Indirect Monitoring Research Articles

Enhancing Machining Efficiency: Real-Time Monitoring of Tool Wear with Acoustic Emission and STFT Techniques

Tool wear in machining is inevitable, and determining the precise moment to change the tool is challenging, as the tool transitions from the steady wear phase to the rapid wear phase, where wear accelerates significantly. If the tool is not replaced correctly, it can result in poor machining performance. On the other hand, changing the tool too early can lead to unnecessary downtime and increased tooling costs. This makes it critical to closely monitor tool wear and utilize predictive maintenance strategies, such as tool condition monitoring systems, to optimize tool life and maintain machining efficiency. Acoustic emission (AE) is a widely used technique for indirect monitoring. This study investigated the use of Short-Time Fourier Transform (STFT) for real-time monitoring of tool wear in machining AISI 4340 steel using carbide tools. The research aimed to identify specific wear mechanisms, such as abrasive and adhesive ones, through AE signals, providing deeper insights into the temporal evolution of these phenomena. Machining tests were conducted at various cutting speeds, feed rates, and depths of cut, utilizing uncoated and AlCrN-coated carbide tools. AE signals were acquired and analyzed using STFT to isolate wear-related signals from those associated with material deformation. The results showed that STFT effectively identified key frequencies related to wear, such as abrasive between 200 and 1000 kHz and crack propagation between 350 and 550 kHz, enabling a precise characterization of wear mechanisms. Comparative analysis of uncoated and coated tools revealed that AlCrN coatings reduced tool wear extending tool life, demonstrating superior performance in severe cutting conditions. The findings highlight the potential of STFT as a robust tool for monitoring tool wear in machining operations, offering valuable information to optimize tool maintenance and enhance machining efficiency.

Experimental modal identification of a pedestrian bridge through drive-by monitoring integrated with shared-mobility vehicles

In the context of improving resilience in transport infrastructure, an emerging approach of indirect structural health monitoring is gaining attention, known as drive-by monitoring, instrumenting a vehicle with sensors to evaluate the bridges it crosses. However, their effectiveness has predominantly been investigated in ideal scenarios, with actual real-world applications being quite scarce. The research presented in this paper explores the feasibility of combining two routes: (1) fleet emissions reduction and (2) transport resilience enhancement, through drive-by monitoring integrated with shared mobility, including electric mobility scooter and bicycle. The information from drive-by data collected from shared mobility can give valuable information on critical transport infrastructure (e.g., bridges) and such a drive-by database has the potential for network level infrastructure condition assessment. In this paper, two different drive-by roadmaps are investigated subject to the flexibility of the bridges, namely partially or fully drive-by monitoring. To validate the proposed roadmaps, a full-scale pedestrian bridge was chosen for drive-by testing, where smartphone sensors and specialised accelerometers are mounted on shared mobility for data acquisition. Experimental results demonstrate that (i) smartphone sensing can provide data with similar accuracy compared to specialised accelerometers, (ii) bridge frequencies can be easily obtained from temporarily parked shared mobility, with a maximum relative error of 1.05%, (iii) both the bridge frequencies and operational deflection shapes are successfully extracted from the moving shared mobility by using variational mode decomposition and filtering techniques, and shared mobility's GPS data along with moving speeds are collected for potential vehicle positioning and drive-by database updating.

A Comparative Study of Measurement Systems Using Machine Learning Techniques for Tool Fault Di-agnosis During Metal Removal in the Milling Process

Despite significant advancements in both measurement systems and machine learning techniques, the integration of these technologies for real-time tool fault diagnosis in milling processes remains under developed. Existing studies tend to focus on a comprehensive comparative analysis that bridges these two areas machine learning algorithms or the application of specific measurement sys-tems. There is also a gap in evaluating the cost-effectiveness and practicality of different measurement systems when integrated with machine learning models for industrial applications. This study addresses these gaps by conducting a detailed comparative analysis of multiple measurement sys-tems and their performance with machine learning techniques in a real-world milling context, aim-ing to provide practical recommendations for industry adoption. Using both traditional and Artificial Intelligence (AI) to define and exploit sensory systems in the milling process, as well as various (direct and indirect) monitoring approaches, are summarised in this study. Machine learning tech-niques SVM, KNN, DT performs better and provide higher accuracy and in feature extraction clas-sification techniques statistical features, wavelet transform with the Holder Exponent (HE) having higher accuracy for diagnosing the tool faults.

Data-driven evaluation of the Paris’ law parameters in polyethylene pipe grades — Increasing the precision of fracture mechanical lifetime estimation

The Paris’ Law parameters A and m are a necessity for predicting lifetimes of structural components under static or fatigue loading that fail due to crack initiation and propagation. Conventional methods require measurements of crack growth kinetics that involve direct or indirect monitoring of physical crack extension during long-term experiments. Usually, measurement series also involve multiple specimens in order to obtain a crack growth controlled failure diagram of an investigated material under relevant load conditions. In this contribution a combination of simple numerical, statistical and analytical approaches is presented to obtain A and m without the need to measure actual crack growth. This is accomplished by reformulating the Paris’ Law to express A as a function of m. The parameter m is varied within a reasonable range to generate an analytical function for A that solves the equation of the Paris’ Law based lifetime for a single specimen. A subsequent superposition of all available specimens reveals an intersection of all A functions at the technically relevant pair of A and m values that are capable of describing the lifetime of all specimens with a minimum error. The obtained best-fitting A and m are in good agreement with literature and are able to predict the lifetime of previously published sample data based upon cyclic Cracked Round Bar test results with an average error of 3.30 ± 2.67%.

The capability of Acoustic Emission features to monitor diamond-coated burr grinding wear and effectiveness

Within manufacturing there is a growing need for autonomous Tool Condition Monitoring (TCM) systems, with the ability to predict tool wear and failure. This need is increased, when using specialised tools such as Diamond-Coated Burrs (DCBs) for grinding high strength ceramics or glass, in which the random nature of the tool, inconsistent manufacturing methods and high wear rates create large variance in tool life. This unpredictable nature leads to a significant fraction of a DCB tool's life being underutilised due to premature replacement. Workpiece surface damage, increased grinding forces and large-scale diamond grain pullout could all be the result of high levels of runout and in-circularity common within electroplated DCBs. As such it is important to not only monitor the overall tool wear but also tool condition. Acoustic Emission (AE) presents as an indirect on-machine sensing method highly suited to grinding applications. The high frequency range of AE, >20 kHz, prevents machine noise from dominating the acquired signals, isolating the micro-scale machining processes within noises machine environments. AE resulting from the grinding process has the potential to monitor not only tool wear but also runout and circularity. A series of DCB wear tests have been conducted, each consisting of the continuous acquisition of AE during grinding and frequent tool surface measurements. Preliminary results demonstrate AE kurtosis can be seen as an indicator of each tool’s runout, representing the fraction of time the tool and workpiece are in contact during a revolution. As a result, an indirect monitoring system capable of monitoring wear and tool state with AE could be utilised within the manufacturing sector.

Dynamic correlation between surface carbon response and underlying emissions from spontaneous combustion goaf: field study of an abandoned coal mine

Abandoned coal mine goaf is affected by air leakages and prone to spontaneous combustion, resulting in environmental pollution and geological disasters. Haizhou Open-pit Mine adopts both underground and open-pit mining methods. During the long-term mining process, the original stable stratum structure is constantly destroyed, and the slope slides, increasing cracks and severe air leakage around the goaf and roadway. The spontaneous combustion of coal is particularly prominent after the mine shut down. At present, there is no suitable indirect monitoring method to effectively explore the spontaneous combustion area in goaf. The study developed an all-weather monitoring plan and conducted multi-point continuous long-term measurements of the spontaneous combustion state in one abandoned coal mine goaf located in the eastern part of the Haizhou Open-pit Mine. We evaluated the dynamic correlation between surface CO2 flux (SCF) and changes in the underground fire areas, determined the scope and evolution trend of the fire areas, and identified the distribution and change laws of SCF. The results show a significant positive correlation between SCF and soil temperature; moreover, the SCF value was found to reflect the CO2 emission intensity of the goaf. The high SCF in the test area showed month-wise expansion and increase, while the CO2 emission gradually increased monthly, and the calculated annual total emission was approximately 7017 t. Hence, the study can further provide guidance for the monitoring of spontaneous combustion in shallow coal seams, goaf and the assessment of CO2 emissions from underground coal fires through the on-site monitoring and analysis results.

Diesel Engine Turbocharger Monitoring by Processing Accelerometric Signals through Empirical Mode Decomposition and Independent Component Analysis

In this study, a method for the monitoring of internal combustion engine operation by vibration signals is proposed. The work falls within the context of the increasingly stringent standards relating to the environmental impact of engines and the development of monitoring and control techniques to ensure increased engine performance as well as fuel saving and reduction of pollutant emissions. Experimentation was performed on a turbocharged light-duty compression ignition direct-injection engine. Two monoaxial accelerometers were installed on the engine compressor case, the speed of which has been demonstrated to be closely related to the engine operation. Vibration measurements of the engine compressor case have been processed by combining the Empirical Mode Decomposition technique with Independent Component Analysis and Short Time Fourier Transform to indirectly estimate the turbocharger speed. The obtained traces have been compared to the direct turbocharger velocity measures during the stationary running of the engine (speed and load conditions varied in the complete engine’s range of operation). The results point out the potentiality of the methodology in algorithms devoted to identifying modifications of the combustion development regarding regular operation via indirect turbocharger speed monitoring.

Phyformer: A degradation physics-informed self-data driven approach to machinery prognostics

Machinery degradation prognostics methods have been suffering the “dilemma” that both physics and data-driven methods have their own limitations. On one hand, constructing accurate close-form mathematical physics models are prohibitively difficult for a complex system. On the other hand, for pure data driven models such as deep learning models, due to the lack of physics guidance, their extrapolation capability decays over time and even yield unexplainable predictions that violate common cognition. Therefore, relaxing the pursuit for the perfect and accurate degradation model, and instead coupling general degradation model with deep learning model, fully leveraging the advantages from both physics-based prognostics and data-driven prognostics, is a promising way to solve the “dilemma”. Driven by this motivation, this paper proposes Phyformer, a general degradation physics-informed self-data-driven method for machinery prognostics. A backbone deep learning model based on auto-correlation and Transformer architecture is developed for time series data prediction. Multiple local physical models that are constructed with data in sliding time windows are embedded into the backbone model in the form of loss function. Only the historical data of the machine itself are used to extrapolate the future, which is significant especially for high-end equipment since their run-to-fail data are scarce. Then the predicted data are mapped to degradation stage by an unsupervised clustering model Gath-Geva adaptively without the requirement of knowing the number of degradation stages in advance. Due to the above nature, Phyformer is high flexibility, easy to be deployed in different applications, working condition-independent and simple. Two typical prognostics tasks are studied, which use direct and indirect condition monitoring data as input, respectively. Comparisons with other state-of-the-art machinery prognostics models show the advances of Phyformer.

Research on charging monitoring method for lithium-ion batteries based on magnetic field sensing

With the rapid and continuous development of new energy vehicles, safety issues during charging have become increasingly prominent. Thus, there is an urgent need to strengthen charging safety management. However, current charging detection methods have limitations, such as difficulty in monitoring local battery pack currents, and indirect monitoring through temperature and gas sensors lead to lag. To address these limitations, this article proposes a safety monitoring method based on an improved lithium-ion battery magnetic field sensing model that covers the entire charging process. Specifically, thermal field analysis was introduced to the existing electric magnetic coupling model, resulting in an electric thermal magnetic multi-physical field coupling model. By conducting experiments, the study verified that the accuracy of the improved model has increased by 15.5 %. Based on this model, the study proposed a method for predicting and warning of abnormal currents using BP neural network and mathematical statistical laws. The results showed that the current prediction accuracy of this method reached 96.4 %. Such high accuracy ensures real-time comprehensive monitoring of the current information of lithium-ion battery packs, thereby eliminating the hidden dangers of charging safety accidents caused by the lack of local current detection.

How Effective Are Ireland's Monitory Mechanisms in Improving Its Child Protection and Welfare Services?

ABSTRACTChild protection and welfare systems are entrusted with significant power by governments and are therefore a significant focus of monitoring activities. Monitoring can help to build a better child protection system and to ensure child safety; track policy and legislative implementation and resource allocation; contribute to preventing systemic ‘failures’; provide data for system, policy, and legislative reforms; and support continuous improvement. The Irish system is worth examining for three reasons. First, Ireland has had a single state‐provided child protection and welfare service called the Child and Family Agency. Second, the system has been through a sustained period of development. Third, Ireland has one dedicated statutory authority called the Health Information and Quality Authority (HIQA). This article provides a case study of the strategies and mechanisms adopted by Ireland to monitor the operation, quality and development of its child protection and welfare system. We will show that there is significant direct and indirect monitoring and focus on child protection from regulators, civil society organisations, government agencies and committees and independent actors. Such monitoring has significantly influenced the development of policy, practice and the development of the child protection system in Ireland.

Touch–based potentiometric sensors for simultaneous detection of urea and ammonium from fingertip sweat

Urea is a biomarker for diagnosis and management of multiple disorders. The measurement of urea could help in the diagnosis of patients with kidney dysfunction. Despite their attractive analytical performances, the existing urea measurement methods are still limited to sophisticated and time–consuming or require invasive blood sampling. Herein, a reliable noninvasive, simple, and disposable touch–based potentiometric sensor is fabricated for rapid monitoring of fingertip sweat urea. Novel touch–based urea detection relies on immobilized urease recognition and a solid–contact ammonium–ion–selective electrode, along with a screen–printed carbon working electrode modified carboxylated carbon nanotubes for enhancing sensor’s sensitivity. The fingertip sweat under sedentary rest conditions can be instantaneously collected on customized porous polyvinyl alcohol hydrogel transporting sweat to working electrode surface where the hydrolysis of urea is catalyzed, generating bicarbonate and ammonium. The resulting ammonium can be detected with ion–selective transducer, enabling indirect monitoring of sweat urea. Potential interference from co–existing ammonium in human sweat is addressed by developing a dual disposable potentiometric sensor, consisting of urea and ammonium sensors on a single strip platform allowing for simultaneous touch–based detection of sweat urea and ammonium. The resulting dual disposable sensing system offers rapid urea and ammonium monitoring within a group of healthy human subjects, along with a good correlation with R2 = 0.974 between the sweat urea and capillary blood urea levels. Additional confirmation among a broad sample of individuals along with the validation of sensor data through centralized laboratory tests will establish the practicality of simultaneously monitoring urea and ammonium levels.

Enhancing Turbine Blade Manufacturing through MEMS-Based Milling Monitoring

Manufacturing steam turbine blades with intricate shapes poses a substantial challenge. These blades significantly impact turbine efficiency, reliability, and productivity. Their precise formation through milling processes demands vigilant monitoring, especially concerning the cutting tool’s condition, which is responsible for shaping the blade profiles and ensuring high-quality outcomes. Direct tool monitoring, however, disrupts productivity, prompting the need for an indirect Tool Condition Monitoring (TCM) system. Such a system becomes essential for detecting tool wear and damage early, ensuring dimensional accuracy and surface smoothness. This experiment monitors tool conditions by analyzing vibrations generated by an endmill cutter while machining martensitic stainless steel (MSS) AISI 420. The setup employs a cost-effective MEMS vibration sensor, the ADXL 345, and Raspberry Pi for real-time online TCM functionality as the signal processor. This study delves into the sensor’s ability to capture vibrations representing actual tool conditions, focusing on the affordability and effectiveness of MEMS-based monitoring. The successful real-time implementation of MEMS-based monitoring could be an easily accessible gateway for manufacturing industries to embrace cloud-based monitoring systems, aligning with current technological trends. This research aims to underscore the viability and potential adoption of affordable MEMS sensors for comprehensive tool condition monitoring, offering a seamless transition toward cloud-integrated monitoring systems for manufacturing industries.

Contact-point response reconstruction for indirect bridge monitoring via Bayesian expectation-maximization based augmented Kalman filter

The drive-by bridge dynamic measurement using an instrumented vehicle during its passage over the bridge can be an indirect bridge structural health monitoring system. The identification of the contact-point (CP) responses between vehicle wheels and bridge structure from vehicle responses offers an efficient and economical way for the modal identification and condition assessment of short- to mid-span road bridges. This paper proposes a novel method for the contact-point displacement response reconstruction of the vehicle-bridge interaction (VBI) system in a joint input-state estimation manner based on Bayesian Expectation-Maximization (BEM). An analytical model of VBI including the road approach in front of a simply-supported bridge is presented firstly. Then the augmented state-space model of the vehicle with the contact-point responses included as variables along with vehicle states is established. A new method by integrating the augmented Kalman filter (AKF) with a BEM strategy is proposed to solve the state estimation problem without knowing the vehicle axle responses. The vehicle states and CP displacement responses are identified simultaneously. The effects of the measurement noise, road surface roughness and vehicle speed on the identified results are investigated using numerical simulations. The experimental tests are used to further verify the feasibility and accuracy of the proposed method. The results show that the proposed method has potential for drive-by bridge condition assessment using on-board vehicle response measurements for a large population of short- to mid-span bridges.

Damage Detection in Bridge Structures through Compressed Sensing of Crowdsourced Smartphone Data

Traditional bridge health monitoring methods that necessitate sensor installation are not only costly but also time-consuming. In contrast, utilizing smartphone data collected from vehicles as they traverse bridges offers an efficient and cost-effective alternative. This paper introduces a cutting-edge damage detection framework for indirect monitoring of bridge structures, leveraging a substantial volume of acceleration data collected from smartphones in vehicles passing over the bridge. Our innovative approach addresses the challenge of collecting and transmitting high-frequency data while preserving smartphone battery life and data plans through the integration of compressed sensing (CS) into the crowdsensing-based monitoring framework. CS employs random sampling and signal recovery from a significantly reduced number of samples compared to the requirements of the Nyquist–Shannon sampling theorem. In the proposed framework, acceleration signals from vehicles are initially acquired using smartphone sensors, undergo compression, and are then transmitted for signal reconstruction. Subsequently, feature extraction and dimensionality reduction are performed using Mel-frequency cepstral coefficients and principal component analysis. Damage indexes are computed based on the dissimilarity between probability distribution functions utilizing the Wasserstein distance metric. The efficacy of the proposed methodology in bridge monitoring has been substantiated through the utilization of numerical models and a lab-scale bridge. Furthermore, the feasibility of implementing the framework in a real-world application has been investigated, leveraging the smartphone data from 102 vehicle trips on the Golden Gate Bridge. The results demonstrate that damage detection using the reconstructed signals obtained through compressed sensing achieves comparable performance to that obtained with the original data sampled at the Nyquist measurement sampling rate. However, it is observed that to retain severity information within the signals for accurate damage severity identification, the compression level should be limited to 20%. These findings affirm that compressed sensing significantly reduces the data collection requirements for crowdsensing-based monitoring applications, without compromising the accuracy of damage detection while preserving essential damage-sensitive information within the dataset.

Comparative deep learning studies for indirect tunnel monitoring with and without Fourier pre-processing

In the last decades, the majority of the existing infrastructure heritage is approaching the end of its nominal design life mainly due to aging, deterioration, and degradation phenomena, threatening the safety levels of these strategic routes of communications. For civil engineers and researchers devoted to assessing and monitoring the structural health (SHM) of existing structures, the demand for innovative indirect non-destructive testing (NDT) methods aided with artificial intelligence (AI) is progressively spreading. In the present study, the authors analyzed the exertion of various deep learning models in order to increase the productivity of classifying ground penetrating radar (GPR) images for SHM purposes, especially focusing on road tunnel linings evaluations. Specifically, the authors presented a comparative study employing two convolutional models, i.e. the ResNet-50 and the EfficientNet-B0, and a recent transformer model, i.e. the Vision Transformer (ViT). Precisely, the authors evaluated the effects of training the models with or without pre-processed data through the bi-dimensional Fourier transform. Despite the theoretical advantages envisaged by adopting this kind of pre-processing technique on GPR images, the best classification performances have been still manifested by the classifiers trained without the Fourier pre-processing.

An evolutionary vehicle scanning method for bridges based on time series segmentation and change point detection

In this paper, the problem of bridge health monitoring is formulated as an unsupervised semantic segmentation where the aim is to identify the onset of structural damage by detecting unexpected irregularities in the time series data. In particular, the main objective of this study is to approach this problem through an indirect sensing framework based on the concept of vehicle-mounted scanning. Thus, the sole source of information is the acceleration responses of the moving vehicle passing over the bridge. This is achieved through automatic feature learning by adopting the concept of matrix profile to conduct an all-pair-similarity search for a given time series, where the location of the contextual change is determined by the corresponding Corrected Arc Crossings (CAC) profile. The performance of the method is first evaluated through numerical investigations for different damage cases where the effect of operational uncertainties such as road roughness, and variability in vehicle properties are considered. Further, the method is experimentally evaluated using a lab-scale bridge model with different damage cases. Through these investigations, it is consistently observed that the location of the change point always coincides with the minimum point of the CAC profile. The proposed framework has three major advantages making it more appealing compared to the competing techniques. First, it has been fully formulated in the context of unsupervised learning where no access to labeled data is required. Second, it is capable of working with a limited amount of data yet providing promising results. Finally, it can be easily applied to online settings where data is streaming in real-time. These successful results demonstrate the potential of the proposed method as a cost-effective and rigorous bridge inspection tool and open up opportunities in the field of indirect bridge health monitoring.

Tool condition monitoring in micro milling of brittle materials

This paper introduces a novel indirect tool condition monitoring (TCM) system for micro-milling brittle materials (glass and silicon) based on acoustic emission (AE) and cutting force signals. The milling experiments are also applied to a ductile material (steel) for comparison. Tool wear calibration is conducted to determine the three tool wear stages. The collected signals are processed in time, frequency, and time-frequency domains. Specific frequency subrange combinations of cutting force signals in three directions after wavelet packet decomposition show strong correlation to the tool wear stages, whilst processed AE signals are the secondary feature to the tool wear. A tool wear prediction model for each material is built based on all the chosen features and back propagation (BP) neural network. The prediction model is simply structured and highly efficient with the highest prediction accuracy of more than 95%.

Mono-spectral hemoglobin detection in the porcine collateral network: Data acquisition for automatic AI-powered indirect spinal cord perfusion monitoring

Mono-spectral hemoglobin detection in the porcine collateral network: Data acquisition for automatic AI-powered indirect spinal cord perfusion monitoring

Long-term indirect monitoring system for short and medium span bridges

Bridge health monitoring has been widely discussed for establishing the rational maintenance strategies of existing bridges. Meanwhile, we need to provide discussions and recommendations on how to establish fully automatic data collection system and how to manage the big data for holistic bridge maintenance decisions. The author developed a vehicle-based (coupling of a moving vehicle and a bridge) bridge health monitoring system by using public buses which is one practical solution to the problem for condition assessment of short and medium span bridges (called “bus-based monitoring system”). The proposed system includes some new ideas about making use of the under rear-wheel spring acceleration sensors installed on an in-service fixed-route bus operating on the public transit system (long-term and indirect methods), define “characteristic deflection” as an indicator that may be useful in efficiently detecting a structural anomaly of a bridge, etc. The main objectives of this paper are 1) to introduce the details of how to assess the bridge condition based on the results of long-term application (total more than 10 years) to actual bridges on Ube City bus network in Japan, as a specific example for verify the system, 2) to discuss what is possible steps in realizing practical application of the system to existing bridges for consideration systematically into the maintenance strategies in the future. This paper is a continuation of previously published ones with new materials from the practical application viewpoint.

Tool wear segmentation in blanking processes with fully convolutional networks based digital image processing

The extend of tool wear significantly affects blanking processes and has a decisive impact on product quality and productivity. For this reason, numerous scientists have addressed their research to wear monitoring systems in order to identify or even predict critical wear at an early stage. Existing approaches are mainly based on indirect monitoring using time series, which are used to detect critical wear states via thresholds or machine learning models. Nevertheless, differentiation between types of wear phenomena affecting the tool during blanking as well as quantification of worn surfaces is still limited in practice. While time series data provides partial insights into wear occurrence and evolution, direct monitoring techniques utilizing image data offer a more comprehensive perspective and increased robustness when dealing with varying process parameters. However, acquiring and processing this data in real-time is challenging. In particular, high dynamics combined with increasing strokes rates as well as the high dimensionality of image data have so far prevented the development of direct image-based monitoring systems. For this reason, this paper demonstrates how high-resolution images of tools at 600 spm can be captured and subsequently processed using semantic segmentation deep learning algorithms, more precisely Fully Convolutional Networks (FCN). 125,000 images of the tool are taken from successive strokes. Selected images are labeled pixel by pixel according to their wear condition and used to train a FCN (U-Net). The approach presented offers the possibility of spatially monitoring wear in high-speed blanking operations in real-time.