Explainable AI Algorithms for Vibration Data-Based Fault Detection: Use Case-Adadpted Methods and Critical Evaluation

Analyzing vibration data using deep neural networks is an effective way to detect damages in rotating machinery at an early stage. However, the black-box approach of these methods often does not provide a satisfactory solution because the cause of classifications is not comprehensible to humans. Therefore, this work investigates the application of the explainable AI (XAI) algorithms to convolutional neural networks for vibration-based condition monitoring. Thus, the three XAI algorithms GradCAM, LRP and LIME with a modified perturbation strategy are applied to classifications based on the Fourier transform as well as the order analysis of the vibration signal. The following visualization as frequency-RPM maps and order-RPM maps allows for an effective assessment of saliency values for variable periodicity of the data, which translates to a varying rotation speed of a real-world machine. To compare the explanatory power of the XAI methods, investigations are first carried out with a synthetic data set with known class-specific characteristics. Both a visual and a quantitative analysis of the resulting saliency maps are presented. Then, a real-world data set for vibration-based imbalance classification on an electric motor, which runs at a broad range of rotation speeds, is used. The results indicate that the investigated algorithms are each partially successful in providing sample-specific saliency maps which highlight class-specific features and omit features which are not relevant for classification.

A Machine-Learning-Based Robust Classification Method for PV Panel Faults

Renewable energy resources have gained considerable attention in recent years due to their efficiency and economic benefits. Their proportion of total energy use continues to grow over time. Photovoltaic (PV) cell and wind energy generation are the least-expensive new energy sources in most countries. Renewable energy technologies significantly contribute to climate mitigation and provide economic benefits. Apart from these advantages, renewable energy sources, particularly solar energy, have drawbacks, for instance restricted energy supply, reliance on weather conditions, and being affected by several kinds of faults, which cause a high power loss. Usually, the local PV plants are small in size, and it is easy to trace any fault and defect; however, there are many PV cells in the grid-connected PV system where it is difficult to find a fault. Keeping in view the aforedescribed facts, this paper presents an intelligent model to detect faults in the PV panels. The proposed model utilizes the Convolutional Neural Network (CNN), which is trained on historic data. The dataset was preprocessed before being fed to the CNN. The dataset contained different parameters, such as current, voltage, temperature, and irradiance, for five different classes. The simulation results showed that the proposed CNN model achieved a training accuracy of 97.64% and a testing accuracy of 95.20%, which are much better than the previous research performed on this dataset.

A Review on Rolling Bearing Fault Signal Detection Methods Based on Different Sensors

As a precision mechanical component to reduce friction between components, the rolling bearing is widely used in many fields because of its slight friction loss, strong bearing capacity, high precision, low power consumption, and high mechanical efficiency. This paper reviews several excellent kinds of study and their relevance to the fault detection of rolling bearings. We summarize the fault location, sensor types, bearing fault types, and fault signal analysis of rolling bearings. The fault signal types are divided into one-dimensional and two-dimensional images, which account for 40.14% and 31.69%, respectively, and their classification is clarified and discussed. We counted the proportions of various methods in the references cited in this paper. Among them, the method of one-dimensional signal detection with external sensors accounted for 3.52%, the method of one-dimensional signal detection with internal sensors accounted for 36.62%, and the method of two-dimensional signal detection with external sensors accounted for 19.72%. The method of two-dimensional signal detection with internal sensors accounted for 11.97%. Among these methods, the highest detection rate is 100%, and the lowest detection rate is more than 70%. The similarities between the different methods are compared. The research results summarized in this paper show that with the progress of the times, a variety of new and better research methods have emerged, which have sped up the detection and diagnosis of rolling bearing faults. For example, the technology using artificial intelligence is still developing rapidly, such as artificial neural networks, convolutional neural networks, and machine learning. Although there are still defects, such methods can quickly discover a fault and its cause, enrich the database, and accumulate experience. More and more advanced techniques are applied in this field, and the detection method has better robustness and superiority.

Multiple Sensor Fault Detection Using Index-Based Method

The research on sensor fault detection has drawn much interest in recent years. Abrupt, incipient, and intermittent sensor faults can cause the complete blackout of the system if left undetected. In this research, we examined the observer-based residual analysis via index-based approaches for fault detection of multiple sensors in a healthy drive. Seven main indices including the moving mean, average, root mean square, energy, variance, first-order derivative, second-order derivative, and auto-correlation-based index were employed and analyzed for sensor fault diagnosis. In addition, an auxiliary index was computed to differentiate a faulty sensor from a non-faulty one. These index-based methods were utilized for further analysis of sensor fault detection operating under a range of various loads, varying speeds, and fault severity levels. The simulation results on a permanent magnet synchronous motor (PMSM) are provided to demonstrate the pros and cons of various index-based methods for various fault detection scenarios.

A Fault Detection Method Based on an Oil Temperature Forecasting Model Using an Improved Deep Deterministic Policy Gradient Algorithm in the Helicopter Gearbox

The main gearbox is very important for the operation safety of helicopters, and the oil temperature reflects the health degree of the gearbox; therefore establishing an accurate oil temperature forecasting model is an important step for reliable fault detection. Firstly, in order to achieve accurate gearbox oil temperature forecasting, an improved deep deterministic policy gradient algorithm with a CNN-LSTM basic learner is proposed, which can excavate the complex relationship between oil temperature and working condition. Secondly, a reward incentive function is designed to accelerate the training time costs and to stabilize the model. Further, a variable variance exploration strategy is proposed to enable the agents of the model to fully explore the state space in the early training stage and to gradually converge in the training later stage. Thirdly, a multi-critics network structure is adopted to solve the problem of inaccurate Q-value estimation, which is the key to improving the prediction accuracy of the model. Finally, KDE is introduced to determine the fault threshold to judge whether the residual error is abnormal after EWMA processing. The experimental results show that the proposed model achieves higher prediction accuracy and shorter fault detection time costs.

Power Equipment Fault Diagnosis Method Based on Energy Spectrogram and Deep Learning

With the development of industrial manufacturing intelligence, the role of rotating machinery in industrial production and life is more and more important. Aiming at the problems of the complex and changeable working environment of rolling bearings and limited computing ability, fault feature information cannot be effectively extracted, and the current deep learning model is difficult to be compatible with lightweight and high efficiency. Therefore, this paper proposes a fault detection method for power equipment based on an energy spectrum diagram and deep learning. Firstly, a novel two-dimensional time-frequency feature representation method and energy spectrum feature map based on wavelet packet transform is proposed, and an energy spectrum feature map dataset is made for subsequent diagnosis. This method can realize multi-resolution analysis, fully extract the feature information contained in the fault signal, and accelerate the convergence of the subsequent diagnosis model. Secondly, a lightweight residual dense convolutional neural network model (LR-DenseNet) is proposed. This model combines the advantages of residual learning and a dense connection, and can not only extract deep features more easily, but can also effectively use shallow features. Then, based on the lightweight residual dense convolutional neural network model, an LR-DenseSENet model is proposed. By introducing the transfer learning strategy and adding the channel domain, an attention mechanism is added to the channel feature fusion layer, with the accuracy of detection up to 99.4%, and the amount of parameter calculation greatly reduced to one-fifth of that of VGG. Finally, through an experimental analysis, it is verified that the fault detection model designed in this paper based on the combination of an energy spectrum feature map and LR-DenseSENet achieves a satisfactory detection effect.

Multi-Fault Detection and Classification of Wind Turbines Using Stacking Classifier

Wind turbines are widely used worldwide to generate clean, renewable energy. The biggest issue with a wind turbine is reducing failures and downtime, which lowers costs associated with operations and maintenance. Wind turbines' consistency and timely maintenance can enhance their performance and dependability. Still, the traditional routine configuration makes detecting faults of wind turbines difficult. Supervisory control and data acquisition (SCADA) produces reliable and affordable quality data for the health condition of wind turbine operations. For wind power to be sufficiently reliable, it is crucial to retrieve useful information from SCADA successfully. This article proposes a new AdaBoost, K-nearest neighbors, and logistic regression-based stacking ensemble (AKL-SE) classifier to classify the faults of the wind turbine condition monitoring system. A stacking ensemble classifier integrates different classification models to enhance the model's accuracy. We have used three classifiers, AdaBoost, K-nearest neighbors, and logistic regression, as base models to make output. The output of these three classifiers is used as input in the logistic regression classifier's meta-model. To improve the data validity, SCADA data are first preprocessed by cleaning and removing any abnormal data. Next, the Pearson correlation coefficient was used to choose the input variables. The Stacking Ensemble classifier was trained using these parameters. The analysis demonstrates that the suggested method successfully identifies faults in wind turbines when applied to local 3 MW wind turbines. The proposed approach shows the potential for effective wind energy use, which could encourage the use of clean energy.

Fault Detection of Wind Turbine Gearboxes Based on IBOA-ERF

As one of the key components of wind turbines, gearboxes are under complex alternating loads for a long time, and the safety and reliability of the whole machine are often affected by the failure of internal gears and bearings. Aiming at the difficulty of optimizing the parameters of wind turbine gearbox fault detection models based on extreme random forest, a fault detection model with extreme random forest optimized by the improved butterfly optimization algorithm (IBOA-ERF) is proposed. The algebraic sum of the false alarm rate and the missing alarm rate of the fault detection model is constructed as the fitness function, and the initial position and position update strategy of the individual are improved. A chaotic mapping strategy is introduced to replace the original population initialization method to enhance the randomness of the initial population distribution. An adaptive inertia weight factor is proposed, combined with the landmark operator of the pigeon swarm optimization algorithm to update the population position iteration equation to speed up the convergence speed and improve the diversity and robustness of the butterfly optimization algorithm. The dynamic switching method of local and global search stages is adopted to achieve dynamic balance between global exploration and local search, and to avoid falling into local optima. The ERF fault detection model is trained, and the improved butterfly optimization algorithm is used to obtain optimal parameters to achieve fast response of the proposed model with good robustness and generalization under high-dimensional data. The experimental results show that, compared with other optimization algorithms, the proposed fault detection method of wind turbine gearboxes has a lower false alarm rate and missing alarm rate.

Miniterm, a Novel Virtual Sensor for Predictive Maintenance for the Industry 4.0 Era

This article introduces a novel virtual sensor for predictive maintenance called mini-term. A mini-term can be defined as the time it takes for a part of the machine to do its job. Being a technical sub-cycle time, its function has been linked to production. However, when a machine or component gets deteriorated, the mini-term also suffers deterioration, allowing it to be a multifunctional indicator for the prediction of machine failures as well as measurement of production. Currently, in Industry 4.0, one of the handicaps is Big Data and Data Analysis. However, in the case of predictive maintenance, the need to install sensors in the machines means that when the proposed scientific solutions reach the industry, they cannot be carried out massively due to the high cost this entails. The advantage introduced by the mini-term is that it can be implemented in an easy and simple way in pre-installed systems since you only need to program a timer in the PLC or PC that controls the line/machine in the production line, allowing, according to the authors' knowledge, to build industrial Big Data on predictive maintenance for the first time, which is called Miniterm 4.0. This article shows evidence of the important improvements generated by the use of Miniterm 4.0 in a factory. At the end of the paper we show the evolution of TAV (Technical availability), Mean Time To Repair (MTTR), EM (Number of Work order (Emergency Orders/line Stop)) and OM (Labour hours in EM) showing a very important improvement as the number of mini-terms was increased and the Miniterm 4.0 system became more reliable. In particular, TAV is increased by 15%, OM is reduced in 5000 orders, MTTR is reduced in 2 h and there are produced 3000 orders less than when mini-terms did not exist. At the end of the article we discuss the benefits and limitations of the mini-terms and we show the conclusions and future works.

An Automated Machine Learning Approach for Real-Time Fault Detection and Diagnosis

This work presents a novel Automated Machine Learning (AutoML) approach for Real-Time Fault Detection and Diagnosis (RT-FDD). The approach's particular characteristics are: it uses only data that are commonly available in industrial automation systems; it automates all ML processes without human intervention; a non-ML expert can deploy it; and it considers the behavior of cyclic sequential machines, combining discrete timed events and continuous variables as features. The capacity for fault detection is analyzed in two case studies, using data from a 3D machine simulation system with faulty and non-faulty conditions. The enhancement of the RT-FDD performance when the proposed approach is applied is proved with the Feature Importance, Confusion Matrix, and F1 Score analysis, reaching mean values of 85% and 100% in each case study. Finally, considering that faults are rare events, the sensitivity of the models to the number of faulty samples is analyzed.

Towards an Accurate Faults Detection Approach in Internet of Medical Things Using Advanced Machine Learning Techniques

Remotely monitoring people's healthcare is still among the most important research topics for researchers from both industry and academia. In addition, with the Wireless Body Networks (WBANs) emergence, it becomes possible to supervise patients through an implanted set of body sensors that can communicate through wireless interfaces. These body sensors are characterized by their tiny sizes, and limited resources (power, computing, and communication capabilities), which makes these devices prone to have faults and sensible to be damaged. Thus, it is necessary to establish an efficient system to detect any fault or anomalies when receiving sensed data. In this paper, we propose a novel, optimized, and hybrid solution between machine learning and statistical techniques, for detecting faults in WBANs that do not affect the devices' resources and functionality. Experimental results illustrate that our approach can detect unwanted measurement faults with a high detection accuracy ratio that exceeds the 99.62%, and a low mean absolute error of 0.61%, clearly outperforming the existing state-of-art solutions.

Enhanced Multiscale Principal Component Analysis for Improved Sensor Fault Detection and Isolation

Multiscale PCA (MSPCA) is a well-established fault-detection and isolation (FDI) technique. It utilizes wavelet analysis and PCA to extract important features from process data. This study demonstrates limitations in the conventional MSPCA fault detection algorithm, thereby proposing an enhanced MSPCA (EMSPCA) FDI algorithm that uses a new wavelet thresholding criterion. As such, it improves the projection of faults in the residual space and the threshold estimation of the fault detection statistic. When tested with a synthetic model, EMSPCA resulted in a 30% improvement in detection rate with equal false alarm rates. The EMSPCA algorithm also relies on the novel application of reconstruction-based fault isolation at multiple scales. The proposed algorithm reduces fault smearing and consequently improves fault isolation performance. The paper will further investigate the use of soft vs. hard wavelet thresholding, decimated vs. undecimated wavelet transforms, the choice of wavelet decomposition depth, and their implications on FDI performance.The FDI performance of the developed EMSPCA method was illustrated for sensor faults. This undertaking considered synthetic data, the simulated data of a continuously stirred reactor (CSTR), and experimental data from a packed-bed pilot plant. The results of these examples show the advantages of EMSPCA over existing techniques.

Dynamic Calibration Method of Sensor Drift Fault in HVAC System Based on Bayesian Inference

Sensor drift fault calibration is essential to maintain the operation of heating, ventilation and air conditioning systems (HVAC) in buildings. Bayesian inference (BI) is becoming more and more popular as a commonly used sensor fault calibration method. However, this method focused mainly on sensor bias fault, and it could be difficult to calibrate drift fault that changes with time. Therefore, a dynamic calibration method for sensor drift fault of HVAC systems based on BI is developed. Taking the drift fault calibration of the chilled water supply temperature sensor of the chiller as an example, the performance of the proposed dynamic calibration method is evaluated. Results show that the combination of the Exponentially Weighted Moving-Average (EWMA) method with high detection accuracy and the proposed BI dynamic calibration method can effectively improve the calibration accuracy of drift fault, and the Mean Absolute Percentage Error (MAPE) value between the calibrated and normal data is less than 5%.

A RUL Estimation System from Clustered Run-to-Failure Degradation Signals

The prognostics and health management disciplines provide an efficient solution to improve a system's durability, taking advantage of its lifespan in functionality before a failure appears. Prognostics are performed to estimate the system or subsystem's remaining useful life (RUL). This estimation can be used as a supply in decision-making within maintenance plans and procedures. This work focuses on prognostics by developing a recurrent neural network and a forecasting method called Prophet to measure the performance quality in RUL estimation. We apply this approach to degradation signals, which do not need to be monotonical. Finally, we test our system using data from new generation telescopes in real-world applications.

Interturn Short Fault Diagnosis Using Magnitude and Phase of Currents in Permanent Magnet Synchronous Machines

With the increased demand for permanent magnet synchronous machines (PMSMs) in various industrial fields, interturn short fault (ITSF) diagnosis of PMSMs is under the limelight. In particular, to prevent accidents caused by PMSM malfunctions, it is difficult and greatly necessary to diagnose slight ITSF, which is a stage before the ITSF becomes severe. In this paper, we propose a novel fault indicator based on the magnitude and phase of the current. The proposed fault indicator was developed using analysis of positive-sequence current (PSC) and negative-sequence current (NSC), which means the degree of the asymmetry of the three-phase currents by ITSF. According to the analysis, as ITSF increases, the phase difference between PSC and NSC decreases and the magnitude of NSC increases. Therefore, the novel fault indicator is suggested as a product of the cosine value of the phase indicator and the magnitude indicator. The magnitude indicator is the magnitude of NSC, and the phase indicator means the phase difference between the PSC and the NSC. The suggested fault indicator diagnoses the degree of ITSF as well as slight ITSFs under various conditions by only measured three-phase currents. Experimental results demonstrate the effectiveness of our proposed method under various torque and speeds.

Lightweight Self-Detection and Self-Calibration Strategy for MEMS Gas Sensor Arrays

With the development of Internet of Things (IoT) and edge computing technology, gas sensor arrays based on Micro-Electro-Mechanical System (MEMS) fabrication technique have broad application prospects in intelligent integrated systems, portable devices, and other fields. In such complex scenarios, the normal operation of a gas sensing system depends heavily on the accuracy of the sensor output. Therefore, a lightweight Self-Detection and Self-Calibration strategy for MEMS gas sensor arrays is proposed in this paper to monitor the working status of sensor arrays and correct the abnormal data in real time. Evaluations on real-world datasets indicate that the strategy has high performance of fault detection, isolation, and data recovery. Furthermore, our method has low computation complexity and low storage resource occupation. The board-level verification on CC1350 shows that the average calculation time and running power consumption of the algorithm are 0.28 ms and 9.884 mW. The proposed strategy can be deployed on most resource-limited IoT devices to improve the reliability of gas sensing systems.

Fuzzy-Logic-Based Recommendation System for Processing in Condition Monitoring

The development of a machine's condition monitoring system is often a challenging task. This process requires the collection of a sufficiently large dataset on signals from machine operation, context information related to the operation conditions, and the diagnosis experience. The two referred problems are today relatively easy to solve. The hardest to describe is the diagnosis experience because it is based on imprecise and non-numerical information. However, it is essential to process acquired data to develop a robust monitoring system. This article presents a framework for a system dedicated to recommending processing algorithms for condition monitoring. It includes a database and fuzzy-logic-based modules composed within the system. Based on the contextual knowledge provided by the user, the procedure suggests processing algorithms. This paper presents the evaluation of the proposed agent on two different parallel gearboxes. The results of the system are processing algorithms with assigned model types. The obtained results show that the algorithms recommended by the system achieve a higher accuracy than those selected arbitrarily. The results obtained allow for an average of 5 to 14.5% higher accuracy.

Machine Learning-Based Anomaly Detection Algorithms to Alert Patients Using Sensor Augmented Pump of Infusion Site Failures

BACKGROUND

Personal insulin pumps have shown to be effective in improving the quality of therapy for people with type 1 diabetes (T1D). However, the safety of this technology is limited by the possible infusion site failures, which are linked with hyperglycemia and ketoacidosis. Thanks to the large availability of collected data provided by modern therapeutic technologies, machine learning algorithms have the potential to provide new way to identify failures early and avert adverse events.

METHODS

A clinical dataset (N = 20) is used to evaluate a novel method for detecting real-time infusion site failures using unsupervised anomaly detection algorithms, previously proposed and developed on in-silico data. An adapted feature engineering procedure is introduced to make the method able to operate in the absence of a closed-loop (CL) system and meal announcements.

RESULTS

In the optimal configuration, we obtained a performance of 0.75 Sensitivity (15 out of 20 total failures detected) and 0.08 FP/day, outperforming previously proposed literature algorithms. The algorithm was able to anticipate the replacement of the malfunctioning infusion sets by ~2 h on average.

CONCLUSIONS

On the considered dataset, the proposed algorithm showed the potential to improve the safety of patients treated with sensor-augmented pump systems.

A Visual Fault Detection Method for Induction Motors Based on a Zero-Sequence Current and an Improved Symmetrized Dot Pattern

Motor faults, especially mechanical faults, reflect eminently faint characteristic amplitudes in the stator current. In order to solve the issue of the motor current lacking effective and direct signal representation, this paper introduces a visual fault detection method for an induction motor based on zero-sequence current and an improved symmetric dot matrix pattern. Empirical mode decomposition (EMD) is used to eliminate the power frequency in the zero-sequence current derived from the original signal. A local symmetrized dot pattern (LSDP) method is proposed to solve the adaptive problem of classical symmetric lattice patterns with outliers. The LSDP approach maps the zero-sequence current to the ultimate coordinate and obtains a more intuitive two-dimensional image representation than the time-frequency image. Kernel density estimation (KDE) is used to complete the information about the density distribution of the image further to enhance the visual difference between the normal and fault samples. This method mines fault features in the current signals, which avoids the need to deploy additional sensors to collect vibration signals. The test results show that the fault detection accuracy of the LSDP can reach 96.85%, indicating that two-dimensional image representation can be effectively applied to current-based motor fault detection.

Condition Monitoring of the Dampers in the Railway Vehicle Suspension Based on the Vibrations Response Analysis of the Bogie

This paper investigates the possibility of developing a new method for fault detection of a damper in the primary suspension of the railway vehicle, based on the analysis of the vertical vibration's response of the bogie. To this purpose, experimental data are used, along with results from numerical simulations regarding the Root Mean Square (RMS) accelerations measured/simulated in four reference bogie points-two points on the chassis, against the suspension, and two points located against the axle boxes. The experimental data are utilized to define the normal area of operating and the damper failure area in the bogie primary suspension, as well as a basis for validating the results of numerical simulations. The numerical simulations are developed on the basis of two original models of the vehicle-track system, rigid-flexible coupled type, which take into account the elasticity of the vehicle carbody and the elasticity of the wheel-rail contact: a reference model with 15 degrees of freedom, for simulating the bogie response to vertical vibrations for the normal operating of the primary suspension dampers, and an extended model with 20 degrees of freedom, for simulating the bogie vibration response to the failure damper of a primary suspension. The presented results show that there are clear premises on the possibilities of developing a fault detection method of any of the four dampers of the primary suspension corresponding to a vehicle bogie, based on the RMS accelerations measured only in two reference points of the bogie.

Condition Monitoring of Ball Bearings Based on Machine Learning with Synthetically Generated Data

Rolling element bearing faults significantly contribute to overall machine failures, which demand different strategies for condition monitoring and failure detection. Recent advancements in machine learning even further expedite the quest to improve accuracy in fault detection for economic purposes by minimizing scheduled maintenance. Challenging tasks, such as the gathering of high quality data to explicitly train an algorithm, still persist and are limited in terms of the availability of historical data. In addition, failure data from measurements are typically valid only for the particular machinery components and their settings. In this study, 3D multi-body simulations of a roller bearing with different faults have been conducted to create a variety of synthetic training data for a deep learning convolutional neural network (CNN) and, hence, to address these challenges. The vibration data from the simulation are superimposed with noise collected from the measurement of a healthy bearing and are subsequently converted into a 2D image via wavelet transformation before being fed into the CNN for training. Measurements of damaged bearings are used to validate the algorithm’s performance.

Improvements in Medical System Safety Analytics for Authentic Measure of Vital Signs Using Fault-Tolerant Design Approach

The study presents a novel design method that improves system availability using fault-tolerant features in a non-invasive medical diagnostic system. This approach addresses the effective detection of functional faults, improves the uninterruptible system operating period with reduced false alarms, and provides an authentic measure of vital cardiac signs using diverse multimodal sensing elements like the photoplethysmogram (PPG) and the ECG. Most systems rely on a 1oo1 (one-out-of-one) design method, which inherently limits accuracy in existing practice. In this proposed approach, the quality of segregated authentic vital sign measured values could tremendously benefit the performance of resourceful nursing with negligible alarm fatigue and predict illness more accurately. The system builds upon the selected 2oo2 (two-out-of-two) safety-related design architecture and is evaluated with implemented functions like the fault detection and identification logic, the correlation coefficient-based safety function, and the fault-tolerant safe degradation switching mechanism for accurate measurements. The system was tested on 50 adults of various age groups. The analyzed captured data showed highly accurate vital sign data in this fault-tolerant approach with reduced false alarms. The proposed design method evaluated safety-related mechanisms along with a combination of the same and diverse sensors in a medical monitoring device, showing more reliable functioning of the system and authentic data for better nursing. This design approach showed a 45–55% increased improvement in system availability, thus allowing for accurate and uninterruptable tracking of vital signs for better nursing during critical times in the ICU.

Providing Fault Detection from Sensor Data in Complex Machines That Build the Smart City

Household appliances, climate control machines, vehicles, elevators, cash counting machines, etc., are complex machines with key contributions to the smart city. Those devices have limited memory and processing power, but they are not just actuators; they embed tens of sensors and actuators managed by several microcontrollers and microprocessors communicated by control buses. On the other hand, predictive maintenance and the capability of identifying failures to avoid greater damage of machines is becoming a topic of great relevance in Industry 4.0, and the large amount of data to be processed is a concern. This article proposes a layered methodology to enable complex machines with automatic fault detection or predictive maintenance. It presents a layered structure to perform the collection, filtering and extraction of indicators, along with their processing. The aim is to reduce the amount of data to work with, and to optimize them by generating indicators that concentrate the information provided by data. To test its applicability, a prototype of a cash counting machine has been used. With this prototype, different failure cases have been simulated by introducing defective elements. After the extraction of the indicators, using the Kullback–Liebler divergence, it has been possible to visualize the differences between the data associated with normal and failure operation. Subsequently, using a neural network, good results have been obtained, being able to correctly classify the failure in 90% of the cases. The result of this application demonstrates the proper functioning of the proposed approach in complex machines.

A Low-Cost Multi-Sensor Data Acquisition System for Fault Detection in Fused Deposition Modelling

Fused deposition modelling (FDM)-based 3D printing is a trending technology in the era of Industry 4.0 that manufactures products in layer-by-layer form. It shows remarkable benefits such as rapid prototyping, cost-effectiveness, flexibility, and a sustainable manufacturing approach. Along with such advantages, a few defects occur in FDM products during the printing stage. Diagnosing defects occurring during 3D printing is a challenging task. Proper data acquisition and monitoring systems need to be developed for effective fault diagnosis. In this paper, the authors proposed a low-cost multi-sensor data acquisition system (DAQ) for detecting various faults in 3D printed products. The data acquisition system was developed using an Arduino micro-controller that collects real-time multi-sensor signals using vibration, current, and sound sensors. The different types of fault conditions are referred to introduce various defects in 3D products to analyze the effect of the fault conditions on the captured sensor data. Time and frequency domain analyses were performed on captured data to create feature vectors by selecting the chi-square method, and the most significant features were selected to train the CNN model. The K-means cluster algorithm was used for data clustering purposes, and the bell curve or normal distribution curve was used to define individual sensor threshold values under normal conditions. The CNN model was used to classify the normal and fault condition data, which gave an accuracy of around 94%, by evaluating the model performance based on recall, precision, and F1 score.

Intelligent Fault Diagnosis Framework for Modular Multilevel Converters in HVDC Transmission

Open circuit failure mode in insulated-gate bipolar transistors (IGBT) is one of the most common faults in modular multilevel converters (MMCs). Several techniques for MMC fault diagnosis based on threshold parameters have been proposed, but very few studies have considered artificial intelligence (AI) techniques. Using thresholds has the difficulty of selecting suitable threshold values for different operating conditions. In addition, very little attention has been paid to the importance of developing fast and accurate techniques for the real-life application of open-circuit failures of IGBT fault diagnosis. To achieve high classification accuracy and reduced computation time, a fault diagnosis framework with a combination of the AC-side three-phase current, and the upper and lower bridges’ currents of the MMCs to automatically classify health conditions of MMCs is proposed. In this framework, the principal component analysis (PCA) is used for feature extraction. Then, two classification algorithms—multiclass support vector machine (SVM) based on error-correcting output codes (ECOC) and multinomial logistic regression (MLR)—are used for classification. The effectiveness of the proposed framework is validated by a two-terminal simulation model of the MMC-high-voltage direct current (HVDC) transmission power system using PSCAD/EMTDC software. The simulation results demonstrate that the proposed framework is highly effective in diagnosing the health conditions of MMCs compared to recently published results.