Real-time Condition Monitoring Research Articles

Stages-Based ECG Signal Analysis From Traditional Signal Processing to Machine Learning Approaches: A Survey

Electrocardiogram (ECG) gives essential information about different cardiac conditions of the human heart. Its analysis has been the main objective among the research community to detect and prevent life threatening cardiac circumstances. Traditional signal processing methods, machine learning and its subbranches, such as deep learning, are popular techniques for analyzing and classifying the ECG signal and mainly to develop applications for early detection and treatment of cardiac conditions and arrhythmias. A detailed literature survey regarding ECG signal analysis is presented in this article. We first introduce a stages-based model for ECG signal analysis where a survey of ECG analysis related work is then presented in the form of this stage-based process model. The model describes both traditional time/frequency-domain and advanced machine learning techniques reported in the published literature at every stage of analysis, starting from ECG data acquisition to its classification for both simulations and real-time monitoring systems. We present a comprehensive literature review of real-time ECG signal acquisition, prerecorded clinical ECG data, ECG signal processing and denoising, detection of ECG fiducial points based on feature engineering and ECG signal classification along with comparative discussions among the reviewed studies. This study also presents a detailed literature review of ECG signal analysis and feature engineering for ECG-based body sensor networks in portable and wearable ECG devices for real-time cardiac status monitoring. Additionally, challenges and limitations are discussed and tools for research in this field as well as suggestions for future work are outlined.

Classification of catenary equipment based on a coupled genetic algorithm-extreme learning machine method

In an online monitoring system for an electrified railway, it is important to classify the catenary equipment successfully. The extreme learning machine (ELM) is an effective image classification algorithm and the genetic algorithm (GA) is a typical optimisation method. In this paper, a coupled genetic algorithm-extreme learning machine (GA-ELM) technique is proposed for the classification of catenary equipment. Firstly, the GA is used to search for optimal features by reducing the initial multi-dimensional features to low-dimensional features. Next, the optimised features are used as the input to the ELM. The ELM algorithm is then used to classify the catenary equipment. In this process, the impacts of the activation function, the number of hidden layer neurons and different models on the performance of the ELM are discussed in turn. Finally, the proposed method is compared with traditional methods in terms of classification accuracy and efficiency. Experimental results show that the number of feature dimensions decreases to 58% of the original number and the computational complexity is greatly decreased. Moreover, the reduced features and the few steps of the ELM improve the classification accuracy and speed. Noticeably, when the performance of the GA-ELM method is compared with that of the ELM method, the classification accuracy rate is 93.33% compared with 85.83% and the time consumption is 2.25 s compared with 8.85 s, respectively. That is to say, the proposed method not only decreases the number of features but also increases the classification accuracy and efficiency. This meets the needs of a real-time online condition monitoring system.

Recent advancements in microfluidics that integrate electrical sensors for whole blood analysis.

Whole blood analysis reveals crucial information about various physiological and pathological conditions, including cancer metastasis, infection, and immune status, among others. Despite this rich information, the complex composition of whole blood usually required multiple sample preparation steps to purify targeted analytes. Traditionally, whole blood preparation processes, including centrifugation, lysis, dilution, or staining, are usually manually operated by well-trained technicians using bench-top instruments. This preparation can require a large blood volume and cannot be directly integrated with detection systems. Recently, various studies have integrated microfluidics with electrical sensors for whole blood analysis, with a focus on cell-based analysis, such as cell type, number, morphology, phenotype, and secreted molecules. These miniaturized systems require less sample and shorter reaction times. Besides, the sample processing and analysis can be fully integrated and automated with minimal operations. We believe these systems can transfer the current whole blood analysis from hospitals or laboratories into clinics or home settings to enable real-time and continuous health condition monitoring in point-of-care settings.

Data-Driven Incipient Fault Prediction for Non-Stationary and Non-Linear Rotating Systems: Methodology, Model Construction and Application

Many researches have been carried out on incipient fault prediction technology for key machine components (such as bearings) based on historical and real-time condition monitoring data. However, there is still lack of well-understood systematic methodologies for detecting incipient fault for rotating machines. Based on machine learning technology, this paper studies an incipient fault prediction model applying with wavelet packet decomposition and dynamic kernel principal component analysis (WPD-DKPCA) to meet the needs of engineering applications. The incipient fault prediction WPD-DKPCA model, which does not require knowledge on equipment structure and failure mechanisms, only requires normal state data of the machine, and incipient fault prediction can be achieved through self-learning. Run-to-failure experimental data and engineering case data have been used to verify the constructed model, and the verification results show that the constructed model can reliably and accurately detect an incipient bearing fault. Comparisons of fault prediction effects prove that using T2 statistic monitoring can detect upcoming faults of machines much earlier than Kurtosis and Root Mean Square (RMS).

Remaining Useful Life Estimation of BLDC Motor Considering Voltage Degradation and Attention-Based Neural Network

Brushless DC motor, also referred to as BLDC motor, has been a widely used electric machine due to its excellent performance over conventional DC motors. Due to complex operating conditions and overloading, several irregularities can take place in a motor. Stator related faults are among the most commonly occurring faults in BLDC motor. With an initial raise in local heating, a fault in the stator can largely reduce motor efficiency and account for the entire system breakdown. In this study, we present a deep learning-based approach to estimate the remaining useful life (RUL) of BLDC motor affected by different stator related faults. To analyze the motor health degradation, we have investigated two types of stator faults namely inter-turn fault (ITF) and winding short-circuit fault (WSC). A generator was coupled with the motor and using an average value rectifier (AVR), generator's output voltage was monitored for the entire lifecycle. A proven neural network for effective sequence modeling, recurrent neural network (RNN) is selected to train the voltage degradation data. For a better estimation of nonlinear trends, long-short term memory (LSTM) with attention mechanism is chosen to make predictions of the motor RUL for both types of faults. The main concern that encourages authors of this paper is the proposed method can be used for the real-time condition monitoring and health state estimation of BLDC motors. Also, the proposed AVR-LSTM method is not affected by environmental influences, making it suitable for diverse operating conditions.

Application of Machine Learning and Vision for real-time condition monitoring and acceleration of product development cycles

Development work within an experimental environment, in which certain properties are investigated and optimized, requires many test runs and is therefore often associated with long execution times, costs and risks. This can affect product, material and technology development in industry and research. New digital driver technologies offer the possibility to automate complex manual work steps in a cost-effective way, to increase the relevance of the results and to accelerate the processes many times over. In this context, this article presents a low-cost, modular and open-source machine vision system for test execution and evaluates it on the basis of a real industrial application. For this purpose a methodology for the automated execution of the load intervals, the process documentation and for the evaluation of the generated data by means of machine learning to classify wear levels. The software and the mechanical structure are designed to be adaptable to different conditions, components and for a variety of tasks in industry and research. The mechanical structure is required for tracking the test object and represents a motion platform with independent positioning by machine vision operators or machine learning. An evaluation of the state of the test object is performed by the transfer learning after the initial documentation run. The manual procedure for classifying the visually recorded data on the state of the test object is described for the training material. This leads to an increased resource efficiency on the material as well as on the personnel side since on the one hand the significance of the tests performed is increased by the continuous documentation and on the other hand the responsible experts can be assigned time efficiently. The presence and know-how of the experts are therefore only required for defined and decisive events during the execution of the experiments. Furthermore, the generated data are suitable for later use as an additional source of data for predictive maintenance of the developed object.

A Real-Time Condition Monitoring and Maintenance Management System for Low Voltage Industrial Motors Using Internet-of-Things

Maintenance and sound operating industrial equipment are critical for any manufacturing company. Standardization of the manufacturing infrastructure and establishment of a systematic maintenance program is essential for this process. However, condition monitoring must also be an integral part of a smart manufacturing program that seeks to improve and optimize the operational efficiency of production systems. Absence of this type of data-driven observations and mindset in manufacturing decision making, may result in safety risks, missing critical signs, or occurrence of unexpected repairs that could bring the equipment to a halt. Moreover, routine maintenance may short the actual useful life of some equipment and when scheduled maintenance shutdowns occur too often, they increase downtime and cost.This paper presents the design and development of a real-time condition monitoring system for managing industrial low voltage motors using internet-of-things. The system can record and monitor vibration and temperature conditions of an industrial motor and transmit the data through a wireless network to a data logging center. The current prototype was developed using open source software and hardware and can successfully identify abnormal motor conditions from sensor input values that exceed predefined setpoints. When a motor is approaching an abnormal condition, the system changes state and informs the user through a mobile alert. The motor management system requires the user to perform an RFID enforced inspection. The alarm system remains active until authorized personnel visits the equipment and scans the dedicated RFID batch. Furthermore, the motor management system can be accessed remotely to allow the user to visualize the current condition of the motor. Finally, we present a pilot experiment that was conducted to test the condition monitoring capabilities of the prototype and we discuss future steps in order to further develop the current system to a smart predictive maintenance system capable of detecting and predicting specific motor faults.Click here to enter text.

Causal Modelling for Predicting Machine Tools Degradation in High Speed Production Process

A dynamic health indicator based on regressive event-tracker algorithm is proposed to accurately interpret the condition of critical components of machine tools in a production system and to predict their potential sudden breakdown based on future trends. Through sensors/actuators data acquisition, the algorithm predicts the causal links between various monitored parameters of the system and offers a diagnosis of the health state of the system. A safety and operational robustness regime determines the acceptable thresholds of the operational boundaries of the electro-mechanical components of the machines. The proposed model takes into account the possibilities of sensor values being a piecewise-linear models or a pair of exponential functions with restricted model parameters, which can predict the runs-to-failure or remaining useful life until a safety threshold. The events caused by sensors passing through sub levels of safety threshold are used as a re-enforcement learning for the models. Each remaining useful life estimation diagnosis and prognosis analysis can be conducted on individual or an interconnected network of components within a machine. The overall health indicator based on individual useful life estimation is calculated by deriving the weights from event-clustering algorithm. The work can be extended to a network of machines representing a process. The outcome of the continuously learning real-time condition monitoring modus-operandi is to accurately measure the remaining useful life of the network of critical components of a machine.

Synchrophasor-Based Condition Monitoring of Instrument Transformers Using Clustering Approach

During the life span of an instrument transformers (IT), non-catastrophic and typically undetected problems may degrades the overall performance of the instrumentation system. Synchrophasor provides high resolution synchronized data compared to the Supervisory Control and Data Acquisition (SCADA) systems. In the proposed non-intrusive method, time-series Phasor Measurement Units (PMUs) data are processed using three step approach for real time condition monitoring of IT at the control center. Bad data are filtered or flagged using MLE based ensemble method before applying density-based spatial clustering of applications with noise (DBSCAN) to detect clusters, changing over time within each time window. In the third step, a decision logic diagram is developed using additional data and system information to categorize the observed anomalies into different types of malfunction, failure, event or degradation. The proposed method has been validated using IEEE test system modeled in the Real Time Digital Simulator (RTDS) interfaced with industrial hardware PMU device. Detecting equipment failures at an early stage will result in improved system reliability and reduced maintenance cost. Several test cases demonstrate superiority of the proposed method for detecting incipient failures in the IT, and provides early warning mechanism for preventing full blown faults resulting in major impact on the power grid.

Photovoltaic Cell/Module Equivalent Electric Circuit Modeling Using Impedance Spectroscopy

This work is focused on the dynamic alternating current equivalent electric circuit (AC-EEC) modeling of the polycrystalline silicon wafer-based photovoltaic cell and module under various operational and fault conditions. The models are drawn from the impedance changes observed using electrochemical impedance spectroscopy. Vital considerations for valid impedance data and appropriate parameter extraction were also discussed. In this work, cell-level tests were carried out at different voltage bias levels under illumination, dark, uniform partial shading, and cell-mismatch conditions. The cell characterization procedure was then used as a basis for developing the photovoltaic (PV) module level model. Normal illumination and partial shading tests were carried out at the module/panel level. The AC-EEC model for both the levels was obtained from the associated Nyquist and Bode plots. These models were then used to extract the internal cell/module parameters and the associated trends were observed. Results obtained demonstrates that the variations of the ac model give several parameters that show distinct variation such that they can be used for inline characterization and real-time condition monitoring of the photovoltaic cells/module under the different operational and fault conditions.

Inverse Thermal Identification of a Thermally Instrumented Induction Machine Using a Lumped-Parameter Thermal Model

Accurate temperature estimation inside an electrical motor is key for condition monitoring, fault detection, and enhanced end-of-life duration. Additionally, thermal information can benefit motor control to improve operational performance. Lumped-parameter thermal networks (LPTNs) for electrical machines are both flexible and cost-effective in computation time, which makes them attractive for use in real-time condition monitoring and integration in motor control. However, the accuracy of these thermal networks heavily depends on the accuracy of its system parameters, some of which are difficult to calculate analytically or even empirically and need to be determined experimentally. In this paper, a methodology for the thermal condition monitoring of long-duration transient and steady-state temperatures in an induction motor is presented. To achieve this goal, a computationally efficient second-order LPTN for a 5.5 kW squirrel-cage induction motor is proposed to apprehend the dominant heat paths. A fully thermally instrumented induction motor has been prepared to collect spatial and temporal temperature information. Using the experimental stator and rotor temperature data collected at different motor operating speeds and torques, the key thermal parameter values in the LPTN are identified by means of an inverse methodology that aligns the simulated temperatures of the stator windings and rotor with the corresponding measured temperatures. Validation results show that the absolute average thermal modelling error does not exceed 1.45 °C with maximum absolute error of 2.10 °C when the motor operates at fixed speed and torque. During intermittent motor-loading operation, a mean (maximum) stator temperature error of 0.38 °C (0.92 °C) was achieved and mean (maximum) rotor errors of 2.11 °C (3.40 °C). These results show the validity of the proposed thermal model but also its ability to predict in real time the temperature variations in stator and rotor for condition monitoring and motor control.

Multi-Channel Real-Time Condition Monitoring System Based on Wideband Vibration Analysis of Motor Shafts Using SAW RFID Tags Coupled with Sensors.

While there is a wide range of approaches to monitor industrial machinery through their static components, rotating components are usually harder to monitor, since sensors are difficult to be mounted on them and continuously read during operation. However, the characteristics of rotating components may provide useful information about the machine condition to be included in monitoring algorithms, specially for long-term data analysis. In this work, wireless vibration monitoring of rotating machine parts is investigated using surface acoustic wave (SAW) radio frequency identification (RFID) tags coupled with sensors. The proposed augmented transponder solution, combined with low-latency interrogation and signal processing, enables real-time identification and wideband vibration sensing. On top of that, a multi-channel interrogation approach is used to compensate motion effects. This approach enhances the signal-to-noise ratio of low-power high-frequency components present on the vibration signatures and enables discriminant information extraction from rotating machine parts. Final feasibility is evaluated with induction motors and vibration measurements on rotating shafts are verified. In addition, a condition classification algorithm is implemented in an experimental setup based on different motor states. The results of this work open the possibility to feed predictive maintenance algorithms using new features extracted in real-time from wideband vibration measurements on rotating components.

Condition Monitoring of Electrolytic Capacitors in Boost Converters by Magnetic Sensors

The aluminum electrolytic capacitors (AEC) are important in filtering and energy storage applications since they are of high energy density and low cost. However, the AECs suffer from low reliability and limited lifetime due to the electrolyte vaporization. The degradation of AECs challenges the reliability and efficiency of the power electronic systems. Therefore, it is essential to monitor the condition of AECs for system reliability and efficiency. In this paper, a real-time condition monitoring technique for input and output AECs based on the estimation of equivalent series resistance (ESR) was developed for the boost converter. The ESRs were estimated by using the capacitor and inductor current obtained through the magnetic-field sensing by magnetic sensors. The influences of the load, duty cycle, inductance, ESR, and capacitance on the ESR estimation were investigated to develop the proposed methodology. The proposed methodology can largely improve the accuracy of the estimated ESR of output AEC in the boost converter. The higher accuracy enables predictive maintenance of AECs and contributes to the reliability and efficiency of power electronic systems. The simulation and experimental results from −25 °C to 100 °C verified that the proposed methodology could be used to estimate the ESR of input and output AECs accurately. The experimental results showed that the compact tunnel magnetoresistance (TMR) sensors were capable of measuring the capacitor and inductor current for the estimation of ESR. This technique is cost-effective and non-invasive since it can be implemented with TMR magnetic sensors and does not require the Hall-effect current transducers with magnetic concentrators of larger volume, higher cost, and limited bandwidth or invasive shunt resistors.

A novel weight coefficient calculation method for the real‐time state monitoring of the lithium‐ion battery packs under the complex current variation working conditions

AbstractA novel real‐time state monitoring method is proposed to realize the real‐time energy management of the lithium‐ion battery packs, which is conducted in the iterative computational calculation process by introducing an improved weighting factor‐unscented Kalman filtering algorithm. The accurate state monitoring treatment is investigated by applying a new iterate calculation thought, in which the improved weight coefficient parameter is constructed and its numerical stability is improved. Meanwhile, the recursive calculation is derived by using the real‐time measured factors, according to which the state‐of‐charge estimation is realized accurately. Aiming to adapt the complex current variation working conditions, the nonlinear treatment is introduced to construct the mathematical unscented transforming function. As can be known from the experimental results, the state‐of‐charge estimation accuracy is 98.34% under the complex current charge‐discharge working conditions. Meanwhile, the effective closed‐circuit voltage trackage is also investigated accurately and its tracking error is within 3.51% in the complex working conditions, which provides a good security guarantee for the reliable energy supply of the lithium‐ion battery packs.

Evaluation of Early Fatigue Signatures in Lightweight Aluminum Alloy 7075

The current work investigates early material changes that manifest in aluminum alloy 7075 subjected to fatigue loading. Tension-tension specimens linearly tapered in width were exposed to cyclic loading with compliance and eddy current measurements made in-situ, while nano-indentation, surface potential, x-ray diffraction, and ultrasonic response were made ex-situ. Early stage damage progression was evaluated at various size scales and used to estimate a remaining life in real time. Baseline, in-situ, and post failure measurements were compared and analyzed as a function of fatigue cycles. Eddy current and compliance measurements showed a strong correlation between signal strength and fatigue life with eddy current providing a more precise signal per specimen. Microscale electrical measurements showed a link between surface potential and regions of the specimen that experienced higher tensile stress. Results are discussed with an emphasis on multiscale damage detection and identification due to fatigue cycling, while highlighting the most promising techniques for improving structural durability and real-time state monitoring.

Online Condition Monitoring for Diagnosis and Prognosis of Insulation Degradation of Inverter-Fed Machines

Real-time winding insulation condition monitoring is becoming an important research topic in response to increase in high availability and reliability demands in modern drives. In this paper, a novel method is proposed to monitor ground-wall insulation of low-voltage inverter-fed machines based on a multifrequency measurement of equivalent insulation ground-wall capacitance and dissipation factor. The monitoring is applied to four machine stator samples subjected to accelerated aging until failure. Insulation degradation is continuously tracked via the capacitance and dissipation parameters. A link between ground-wall insulation capacitance and final lifetime is established. The relationship between capacitance progression and its value at the final machine failure is used to develop a method for the prognosis of the final failure time.

Real time condition monitoring of hydraulic brake system using naive bayes and bayes net algorithms

The vehicles usage is increasing day by day due to the recent, technological development in the automotive field. In a competitive global market in order to survive, the reliability needs to be ensured, through a proper monitoring system. Brake system is one such control component, in which much focus is very much essential. An efficient brake system should provide reliable and effective performance in order to ensure the safety. If it is not properly monitored, it may lead to a serious catastrophic effect such as accidents, brake down, etc. Hence, the brake system needs to be monitored continuously. In this study, an experimental investigation was carried out for monitoring the brake system using vibration signals. An experimental setup which resembles the brake system was fabricated. The vibration signals were acquired under various brake condition such as good and faulty. From the acquired vibration signals, the features were extracted using statistical and histogram feature extraction techniques and feature selection was carried out. The selected features were then classifieds using a Naive Bayes and Bayes Net a classifier. The classification accuracy of all the algorithms were compared for finding the best feature classifier model for monitoring the brake condition.

PERFORMANCE OPTIMIZATION OF 500MW STEAM TURBINE BY CONDITION MONITORING TECHNIQUE USING VIBRATION ANALYSIS METHOD

In this paper, description of vibration analysis method used for condition monitoring of 500MW steam turbine (Kraftwork Union, Germany) has been discussed. The importance and critical role of this technique in predictive maintenance used in thermal power plant is evaluated based on possible type of failures that can be detected at early stage before any unexpected, unscheduled breakdown during operation of steam turbine. Efforts are being made for most of the steam turbines in power generation thermal plants to run or operate them at Ideal condition by regular technical consultations resulting in continuous generation of power and reasonable increase in operational life span of these turbines. This paper deals with one such effort. In this paper real time condition monitoring of 500MW turbine, which was carried out by them at NTPC-Ramagundum power plant, using vibration analysis, has been discussed without disturbing its operational working condition and the causes for increase in the vibrations and its practical diagnosis by spectrum analysis, Interpretations and recommended steps to be taken to minimize the vibrations.

An Industrial Digitalization Platform for Condition Monitoring and Predictive Maintenance of Pumping Equipment.

This paper is concerned with the implementation and field-testing of an edge device for real-time condition monitoring and fault detection for large-scale rotating equipment in the UK water industry. The edge device implements a local digital twin, processing information from low-cost transducers mounted on the equipment in real-time. Condition monitoring is achieved with sliding-mode observers employed as soft sensors to estimate critical internal pump parameters to help detect equipment wear before damage occurs. The paper describes the implementation of the edge system on a prototype microcontroller-based embedded platform, which supports the Modbus protocol; IP/GSM communication gateways provide remote connectivity to the network core, allowing further detailed analytics for predictive maintenance to take place. The paper first describes validation testing of the edge device using Hardware-In-The-Loop techniques, followed by trials on large-scale pumping equipment in the field. The paper concludes that the proposed system potentially delivers a flexible and low-cost industrial digitalization platform for condition monitoring and predictive maintenance applications in the water industry.

The real-time condition monitoring system of gapless arrester based on ZigBee protocol and third harmonic leakage current as indicator parameters

Monitoring the condition of arrester in service have been presented in the past. Several methods of monitoring the condition of surge arrester were introduced by many researchers. Some and almost all the methods are based on measuring the leakage current. The goal of this study is to design a portable arrester condition monitoring system which is capable of performing arrester health analysis without the need for a voltage source reference — the design based on the previous research which introduces the new method to discriminant the resistive leakage current from total leakage current. The method for diagnosing arrester health is only with the leakage current waveform. The advantage of not needing the voltage reference is that the system can ultimately be used in the field where access a voltage reference about 150 kV is not practical. The design employs a ZigBee wireless protocol as well as a microprocessing based system utilizing an Arduino microcontroller. The ground side current from the arrester is fed through a resistor, which can then be used as a read voltage. The LabVIEW based graphical user interface was prepared to analyze an arrester’s data. The user connects the GUI to the Arduino via ZigBee, samples the waveform from an arrester and then additional data processing is executed. The main part of this processing involves the modified shifted current method on the sampled waveform. Upon conclusion, the user is left with only the resistive component of the arrester leakage current, which is displayed in both the time and frequency domains for analysis.