Predictive Maintenance Techniques Research Articles

Improving linear transport infrastructure efficiency by automated learning and optimised predictive maintenance techniques (INFRALERT)

The on-going H2020 project INFRALERT aims to increase rail and road infrastructure capacity in the current framework of increased transportation demand by developing and deploying solutions to optimise maintenance interventions planning. It includes two real pilots for road and railways infrastructure. INFRALERT develops an ICT platform (the expert-based Infrastructure Management System, eIMS) which follows a modular approach including several expert-based toolkits. This paper presents the methodologies and preliminary results of the toolkits for i) nowcasting and forecasting of asset condition, ii) alert generation, iii) RAMS & LCC analysis and iv) decision support. The results of these toolkits in a meshed road network in Portugal under the jurisdiction of Infraestruturas de Portugal (IP) are presented showing the capabilities of the approaches.

Predictive maintenance based on event-log analysis: A case study

Predictive maintenance techniques are designed to help anticipate equipment failures to allow for advance scheduling of corrective maintenance, thereby preventing unexpected equipment downtime, improving service quality for customers, and also reducing the additional cost caused by over-maintenance in preventative maintenance policies. Many types of equipment--e.g., automated teller machines (ATMs), information technology equipment, medical devices, etc.--track run-time status by generating system messages, error events, and log files, which can be used to predict impending failures. Aiming at these types of equipment, we present a general classification-based failure prediction method. In our parameterized model, we systematically defined four categories of features to try to cover all possibly useful features, and then used feature selection to identify the most important features for model construction. The general solution is sufficiently flexible and complex to address failure prediction for target equipment types. We chose ATMs as the example equipment and used real ATM run-time event logs and maintenance records as experimental data to evaluate our method on the feasibility and effectiveness. In this paper, we also share insights on how to optimize the model parameters, select the most effective features, and tune classifiers to build a high-performance prediction model.

Estudio del impacto logístico – técnico que genera el mantenimiento predictivo en las PYMES de Milagro, Ecuador

This article is to discern the organizational culture regarding the acceptance of monitoring techniques and predictive inspections that help reduce the problems arising in the infrastructure of production processes within an industrial plant and thus optimize downtime that they occur due to ignorance or little attention and in some cases due to lack of awareness of operators or direct users of these computers. SMEs currently are optimizing their processes, automating their production line which implies that there should be better controls, to be more competitive because the increase in technology, greatly improves productivity. Before the technology boom happening in terms of machinery and processes, corrective measures were used to eliminate problems and these generated high rates in terms of costs for equipment maintenance, in the XXI century predictive maintenance techniques, consisting develops diagnose and maintain installed both in the electrical part and mechanical non-stop operational teams, allowing the reduction of lost production time and costs, costs of obtaining the most competitive end products in the market infrastructure. Also analyze the type of maintenance they currently possess SMEs in the city of Milagro-Ecuador and how it is influencing their productivity in order to develop proposals for improvements and recommendations in order to optimize their processes and increase profitability.

Predictive maintenance, its implementation and latest trends

This study covers new trends and techniques in the field of predictive maintenance, which has been superseding traditional management policies, at least in part. It also presents suggestions for how to implement a predictive maintenance programme in a factory/premise and so on. Predictive maintenance primarily involves foreseeing breakdown of the system to be maintained by detecting early signs of failure in order to make maintenance work more proactive. In addition to the aim of acting before failure, it also aims to attend to any fault, even if there is no immediate danger of failure, to ensure smooth operation and reduce energy consumption. Predictive maintenance has been adopted by various sectors in manufacturing and service industries in order to improve reliability, safety, availability, efficiency and quality as well as to protect the environment. It also has created a separate sector, which specializes in developing predictive maintenance instruments, offering dedicated predictive maintenance solutions and training predictive maintenance experts. Predictive maintenance techniques are closely associated with sensor technologies but for efficient predictive maintenance applications, a comprehensive approach, which integrates sensing with subsequent maintenance activities, is needed to be adapted in accordance with the needs of the particular organization. Recent advances in information, communication and computer technologies, such as Internet of Things and radio-frequency identifications, have been enabling predictive maintenance applications to be more efficient, applicable, affordable, and consequently more common and available for all sorts of industries. Researches on remote maintenance and e-maintenance have been supporting predictive maintenance activities especially in unsafe working environments and scattered locations.

Análisis de Ruido en el Diseño y Construcción de una Cámara Semianecoica para Diagnóstico de Motores de Combustión Interna

The results of the design, implementation and evaluation of a semi-anechoic chamber to guarantee reliability in data acquisition for diagnosis of mechanical systems are presented. Noise analysis is one of the predictive maintenance techniques that detect faults based on the monitoring of characteristic parameters of proper functioning of a system. To do this, the characteristics of insulating and absorbent material were determined, and their arrangement inside the chamber. The evaluation was performed using the analysis of acoustic signals emitted by a source (internal combustion Diesel engine) under its normal and failure operation. These signals were acquired using a sound level meter and analyzed in the frequency domain using fast Fourier transform.

AI Fitness Checks, Maintenance and Monitoring on Systems Managing Content & Data: A Study on CMS World

Artificial intelligence health checks, monitoring, and maintenance on system managing content in the CMS world are essential to prevent damage and avoidable expenses. To also avoid loss of time from massive downtime, AI has been employed for maintenance functions. Servers which are the houses and homes for data and content from computers, have to be maintained to allow the system to function at an optimal level. The use of AI for this maintenance involves a lot of factors. This work discusses these factors and how the various aspects work to monitor and maintain systems managing content and data. AI is programmed using health monitoring scripts which are commands executed by a particular programming language. This scripts program the AI to check and monitor servers for maintenance function. This maintenance aims to track CPU optimization and allow for the best functioning of the CPU using preventive and predictive maintenance techniques.

Sensor Based Intelligent Systems for Detection and Diagnosis

Sensors are very commonly used as monitoring and observing elements in real world applications. However, sensors cannot provide a solution for systems, because data obtained with sensors should be processed for object functions. There are many approaches for processing and decisionmaking methods in industrial applications. In particular, it is very important for intelligent systems which used condition monitoring, detection, diagnosis, prognosis, and predictive maintenance. The predictive maintenance basically consists of four components. These components are the condition monitoring, the detection, the diagnosis, and the prognosis for identification of anomalies and faulty conditions. The condition monitoring is monitoring of one or several parameters to identify an unexpected situation. The detection is indicating if there is an abnormal status using observable effects in a monitored system. The diagnosis is the process of detecting and determining where that anomaly is. The prognosis is predicting and estimating useful and healthy status of a system. A lot of predictive maintenance techniques have been improved for detection, diagnosis, and prognosis of faults in the literature. The detection, diagnosis, and prognosis methods use measurements and preprocessed data using the condition monitoring. Then, the predictive maintenance methods provide a result for health of system. But systems have many uncertainties and they affect the performance of the predictive maintenance method. Therefore, intelligent systems are used to avoid this disadvantage. Process flow of the predictive maintenance is as follows. Firstly, actual functions or signals in any system for the condition monitoring aremeasured. Secondly, normal operation or faulty operation of the system for early recognition is determined in the anomaly detection.Thirdly, anomaly classification for correct assignment can be carried out for the diagnosis. Next, the prognosis stage can be used to estimate or predict anomalies. Finally, a result is obtained with decision-making. An efficient predictive maintenance method should have some specifications. First of all, minimum information and sensor should be required. Other specifications are accurate and confident prediction, fast response, and robustness. Optimization algorithms such as genetic algorithms, artificial immune systems, and advancedMonte Carlomethods, learning algorithms such as artificial neural networks and support vector machines, and reasoning algorithms such as fuzzy logic systems, clustering algorithms, particle filtering algorithms,wavelet analysis algorithms, and principal component analysis algorithm can be used for the predictivemaintenance methods. These methods can be used in data preprocessing, feature extraction, and feature selection too. Finally, the predictive maintenance methods using computational intelligence techniques are very important for improving production and manufacturing and product quality. If the faulty condition in a system can be determined with the detection, the diagnosis, or the prognosis in early stage, most problems can be repaired at this time.

Implementation of a novel double-side technique for partial discharge detection and location in covered conductor overhead distribution networks

Partial discharge (PD) detection has proven to be one of the most acceptable techniques for on-line condition monitoring and predictive maintenance of power apparatus. A powerful tool for detecting PD in covered-conductor (CC) lines is urgently needed to improve the asset management of CC overhead distribution lines. In this paper, an appropriate, portable and simple system designed to detect PD activity in CC lines and ultimately pinpoint the PD source is developed and tested. The system is based on a novel double-side synchronised PD measurement technique driven by pulse injection. Emphasis is placed on the proposed PD-location mechanism and hardware structure, with descriptions of the pulse-injection process, detection device, synchronisation principle and PD-location algorithm. The system is simulated using ATP-EMTP, and the simulated results are found to be consistent with the actual simulation layout. For further validation, the capability of the system is tested in a high-voltage laboratory experiment using a 10-kV CC line with cross-linked polyethylene insulation.

Estratégia de manutenção preditiva no departamento gráfico de uma empresa do ramo fumageiro

O objetivo deste artigo foi descrever a implantação de uma estratégia de Manutenção Preditiva no departamento gráfico de uma empresa da indústria fumageira. O método de pesquisa foi a pesquisa-ação. Foram revisadas e usadas as técnicas de Analise de Vibração, Termografia, Ferrografia e Inspeção Sensitiva. Foi construída e aplicada uma estratégia de Manutenção Preditiva, integrada à estratégia de produção do departamento gráfico. Na pesquisa, foram levantados, no sistema de informação da empresa, o índice de quebra de equipamentos (downtime), custos com desperdícios de energéticos, custos de manutenção e disponibilidade de H/h (homem/hora) e disponibilidade de equipamentos para a produção no período anterior a implantação de uma gestão de manutenção preditiva consistente e comparada com resultados após esta implantação. A partir dos resultados obtidos, ficou evidente que somente a manutenção preventiva, associada a um modelo parcial de manutenção preditiva, não foi suficiente para obter e manter a confiabilidade desejada em um departamento crítico para a empresa. A pesquisa concluiu que a implantação de uma estratégia de manutenção preditiva pode trazer resultados positivos em indicadores estratégicos que capturam objetivos compartilhados com outros setores da empresa, principalmente a produção.

SIMULATION OF THERMAL DECOMPOSITION OF MINERAL INSULATING OIL

Dissolved gas analysis (DGA) has been applied for decades as the main predictive maintenance technique for diagnosing incipient faults in power transformers since the decomposition of the mineral insulating oil (MIO) produces gases that remain dissolved in the liquid phase. Nevertheless, the most known diagnostic methods are based on findings of simplified thermodynamic and compositional models for the thermal decomposition of MIO, in addition to empirical data. The simulation results obtained from these models do not satisfactorily reproduce the empirical data. This paper proposes a flexible thermodynamic model enhanced with a kinetic approach and selects, among four compositional models, the one offering the best performance for the simulation of thermal decomposition of MIO. The simulation results obtained from the proposed model showed better adequacy to reported data than the results obtained from the classical models. The proposed models may be applied in the development of a phenomenologically-based diagnostic method.

Development and evaluation of a new DGA diagnostic method based on thermodynamics fundamentals

The dissolved gas analysis (DGA) is a well-accepted power transformer predictive maintenance technique. However, despite the importance of the technique, classical diagnostic methods were developed from observations of simplified thermodynamic and compositional models. This paper presents the development of a DGA diagnostic method based on a thermodynamic model that includes a contribution inspired on the kinetics of the process. The performances of the new phenomenological proposition and of classical DGA diagnostic methods are compared and discussed. Also, a general procedure for the application of the new diagnostic method is described.

Influence of Field of View of Thermal Imagers and Angle of View on Temperature Measurements by Infrared Thermovision

Temperature measurement by infrared thermography is a powerful technique for predictive maintenance. To make accurate diagnoses, errors associated with temperature measurement results should be minimized because the diagnosis is obtained by applying temperature difference criteria in consecutive measurements or in similar equipment. Temperature measurement by infrared thermography is susceptible to errors caused by many things. We study and assess the effectiveness obtained in use of experimental models for error correction in temperature measurement by thermal imagers due to the influence of the field of view of the imager's lens, combined with varying viewing angles between the analyzed object and the imager. The results indicate that these factors can introduce errors that compromise the diagnosis. Experimental models based on artificial neural network and on regression were developed to correct such errors.

Predictive maintenance of a worm reducer with an unbalanced load

Among all predictive maintenance techniques the oil analysis and vibration analysis are the most important for monitoring some mechanical systems. The integration of these techniques has potential to improve industrial maintenance practices and provide a better economic gain for industries. To study the integration of these two techniques, a test rig was set up to obtain an extreme working condition for the worm reducer used in this paper. The test rig was composed by a motor connected to a reducer through a flexible coupling and with an unbalanced load. The analysis of the results carried out by using a sample of the oil recommended by the manufacturer in extreme conditions, and using liquid contaminant is presented. From the results it was observed that if there is an abnormal instantaneous load in a system, the subsequent vibration analysis may not perceive what occurred if there was no permanent damage, which is not the case with the lubricant analysis.

Key Equipment Identification model for correcting milepost errors of track geometry data from track inspection cars

Track geometry data from track inspection cars is an important data source for maintenance-of-way departments to evaluate track geometry irregularities. However, there are errors in mileposts of track geometry data. The milepost errors not only increase work intensity of track maintenance workers and decrease track maintenance window utilization, but also have negative influences on predictive track maintenance technique researches currently carried out around the world. To reduce milepost errors, milepost correction systems employing modern positioning technologies, such as Global Positioning System (GPS), Differential Global Positioning System (DGPS) and Radio Frequency Identification (RFID), have been incorporated into track geometry measurement systems on the track inspection cars. Due to various factors peculiar to Chinese railroad systems, these correcting systems don’t seem to offer satisfactory solutions to the milepost correction problem. Field investigations have revealed that the milepost position could be off up to 200m. To address the issue, a model named Key Equipment Identification (KEI) has been developed in this article using maintenance-of-way infrastructure data and track geometry data. KEI can automatically identify curves in main tracks and sidetracks, and diverging tracks of turnouts. These three kinds of equipment are referred to as Key Equipment (KE). This identified equipment list provides accurate references for correcting mileposts of track geometry data. Following KEI development, performance of KEI and milepost errors of corrected track geometry data are analyzed using maintenance-of-way infrastructure data and track geometry data collected from the Jinan bureau of China Railroads, one of 18 railroad bureaus in China. The analysis results show that all passed KEs in the course of inspections are accurately identified from track geometry data and are precisely positioned in the inspection data. After mileposts of track geometry data are corrected according to outputs of KEI, differences between actual mileposts and corrected mileposts are far below 5m and milepost differences of sampling points between two corrected inspections fall in the range of 1m.

Wind turbine condition monitoring: technical and commercial challenges

ABSTRACTDeployment of larger scale wind turbine systems, particularly offshore, requires more organized operation and maintenance strategies to ensure systems are safe, profitable and cost‐effective. Among existing maintenance strategies, reliability centred maintenance is regarded as best for offshore wind turbines, delivering corrective and proactive (i.e. preventive and predictive) maintenance techniques enabling wind turbines to achieve high availability and low cost of energy. Reliability centred maintenance analysis may demonstrate that an accurate and reliable condition monitoring system is one method to increase availability and decrease the cost of energy from wind. In recent years, efforts have been made to develop efficient and cost‐effective condition monitoring techniques for wind turbines. A number of commercial wind turbine monitoring systems are available in the market, most based on existing techniques from other rotating machine industries. Other wind turbine condition monitoring reviews have been published but have not addressed the technical and commercial challenges, in particular, reliability and value for money. The purpose of this paper is to fill this gap and present the wind industry with a detailed analysis of the current practical challenges with existing wind turbine condition monitoring technology. Copyright © 2012 John Wiley & Sons, Ltd.

Novel wide‐area fault location and predictive maintenance technique for distribution system

AbstractLocating faults in a power network has been a common chore for utility personnel. However, with the increasing demand in consumer expectations nowadays, fault location time needs to be shortened so that supply can be restored as soon as possible. Considerable amount of time can be saved if an automated mechanism is available to direct the repair crew to the region of fault. Transmission system has already adopted the automated fault location algorithm in their relay. However, implementing such an approach in wide‐area distribution network would require huge capital expenditure and resources. This paper presents a reliable, automated fault location algorithm within a wide‐area distribution network using a limited number of monitoring devices. In addition, the proposed algorithm is also capable of offering a means of predictive maintenance strategy for the utilities. The algorithm is able to identify the faulted branches in a radial distribution system with high accuracy. In most cases, the calculated fault location is very close to the actual one. This algorithm has also overcome the drawbacks of existing techniques for automated fault location in wide‐area distribution network. © 2011 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

State-of-the-Art Predictive Maintenance Techniques*

Condition-based maintenance techniques for industrial equipment and processes are described in this paper together with examples of their use and discussion of their benefits. These techniques are divided here into three categories. The first category uses signals from existing process sensors, such as resistance temperature detectors (RTDs), thermocouples, or pressure transmitters, to help verify the performance of the sensors and process-to-sensor interfaces and also to identify problems in the process. The second category depends on signals from test sensors (e.g., accelerometers) that are installed on plant equipment (e.g., rotating machinery) in order to measure such parameters as vibration amplitude. The vibration amplitude is then trended to identify the onset of degradation or failure. This second category also includes the use of wireless sensors to provide additional points for collection of data or allow plants to measure multiple parameters to cover not only vibration amplitude but also ambient temperature, pressure, humidity, etc. With each additional parameter that can be measured and correlated with equipment condition, the diagnostic capabilities of the category can increase exponentially. The first and second categories just mentioned are passive, which means that they do not involve any perturbation of the equipment or the process being monitored. In contrast, the third category is active. That is, the third category involves injecting a test signal into the equipment (sensors, cables, etc.) to measure its response and thereby diagnose its performance. For example, the response time of temperature sensors (RTDs and thermocouples) can be measured by the application of the step current signal to the sensor and analysis of the sensor response to the application of the step current. Cable anomalies can be located by a similar procedure referred to as the time domain reflectometry (TDR). This test involves a signal that is sent through the cable to the end device. Its reflection is then recorded and compared to a baseline to identify impedance changes along the cable and thereby identify and locate anomalies. Combined with measurement of cable inductance (L), capacitance (C), and loop resistance (R), or LCR testing, the TDR method can identify and locate anomalies along a cable, identify moisture in a cable or end device, and even reveal gross problems in the cable insulation material. There are also frequency domain reflectometry (FDR) methods, reverse TDR, trending of insulation resistance (IR) measurement, and other techniques which can be used in addition to or instead of TDR and LCR to provide a wide spectrum of tools for cable condition monitoring. The three categories of techniques described in this paper are the subject of current research and development projects conducted by the author and his colleagues at the AMS Corporation with funding from the U.S. Department of Energy (DOE) under the Small Business Innovation Research (SBIR) program.

Predictive maintenance of a reducer with contaminated oil under an excentrical load through vibration and oil analysis

Among all Predictive Maintenance techniques the oil analysis and vibration analysis are the most important for monitoring some systems. The integration of these techniques has the potential to revolutionize industrial practices and provide a large economic gain for industries. To study the integration of both techniques a bench test was set up and put to work to the extreme limit of use. Tests were carried out with the lubricant recommended by the manufacturer of the equipment, using lubricants supplemented with various percentages of liquid contaminant and lubricants supplemented with several percentages of solid contaminant. This paper presents the results of the first test, that is, with the oil recommended by the manufacturer in extreme conditions. From the results it was observed that if in a system an abnormal occurrence takes place, for example an extra load during a certain period of time, the lubricant analysis can be used together with the vibration analysis to complement it.

State-of-the-Art Predictive Maintenance Techniques

Condition-based maintenance techniques for industrial equipment and processes are described in this paper together with examples of their use and discussion of their benefits. These techniques are divided here into three categories. The first category uses signals from existing process sensors, such as resistance temperature detectors (RTDs), thermocouples, or pressure transmitters, to help verify the performance of the sensors and process-to-sensor interfaces and also to identify problems in the process. The second category depends on signals from test sensors (e.g., accelerometers) that are installed on plant equipment (e.g., rotating machinery) in order to measure such parameters as vibration amplitude. The vibration amplitude is then trended to identify the onset of degradation or failure. This second category also includes the use of wireless sensors to provide additional points for collection of data or allow plants to measure multiple parameters to cover not only vibration amplitude but also ambient temperature, pressure, humidity, etc. With each additional parameter that can be measured and correlated with equipment condition, the diagnostic capabilities of the category can increase exponentially. The first and second categories just mentioned are passive, which means that they do not involve any perturbation of the equipment or the process being monitored. In contrast, the third category is active. That is, the third category involves injecting a test signal into the equipment (sensors, cables, etc.) to measure its response and thereby diagnose its performance. For example, the response time of temperature sensors (RTDs and thermocouples) can be measured by the application of the step current signal to the sensor and analysis of the sensor response to the application of the step current. Cable anomalies can be located by a similar procedure referred to as the time domain reflectometry (TDR). This test involves a signal that is sent through the cable to the end device. Its reflection is then recorded and compared to a baseline to identify impedance changes along the cable and thereby identify and locate anomalies. Combined with measurement of cable inductance (L), capacitance (C), and loop resistance (R), or LCR testing, the TDR method can identify and locate anomalies along a cable, identify moisture in a cable or end device, and even reveal gross problems in the cable insulation material. There are also frequency domain reflectometry (FDR) methods, reverse TDR, trending of insulation resistance (IR) measurement, and other techniques which can be used in addition to or instead of TDR and LCR to provide a wide spectrum of tools for cable condition monitoring. The three categories of techniques described in this paper are the subject of current research and development projects conducted by the author and his colleagues at the AMS Corporation with funding from the U.S. Department of Energy (DOE) under the Small Business Innovation Research (SBIR) program.

Applications of predictive maintenance techniques in industrial systems

Prognostic methods represent a new methodology for system maintenance which offers significant time and cost savings. The paper offers a short overview of the available prognosis techniques and proposes the implementation of one model-based and one data-driven method. As a representative of the model-based methods the autoregressive moving average (ARMA) modeling approach is chosen. The estimated model parameters are further used for implementing the early change detector which is realized as a Neyman-Pearson hypothesis test. On the other hand, hidden Markov model (HMM) based prognosis illustrates the use of data-driven techniques. Using the cross-correlation input-output functions, HMM prognosis algorithm is proposed, as a suitable way of timely detection. Both techniques were implemented to detect performance changes of the water level sensor in a steam separator system in thermal power plants.