Industrial Monitoring Research Articles

On the Stability and Homogeneous Ensemble of Feature Selection for Predictive Maintenance: A Classification Application for Tool Condition Monitoring in Milling.

Feature selection (FS) represents an essential step for many machine learning-based predictive maintenance (PdM) applications, including various industrial processes, components, and monitoring tasks. The selected features not only serve as inputs to the learning models but also can influence further decisions and analysis, e.g., sensor selection and understandability of the PdM system. Hence, before deploying the PdM system, it is crucial to examine the reproducibility and robustness of the selected features under variations in the input data. This is particularly critical for real-world datasets with a low sample-to-dimension ratio (SDR). However, to the best of our knowledge, stability of the FS methods under data variations has not been considered yet in the field of PdM. This paper addresses this issue with an application to tool condition monitoring in milling, where classifiers based on support vector machines and random forest were employed. We used a five-fold cross-validation to evaluate three popular filter-based FS methods, namely Fisher score, minimum redundancy maximum relevance (mRMR), and ReliefF, in terms of both stability and macro-F1. Further, for each method, we investigated the impact of the homogeneous FS ensemble on both performance indicators. To gain broad insights, we used four (2:2) milling datasets obtained from our experiments and NASA's repository, which differ in the operating conditions, sensors, SDR, number of classes, etc. For each dataset, the study was conducted for two individual sensors and their fusion. Among the conclusions: (1) Different FS methods can yield comparable macro-F1 yet considerably different FS stability values. (2) Fisher score (single and/or ensemble) is superior in most of the cases. (3) mRMR's stability is overall the lowest, the most variable over different settings (e.g., sensor(s), subset cardinality), and the one that benefits the most from the ensemble.

Innovative cost-effective pendulum tiltmeter based on the moiré technique

A newly constructed pendulum tiltmeter has been described based on the moiré technique, which is specified as a series of dark and light patterns configured by the superposition of two regular gratings. This tiltmeter chiefly consists of two frames that can move with respect to each other, and tilting movements are recorded by using the moiré technique. Double similar gratings are attached to the frames. The gratings are installed in parallel, having no physical contact, so that their lines make a small angle with respect to each other. A diode laser beam passes through the formed moiré pattern, and a narrow slit then hits on a light detector. Due to external stress and making a tilt or incline in the moiré tiltmeter, one of the gratings is displaced respect to the other one, leading to the moiré fringe movement. So, the light intensity on the detector changes and is registered as the output voltage. A digital signal processor samples the voltage and generates a record of the tiltmeter. The recorded signal can be used to measure the tilt. We have investigated the experimental results of the tiltmeter, and the outcomes show that the sensor is quite viable. In total, the proposed tiltmeter is cost-effective, easy to install, and able to detect tilts of the order of 0.00001 rad. It is also insensitive to electromagnetic noise and environmental conditions, such as temperature fluctuations. Therefore, the applications of this device vary from civil engineering and geological studies to industrial monitoring and experimental ones.

Deep transfer learning methods for typical supervised tasks in industrial monitoring: state-of-the-art, challenges, and perspectives

Deep transfer learning methods for typical supervised tasks in industrial monitoring: state-of-the-art, challenges, and perspectives

Nonstationary Industrial Process Monitoring Based on Stationary Projective Dictionary Learning

Industrial data generally exhibit nonstationary features due to load changes, external disturbances, etc., which raise great challenges for process monitoring. Most existing methods for nonstationary process monitoring use a two-stage strategy, where stationary series are obtained via preprocessing, followed by feature extraction and modeling for the stationary series. However, the two-stage strategy treats preprocessing and feature extraction separately, which would obtain suboptimal features. This article develops the stationary projective dictionary learning (SPDL) method for monitoring nonstationary industrial processes. Specifically, nonstationary data are projected into a stationary subspace, overcoming the disadvantages of nonstationarity and improving the representation ability of the dictionary. Then, considering process data usually have temporal correlations, a novel self-expression constraint term is added to optimize the features so that it can describe the process data accurately. Finally, an iterative approach is proposed for joint optimization of the projection matrix and the dictionary, and the formulas for parameter estimation are derived in detail. After the projection matrix and the dictionary are obtained, the nonstationary data are projected and reconstructed, and the reconstruction error is adopted to design the monitoring statistic. Case studies on a numerical example, a nonstationary continuous stirred tank reactor (CSTR), and an industrial roasting process reveal that the proposed method is suitable for nonstationary industrial process monitoring.

Evaluation of Spatial Gas Temperature and Water Vapor Inhomogeneities in TDLAS in Circular Multipass Absorption Cells Used for the Analysis of Dynamic Tube Flows.

The use of optical circular multipass absorption cells (CMPAC) in an open-path configuration enables the sampling free analysis of cylindrical gas flows with high temporal resolution and only minimal disturbances to the sample gas in the pipe. Combined with their robust unibody design, CMPACs are a good option for many applications in atmospheric research and industrial process monitoring. When deployed in an open-path configuration, the effects of inhomogeneities in the gas temperature and composition have to be evaluated to ensure that the resulting measurement error is acceptable for a given application. Such an evaluation needs to consider the deviations caused by spectroscopic effects, e.g., nonlinear effects of temperature variations on the intensity of the spectral line, as well as the interaction of the temperature and concentration field with the characteristic laser beam pattern of the CMPAC. In this work we demonstrate this novel combined evaluation approach for the CMPAC used as part of the tunable diode laser absorption spectroscopy (TDLAS) reference hygrometer in PTB's dynH2O setup for the characterization of the dynamic response behavior of hygrometers. For this, we measured spatially resolved, 2D temperature and H2O concentration distributions, and combined them with spatially resolved simulated spectra to evaluate the inhomogeneity effects on the line area of the used H2O spectral line at 7299.43 cm-1. Our results indicate that for dynH2O, the deviations caused by the interaction between large concentration heterogeneities and the characteristic sampling of the beam pattern of the CMPAC are three orders of magnitude larger than deviations caused by small temperature heterogeneity induced spectroscopic effects. We also deduce that the assumption that the "path-integrated" H2O concentration derived with the open-path CMPAC setup represents an accurate H2O area average in the flow section covered by the CMPAC in fact shows significant differences of up to 16% and hence does not hold true when large H2O concentration gradients are present.

Improving self-supported nanowire arrays using response surface methodology for the synthesis of a H2O2 nanostructured sensor

In this paper, we introduce a data-driven approach for optimizing the design of highly sensitive hydrogen peroxide (H2O2) sensors based on self-supported nickel nanowire (Ni NW) arrays using response surface methodology (RSM). Our goal is to enhance the performance of the sensor by customizing the length of Ni NW while optimizing the working conditions, including the H2O2 concentration and applied potential. Focusing solely on the commercially available porous membranes, we successfully employ RSM to identify the optimal conditions for the sensor: a NW length of 2.64 μm, an optimal concentration of 3.25 mM H2O2, and a detection potential of 0.02 V. We fabricate a sensor with the optimal NW length and validate its performance through comparison to sensors with different NW lengths, such as a planar Ni sensor and a previously investigated nanostructured sensor. Our optimized sensor achieves a 50% reduction in the limit of detection (LOD) and an 18% increase in sensitivity compared with the previously investigated nanostructured sensor. Moreover, the optimized sensor is at least 35 times more sensitive for H2O2 detection than sensors with planar geometries, which are standard in commercial applications. Our results highlight the potential of RSM as a powerful, cost-effective, and theoretically agnostic tool for optimizing nanostructured sensors and accelerating the design-build-test cycle. The optimized H2O2 sensor can be applied in various fields, including the food industry, medical diagnostics, and environmental monitoring. This sensor can significantly enhance glucose detection in non-invasive samples such as tears and saliva when combined with immobilized glucose oxidase enzymes, which is crucial for continuous glucose monitoring in diabetes mellitus patients. The methodology presented in this study allows further extensions to investigate other design variables and sensing materials as well as optimize sensors for detecting different substances, fostering advancements in sensor technologies.

Non-Hermitian Topolectrical Circuit Sensor with High Sensitivity.

Electronic sensors play important roles in various applications, such as industry and environmental monitoring, biomedical sample ingredient analysis, wireless networks and so on. However, the sensitivity and robustness of current schemes are often limited by the low quality-factors of resonators and fabrication disorders. Hence, exploring new mechanisms of the electronic sensor with a high-level sensitivity and a strong robustness is of great significance. Here, a new way to design electronic sensors with superior performances based on exotic properties of non-Hermitian topological physics is proposed. Owing to the extreme boundary-sensitivity of non-Hermitian topological zero modes, the frequency shift induced by boundary perturbations can show an exponential growth trend with respect to the size of non-Hermitian topolectrical circuit sensors. Moreover, such an exponential growth sensitivity is also robust against disorders of circuit elements. Using designed non-Hermitian topolectrical circuit sensors, the ultrasensitive identification of the distance, rotation angle, and liquid level is further experimentally verified with the designed capacitive devices. The proposed non-Hermitian topolectrical circuit sensors can possess a wide range of applications in ultrasensitive environmental monitoring and show an exciting prospect for next-generation sensing technologies.

Semi-quantitative microfluidic paper-based analytical device for ionic silica detection.

Silicate ions (SiO32-), or ionic silica, are known to cause silica scaling in industrial water applications when excess amounts are present; hence, concentrations must be monitored and kept at a constant low level. Ionic silica is conventionally measured by spectrophotometry in the form of its silicomolybdic complex based on the molybdenum blue reaction, but the operation process is complicated and not suitable for on-site detection. To solve these issues, microfluidic paper-based analytical devices (µPADs) have been gaining attention as portable, low-cost analytical devices suitable for on-site detection. The foldable origami type device described in this work enabled silica detection based on the molybdenum blue reaction, in the range of 50-1000mg/L, with a practically detectable lowest concentration of 50mg/L. The device showed selectivity for silicate ions and stability over 21days when stored at 4°C. The semi-quantitative analytical performance makes the proposed paper-based device attractive for on-site industrial monitoring.

Sistem Monitoring Dan Kontrol Motor Listrik Industri Menggunakan Internet Of Things (IoT)

The electric motor plays an important role in an industry as a driver in a particular system, for example, rotating a pump impeller, fan or blower, moving a compressor, and lifting materials. The electric motor is sometimes called the "work horse" of industry. It is estimated that motors use about 70% of the total electrical energy in the industry. In making industrial motor monitoring and control systems using IoT with the blynk application, it works in showing the parameters of voltage, current, temperature, and RPM. The module consists of an electric motor, a DHT11 sensor, a PZEM004T sensor, an RPM sensor, and an ESP 8266 microcontroller. In the research conducted, the system made is capable of monitoring and controlling many modules. So that the overall monitoring and control system of electric motors in the company can be integrated into a system. In testing the voltage sensor with a measuring instrument there is an error value of 0.04%, while the current sensor has an error of 2.5%, followed by temperature testing there is an error of 1.8%. the RPM sensor parameter has a value of 1.17%. Remote manual control can be controlled using a smartphone with the blynk application so that the operator can immediately turn off the motor remotely if there is a problem or damage to the electric motor.

Perspectives of Polymers in Forensic Analysis

Polymeric materials have recently attracted a lot of attention due to their potential applications in many fields, ranging from biomedicine, the food industry and environmental monitoring to electronic, energy storage and sensing devices. Their versatility, functionalization capability, chemical/physical stability, reusability, long shelf-life, as well as good mechanical and thermal properties, also make them idoneous candidates for use in forensic sciences, which deal with the investigation of crimes, finding relations between evidence and criminals. In particular, molecularly imprinted polymers (MIPs), designed based on the principle of generating template-specific polymeric cavities fitted to the target molecules in the presence of selected chemicals via non-covalent or covalent interactions, are highly suitable for forensic analysis. In addition, their combination with other compounds such as carbon nanomaterials can provide composites with improved properties to be used in the analysis of illicit drugs, doping substances, biological agents, toxins and so forth. In this article, recent applications of polymeric materials in the field of forensic analysis are discussed. The goal is to summarize their current uses and put forth a projection of their potential as promising alternatives for standard competitors.

Phase Distribution Measurement of Liquid–Solid Media Using Multimode Ultrasound

Ultrasound tomography (UT) is an emerging technique for phase distribution measurement in industrial multiphase flow monitoring. A multimode UT combining the ultrasound transmission and reflection modes is proposed to measure stratified phase distribution and sound speed of liquid–solid mixtures in the horizontal pipe. First, the acoustic attenuation is used to reconstruct the phase distribution with a modified matrix regularization algorithm, and the time-of-flight (TOF) of ultrasound is used to extract the boundary of multiphase layers with the weight Akaike information criterion algorithm. Second, a wavelet fusion method based on the principal component averaging is proposed to fuse the reconstruction images of UT transmission and reflection. Third, numerical investigations are carried out to validate the feasibility of the proposed method. According to the quantitative indexes, the proposed method can effectively reconstruct the phase distribution of multiphase media. Finally, an experimental test pipeline Section is established, and sand–water and oil–water–sand liquid–solid media are investigated. The minimum relative error (RE) and the maximum correlation coefficient (CC) of experimental fusion results can reach 0.33 and 0.94, respectively, and the average measurement error (ME) of the layer height is 3.80%.

Transformation to a smart factory using NodeMCU with Blynk platform

Incorporating internet of things (IoT) in industrial systems prompted the development of industrial internet of things (IIoT) systems, which in turn enable the automation of intelligent devices to gather, analyze, and transmit data from industrial systems in real-time. This paper develops a low-cost and smart industrial remote monitoring and control system based on NodeMCU microcontrollers and Blynk server platform. It is deployed to remotely monitor manufacturer's environment and industrial equipment and control them autonomously. Also, it protects the manufacturer's employees from fire catastrophe by warning them using a buzzer and notification. The system comprises two main parts, sensing and actuation. The sensing part consists of three subsystems that measure temperature and humidity, water flow, and flame. The actuation part consists of a water pump, light, and fan. A powerful user interface is developed based on the Blynk platform. The proposed system controls the water pump by sensing water flow autonomously. In addition, based on a fire detected, a protection system is implemented to shut down the electricity from load in case of fire event occurs. Several testing scenarios were carried on to check the response of the system, and the result shows successful implementation of the proposal to handle different situations.

Acoustic Backscatter Communication and Power Transfer for Batteryless Wireless Sensors

Sensors for industrial and structural health monitoring are often in shielded and hard-to-reach places. Acoustic wireless power transfer (WPT) and piezoelectric backscatter enable batteryless sensors in such scenarios. Although the low efficiency of WPT demands power-conserving sensor nodes, backscatter communication, which consumes near-zero power, has not yet been combined with WPT. This study reviews the available approaches to acoustic WPT and active and passive acoustic through-metal communication. We design a batteryless and backscattering tag prototype from commercially available components. Analysis of the prototypes reveals that low-power hardware poses additional challenges for communication, i.e., unstable and inaccurate oscillators. Therefore, we implement a software-defined receiver using digital phase-locked loops (DPLLs) to mitigate the effects of oscillator instability. We show that DPLLs enable reliable backscatter communication with inaccurate clocks using simulation and real-world measurements. Our prototype achieves communication at 2 kBs-1 over a distance of 3 m. Furthermore, during transmission, the prototype consumes less than 300 μW power. At the same time, over 4 mW of power is received through wireless transmission over a distance of 3 m with an efficiency of 2.8%.

Industrial Disaster Control Systems in Unmanned Aerial Vehicles

The frequency of catastrophes is rising. Whether a disaster is caused by anything man-made or something natural, the effects are always serious and will take a long time to resolve with human strength alone. To cut down on time and human exposure, robotic technologies are being used more and more in search and rescue operations. It has suffered from a lack of environmental adaptability.Every robot is built differently in order to carry out the tasks assigned to it more effectively and efficiently. For continuous industry monitoring without human interaction, our concept suggests a hanging robot. With the use of forward and backward motors, a hung robot may travel along a rope path from one location to another.

A new distribution-free scheme for simultaneous Phase-I analysis of four process aspects and its application in monitoring customers’ waiting times

In industrial process monitoring and control, Phase-I or retrospective analysis plays an essential role in assessing historical information and setting up a benchmark for future quality practices. It is vital to analyse four critical aspects of a process, including location, scale, skewness and kurtosis. Assuming or fitting a parametric distribution and estimating process parameters are not recommended before conducting a Phase-I analysis. Otherwise, extreme values may influence the estimation, and the fit could be misleading. However, no distribution-free method can monitor the four aspects in tandem. In recent years, the RS/P schemes, and some rank-based schemes, including Lepage-type combined statistics for location and scale shifts, have been introduced for retrospective study in industrial operations. However, they are not designed to handle complex process shifts involving skewness and kurtosis. To this end, we propose a new scheme using subgroup samples, the Legendre polynomial and an omnibus rank globe statistic, referred to as the Phase-I globe chart and abbreviated as the GLC scheme. The GLC scheme can recognise out-of-control subsamples in respect of four aspects of a process, and hence it may be looked upon as a quad-aspect scheme which is a further improvement over the existing bi and tri-aspect schemes. Computer-intensive Monte Carlo simulations are used to evaluate charting constants for a given false alarm probability. The correct alarm probability of our scheme is compared with that of various existing distribution-free Phase-I schemes. Various shift patterns in one or more subsamples from different various out-of-control situations in a class of in-control distributions are considered in our investigation. The proposed Phase-I scheme is found to have an excellent performance in a wide range of scenarios. Some applications in monitoring service processes are also discussed.

Global Mechanical Response Sensing of Corrugated Compensators Based on Digital Twins

The corrugated compensators are important components in the piping system, absorbing mechanical deformation flexibly. To reduce the risk of the piping system with corrugated compensators and improve the safety and stability of industrial equipment, condition monitoring and fault diagnosis of bellows is necessary. However, the stress monitoring method of corrugated compensators with limited localized sensors lack real-time and full-domain sensing. Therefore, this paper proposes a digital twin construction method for global mechanical response sensing of corrugated compensators, combining Gaussian process regression in machine learning and finite element analysis. The sensing data of three types of displacements are used as the associated information of a finite element model with 19,800 elements and its digital twin. The results show that the values of performance metrics correlation of determination R2 and standardized average leave-one-out cross-validation CVavg of the digital twin satisfy the recommended threshold, which indicates that the digital twin has excellent predictive performance. The single prediction time of the digital twin is 0.76% of the time spent on finite element analysis, and the prediction result has good consistency with the true response under dynamic input, indicating that the digital twin can achieve fast and accurate stress field prediction. The important state information hidden in the multi-source data obtained by limited sensors is effectively mined to achieve the real-time prediction of the stress field. This paper provides a new approach for intelligent sensing and feedback of corrugated compensators in the piping system.

Multifunctional and highly sensitive humidity sensor based on KCl/Sm2O3 nanoflowers with ultrafast response

With the emerging “Internet of Things” technology, humidity sensors are becoming indispensable in agriculture, industry, medicine, and environmental monitoring. However, the development of high-performance and multifunctional humidity sensors in the field of human-computer interaction still remains a challenge. Herein, Sm2O3 nanoflowers with hierarchical structure were synthesized by modulating the material morphology. The unique hierarchical structure could provide more active sites, which is beneficial for the adsorption and desorption of water molecules. Furthermore, by compounding with KCl, the KCl/Sm2O3 composite demonstrated enhanced humidity sensitivity and rapid response in diverse humidity conditions on account of that the introduced K+ and Cl- could participate in the conduction process. The humidity sensor based on KCl/Sm2O3 showed extremely high response value of 127121 (60 times higher than that of pure Sm2O3) with ultrafast response (0.1 s) at 11%−95% RH. Besides, the sensor exhibited excellent application potential for human health, activity monitoring, non-contact sensing, and alcohol count measurement. This work proves that KCl/Sm2O3 is a promising candidate for high-performance humidity sensing in the application of intelligent perception.

A Survey of Drones Using IOT Technology

A Drone is a new invention or era in last recent years which is very effective; Drones are commonly known as Unmanned Aircraft System (UAV) or Flying Robots. Previously Unmanned Aircraft System (UAV) was used by militaries all around the world for different kind of technologies like Intelligence, Surveillance and Reconnaissance Operations. The use of Drones has been using continuously in different and wide areas, like Cargo, Parcel Delivery, Internet provision in remote areas, Emergency Aid, Industrial Asset monitoring pipelines, surveillance in crowded areas, providing food in flooded area and many more. Drones are a microcosm of the Internet of Things(IOT), also it has been realized that Drones have potential to increase the business by using the Technology for their Organizations.

The Versatility of Piezoelectric Composites

Piezoelectric materials possess the capability to interchangeably convert electrical energy into a mechanical response. While current piezoelectric materials exhibit strong properties, known limitations have inhibited further development. This review describes the ability to combine different piezoelectric materials into a composite to create well-rounded properties. The different types of connectivity classes are described as well as important design considerations and theoretical models. The contributions from the active and passive phases are outlined, focusing primarily on ferroelectric ceramics and polymer-based composites. The key advantage of piezoelectric composites is their ability to combine the flexibility of polymers with the high electromechanical coupling and piezoelectric coefficients of ferroelectric ceramics or single crystals appropriate for a variety of applications. Composites are prominent in medical ultrasound imaging and therapy, underwater acoustic sensing, industrial structural health monitoring, energy harvesting, and numerous other emerging applications.

Cosmetic Surveillance: An update and comprehensive Review

Surveillance has recently been expanded to include the safety of herbal and cosmetic products. A new phrase has emerged characterizing industry monitoring known as “Kosmetonadzor” to ensure the safety of cosmetic products. Wigan (1997) was the first in the literature to use the term to refer to the safety management of cosmetic products. It is now recognized as a public health concept worldwide. 
 As there is no systematic reporting system, side effects or side effects of cosmetics are relatively insignificant or overlooked. There is a follow-up management system for preparation. In India, market surveillance is usually focused on drug side effects. As a result, more emphasis has been placed on medical devices, blood products, biologics, natural products, and special diets, and less on related side effects. To sum up, Cosmetology refers to the monitoring of cosmetic products that have recently entered the market.
 Conclusions
 Cosmetic surveillance is a new approach to regulating the safety of cosmetic products. It is an important component of public health initiatives. As post-sale surveillance of cosmetics becomes more and more common worldwide, malfunctions in these items can be detected and corrected, so safety can be achieved. Physicians Family doctors and general practitioners play an important role in detecting ADRs caused by cosmetic items and, as a result, urge patients to report ADRs. Raising awareness of this new notion will be an important contribution to global public health. 
 In general, the Cosmetovigilance system can avoid both significant and minor adverse effects. Because the notification procedure is critical to the system, healthcare workers must be taught and supported as part of the Cosmetovigilance feedback system. Some regulation modifications are required to keep up with Turkey's developing cosmetics business.
 In general, cosmetic monitoring systems can prevent both serious and minor side effects. As the reporting process is a key element of the system, medical professionals must be educated and supported in terms of cosmetic oversight within the feedback system. Some regulatory updates are needed to keep pace with the changing Turkish cosmetics market.