Towards a Distributed Digital Twin Framework for Predictive Maintenance in Industrial Internet of Things (IIoT)

This study uses a wind turbine case study as a subdomain of Industrial Internet of Things (IIoT) to showcase an architecture for implementing a distributed digital twin in which all important aspects of a predictive maintenance solution in a DT use a fog computing paradigm, and the typical predictive maintenance DT is improved to offer better asset utilization and management through real-time condition monitoring, predictive analytics, and health management of selected components of wind turbines in a wind farm. Digital twin (DT) is a technology that sits at the intersection of Internet of Things, Cloud Computing, and Software Engineering to provide a suitable tool for replicating physical objects in the digital space. This can facilitate the implementation of asset management in manufacturing systems through predictive maintenance solutions leveraged by machine learning (ML). With DTs, a solution architecture can easily use data and software to implement asset management solutions such as condition monitoring and predictive maintenance using acquired sensor data from physical objects and computing capabilities in the digital space. While DT offers a good solution, it is an emerging technology that could be improved with better standards, architectural framework, and implementation methodologies. Researchers in both academia and industry have showcased DT implementations with different levels of success. However, DTs remain limited in standards and architectures that offer efficient predictive maintenance solutions with real-time sensor data and intelligent DT capabilities. An appropriate feedback mechanism is also needed to improve asset management operations.

Prediction of Degraded Infrastructure Conditions for Railway Operation

In the railway sector, rolling stock and infrastructure must be maintained in perfect condition to ensure reliable and safe operation for passengers. Climate change is affecting the urban and regional infrastructure through sea level rise, water accumulations, river flooding, and other increased-frequency extreme natural situations (heavy rains or snows) which pose a challenge to maintenance. In this paper, the use of artificial intelligence based on predictive maintenance implementation is proposed for the early detection of degraded conditions of a bridge due to extreme climatic conditions. For this prediction, continuous monitoring is proposed, with the aim of establishing alarm thresholds to detect dangerous situations, so restrictions could be determined to mitigate the risk. However, one of the main challenges for railway infrastructure managers nowadays is the high cost of monitoring large infrastructures. In this work, a methodology for monitoring railway infrastructures to define the optimal number of transductors that are economically viable and the thresholds according to which infrastructure managers can make decisions concerning traffic safety is proposed. The methodology consists of three phases that use the application of machine learning (Random Forest) and artificial cognitive systems (LSTM recurrent neural networks).

Incremental Clustering for Predictive Maintenance in Cryogenics for Radio Astronomy

In a cooling system for radio astronomy receivers, maintaining cold heads and compressors is essential for consistent performance. This project focuses on monitoring the power currents of the cold head's motor to address potential mechanical deterioration, which could jeopardize the overall functionality of the system. Using Hall effect sensors, a microcontroller-based electronic board, and artificial intelligence, the system detects and predicts anomalies. The model operates using an unsupervised approach based on incremental clustering. Since potential fault scenarios can be multiple and often challenging to simulate or identify during training, the system is initially trained using known operational categories. Over time, the system adapts and evolves by incorporating new data, which can be assigned to existing categories or, in the case of new anomalies, form new categories. This incremental approach enables the system to enhance its performance over the years, adapting to new anomaly scenarios and ensuring precise and reliable monitoring of the cold head's health.

A Vibration Analysis for the Evaluation of Fuel Rail Pressure and Mass Air Flow Sensors on a Diesel Engine: Strategies for Predictive Maintenance

This research focuses on the analysis of vibration of a compression ignition engine (CIE), specifically examining potential failures in the Fuel Rail Pressure (FRP) and Mass Air Flow (MAF) sensors, which are critical to combustion control. In line with current trends in mechanical system condition monitoring, we are incorporating information from these sensors to monitor engine health. This research proposes a method to validate the correct functioning of these sensors by analysing vibration signals from the engine. The effectiveness of the proposal is confirmed using real data from a Common Rail Direct Injection (CRDi) engine. Simulations using a GT 508 pressure simulator mimic FRP sensor failures and an adjustable potentiometer manipulates the MAF sensor signal. Vibration data from the engine are processed in MATLAB using frequency domain techniques to investigate the vibration response. The results show that the proposal provides a basis for an efficient predictive maintenance strategy for the MEC engine. The early detection of FRP and MAF sensor problems through a vibration analysis improves engine performance and reliability, minimizing downtime and repair costs. This research contributes to the advancement of monitoring and diagnostic techniques in mechanical engines, thereby improving their efficiency and durability.

Remaining Useful Life Estimation of MoSi2 Heating Element in a Pusher Kiln Process

The critical challenge of estimating the Remaining Useful Life (RUL) of MoSi2 heating elements utilized in pusher kiln processes is to enhance operational efficiency and minimize downtime in industrial applications. MoSi2 heating elements are integral components in high-temperature environments, playing a pivotal role in achieving optimal thermal performance. However, prolonged exposure to extreme conditions leads to degradation, necessitating precise RUL predictions for proactive maintenance strategies. Since insufficient failure experience deals with Predictive Maintenance (PdM) in real-life scenarios, a Generative Adversarial Network (GAN) generates specific training data as failure experiences. The Remaining Useful Life (RUL) is the duration of the equipment's operation before repair or replacement, often measured in days, miles, or cycles. Machine learning models are trained using historical data encompassing various operational scenarios and degradation patterns. The RUL prediction model is determined through training, hyperparameter tuning, and comparisons based on the machine-learning model, such as Long Short-Term Memory (LSTM) or Support Vector Regression (SVR). As a result, SVR reflects the actual resistance variation, achieving the R-Square (R2) of 0.634, better than LSTM. From a safety perspective, SVR offers high prediction accuracy and sufficient time to schedule maintenance plans.

Cloud-Based Machine Learning Methods for Parameter Prediction in Textile Manufacturing

In traditional textile manufacturing, downstream manufacturers use raw materials, such as Nylon and cotton yarns, to produce textile products. The manufacturing process involves warping, sizing, beaming, weaving, and inspection. Staff members typically use a trial-and-error approach to adjust the appropriate production parameters in the manufacturing process, which can be time consuming and a waste of resources. To enhance the efficiency and effectiveness of textile manufacturing economically, this study proposes a query-based learning method in regression analytics using existing manufacturing data. Query-based learning allows the model training to evolve its decision-making process through dynamic interactions with its solution space. In this study, predefined target parameters of quality factors were first used to validate the training results and create new training patterns. These new patterns were then imported into the solution space of the training model. In predicting product quality, the results show that the proposed query-based regression algorithm has a mean squared error of 0.0153, which is better than those of the original regression-related methods (Avg. mean squared error = 0.020). The trained model was deployed as an application programing interface (API) for cloud-based analytics and an extensive auto-notification service.

Monitoring of Rubber Belt Material Performance and Damage

Conveyors play a very important role in modern manufacturing processes, and one of the most popular types is the belt conveyor. The main elements of a conveyor include a conveyor belt, roller sets, a supporting frame and a drive and control system. The reliable operation of the conveyor depends on the strength and durability of individual elements (especially the belt). Conveyor belts are made from various materials and have received a lot of attention in the scientific and research community. This article presents tests of the strength of the rubber belt material and its damage under load. The belt consists of two internal layers covered with a PVC coating on the outside, and the nominal belt thickness was 2 mm. In the experiment, various configurations of longitudinal and transverse damage were verified, and statistical methods were used to analyze the results. The obtained test results provided a new understanding of the propagation of conveyor belt damage and helped to improve the strain gauge-based monitoring system.

An In-Depth Study of Vibration Sensors for Condition Monitoring

Heavy machinery allows for the efficient, precise, and safe management of large-scale operations that are beyond the abilities of humans. Heavy machinery breakdowns or failures lead to unexpected downtime, increasing maintenance costs, project delays, and leading to a negative impact on personnel safety. Predictive maintenance is a maintenance strategy that predicts possible breakdowns of equipment using data analysis, pattern recognition, and machine learning. In this paper, vibration-based condition monitoring studies are reviewed with a focus on the devices and methods used for data collection. For measuring vibrations, different accelerometers and their technologies were investigated and evaluated within data collection contexts. The studies collected information from a wide range of sources in the heavy machinery. Throughout our review, we came across some studies using simulations or existing datasets. We concluded in this review that due to the complexity of the situation, we need to use more advanced accelerometers that can measure vibration.

Enhancing Aircraft Safety through Advanced Engine Health Monitoring with Long Short-Term Memory

Predictive maintenance holds a crucial role in various industries such as the automotive, aviation and factory automation industries when it comes to expensive engine upkeep. Predicting engine maintenance intervals is vital for devising effective business management strategies, enhancing occupational safety and optimising efficiency. To achieve predictive maintenance, engine sensor data are harnessed to assess the wear and tear of engines. In this research, a Long Short-Term Memory (LSTM) architecture was employed to forecast the remaining lifespan of aircraft engines. The LSTM model was evaluated using the NASA Turbofan Engine Corruption Simulation dataset and its performance was benchmarked against alternative methodologies. The results of these applications demonstrated exceptional outcomes, with the LSTM model achieving the highest classification accuracy at 98.916% and the lowest mean average absolute error at 1.284%.

Machine Learning for the Detection and Diagnosis of Anomalies in Applications Driven by Electric Motors

Manufacturing systems are becoming increasingly flexible, necessitating the adoption of new technologies that allow adaptations to a turbulent and complex modern market. Consequently, modern concepts of production systems require horizontal and vertical integration, extending across value networks and within a factory or production shop. The integration of these environments enables the acquisition of a substantial amount of data containing information pertaining to production, processes, and equipment located on the shop floor. When these data and information are processed and analyzed, they have the potential to reveal valuable insights and knowledge about the manufacturing systems, offering interpretive outcomes for strategic decision making. One of the opportunities presented in this context includes the implementation of predictive maintenance (PdM). However, industrial adoption of PdM is still relatively low. In this paper, the aim is to propose a methodology for selecting the main attributes (variables) to be considered in the instrumentation setup of rotating machines driven by electric motors to decrease the associated costs and the time spent defining them. For this, the most well-known data science and machine learning algorithms are investigated to choose the one most adequate for this task. For the experiments, different testing scenarios were proposed to detect the different possible types of anomalies, such as uncoupled, overloaded, unbalanced, misaligned, and normal. The results obtained show how these algorithms can be effective in classifying the different types of anomalies and that the two models that presented the best accuracy values were k-nearest neighbor and multi-layer perceptron.

A Robot-Operation-System-Based Smart Machine Box and Its Application on Predictive Maintenance

Predictive maintenance is a proactive approach to maintenance in which equipment and machinery are monitored and analyzed to predict when maintenance is needed. Instead of relying on fixed schedules or reacting to breakdowns, predictive maintenance uses data and analytics to determine the appropriate time to perform maintenance activities. In industrial applications, machine boxes can be used to collect and transmit the feature information of manufacturing machines. The collected data are essential to identify the status of working machines. This paper investigates the design and implementation of a machine box based on the ROS framework. Several types of communication interfaces are included that can be adopted to different sensor modules for data sensing. The collected data are used for the application on predictive maintenance. The key concepts of predictive maintenance include data collection, a feature analysis, and predictive models. A correlation analysis is crucial in a feature analysis, where the dominant features can be determined. In this work, linear regression, a neural network, and a decision tree are adopted for model learning. Experimental results illustrate the feasibility of the proposed smart machine box. Also, the remaining useful life can be effectively predicted according to the trained models.

Low-Latency Collaborative Predictive Maintenance: Over-the-Air Federated Learning in Noisy Industrial Environments

The emergence of Industry 4.0 has revolutionized the industrial sector, enabling the development of compact, precise, and interconnected assets. This transformation has not only generated vast amounts of data but also facilitated the migration of learning and optimization processes to edge devices. Consequently, modern industries can effectively leverage this paradigm through distributed learning to define product quality and implement predictive maintenance (PM) strategies. While computing speeds continue to advance rapidly, the latency in communication has emerged as a bottleneck for fast edge learning, particularly in time-sensitive applications such as PM. To address this issue, we explore Federated Learning (FL), a privacy-preserving framework. FL entails updating a global AI model on a parameter server (PS) through aggregation of locally trained models from edge devices. We propose an innovative approach: analog aggregation over-the-air of updates transmitted concurrently over wireless channels. This leverages the waveform-superposition property in multi-access channels, significantly reducing communication latency compared to conventional methods. However, it is vulnerable to performance degradation due to channel properties like noise and fading. In this study, we introduce a method to mitigate the impact of channel noise in FL over-the-air communication and computation (FLOACC). We integrate a novel tracking-based stochastic approximation scheme into a standard federated stochastic variance reduced gradient (FSVRG). This effectively averages out channel noise's influence, ensuring robust FLOACC performance without increasing transmission power gain. Numerical results confirm our approach's superior communication efficiency and scalability in various FL scenarios, especially when dealing with noisy channels. Simulation experiments also highlight significant enhancements in prediction accuracy and loss function reduction for analog aggregation in over-the-air FL scenarios.

Virtual Sensor of Gravity Centres for Real-Time Condition Monitoring of an Industrial Stamping Press in the Automotive Industry

This article proposes the development of a novel tool that allows real-time monitoring of the balance of a press during the stamping process. This is performed by means of a virtual sensor that, by using the tonnage information in real time, allows us to calculate the gravity centre of a virtual load that moves the slide up and down. The present development follows the philosophy shown in our previous work for the development of industrialised predictive systems, that is, the use of the information available in the system to develop IIoT tools. This philosophy is defined as I3oT (industrializable industrial Internet of Things). The tonnage data are part of a set of new criteria, called Criterion-360, used to obtain this information. This criterion stores data from a sensor each time the encoder indicates that the position of the main axis has rotated by one degree. Since the main axis turns in a complete cycle of the press, this criterion allows us to obtain information on the phases of the process and easily shows where the measured data are in the cycle. The new system allows us to detect anomalies due to imbalance or discontinuity in the stamping process by using the DBSCAN algorithm, which allows us to avoid unexpected stops and serious breakdowns. Tests were conducted to verify that our system actually detects minimal imbalances in the stamping process. Subsequently, the system was connected to normal production for one year. At the end of this work, we explain the anomalies detected as well as the conclusions of the article and future works.

Implementation of a Sequence-to-Sequence Stacked Sparse Long Short-Term Memory Autoencoder for Anomaly Detection on Multivariate Timeseries Data of Industrial Blower Ball Bearing Units

The advent of Industry 4.0 introduced new ways for businesses to evolve by implementing maintenance policies leading to advancements in terms of productivity, efficiency, and financial performance. In line with the growing emphasis on sustainability, industries implement predictive techniques based on Artificial Intelligence for the purpose of mitigating machine and equipment failures by predicting anomalies during their production process. In this work, a new dataset that was made publicly available, collected from an industrial blower, is presented, analyzed and modeled using a Sequence-to-Sequence Stacked Sparse Long Short-Term Memory Autoencoder. Specifically the right and left mounted ball bearing units were measured during several months of normal operational condition as well as during an encumbered operational state. An anomaly detection model was developed for the purpose of analyzing the operational behavior of the two bearing units. A stacked sparse Long Short-Term Memory Autoencoder was successfully trained on the data obtained from the left unit under normal operating conditions, learning the underlying patterns and statistical connections of the data. The model was evaluated by means of the Mean Squared Error using data from the unit's encumbered state, as well as using data collected from the right unit. The model performed satisfactorily throughout its evaluation on all collected datasets. Also, the model proved its capability for generalization along with adaptability on assessing the behavior of equipment similar to the one it was trained on.

The Remaining Useful Life Prediction Method of a Hydraulic Pump under Unknown Degradation Model with Limited Data

This study proposes a remaining useful life (RUL) prediction method using limited degradation data with an unknown degradation model for hydraulic pumps with long service lives and no failure data in turbine control systems. The volumetric efficiency is calculated based on real-time monitoring signal data, and it is used as the degradation indicator. The optimal degradation curve is established using the degradation trajectory model, and the optimal probability distribution model is selected via the K-S test. The above process was repeated to optimize the degradation model and update parameters in different performance degradation stages of the hydraulic pump, providing quantification of the prediction uncertainty and enabling accurate online prediction of the hydraulic pump's RUL. Finally, an RUL test bench for hydraulic pumps is built for verification. The results show that the proposed method is convenient, efficient, and has low model complexity. The method enables online accurate prediction of the RUL of hydraulic pumps using only limited degradation data, with a prediction accuracy of over 85%, which meets practical application requirements.

Quantifying the Displacement of Data Matrix Code Modules: A Comparative Study of Different Approximation Approaches for Predictive Maintenance of Drop-on-Demand Printing Systems

Drop-on-demand printing using colloidal or pigmented inks is prone to the clogging of printing nozzles, which can lead to positional deviations and inconsistently printed patterns (e.g., data matrix codes, DMCs). However, if such deviations are detected early, they can be useful for determining the state of the print head and planning maintenance operations prior to reaching a printing state where the printed DMCs are unreadable. To realize this predictive maintenance approach, it is necessary to accurately quantify the positional deviation of individually printed dots from the actual target position. Here, we present a comparison of different methods based on affinity transformations and clustering algorithms for calculating the target position from the printed positions and, subsequently, the deviation of both for complete DMCs. Hence, our method focuses on the evaluation of the print quality, not on the decoding of DMCs. We compare our results to a state-of-the-art decoding algorithm, adopted to return the target grid positions, and find that we can determine the occurring deviations with significantly higher accuracy, especially when the printed DMCs are of low quality. The results enable the development of decision systems for predictive maintenance and subsequently the optimization of printing systems.

Predicting Network Hardware Faults through Layered Treatment of Alarms Logs

Maintaining and managing ever more complex telecommunication networks is an increasingly difficult task, which often challenges the capabilities of human experts. There is a consensus both in academia and in the industry on the need to enhance human capabilities with sophisticated algorithmic tools for decision-making, with the aim of transitioning towards more autonomous, self-optimizing networks. We aimed to contribute to this larger project. We tackled the problem of detecting and predicting the occurrence of faults in hardware components in a radio access network, leveraging the alarm logs produced by the network elements. We defined an end-to-end method for data collection, preparation, labelling, and fault prediction. We proposed a layered approach to fault prediction: we first detected the base station that is going to be faulty and at a second stage, and using a different algorithm, we detected the component of the base station that is going to be faulty. We designed a range of algorithmic solutions and tested them on real data collected from a major telecommunication operator. We concluded that we are able to predict the failure of a network component with satisfying precision and recall.

An Online Anomaly Detection Approach for Fault Detection on Fire Alarm Systems

The early detection of fire is of utmost importance since it is related to devastating threats regarding human lives and economic losses. Unfortunately, fire alarm sensory systems are known to be prone to failures and frequent false alarms, putting people and buildings at risk. In this sense, it is essential to guarantee smoke detectors' correct functioning. Traditionally, these systems have been subject to periodic maintenance plans, which do not consider the state of the fire alarm sensors and are, therefore, sometimes carried out not when necessary but according to a predefined conservative schedule. Intending to contribute to designing a predictive maintenance plan, we propose an online data-driven anomaly detection of smoke sensors that model the behaviour of these systems over time and detect abnormal patterns that can indicate a potential failure. Our approach was applied to data collected from independent fire alarm sensory systems installed with four customers, from which about three years of data are available. For one of the customers, the obtained results were promising, with a precision score of 1 with no false positives for 3 out of 4 possible faults. Analysis of the remaining customers' results highlighted possible reasons and potential improvements to address this problem better. These findings can provide valuable insights for future research in this area.

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

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

Boosted Convolutional Neural Network Algorithm for the Classification of the Bearing Fault form 1-D Raw Sensor Data

Renewable energy sources are a growing branch of industry. One such source is wind farms, which have significantly increased their number over recent years. Alongside the increased number of turbines, maintenance problems are growing. There is a need for newer and less intrusive predictive maintenance methods. About 40% of all turbine failures are due to bearing failure. This paper presents a modified neural direct classifier method using raw accelerometer measurements as input. This proprietary platform allows for better damage prediction results than convolutional networks in vibration spectrum image analysis. It operates in real time and without signal processing methods converting the signal to a time-frequency spectrogram. Image processing methods can extract features from a set of preset features and based on their importance. The proposed method is not based on feature extraction from image data but on automatically finding a set of features from raw tabular data. This fact significantly reduces the computational cost of detection and improves the failure detection accuracy compared to the classical methods. The model achieved a precision of 99.32% on the validation set, and 96.3% during bench testing. These results were an improvement over the method that classifies time-frequency spectrograms of 97.76% for the validation set and 90.8% for the real-world tests, respectively.

Evaluation of Smart Sensors for Subway Electric Motor Escalators through AHP-Gaussian Method

This paper proposes the use of the AHP-Gaussian method to support the selection of a smart sensor installation for an electric motor used in an escalator in a subway station. The AHP-Gaussian methodology utilizes the Analytic Hierarchy Process (AHP) framework and is highlighted for its ability to save the decision maker's cognitive effort in assigning weights to criteria. Seven criteria were defined for the sensor selection: temperature range, vibration range, weight, communication distance, maximum electric power, data traffic speed, and acquisition cost. Four smart sensors were considered as alternatives. The results of the analysis showed that the most appropriate sensor was the ABB Ability smart sensor, which scored the highest in the AHP-Gaussian analysis. In addition, this sensor could detect any abnormalities in the equipment's operation, enabling timely maintenance and preventing potential failures. The proposed AHP-Gaussian method proved to be an effective approach for selecting a smart sensor for an electric motor used in an escalator in a subway station. The selected sensor was reliable, accurate, and cost-effective, contributing to the safe and efficient operation of the equipment.

Overcoming Challenges Associated with Developing Industrial Prognostics and Health Management Solutions

The development of prognostics and health management solutions in the manufacturing industry has lagged behind academic advances due to a number of practical challenges. This work proposes a framework for the initial development of industrial PHM solutions that is based on the system development life cycle commonly used for software-based applications. Methodologies for completing the planning and design stages, which are critical for industrial solutions, are presented. Two challenges that are inherent to health modeling in manufacturing environments, data quality and modeling systems that experience trend-based degradation, are then identified and methods to overcome them are proposed. Additionally included is a case study documenting the development of an industrial PHM solution for a hyper compressor at a manufacturing facility operated by The Dow Chemical Company. This case study demonstrates the value of the proposed development process and provides guidelines for utilizing it in other applications.

Predictive Maintenance of Norwegian Road Network Using Deep Learning Models

Industry 4.0 has revolutionized the use of physical and digital systems while playing a vital role in the digitalization of maintenance plans for physical assets in an optimal way. Road network conditions and timely maintenance plans are essential in the predictive maintenance (PdM) of a road. We developed a PdM-based approach that uses pre-trained deep learning models to recognize and detect the road crack types effectively and efficiently. We, in this work, explore the use of deep neural networks to classify roads based on the amount of deterioration. This is done by training the network to identify various types of cracks, corrugation, upheaval, potholes, and other types of road damage. Based on the amount and severity of the damage, we can determine the degradation percentage and have a PdM framework where we can identify the intensity of damage occurrence and, thus, prioritize the maintenance decisions. The inspection authorities and stakeholders can make maintenance decisions for certain types of damages using our deep learning-based road predictive maintenance framework. We evaluated our approach using precision, recall, F1-score, intersection-over-union, structural similarity index, and mean average precision measures, and found that our proposed framework achieved significant performance.

Online SFRA for Reliability of Power Systems: Characterization of a Batch of Healthy and Damaged Induction Motors for Predictive Maintenance

Asynchronous motors represent a large percentage of motors used in the electrical industry. Suitable predictive maintenance techniques are strongly required when these motors are critical in their operations. Continuous non-invasive monitoring techniques can be investigated to avoid the disconnection of the motors under test and service interruption. This paper proposes an innovative predictive monitoring system based on the online sweep frequency response analysis (SFRA) technique. The testing system applies variable frequency sinusoidal signals to the motors and then acquires and processes the applied and response signals in the frequency domain. In the literature, SFRA has been applied to power transformers and electric motors switched off and disconnected from the main grid. The approach described in this work is innovative. Coupling circuits allow for the injection and acquisition of the signals, while grids feed the motors. A comparison between the transfer functions (TFs) of healthy motors and those with slight damage was performed with a batch of 1.5 kW, four-pole induction motors to investigate the technique's performance. The results show that the online SFRA could be of interest for monitoring induction motors' health conditions, especially for mission-critical and safety-critical applications. The overall cost of the whole testing system, including the coupling filters and cables, is less than EUR 400.

Damage Caused by Material Defects of Carbon Composites Used on Various Types of Railway Pantographs

Mainstream materials of the railway pantograph strips are carbon composites. They are subject to wear during use, as well as various types of damage. It is important that their operation time is as long as possible and that they are not damaged, as it may damage the remaining elements of the pantograph and the overhead contact line. As part of the article, three types of pantographs were tested: AKP-4E, 5ZL, and 150 DSA. They had carbon sliding strips made of MY7A2 material. By testing the same material on different types of current collectors, it was possible to check what impact the wear and damage of the sliding strips has on (among others) the method of their installation, i.e., whether the damage to the strips depends on the type of current collector and what is the participation of damage caused by material defects. As a result of the research, it was found that the type of pantograph on which it is used has an undoubted influence on the damage characteristics of the carbon sliding strips, whereas the damage caused by material defects can be classified as a more general group-the group of damage of a sliding strip, which also includes overburning of a carbon sliding strip.

Hot-Pressing Furnace Current Monitoring and Predictive Maintenance System in Aerospace Applications

This research combines the application of artificial intelligence in the production equipment fault monitoring of aerospace components. It detects three-phase current abnormalities in large hot-pressing furnaces through smart meters and provides early preventive maintenance. Different anomalies are classified, and a suitable monitoring process algorithm is proposed to improve the overall monitoring quality, accuracy, and stability by applying AI. We also designed a system to present the heater's power consumption and the hot-pressing furnace's fan and visualize the process. Combining artificial intelligence with the experience and technology of professional technicians and researchers to detect and proactively grasp the health of the hot-pressing furnace equipment improves the shortcomings of previous expert systems, achieves long-term stability, and reduces costs. The complete algorithm introduces a model corresponding to the actual production environment, with the best model result being XGBoost with an accuracy of 0.97.

Prediction of Technical State of Mechanical Systems Based on Interpretive Neural Network Model

A classic problem in prognostic and health management (PHM) is the prediction of the remaining useful life (RUL). However, until now, there has been no algorithm presented to achieve perfect performance in this challenge. This study implements a less explored approach: binary classification of the state of mechanical systems at a given forecast horizon. To prove the effectiveness of the proposed approach, tests were conducted on the C-MAPSS sample dataset. The obtained results demonstrate the achievement of an almost maximal performance threshold. The explainability of artificial intelligence (XAI) using the SHAP (Shapley Additive Explanations) feature contribution estimation method for classification models trained on data with and without a sliding window technique is also investigated.

Condition-Based Failure-Free Time Estimation of a Pump

Reliable and continuous operation of the equipment is expected in the wastewater treatment plant, as any perturbations can lead to environmental pollution and the need to pay penalties. Optimization and minimization of operating costs of the pump station cannot, therefore, lead to a reduction in reliability but rather should be based on preventive works, the necessity of which should be foreseen. The purpose of this paper is to develop an accurate model to predict a pump's mean time to failure, allowing for rational planning of maintenance. The pumps operate under the supervision of the automatic control system and SCADA, which is the source of historical data on pump operation parameters. This enables the research and development of various methods and algorithms for optimizing service activities. In this case, a multiple linear regression model is developed to describe the impact of historical data on pump operation for pump maintenance. In the literature, the least squares method is used to estimate unknown regression coefficients for this data. The original value of the paper is the application of the genetic algorithm to estimate coefficient values of the multiple linear regression model of failure-free time of the pump. Necessary analysis and simulations are performed on the data collected for submersible pumps in a sewage pumping station. As a result, an improvement in the adequacy of the presented model was identified.

Transfer-Learning-Based Estimation of the Remaining Useful Life of Heterogeneous Bearing Types Using Low-Frequency Accelerometers

Deep learning approaches are becoming increasingly important for the estimation of the Remaining Useful Life (RUL) of mechanical elements such as bearings. This paper proposes and evaluates a novel transfer learning-based approach for RUL estimations of different bearing types with small datasets and low sampling rates. The approach is based on an intermediate domain that abstracts features of the bearings based on their fault frequencies. The features are processed by convolutional layers. Finally, the RUL estimation is performed using a Long Short-Term Memory (LSTM) network. The transfer learning relies on a fixed-feature extraction. This novel deep learning approach successfully uses data of a low-frequency range, which is a precondition to use low-cost sensors. It is validated against the IEEE PHM 2012 Data Challenge, where it outperforms the winning approach. The results show its suitability for low-frequency sensor data and for efficient and effective transfer learning between different bearing types.

On-Device Intelligence for Malfunction Detection of Water Pump Equipment in Agricultural Premises: Feasibility and Experimentation

The digital transformation of agriculture is a promising necessity for tackling the increasing nutritional needs on Earth and the degradation of natural resources. Toward this direction, the availability of innovative electronic components and of the accompanying software programs can be exploited to detect malfunctions in typical agricultural equipment, such as water pumps, thereby preventing potential failures and water and economic losses. In this context, this article highlights the steps for adding intelligence to sensors installed on pumps in order to intercept and deliver malfunction alerts, based on cheap in situ microcontrollers, sensors, and radios and easy-to-use software tools. This involves efficient data gathering, neural network model training, generation, optimization, and execution procedures, which are further facilitated by the deployment of an experimental platform for generating diverse disturbances of the water pump operation. The best-performing variant of the malfunction detection model can achieve an accuracy rate of about 93% based on the vibration data. The system being implemented follows the on-device intelligence approach that decentralizes processing and networking tasks, thereby aiming to simplify the installation process and reduce the overall costs. In addition to highlighting the necessary implementation variants and details, a characteristic set of evaluation results is also presented, as well as directions for future exploitation.

Anomaly Detection and Inter-Sensor Transfer Learning on Smart Manufacturing Datasets

Smart manufacturing systems are considered the next generation of manufacturing applications. One important goal of the smart manufacturing system is to rapidly detect and anticipate failures to reduce maintenance cost and minimize machine downtime. This often boils down to detecting anomalies within the sensor data acquired from the system which has different characteristics with respect to the operating point of the environment or machines, such as, the RPM of the motor. In this paper, we analyze four datasets from sensors deployed in manufacturing testbeds. We detect the level of defect for each sensor data leveraging deep learning techniques. We also evaluate the performance of several traditional and ML-based forecasting models for predicting the time series of sensor data. We show that careful selection of training data by aggregating multiple predictive RPM values is beneficial. Then, considering the sparse data from one kind of sensor, we perform transfer learning from a high data rate sensor to perform defect type classification. We release our manufacturing database corpus (4 datasets) and codes for anomaly detection and defect type classification for the community to build on it. Taken together, we show that predictive failure classification can be achieved, paving the way for predictive maintenance.

Operational State Recognition of a DC Motor Using Edge Artificial Intelligence

Edge artificial intelligence (EDGE-AI) refers to the execution of artificial intelligence algorithms on hardware devices while processing sensor data/signals in order to extract information and identify patterns, without utilizing the cloud. In the field of predictive maintenance for industrial applications, EDGE-AI systems can provide operational state recognition for machines and production chains, almost in real time. This work presents two methodological approaches for the detection of the operational states of a DC motor, based on sound data. Initially, features were extracted using an audio dataset. Two different Convolutional Neural Network (CNN) models were trained for the particular classification problem. These two models are subject to post-training quantization and an appropriate conversion/compression in order to be deployed to microcontroller units (MCUs) through utilizing appropriate software tools. A real-time validation experiment was conducted, including the simulation of a custom stress test environment, to check the deployed models' performance on the recognition of the engine's operational states and the response time for the transition between the engine's states. Finally, the two implementations were compared in terms of classification accuracy, latency, and resource utilization, leading to promising results.

MLC performance prognosis using a degradation model based on trajectory log data from a daily test

PURPOSE

This paper investigates the feasibility of implementing a predictive maintenance program for a multileaf collimator (MLC) based on data collected in trajectory logs (TLs) obtained by conducting a simple daily test, with the aim of minimizing unscheduled downtime.

METHODS

A dynamic field test was designed, and the TLs generated in the course of daily administration in a linear accelerator were collected to evaluate trajectory deviations of the MLC leaves as well as interlocks (COL 420219/20, COL 420207/08) reported by the machine. During this evaluation, we observed that the trajectory deviations of some leaves increased up to a threshold value beyond which certain interlocks began to appear in treatment fields in those leaves. An exponential degradation model was therefore developed to predict this drift and determine each leaf's remaining useful life (RUL). Once the applicability of the model was confirmed, we added a second accelerator equipped with an MLC with the same configuration to validate the model.

RESULTS

The model was able to predict certain COL 420219/20 interlocks resulting from primary readout/expected position discrepancies and to estimate each leaf's RUL. In total, 11 cases (8 interlocks + 3 potential interlocks avoided due to service interventions [27.3% of the total]) were detected over 7 days in advance, with no false positive results. Scheduling of service interventions several days prior to MLC failure would therefore be possible. When these types of interlocks were not predicted by the model, they were always generated by leaf motor failure. Consequently, intervention time could also be optimized by directly replacing the motor. During the study period, for these types of interlocks, our approach would have reduced downtime from 35.25 to 4.00 h (88.7%) and from 34.75 to 22.83 h (34.3%) for each accelerator, respectively. For COL 420207/08 interlocks, which are generated by primary/secondary readout discrepancies, no correlation with leaf trajectory deviation increases recorded in the TLs was found. Throughout the study period, these types of interlocks requiring service intervention, also mainly for motor replacement, represented a downtime of 9.50 h for the first accelerator (21.2% of total downtime) and by 4.33 h (11.1% of total downtime) for the second accelerator.

CONCLUSION

This study demonstrates that by applying a predictive MLC maintenance program based on information collected in TLs, it is possible to predict certain interlocks and therefore schedule preemptive interventions to avoid their occurrence. This could optimize health-care delivery performance and minimize the loss of treatment sessions.

Developing an Improved Ensemble Learning Approach for Predictive Maintenance in the Textile Manufacturing Process

With the rapid development of digital transformation, paper forms are digitalized as electronic forms (e-Forms). Existing data can be applied in predictive maintenance (PdM) for the enabling of intelligentization and automation manufacturing. This study aims to enhance the utilization of collected e-Form data though machine learning approaches and cloud computing to predict and provide maintenance actions. The ensemble learning approach (ELA) requires less computation time and has a simple hardware requirement; it is suitable for processing e-form data with specific attributes. This study proposed an improved ELA to predict the defective class of product data from a manufacturing site's work order form. This study proposed the resource dispatching approach to arrange data with the corresponding emailing resource for automatic notification. This study's novelty is the integration of cloud computing and an improved ELA for PdM to assist the textile product manufacturing process. The data analytics results show that the improved ensemble learning algorithm has over 98% accuracy and precision for defective product prediction. The validation results of the dispatching approach show that data can be correctly transmitted in a timely manner to the corresponding resource, along with a notification being sent to users.

Predictive Diagnosis Based on Predictor Symptoms for Isolated Photovoltaic Systems Using MPPT Charge Regulators

In this work, new results are presented on the implementation of predictive diagnosis techniques on isolated photovoltaic (PV) systems and installations. The novelties introduced in this research focus on the additional advantages obtained from the point of view of predictive diagnosis of faults caused by partial shading in isolated PV installations using maximum power point tracking (MPPT) regulators. MPPT regulators are comparatively more appropriate than pulse width modulation (PWM) solar regulators in order to implement fault diagnosis systems. MPPT regulators have a physical separation between the electrical parameters belonging to the part of the solar panel with respect to the batteries part. Therefore, these electrical parameters can be used to obtain early predictive symptoms of the effects of partial shading with a greater level of observation and sensitivity. Additionally, modifications are proposed in the PV system assembly to obtain greater homogeneity of all the panels regarding the solar irradiance reception angle.

Acoustic Limescale Layer and Temperature Measurement in Ultrasonic Flow Meters

Guided acoustic waves are commonly used in domestic water meters to measure the flow rate. The accuracy of this measurement method is affected by factors such as variations in temperature and limescale deposition inside of the pipe. In this work, a new approach using signals from different sound propagation paths is used to determine these quantities and allow for subsequent compensation. This method evaluates the different propagation times of guided Lamb waves in flow measurement applications. A finite element method-based model is used to identify the calibration curves for the device under test. The simulated dependencies on temperature and layer thickness are validated by experimental data. Finally, a test on simulated data with varying temperatures and limescale depositions proves that this method can be used to separate both effects. Based on these values, a flow measurement correction scheme can be derived that provides an improved resolution of guided acoustic wave-based flow meters.

A Digital Twin-Driven Life Prediction Method of Lithium-Ion Batteries Based on Adaptive Model Evolution

Accurate life prediction and reliability evaluation of lithium-ion batteries are of great significance for predictive maintenance. In the whole life cycle of a battery, the accurate description of the dynamic and stochastic characteristics of life has always been a key problem. In this paper, the concept of the digital twin is introduced, and a digital twin for reliability based on remaining useful cycle life prediction is proposed for lithium-ion batteries. The capacity degradation model, stochastic degradation model, life prediction, and reliability evaluation model are established to describe the randomness of battery degradation and the dispersion of the life of multiple cells. Based on the Bayesian algorithm, an adaptive evolution method for the model of the digital twin is proposed to improve prediction accuracy, followed by experimental verification. Finally, the life prediction, reliability evaluation, and predictive maintenance of the battery based on the digital twin are implemented. The results show the digital twin for reliability has good accuracy in the whole life cycle. The error can be controlled at about 5% with the adaptive evolution algorithm. For battery L1 and L6 in this case, predictive maintenance costs are expected to decrease by 62.0% and 52.5%, respectively.

Anomaly Detection Using Autoencoder Reconstruction upon Industrial Motors

Rotary machine breakdown detection systems are outdated and dependent upon routine testing to discover faults. This is costly and often reactive in nature. Real-time monitoring offers a solution for detecting faults without the need for manual observation. However, manual interpretation for threshold anomaly detection is often subjective and varies between industrial experts. This approach is ridged and prone to a large number of false positives. To address this issue, we propose a machine learning (ML) approach to model normal working operations and detect anomalies. The approach extracts key features from signals representing a known normal operation to model machine behaviour and automatically identify anomalies. The ML learns generalisations and generates thresholds based on fault severity. This provides engineers with a traffic light system where green is normal behaviour, amber is worrying and red signifies a machine fault. This scale allows engineers to undertake early intervention measures at the appropriate time. The approach is evaluated on windowed real machine sensor data to observe normal and abnormal behaviour. The results demonstrate that it is possible to detect anomalies within the amber range and raise alarms before machine failure.

MEDEP: Maintenance Event Detection for Multivariate Time Series Based on the PELT Approach

Predictive Maintenance (PdM) is one of the most important applications of advanced data science in Industry 4.0, aiming to facilitate manufacturing processes. To build PdM models, sufficient data, such as condition monitoring and maintenance data of the industrial application, are required. However, collecting maintenance data is complex and challenging as it requires human involvement and expertise. Due to time constraints, motivating workers to provide comprehensive labeled data is very challenging, and thus maintenance data are mostly incomplete or even completely missing. In addition to these aspects, a lot of condition monitoring data-sets exist, but only very few labeled small maintenance data-sets can be found. Hence, our proposed solution can provide additional labels and offer new research possibilities for these data-sets. To address this challenge, we introduce MEDEP, a novel maintenance event detection framework based on the Pruned Exact Linear Time (PELT) approach, promising a low false-positive (FP) rate and high accuracy results in general. MEDEP could help to automatically detect performed maintenance events from the deviations in the condition monitoring data. A heuristic method is proposed as an extension to the PELT approach consisting of the following two steps: (1) mean threshold for multivariate time series and (2) distribution threshold analysis based on the complexity-invariant metric. We validate and compare MEDEP on the Microsoft Azure Predictive Maintenance data-set and data from a real-world use case in the welding industry. The experimental outcomes of the proposed approach resulted in a superior performance with an FP rate of around 10% on average and high sensitivity and accuracy results.

Monitoring and Predictive Maintenance of Centrifugal Pumps Based on Smart Sensors

Centrifugal pumps have a wide range of applications in industrial and municipal water affairs. During the use of centrifugal pumps, failures such as bearing wear, blade damage, impeller imbalance, shaft misalignment, cavitation, water hammer, etc., often occur. It is of great importance to use smart sensors and digital Internet of Things (IoT) systems to monitor the real-time operating status of pumps and predict potential failures for achieving predictive maintenance of pumps and improving the intelligence level of machine health management. Firstly, the common fault forms of centrifugal pumps and the characteristics of vibration signals when a fault occurs are introduced. Secondly, the centrifugal pump monitoring IoT system is designed. The system is mainly composed of wireless sensors, wired sensors, data collectors, and cloud servers. Then, the microelectromechanical system (MEMS) chip is used to design a wireless vibration temperature integrated sensor, a wired vibration temperature integrated sensor, and a data collector to monitor the running state of the pump. The designed wireless sensor communicates with the server through Narrow Band Internet of Things (NB-IoT). The output of the wired sensor is connected to the data collector, and the designed collector can communicate with the server through 4G communication. Through cloud-side collaboration, real-time monitoring of the running status of centrifugal pumps and intelligent diagnosis of centrifugal pump faults are realized. Finally, on-site testing and application verification of the system was conducted. The test results show that the designed sensors and sensor application system can make good use of the centrifugal pump failure mechanism to automatically diagnose equipment failures. Moreover, the diagnostic accuracy rate is above 85% by using the method of wired sensor and collector. As a low-cost and easy-to-implement solution, wireless sensors can also monitor gradual failures well. The research on the sensors and pump monitoring system provides feasible methods and an effective means for the application of centrifugal pump health management and predictive maintenance.

Knowledge Graph Based Hard Drive Failure Prediction

The hard drive is one of the important components of a computing system, and its failure can lead to both system failure and data loss. Therefore, the reliability of a hard drive is very important. Realising this importance, a number of studies have been conducted and many are still ongoing to improve hard drive failure prediction. Most of those studies rely solely on machine learning, and a few others on semantic technology. The studies based on machine learning, despite promising results, lack context-awareness such as how failures are related or what other factors, such as humidity, influence the failure of hard drives. Semantic technology, on the other hand, by means of ontologies and knowledge graphs (KGs), is able to provide the context-awareness that machine learning-based studies lack. However, the studies based on semantic technology lack the advantages of machine learning, such as the ability to learn a pattern and make predictions based on learned patterns. Therefore, in this paper, leveraging the benefits of both machine learning (ML) and semantic technology, we present our study, knowledge graph-based hard drive failure prediction. The experimental results demonstrate that our proposed method achieves higher accuracy in comparison to the current state of the art.

Solar Panels String Predictive and Parametric Fault Diagnosis Using Low-Cost Sensors

This work proposes a method for real-time supervision and predictive fault diagnosis applicable to solar panel strings in real-world installations. It is focused on the detection and parametric isolation of fault symptoms through the analysis of the Voc-Isc curves. The method performs early, systematic, online, automatic, permanent predictive supervision, and diagnosis of a high sampling frequency. It is based on the supervision of predictive electrical parameters easily accessible by the design of its architecture, whose detection and isolation precedes with an adequate margin of maneuver, to be able to alert and stop by means of automatic disconnection the degradation phenomenon and its cumulative effect causing the development of a future irrecoverable failure. Its architecture design is scalable and integrable in conventional photovoltaic installations. It emphasizes the use of low-cost technology such as the ESP8266 module, ASC712-5A, and FZ0430 sensors and relay modules. The method is based on data acquisition with the ESP8266 module, which is sent over the internet to the computer where a SCADA system (iFIX V6.5) is installed, using the Modbus TCP/IP and OPC communication protocols. Detection thresholds are initially obtained experimentally by applying inductive shading methods on specific solar panels.

Sensor-Based Predictive Maintenance with Reduction of False Alarms-A Case Study in Heavy Industry

In this paper, the problem of the identification of undesirable events is discussed. Such events can be poorly represented in the historical data, and it is predominantly impossible to learn from past examples. The discussed issue is considered in the work in the context of two use cases in which vibration and temperature measurements collected by wireless sensors are analysed. These use cases include crushers at a coal-fired power plant and gantries in a steelworks converter. The awareness, resulting from the cooperation with industry, of the need for a system that works in cold start conditions and does not flood the machine operator with alarms was the motivation for proposing a new predictive maintenance method. The proposed solution is based on the methods of outlier identification. These methods are applied to the collected data that was transformed into a multidimensional feature vector. The novelty of the proposed solution stems from the creation of a methodology for the reduction of false positive alarms, which was applied to a system identifying undesirable events. This methodology is based on the adaptation of the system to the analysed data, the interaction with the dispatcher, and the use of the XAI (eXplainable Artificial Intelligence) method. The experiments performed on several data sets showed that the proposed method reduced false alarms by 90.25% on average in relation to the performance of the stand-alone outlier detection method. The obtained results allowed for the implementation of the developed method to a system operating in a real industrial facility. The conducted research may be valuable for systems with a cold start problem where frequent alarms can lead to discouragement and disregard for the system by the user.

A Method of Predicting Wear and Damage of Pantograph Sliding Strips Based on Artificial Neural Networks

The impact of the pantograph of a rail vehicle on the overhead contact line depends on many factors. Among other things, the type of pantograph, i.e., the material of the sliding strip, influences the wear and possible damage to the sliding strip. The possibility of predicting pantograph failures may make it possible to reduce the number of these kinds of failures. This article presents a method for predicting the technical state of the pantograph by using artificial neural networks. The presented method enables the prediction of the wear and damage of the pantograph, with particular emphasis on carbon sliding strips. The paper compares 12 predictive models based on regression algorithms, where different training algorithms and activation functions were used. Two different types of training data were also used. Such a distinction made it possible to determine the optimal structure of the input and output data teaching the neural network, as well as the determination of the best structure and parameters of the model enabling the prediction of the technical condition of the current collector.

AI-Enabled Predictive Maintenance Framework for Autonomous Mobile Cleaning Robots

Vibration is an indicator of performance degradation or operational safety issues of mobile cleaning robots. Therefore, predicting the source of vibration at an early stage will help to avoid functional losses and hazardous operational environments. This work presents an artificial intelligence (AI)-enabled predictive maintenance framework for mobile cleaning robots to identify performance degradation and operational safety issues through vibration signals. A four-layer 1D CNN framework was developed and trained with a vibration signals dataset generated from the in-house developed autonomous steam mopping robot 'Snail' with different health conditions and hazardous operational environments. The vibration signals were collected using an IMU sensor and categorized into five classes: normal operational vibration, hazardous terrain induced vibration, collision-induced vibration, loose assembly induced vibration, and structure imbalanced vibration signals. The performance of the trained predictive maintenance framework was evaluated with various real-time field trials with statistical measurement metrics. The experiment results indicate that our proposed predictive maintenance framework has accurately predicted the performance degradation and operational safety issues by analyzing the vibration signal patterns raised from the cleaning robot on different test scenarios. Finally, a predictive maintenance map was generated by fusing the vibration signal class on the cartographer SLAM algorithm-generated 2D environment map.

Large-Scale Internet of Things Multi-Sensor Measurement Node for Smart Grid Enhancement

Electric power infrastructure has revolutionized our world and our way of living has completely changed. The necessary amount of energy is increasing faster than we realize. In these conditions, the grid is forced to run against its limitations, resulting in more frequent blackouts. Thus, urgent solutions need to be found to meet this greater and greater energy demand. By using the internet of things infrastructure, we can remotely manage distribution points, receiving data that can predict any future failure points on the grid. In this work, we present the design of a fully reconfigurable wireless sensor node that can sense the smart grid environment. The proposed prototype uses a modular developed hardware platform that can be easily integrated into the smart grid concept in a scalable manner and collects data using the LoRaWAN communication protocol. The designed architecture was tested for a period of 6 months, revealing the feasibility and scalability of the system, and opening new directions in the remote failure prediction of low voltage/medium voltage switchgears on the electric grid.

Online Frequency Response Analysis of Electric Machinery through an Active Coupling System Based on Power Electronics

This paper presents an innovative concept for the online application of Frequency Response Analysis (FRA). FRA is a well known technique that is applied to detect damage in electric machinery. As an offline technique, the machine under testing has to be removed from service-which may cause loss of production. Experimental adaptations of FRA to online operation are usually based on the use of passive high pass coupling-which, ideally, should provide attenuation to the grid voltage, and at the same time, allow the high frequency FRA signals to be injected at the machine. In practice, however, the passive coupling results in a trade-off between the required attenuation and the useful area obtained at the FRA spectra. This paper proposes the use of an active coupling system, based on power electronics, in order to cancel the grid voltage at the terminals of FRA equipment and allow its safe connection to an energized machine. The paper presents the basic concepts of FRA and the issue of online measurements. It also presents basic concepts about power electronics converters and the operating principles of the Modular Multilevel Converter, which enables the generation of an output voltage with low THD, which is important for tracking the grid voltage with minimum error.

Predictive Maintenance: An Autoencoder Anomaly-Based Approach for a 3 DoF Delta Robot

Performing predictive maintenance (PdM) is challenging for many reasons. Dealing with large datasets which may not contain run-to-failure data (R2F) complicates PdM even more. When no R2F data are available, identifying condition indicators (CIs), estimating the health index (HI), and thereafter, calculating a degradation model for predicting the remaining useful lifetime (RUL) are merely impossible using supervised learning. In this paper, a 3 DoF delta robot used for pick and place task is studied. In the proposed method, autoencoders (AEs) are used to predict when maintenance is required based on the signal sequence distribution and anomaly detection, which is vital when no R2F data are available. Due to the sequential nature of the data, nonlinearity of the system, and correlations between parameter time-series, convolutional layers are used for feature extraction. Thereafter, a sigmoid function is used to predict the probability of having an anomaly given CIs acquired from AEs. This function can be manually tuned given the sensitivity of the system or optimized by solving a minimax problem. Moreover, the proposed architecture can be used for fault localization for the specified system. Additionally, the proposed method can calculate RUL using Gaussian process (GP), as a degradation model, given HI values as its input.

Beamforming Applied to Ultrasound Analysis in Detection of Bearing Defects

The bearings of rotating machinery often fail, leading to unforeseen downtime of large machines in industrial plants. Therefore, condition monitoring can be a powerful tool to aid in the quick identification of these faults and make it possible to plan maintenance before the fault becomes too drastic, reducing downtime and cost. Predictive maintenance is often based on information gathered from accelerometers. However, these sensors are contact-based, making them less attractive for use in an industrial plant and more prone to breakage. In this paper, condition monitoring based on ultrasound is researched, where non-invasive sensors are used to record the noise originating from different defects of the Machinery Fault Simulator. The acoustic data are recorded using a sparse microphone array in a lab environment. The same array was used to record real spatial noise in a fully operational plant which was later added to the acoustic data containing the different defects with a variety of Signal To Noise ratios. In this paper, we compare the classification results of the noisy acoustic data of only one microphone to the beamformed acoustic data. We do this to investigate how beamforming could improve the classification process in an ultrasound condition-monitoring application in a real industrial plant.

Model of Production System Evaluation with the Influence of FDM Machine Reliability and Process-Dependent Product Quality

This paper investigates the Job Shop Scheduling Problem (JSSP) with FDM (Fused Deposition Modeling) machine unavailability constraints due to Predictive Maintenance (PdM) tasks, under the objective of minimizing the makespan, total tardiness and machine idle time. The Ant-Colony Optimization (ACO) algorithm is elaborated to deal with the JSSP. The reliability characteristics of the critical machine (FDM) influence the product as well as the production system quality. PdM periods are estimated based on historical data on failure-free times of the FDM machine components and deviations from the standards established for the key process parameters: infill density, layer thickness and extruder temperature. The standards for the key process parameters are identified based on investigation of the mechanical properties of printed elements. The impact of failure time and the number of nonstandard measurements of parameters on the quality of the Job Shop System (JSS) are observed. Failure rate of the FDM machine is corrected with the probability of a stoppage in the future period due to the “outlier” in measurements of any key parameters of the additive process. The quality robustness of production schedules increases with the disturbance-free operation of the FDM up to the peak value. After reaching the peak value the quality robustness decreases. The original issue of this paper is a model of scheduling production and maintenance tasks in a job shop system with an FDM machine as a bottleneck using ACO. Additionally, an original FDM-reliability model is also proposed. The model is based on weighted p-moving averages of the observed number of deviations from the norms, established for key process parameters such as fill density, layer thickness and extruder temperature.