompatible interface to the surrounding tissue and, thereby, reduction of thrombogenicity.

Comprehensive Monitoring of Electrical Machine Parameters Using an Integrated Fiber Bragg Grating-Based Sensor System

In this paper, a multiparameter, multisensor system for comprehensive electrical machine condition monitoring has been developed and the results of an evaluation reported. The fiber Bragg grating-based system developed allows for the simultaneous monitoring of key parameters including machine vibration, rotor speed, torque, spin direction, temperature distribution along the stator windings and on the rotor surface as well as the stator wave frequency. This all-optical sensing solution has been designed to be compatible with being fitted in the tight confines of an electric motor and uses the optical nature of the measurement and the insulating nature of the sensor material to avoid problems of electrical interference. The system reduces the component count over conventional sensor systems, i.e., all sensing elements are contained within the machine and operated by a single sensing interrogation unit, thereby reducing cost and allowing for a convenient interface for the user. The design of the system is presented, as are results on the testing and evaluation of the device the sensing system has been successfully integrated into and tested on a permanent magnet motor prototype.

Aerobic Exercise Recognition Through Sparse Representation Over Learned Dictionary by Using Wearable Inertial Sensors

Aerobic exercise is conducive to reducing the risks of cardiovascular disease and central arterial stiffness. However, it can also cause some health hazards (such as tissue oxidative damage), especially for the elderly. It is essential to recognize and monitor different aerobic exercises for the health of exercisers. In this paper, a multi-sensor monitoring system is established for aerobic exercise recognition, and a novel recognition algorithm based on dictionary learning algorithm and sparse representation is proposed. Eight volunteers are invited to carry out ten activities, and five wireless inertial sensor nodes are used to collect the sensor data. Several experiments are implemented to verify the effectiveness of the recognition algorithm proposed in the paper. According to the experimental results, our method achieves the best performance than four other recognition algorithms including decision tree C4.5, naive Bayes, support vector machine and sparse representation. Besides, the other two aspects are also studied in the paper, one is the effect of different binding positions of sensors on classification results, and the other is the effect of selecting different features. The results of the experiments show that two sensor nodes attached to the right wrist and the left thigh achieve better result, and the feature “correlation coefficient” is not important to recognize different aerobic exercises that are investigated in our paper.

Design of Multi-sensor Monitoring System for Logistics Yard

With the development of e-commerce, the scale of logistics industry appears rapid growth, which brings great pressure on the operation for the logistics yard. How to ensure the order of operations, and the security of staffs and goods, are the most important and urgent. In this paper, a safety monitoring system embedded multiple sensors for logistics yard is designed, which consists of a management center and a plurality of terminal monitoring nodes. The management center is composed of group servers, which is in charge of analyzing and processing the various environmental data, communicating with each terminal monitoring node, and human–computer interaction. The terminal monitor is responsible for acquiring the monitoring data and transmitting that to the management center in real time. The system presented in this paper can effectively supervise the yard working environment and immediately alert warning signs when anomalies or dangers occur, to ensure the security of staffs and goods to the greatest extent.