Condition Monitoring Research Articles

Vibrational Analysis on Cooling Fan with Induced Mechanical Faults

Abstract This paper reports the investigation of different mechanical faults such as unbalanced loading, broken blades, and deformed blades induced in a fan using vibration analysis. The fan used in this study was a CPU cooling fan as a representation of a basic fan system giving an idea of vibration signatures when it is subjected to faults conditions. This study assists in developing the techniques of condition monitoring for the diagnosis of failures and faults in fan systems while logging the operating status and trend in performance. The investigation is carried out in the turbine testing lab using a Tri-axial accelerometer and a commercial Intel® CPU cooling fan consisting of 7 blades. The fan was run at different RPMs with normal and fault conditions like unbalanced loading conditions induced by adding mass of different weights, breakage of blades, deformation of blades, and holes in the blades. These faults are induced to simulate various mechanical faults that frequently occur in fan-based systems. The vibration data obtained from the accelerometer are filtered and the time series of acceleration values are plotted using MATLAB. For further analysis of data, the Fast Fourier Transform is done to evaluate the peaks of vibration signals in normal and fault conditions. Vibrational amplitude in fault conditions like Hole-2 has 4.5 times higher amplitude than normal and Cut-2 has 31.5 times higher amplitude.

DecouplingNet: A Stable Knowledge Distillation Decoupling Net for Fault Detection of Rotating Machines Under Varying Speeds.

Fault detection, also known as anomaly detection (AD), is at the heart of prediction and health management (PHM), which plays a vital role in ensuring the safe operation of mechanical equipment. Nonetheless, the lack of anomaly data creates a significant obstacle to the AD of the mechanical system. In particular, the complex modulation effects induced by time-varying speeds make AD much more challenging. For rapid and accurate AD, a stable knowledge distillation decoupling net (DecouplingNet) is provided to overcome these difficulties. First, an adversarial network consisting of an encoder, a decoder, and an encoder-discriminator is developed to model normal samples well by imposing constraints on the latent space. Then, a causal decoupling framework is suggested to disentangle equipment state-related information from operating conditions-related features, enabling stable condition monitoring at varying speeds. Finally, feature-based knowledge distillation is employed to boost the efficiency of AD while maintaining the detection accuracy. The proposed method is tested on two experimental scenarios and compared with some typical AD methods. The finding demonstrates that the net outperforms others in terms of accuracy and efficiency when it comes to detecting anomalies in the mechanical equipment that runs under varying speeds.

A comparative study of principal component analysis and kernel principal component analysis for photogrammetric shape-based turbine blade damage analysis

Photogrammetry is popular as a non-contact full-field data acquisition technique for machine condition monitoring purposes. Two popular implementations thereof are Digital Image Correlation (DIC) and 3D Point Tracking (3DPT). Since these two approaches require surface preparation in the form of applying speckle patterns or discrete markers, their use is negated in several industrial applications, such as the online condition monitoring and/or assessment of horizontal axis wind turbines. A shape-based photogrammetric approach that focuses on analysing changes in the form of a rotating blade’s boundary contour is a viable alternative that does not require any surface preparation. 3D shapes of a blade at a particular instance can be extracted from a pair of stereoscopic images of the blade. Shape Principal Component Descriptors (SPCDs) determined from applying Principal Component Analysis (PCA) to Fourier coefficients of chain-code shape signatures of the blade contour can be determined. These can be regarded as indicators of the form of the shape, and as the blade rotates, variations in the SPCDs can be investigated to better understand the dynamics of the blades. This paper focuses on the application and performance evaluation of data reduction and classification techniques for a novel shape-based photogrammetry condition monitoring technique. Investigations are conducted to show that applying data reduction methods to raw time domain SPCDs can result in successful classification of differently damaged blades. Post-processing strategies are developed to ensure a more robust shape based condition monitoring tool for analysing turbine blades. Kernel Principal Component Analysis (KPCA) is implemented and it is shown that it out-performs the conventional PCA in terms of classifying different blade damage modes. The feasibility of using multi-domain statistical features as feature vectors to which PCA or KPCA is applied for classification purposes is also investigated. Results indicating how well differently damaged blades can be distinguished are provided, and it is clearly illustrated that multi-domain statistical features are more robust to noise contamination in the signals compared to using the raw SPCDs data.

Research on degradation analysis and health condition assessment method of phase shifter

Abstract In high-power, high-reliability power supply systems, the switching operation of power devices is driven by phase shifters. Degradation or failure of the phase shifters leads to power device shoot-through and other failures, and reduces the power devices’ service lifetime. Aiming at the problem of phase shifter condition monitoring, a degradation model is proposed by analysing the degradation mechanism and aging process of the sensitive key components of the phase shifter, and then a health condition monitoring method with multi-dimensional features is presented in this paper. The multidimensional feature vectors in the time domain are extracted from the output voltage signals of different aging stages, and the time-domain feature separability of different health conditions is verified. Then the deep neural networks identification model based on deep learning technology is proposed to recognize the health condition of phase shifter. At last, the performance degradation simulation and experimental evaluation show that the proposed method can identify the phase shifter health conditions more accurately, and the accuracy rate can reach 95.892%.

Highly Robust and Self-Adhesive Soft Strain Gauge via Interface Design Engineering.

Highly robust soft strain gauges are rapidly emerging as a promising candidate in the fields of vital signs and machine conditions monitoring. However, it is still a key challenge to achieve high-performance strain sensing in these sensors with mechanical/electrical robustness for long-term usage. The multilayer structural design of sensors enhances sensing performance while the interfacial connection of heterogeneous materials between different layers is weak. Herein, inspired by the efficient perception mechanism of scorpion slit sensilla with tough interface interconnections, the synergy of ultra-high electrical performance and mechanical robustness is successfully achieved via interface design engineering. The developed multilayer soft strain gauge (MSSG) exhibits a strain sensitivity beyond 105, a lower detection limit of 8.3µm, a frequency resolution within 0.1Hz, and cyclic stability over 63000 strain cycles. Also, the tough interface improves the level of heterogeneous integration in the MSSG which allows to endure different stresses. Furthermore, an MSSG-based wireless strain monitoring system is developed that enables applications on different complex dynamic surfaces, including accurate identification of human throat activity and monitoring of rolling bearing conditions.

Harvesting Knowledge: Data Science and Machine Learning Techniques for Evaluating Pesticide Impact in Vegetable Organic Farming

Abstract: The integration of data science and machine learning is revolutionizing the assessment of pesticide impact in organic vegetable farming. This review explores methodologies, applications, and research examples showcasing the transformative potential of data-driven approaches. Remote sensing, including satellite imagery and drones, is essential for monitoring crop health and detecting pesticide impacts on vegetable crops like tomatoes, lettuce, and red peppers. By synthesizing research and trends, the review underscores technology's significance in informed decision-making for sustainable vegetable organic farming practices. Spectral analysis and vegetation indices quantify changes in crop health, informing pesticide efficacy and environmental impact. Sensor networks and IoT devices allow real-time monitoring of environmental conditions and pesticide dynamics, optimizing application practices to minimize contamination while maximizing yield. Machine learning, particularly decision tree-based models like random forests, predicts and mitigates pesticide impacts by analyzing complex datasets. Incorporating variables such as soil type and climate, these models accurately forecast pesticide fate, aiding in targeted mitigation strategies. Deep learning, such as convolutional neural networks (CNNs), identifies pesticide stress symptoms from digital images of vegetable leaves, facilitating rapid intervention. Challenges like data integration and model interpretability persist, yet ongoing research addresses these through data fusion and explainable AI. This review emphasizes the progress in leveraging data science and machine learning for pesticide impact evaluation in organic vegetable farming. By synthesizing research and trends, it offers insights for future sustainable agriculture applications.

Pan-europski kriteriji i indikatori održivog gospodarenja šumama: Institucionalni aspekti i mogućnost primjene u šumarstvu Federacije Bosne i Hercegovine

Overexploitation of natural resources have caused serious environmental issues, including climate change, biodiversity loss, habitat degradation, and soil and water pollution. The Pan-European criteria and indicators for sustainable forest management offer a tool to monitor progress and report on the sustainability of forest resources at sub-national, national, and regional levels. This paper examines the institutional aspects and possibilities of application of these criteria and indicators in the forestry sector of the Federation of Bosnia and Herzegovina (FBiH), aiming to improve forestry conditions and develop a coherent forest policy. A survey of 360 forestry experts from FBiH revealed that the majority support the positive impact of implementing the Pan-European criteria and indicators in the forestry sector. It was determined that the Ministry of Agriculture, Water Management, and Forestry of FBiH should be responsible for developing and collecting data related to these criteria and indicators. However, the primary barriers to implementation include a lack of financial resources, expertise, and commitment. While the public forest administration is formally prepared to apply these criteria, its current capacities are inadequate for effective implementation. Strengthening the capacity of the public forest administration is crucial to ensure the application of the Pan-European criteria and indicators for sustainable forest management. This would enable consistent and systematic monitoring and improvement of forest resource conditions and the overall state of the forestry sector in FBiH.

A novel deep learning framework for rolling bearing fault diagnosis enhancement using VAE-augmented CNN model

In the context of burgeoning industrial advancement, there is an increasing trend towards the integration of intelligence and precision in mechanical equipment. Central to the functionality of such equipment is the rolling bearing, whose operational integrity significantly impacts the overall performance of the machinery. This underscores the imperative for reliable fault diagnosis mechanisms in the continuous monitoring of rolling bearing conditions within industrial production environments. Vibration signals are primarily used for fault diagnosis in mechanical equipment because they provide comprehensive information about the equipment's condition. However, fault data often contain high noise levels, high-frequency variations, and irregularities, along with a significant amount of redundant information, like duplication, overlap, and unnecessary information during signal transmission. These characteristics present considerable challenges for effective fault feature extraction and diagnosis, reducing the accuracy and reliability of traditional fault detection methods. This research introduces an innovative fault diagnosis methodology for rolling bearings using deep convolutional neural networks (CNNs) enhanced with variational autoencoders (VAEs). This deep learning approach aims to precisely identify and classify faults by extracting detailed vibration signal features. The VAE enhances noise robustness, while the CNN improves signal data expressiveness, addressing issues like gradient vanishing and explosion. The model employs the reparameterization trick for unsupervised learning of latent features and further trains with the CNN. The system incorporates adaptive threshold methods, the "3/5" strategy, and Dropout methods. The diagnosis accuracy of the VAE-CNN model for different fault types at different rotational speeds typically reaches more than 90 %, and it achieves a generally acceptable diagnosis result. Meanwhile, the VAE-CNN augmented fault diagnosis model, after experimental validation in various dimensions, can achieve more satisfactory diagnosis results for various fault types compared to several representative deep neural network models without VAE augmentation, significantly improving the accuracy and robustness of rolling bearing fault diagnosis.

PENGEMBANGAN PERANGKAT MULTI SENSOR UNTUK PRECISION FARMING MONITORING MEDIA TANAM

The development of a multi-sensor device and the implementation of a data transmission system to a database for predicting plant suitability parameters are the focus of this paper. This technology enables accurate and continuous data collection for agricultural condition monitoring. The study utilizes a DHT22 sensor for air temperature and humidity, a pH sensor for soil pH, a soil moisture sensor for soil moisture, a rain gauge sensor for rainfall, and a raindrop sensor for rain detection. The data from these sensors are managed by an ESP32 microcontroller and transmitted to a Firebase database. The multi-sensor device was functionally tested and compared with valid reference sensor readings. The results show that all the multi-sensor devices and the recording system in the Firebase database functioned well. The average measurement error for the DHT22 temperature sensor was 1.15°C, with 2.9% for humidity, 0.72% for the pH sensor, and the soil moisture sensor showed 100% accuracy. This system can be applied as a parameter for predicting the suitability of plant types with soil and environmental conditions.

Cyclostationarity blind deconvolution via eigenvector screening and its applications to the condition monitoring of rotating machinery

Maximum second-order cyclostationarity blind deconvolution (CYCBD) is accomplished by maximizing the second-order cyclostationarity of signals through the indicator of second-order cyclostationarity (ICS2). It is a significant method for extracting weak periodic pulses related to bearing faults. However, since the interference spectral lines generated by the interference signals in squared envelope spectrum do not be considered by ICS2, the method can be invalid for certain applications. In addition, when the detected signal contains compound faults, only one fault can be enhanced, and misdiagnosis is easily caused. This article proposes cyclostationarity blind deconvolution via eigenvector screening, abbreviated as ESCYCBD, for fault feature enhancement and extraction of rotating machinery in order to solve these problems. Different from existing deconvolution methods which use the maximum value of the index as a criterion to select the filter coefficients, this method proposes the concept of eigenvector subspace, which contains the optimal eigenvector as filter coefficients that are ignored by CYCBD. Subsequently, the Fourier coefficients related to a series of cyclic frequencies are extended for compound fault signals. Besides, by using the harmonic characteristics of fault features in the envelope spectrum, fault feature recognition and optimal selection are carried out in eigenvector subspaces. At the same time, the concept of signals set is created, by which means different feature modes of different fault can be extracted at one time. The analysis results of simulation signals and experimental data show that the proposed ESCYCBD has robustness and accuracy in extracting fault feature modes, whether it is single fault diagnosis or compound fault diagnosis.

Intelligent diagnosis of magnetorheological clutch using image processing techniques

The clutch is one of the important components of a vehicle system. Therefore, it is crucial to keep an eye on them and spot any faults while they are operating so that the operators can be informed and possible issues can be solved before they arise. In general, a fault in the clutch causes it to get eccentricity in the disc and the contact between housing and disc during its operation which impacts the non-linear dynamic behaviour of the clutch mechanism and hence it affects the performance of the same. For condition monitoring, infrared imaging is essential since the thermographic patterns change based on the defect or machine condition. In this study, novel attempts have been made to present an intelligent tribological diagnosis of magnetorheological clutch under conditions such as eccentricity of the disc due to misalignment and the contact between housing and disc with the help of an image obtained from the thermal camera. The study for intelligent tribological diagnosis has been done by image processing algorithms in the MATLAB environment. The image segmentation for diagnosis is done by the thresholding technique followed by the feature extraction technique by the Circular Hough transform algorithm. The results successfully determine whether the eccentricity and the contact between the housing and disc due to misalignment are present or not, with the help of circular features in red and blue colour.

Carbon fibre detection in polymer composites reinforced by chopped carbon fibres through digital image processing and machine learning

Mechanical and thermodynamical properties and thus machinability of carbon fibre reinforced polymer composites significantly depend on the fibre orientation relative to the load direction. However, the orientations of the fibre groups in polymer composites reinforced by chopped carbon fibres are stochastic; therefore, the properties and machinability of such composites are challenging to plan, predict and optimise. We developed four different and novel approaches for fibre detection in polymer composites reinforced by chopped carbon fibres: (i) detecting the fibres through naked eye supported manual drawing, (ii) digital image processing of optical images, (iii) machine learning-based fibre detection, and (iv) rectangle fitting on the outputs of the automated processes using the Chaudhuri and Samal method. The applicability of the novel approaches was tested through optically captured images of polymer composites reinforced by chopped carbon fibres. The developed methods are each capable of detecting fibre groups at the top and bottom of the composite plate with certain limitations. The rectangle fitting approaches performed the best from the point of view of correctly identifying of fibre groups, followed by the machine learning-based and the conventional digital image processed, respectively. As a result of this study, the machining process planning and condition monitoring of polymer composites reinforced by chopped carbon fibres is more deeply supported.

Tool Condition Monitoring in the Milling Process Using Deep Learning and Reinforcement Learning

Tool condition monitoring (TCM) is crucial in the machining process to confirm product quality as well as process efficiency and minimize downtime. Traditional methods for TCM, while effective to a degree, often fall short in real-time adaptability and predictive accuracy. This research work aims to advance the state-of-the-art methods in predictive maintenance for TCM and improve tool performance and reliability during the milling process. The present work investigates the application of Deep Learning (DL) and Reinforcement Learning (RL) techniques to monitor tool conditions in milling operations. DL models, including Long Short-Term Memory (LSTM) networks, Feed Forward Neural Networks (FFNN), and RL models, including Q-learning and SARSA, are employed to classify tool conditions from the vibration sensor. The performance of the selected DL and RL algorithms is evaluated through performance metrics like confusion matrix, recall, precision, F1 score, and Receiver Operating Characteristics (ROC) curves. The results revealed that RL based on SARSA outperformed other algorithms. The overall classification accuracies for LSTM, FFNN, Q-learning, and SARSA were 94.85%, 98.16%, 98.50%, and 98.66%, respectively. In regard to predicting tool conditions accurately and thereby enhancing overall process efficiency, SARSA showed the best performance, followed by Q-learning, FFNN, and LSTM. This work contributes to the advancement of TCM systems, highlighting the potential of DL and RL techniques to revolutionize manufacturing processes in the era of Industry 5.0.

Improving Resilience in Water Distribution Systems: An Application of the House of Risk Method at PDAM Gowa Unit Tompobulu

PDAM Gowa, a local government-owned company, plays a vital role in supplying clean water to the Tompobulu community in Gowa District. However, ensuring efficient and reliable water distribution poses challenges at the local and regional levels. This research focuses on enhancing the resilience of the water distribution system to mitigate various risks. The study utilizes the House of Risk (HOR) method for effective risk management and developing suitable mitigation strategies within PDAM's framework. The research identifies 10 risk events and 17 risk factors that affect water distribution in PDAM. Key risks include pipe leakage and supply disruptions. To mitigate these risks, the study proposes nine actionable recommendations for PDAM Gowa Unit Tompobulu. These recommendations involve regular maintenance of distribution infrastructure and equipment to ensure system quality and reliability. Additionally, improving pipe network condition monitoring, implementing leakage control policies, and fostering collaboration with stakeholders to address water supply disruptions are crucial mitigation measures. This research significantly contributes to understanding the risk management associated with water distribution in PDAM. By implementing the recommended mitigation strategies, PDAM Gowa Unit Tompobulu can reduce water distribution risks, enhance system efficiency, and provide improved clean water services to consumers. Moreover, the study has the potential to drive scientific progress and promote the development of best practices and technologies in the drinking water industry.

Time-Series Feature Extraction by Return Map Analysis and Its Application to Bearing-Fault Detection

Developing new features for time-series characterization is a current challenge in data science and machine learning. In this paper, we propose a new metric based on a simple and efficient algorithm, namely, the return map. The return map analysis is well established in the field of non-linear dynamics, in particular, for fitting the parameters of a chaotic system from a waveform, or to attack a chaotic communication channel. We show that our metrics work well for both non-linear dynamics and time-series feature extraction problems in the field of machine learning. In an experiment aiming to classify vibration signals of normal and damaged bearings, we compare our method with two other methods that reported to have excellent accuracy, based on entropy and statistical feature distribution, respectively. We show that our method achieves higher accuracy with almost the lowest time costs, which was confirmed in experiments with two different datasets containing three main classes of bearings: normal, with inner race faults, and with outer race faults, having different damage origins and recorded in various conditions. In particular, for the dataset supplied by Case Western Reserve University, our method reached an accuracy of 100% at signals of 5000 sample points length, with a total time of 0.4 s required for feature estimation, while the entropy-based method reached an accuracy of 95% with a time of 100 s, and a statistical feature distribution method reached an accuracy of 93% with a total time of 1.9 s. Results show that the developed method is better suited to real-time bearing condition monitoring applications than most of the methods reported to date.

Low-Cost Microcontroller-Based System for Condition Monitoring of Permanent-Magnet Synchronous Motor Stator Windings

Low-Cost Microcontroller-Based System for Condition Monitoring of Permanent-Magnet Synchronous Motor Stator Windings

Acoustic signals analysis from an innovative UAV inspection system for wind turbines

Wind energy is emerging as a leading renewable energy source, with the deployment of large and more innovative wind turbines (WTs). This expansion requires new condition monitoring systems (CMS) and diagnostic techniques to reach competitiveness, improve reliability and availability, and minimize maintenance costs associated with WT operations. This research proposes a novel CMS based on acoustic analysis of WTs, combined with advanced analytics for pattern recognition. An acoustic CMS embedded in an unmanned aerial vehicle is developed to capture, send, and process the sound emitted in the nacelle to an acoustic receiver by a ground station for further analysis to assess the viability of the methodology. The article presents initial results from the laboratory using the fast Fourier transform algorithm, studying the signals in the time-frequency domain aspect and measuring the energy. An advanced signal processing method is presented to filter and define patterns that identify the state and condition of different proposed scenarios. The methodology is tested in a working WT, and the results demonstrate that the acoustic analysis is suitable for maintenance management in WTs.

On the condition monitoring of bolted joints through acoustic emission and deep transfer learning: generalization, ordinal loss, and super-convergence

On the condition monitoring of bolted joints through acoustic emission and deep transfer learning: generalization, ordinal loss, and super-convergence

Adaptive filter design for cyclostationary processes associated with sine-wave extraction operation

Cyclostationary signals have been generally used in condition monitoring and fault diagnosis of bearings, for which the adaptive filter is important in obtaining the signal-of-interest cyclostationary process from a disturbed signal. However, due to the nonstationarity of cyclostationary signals, their adaptive filters are usually time-varying and difficult to be implemented. In this study, a new optimum filter is obtained by converting the time-varying optimization function to be time-invariant using a sine-wave extraction (SE) operator, for which the adaptive implementation is designed and analyzed. The time-varying linear minimum mean-square error is converted to be time invariant using the SE operator, and its solution is time invariant but without explicit expression. Subsequently, an adaptive implementation, called the SE least mean square (SE-LMS), is introduced, for which the mean and weighted mean-square-deviation (MSD) are explicitly expressed. The scope of step size is determined to obtain a mean-convergent and mean-square-stable SE-LMS. The steady-state SE-operated excess mean-square-error is analyzed, and its explicit expression is obtained. The SE-LMS algorithm uses only a single cyclic frequency, based on which, the combined SE-LMS algorithm is introduced for a cyclostationary signal with multiple cyclic frequencies. Finally, several simulations are designed to verify the superiority of the proposed estimators over the time-average operator-based estimator.

Implementation of Condition Monitoring and Predictive Maintenance for Building Sustainability

Implementation of Condition Monitoring and Predictive Maintenance for Building Sustainability