Development Of Condition Monitoring System Research Articles

A framework for the practical development of condition monitoring systems with application to the roller compactor

Implementing a condition-based maintenance strategy requires an effective condition monitoring (CM) system that can be complicated to develop and even harder to maintain. In this paper, we review the main complexities of developing condition monitoring systems and introduce a four-stage framework that can address some of these difficulties. The framework achieves this by first using process knowledge to create a representation of the process condition. This representation can be broken down into simpler modules, allowing existing monitoring systems to be mapped to their corresponding module. Data-driven models such as machine learning models could then be used to train the modules that do not have existing CM systems. Even though data-driven models tend to not perform well with limited data, which is commonly the case in the early stages of pharmaceutical process development, application of this framework to a pharmaceutical roller compaction unit shows that the machine learning models trained on the simpler modules can make accurate predictions with novel fault detection capabilities. This is attributed to the incorporation of process knowledge to distill the process signals to the most important ones vis-à-vis the faults under consideration. Furthermore, the framework allows the holistic integration of these modular CM systems, which further extend their individual capabilities by maintaining process visibility during sensor maintenance.

Generation of reproducible and scalable local damage to rolling bearings for the development of condition monitoring systems through the use of laser technology

Abstract In the context of the fourth industrial revolution, condition monitoring and predictive maintenance of machines are becoming increasingly important, as unplanned machine failures can lead to high costs for the operator. Many of these machine failures can be traced back to bearing damage, which must be prevented. In order to be able to make reliable statements about the remaining service life of the bearings, a condition monitoring system is required. For the development of these systems, bearings with a defined level of damage are needed. Since there are not enough reproducible methods for bearing damage so far, the defined generation of reproducible, local bearing damage is researched by using laser technology. The damage caused is checked and evaluated by means of vibration analysis and white light interferometry. The results confirm a reproducibility and scalability of the bearing damages by means of laser technology. The damaged bearings can be used for research-based development as well as in industrial implementations of condition monitoring systems.

A Comprehensive Framework for the Modular Development of Condition Monitoring Systems for a Continuous Dry Granulation Line

The development of condition monitoring systems often follows a modular scheme where some systems are already embedded in certain equipment by their manufacturers, and some are distributed across various equipment and instruments. This work introduces a framework for guiding the modular development of monitoring systems and integrating them into a comprehensive model that can handle uncertainty of predictions from the constituent modules. Furthermore, this framework improves the robustness of the modular condition monitoring systems as it provides a methodology for maintaining quality assurance and preventing unnecessary shutdowns in the event of some modules going off-line due to condition-based maintenance interventions.

A Model for Research of Transitions among Different Care Settings in Smart Lifetime Neighbourhoods

Neighbourhoods in EU member states are ageing, increasingly becoming the subjects of geo-gerontology and supply network analyses. Geo-gerontological aspects of the care and supply networks are becoming essential in the processes of deinstitutionalisation of long-term care and services and are crucial for the successful development and implementation of Ambient Assisted Living technologies (AAL_T). A growing number of older adults suffer from declining functional capacities and, thereby, experience decreased mobility and increasing frailty, which can lead to difficulties in carrying out essential activities of daily living as well as increased exposure to the risk of falls. Aid in managing one's daily life needs to be provided on time and in a properly built environment. Ambient intelligence for real-time monitoring of daily activities and supporting older residents with AAL_T will become essential means by which ageing cities will be able to maintain and improve the quality of life for older adults with declining functional capacities. This development will further influence the delineation of service areas and the timing of services. AAL_T and ambient intelligence will enable older adults to live longer in their dwellings and specialised facilities in the community and postpone or even avoid relocation to a nursing home altogether. Development in AAL_T has raised the possibility of developing more efficient services if the functional area of service networks is optimised. Based on these developments, we propose the adaptation of a multi-state competing risk model based on time to failure as modelled in reliability engineering to model and control the trajectories of declining functional capacities and related events that lead to the ill-health and disabilities. In this model, the built environment and timing in service networks are supposed to be embedded. The final goal of such a model is to support decision making that would extend the physically and mentally active life of seniors and enable them to live autonomously and be engaged with their communities. This paper aims to review the development of condition monitoring systems in ambient technologies, shape the model and propose a future research agenda.

Condition Monitoring using Convolutional Neural Network in Agricultural Machinery - Use Case: Disc Mower

Recent advances in sensing technologies and increasing computation power have accelerated the development of condition monitoring systems based on different approaches. There has been intensive research to automate the detection of anomalies in machines and processes by monitoring the changes in collected sensor data. Especially, a disc mower is prone to damage if it is frequently deployed in places where it might hit solid objects such as boulders and old fence posts. These anomalies cannot be easily recognized by the operator and may cause suboptimal results. In this paper, two deep learning models for intelligent condition monitoring in a disc mower are investigated to notify the machine operator when a failure occurs. For this, a basic convolutional neural network (CNN) and a residual neural network (ResNet) were trained, evaluated and the preliminary results are presented.

A neural network-based model for MCSA of inter-turn short-circuit faults in induction motors and its power hardware in the loop simulation

Induction motors (IMs) are one of the most commonly used rotating machines in industry. In order to avoid downtimes and economical losses, development of condition monitoring systems is of paramount importance. Inter-turn short circuit (ITSC) faults are one of the most commonly occurring faults in IMs. In this regard, motor current signature analysis (MCSA) is a low cost and noninvasive technique, requiring only the current signal to perform the condition monitoring. Also, the development of tools that emulate IM faults, allow the design, calibration and validation of MCSA based techniques. In this work, a multilayer neural network-based model to reproduce the current signatures associated with ITSC fault conditions is presented. The model considers both five severities of ITSC faults and four torque levels. This model is also implemented on a power hardware in the loop scheme to provide a novel tool to test condition monitoring systems of IMs.

Development of Condition Monitoring System for Electric Resistance Spot Welding Used to Manufacture Railway Car Bodies

Stainless steel railway car bodies are assembled by joining the outer plates and the pillar materials using resistance spot welding. In recent years, more and more car bodies are being assembled using laser welding in addition to the resistance spot welding. For this laser welding system, we developed a condition monitoring system considering the processes before and after laser welding as a single system, and obtained and put into practical use an appropriate condition that suppresses spatter generation during laser welding. On the other hand, in resistance spot welding, the current, weld time, electrode load, and electrode tip shape are the main factors that determine the welding quality. Therefore, the configuration of the equipment is less complicated than that of laser welding system, and the condition monitoring is easier than that of the laser welding. In this study, by transferring the concept of the condition monitoring system developed for laser welding to resistance spot welding, we achieved a reduction of more than 60% in terms of electricity consumption and improved the appearance of the car body by optimizing the indentation shape. In addition to this technical achievement, we also present in this paper a case study showing the opportunity for innovation by restructuring the technological paradigm of the resistance spot welding in the production of stainless steel car body shells.

Train Type Identification at S&C

The presented paper concerns the development of condition monitoring system for railroad switches and crossings that utilizes vibration data. Successful utilization of such system requires a robust and efficient train type identification. Given the complex and unique dynamical response of any vehicle track interaction, the machine learning was chosen as a suitable tool. For design and validation of the system, real on-site acceleration data were used. The resulting theoretical and practical challenges are discussed.

Design and Development of Condition Monitoring System for Wind Turbines Based on Generator Output Voltages

Design and Development of Condition Monitoring System for Wind Turbines Based on Generator Output Voltages

A systematic Failure Mode Effects and Criticality Analysis for offshore wind turbine systems towards integrated condition based maintenance strategies

Condition-based maintenance is applied in various industries to monitor and control critical assets and to optimize maintenance efforts. Its applicability to the offshore wind energy industry has been considered for almost 20 years and has resulted in the development and implementation of solutions that have contributed to lower cost of maintenance and increased asset availability. However, there is currently no public domain guidance available that provides the information required to (i) prioritize systems for which condition monitoring would generate highest value and to (ii) understand the parameters that need to be monitored by a specific system from failure cause to failure mode. Both items are addressed in this paper, providing a clearly structured, risk-based assessment methodology and corresponding results for state-of-the-art offshore wind turbines. A total of 337 failure modes have been identified and analysed by experts representing approximately 70% of the European offshore wind market to assess potential benefits of condition monitoring systems. Results may be used to target the development of condition monitoring systems focusing on critical systems and to find optimal O&M strategies by understanding failure paths of main offshore wind turbine systems resulting in a lower cost of energy and a more optimal risk-return balance.

An unsupervised spatiotemporal graphical modeling approach for wind turbine condition monitoring

Abstract The vast installment of wind turbines and the development of condition monitoring system provides large amounts of operational data for condition monitoring and health management, while the lack of labeled data becomes one of the major challenges for the data analytics. To address this issue, this work presents an unsupervised anomaly detection approach for wind turbine condition monitoring, where a spatiotemporal graphical modeling method, spatiotemporal pattern network (STPN), is applied to extract the spatial and temporal features between the variables in the system, and an energy-based model, stacked Restricted Boltzmann Machine (RBM) is used to capture the system-wide patterns and then applied for condition monitoring. Case studies on three data sets are carried out including: (1) anomaly detection on a benchmark model for fault detection and isolation, (2) anomaly detection on an experimental data set with the normal condition and 11 fault conditions and (3) online condition monitoring using real data from a wind farm in northwest China. The results show that the proposed approach is capable of detecting the anomalies without the need for labeling data.

Association rules discovery from diagnostic data-application to gearboxes used in mining industry

One of the key issues encountered in development of condition monitoring systems for industry is definition of decision rules in diagnostic system for determined diagnostic features. In practice, it appears very often that proposed algorithm is not effective for all technical assets of machinery park. The major cause is usually related to smaller or higher diversity of objects, mainly in terms of design features, operating conditions and wear level. These factors directly influence the profile of measured vibration signals, diagnostic features, thresholds, decision rules and so on. In this paper authors propose the usage of Generalized Rule Induction (GRI) algorithm for association rules discovery from data base of the Computerized Maintenance Management System (CMMS) - patterns hidden in data reflecting existing processes phenomena, regularities, and expresses relationships between them. Such approach provides better interpretation of signals, and consequently, much more effective decision rules.

Development of Condition Monitoring System for Laser Welding System of Stainless Steel Sheets

The method of the quality control in the laser welding has various techniques. The laser welding monitor system is one of the many kinds of technique concerning of quality control, and is marketed. However, it is necessary not only to manage in-process in the welding system but also to manage pre-process and post-process to control welding quality. Therefore, the management of the welding quality needs the monitoring in the above method. As the first, the condition monitoring technique uses the waveform analysis of the optical intensity in in-process of the laser welding process. In this report, the application experience to the half penetration welding with the YAG laser for thin stainless steel sheets is discussed.