Online Condition Monitoring Systems Research Articles

SUSTAINABLE DEVELOPMENT AND ADAPTIVE MANAGEMENT CONCEPTS IN THE CEMENT ENTERPRISES OF THE INDUSTRY 4.0

Sustainable development is one the main strategies for advanced cement enterprises in the industry 4.0. There are applied modern technologies and methods resulted in sustainable development and competitive advantage of the Holcim Group. According to the objectives of the Holcim Group it is announced to introduce a new reference for the sustainable construction industry by 2030 and 2050. These goals have been validated by the Science Based Targets initiative. [6] Therefore, managers of cement enterprises introduce new advanced methods of adaptive management to implement the above-mentioned goals to survive in the market. In order to be successful in these tasks there are applied advanced products and solutions to accelerate the transition to green building process. On the basis of these goals there are demanded the advanced techniques and technologies like Computerized Maintenance Management Systems, Fuzzy Recurrent Neural Network, Condition-Based Maintenance, online Condition Monitoring Systems in the cement enterprises of the Holcim Group. These advanced techniques and technologies will ensure successful implementation of the methods of adaptive management to achieve the sustainability. Keywords: sustainable development, adaptive management, Science Based Targets initiative (SBTi), Environmental, Social and Governance (ESG), Construction and Demolition Waste (CDW), Biodiversity Indicator Reporting System (BIRS), International Union for Conservation of Nature (IUCN), Computerized Maintenance Management Systems (CMMS), Fuzzy Recurrent Neural Network (RNN), Condition-Based Maintenance (CBM), online Condition Monitoring Systems (CMS).

Partial Discharge Analysis of SF6/N2 Gas Mixture Filled MV Circuit Breaker Using Time–Frequency Map Clustering Technique

Breakdown of gas insulated switchgears due to partial discharges (PD) is a major threat for the safe and reliable operation of power system network. In recent times, PD based diagnostics has been widely adopted by electrical utilities in order to evaluate the health of high voltage insulation system. However, very few reports are only available on the PD characteristics of gas insulated medium voltage (MV) circuit breaker prototypes, in particular with mixture of gases, over a wide range of pressure and voltage stress. Deep analysis of nature of PD signals in mixture of gas insulation and its time–frequency characteristics is important in order to develop efficient online condition monitoring systems for circuit breakers. This paper is aimed at the analysis of PD characteristics of a 22 kV MV circuit breaker prototype filled with SF6/N2 gas mixtures using time–frequency (T-F) mapping and clustering technique. PD tests are carried out at different gas pressures (1–5 bar) and at different gas mixture ratios. Important PD parameters such as PD inception voltage, phase resolved PD (PRPD) pattern and PD repetition rate are evaluated. Clustering analysis of PD signals from the time–frequency mapping is carried out and formation of three different clusters consisting of short, long and repetitive PD pulses is identified. Online real time monitoring of individual cluster density will be helpful to understand the degradation of gas insulation of circuit breakers. Keeping in mind the minimal usage of SF6 gas, from the results, it is noted that 10% SF6 gas can be mixed with 90% N2 at 3 bar pressure to get satisfactory PD performance of a 22 kV medium voltage circuit breaker.

Early stage crack detection in mechanically joined steel/aluminum joints by condition monitoring

Abstract Monitoring systems for machines, plants, materials and equipment are increasingly used in production processes. These online condition monitoring systems can detect damage or excessive loads at an early stage and can drastically reduce or prevent long downtimes of plants and machines as well as high repair and maintenance costs. This paper depicts a method for online crack detection with pattern recognition methods for specimens joined by self-pierce riveting under cyclic load in fatigue tests (laboratory application). A software specially conceived for this application was developed. This software, AnrissMF, uses active acoustic testing with a structure-borne sensor to detect cracks in the joints at a very early stage. It is shown in this paper that this software can detect cracks much earlier than classical failure criteria for joints (i. e. before any drop in stiffness or frequency is observed). Furthermore, the successful application of software AnrissMF for online crack detection during the fatigue strength test is presented.

Early stage crack detection in mechanically joined steel/aluminum joints by condition monitoring

Abstract Monitoring systems for machines, plants, materials and equipment are increasingly used in production processes. These online condition monitoring systems can detect damage or excessive loads at an early stage and can drastically reduce or prevent long downtimes of plants and machines as well as high repair and maintenance costs. This paper depicts a method for online crack detection with pattern recognition methods for specimens joined by self-pierce riveting under cyclic load in fatigue tests (laboratory application). A software specially conceived for this application was developed. This software, AnrissMF, uses active acoustic testing with a structure-borne sensor to detect cracks in the joints at a very early stage. It is shown in this paper that this software can detect cracks much earlier than classical failure criteria for joints (i. e. before any drop in stiffness or frequency is observed). Furthermore, the successful application of software AnrissMF for online crack detection during the fatigue strength test is presented.

A Novelty Detector and Extreme Verification Latency Model for Nonstationary Environments

Safe and reliable operation of systems relies on the use of online condition monitoring and diagnostic systems that aim to take immediate actions upon the occurrence of a fault. Model-based solutions are often not practical in nonstationary environments. Thus, the evolving data stream requires the data-driven model to be adaptive. In this paper, we propose a framework for the fault detection and classification that is accomplished on the data stream with both the gradual and abrupt drifts. The framework is only provided with prior information about the possible faults at the initial step; however, despite this, the framework can still detect the novel faults without receiving any update. Furthermore, an efficient fault classification algorithm is presented to maximize the efficiency of the proposed framework. Finally, the proposed framework is applied for diagnosing bearing defects in the induction motors to demonstrate its feasibility for industrial applications.

Novel engineering techniques to overcoming traditional challenges in online condition monitoring systems for power transformer

Traditional challenges in online condition monitoring systems for power transformer are usually due to discrepant results between online condition monitoring system and lab-based measurements (e.g....

On-line condition monitoring importance and evolution

In recent years there has been an increase in the availability of on-line condition monitoring devices and systems and of monitoring installations in utility networks. Their installation is judged considering a lot of factors including: general utility strategy, expected value from such an installation, asset and risk management, utility best practices, economic, environmental or regulatory pressures, and will vary from one utility to another. What is the present experience, the value from on-line condition monitoring and what are the opportunities and challenges in future grids and considering modern technologies? The paper intends also to position on-line condition monitoring in life strategies and point to evolution and international activities related to the topic.

Multi‐DGAS: A pattern based educational framework design for power transformers faults interpretation and comparative performance analysis

AbstractTaking necessary measures in advance for early prediction of the transformer failures are very crucial. Dissolved gas in oil analysis (DGA) techniques that used in electrical power systems has become an internationally accepted standard in condition based maintenance (CBM) studies have high complexity and increasing demands of the quick‐service‐ready paradigm in power transformers. This standard includes the implementation steps of various complex DGA methods named basic gas, Rogers, Doernenburg, Duval triangle, and pentagon. Mentioned methods are quite complicated and their application steps are strict rules defined in detailed procedures. The adaptation of these methods to transformer maintenance procedures that prepared according to standards and tested for correctness creates a valuable tool. In this study, a pattern based educational framework has been designed to integrate state of arts DGA techniques help for trainers, graduate researchers working on DGA techniques, and also for maintenance teams while taking critical decisions on power transformers. Developed Multi‐DGAS educational framework is designed for modular construction that can be integrated with online condition monitoring and diagnostic systems for electric power systems. Success of the proposed framework revealed with newly proposed pentagon method in regard with comparison of classical methods by using oil samples which have been taken under the energized condition of different 135 transformers within multi‐phase which failure histories already known. Analysis of dissolved gases in transformer oil is performed according to ASTM‐D 3612 procedure using 3800 GC Varian gas chromatograph apparatus.

Estimate of Correlation between Transformer Diagnostic Variables with a Time Lag

An essential pre-requisite of faultless operation of electric steel-making furnaces is furnishing furnace transformers with on-line condition monitoring systems. The known monitoring systems are operated mostly in a fault indication mode. Efficiency of fault detection may be rationally improved with condition diagnostics based on total controlled parameters represented as time series. The paper substantiates and develops methods estimating correlation between transformer parameters with an available time lag. It provides an example of estimate of correlated variables: temperature and humidity concentration in oil of the furnace transformer. The authors give recommendation for expanded application of the developed software implementing the proposed algorithm.

Estimating uncertainty when using transient data in steady-state calculations

Abstract When using measurement data for monitoring there is often a desire for steady-state analysis. On-line condition monitoring and fault detection systems are typical applications where the traditional way of treating transient data is to remove it using methods that require tuning using thresholds. This paper suggests an alternative approach where the uncertainty estimate in a particular variable is increased in response to the presence of transients and through propagation, varies the uncertainty in the result accordingly. The formulation of the approach is described and applied to two examples from building HVAC systems. The approach is demonstrated to be a pragmatic tool that can be used to increase the robustness of calculations from time series data.

Możliwości zastosowania sieci neuronowej RBF do diagnostyki zużycia ostrza w systemie online

Możliwości zastosowania sieci neuronowej RBF do diagnostyki zużycia ostrza w systemie online

Lessons Learned in Fleetwide Asset Monitoring of Gas Turbines and Supporting Equipment in Power Generation Applications

Condition monitoring remains an important technology for equipment life cycle management. Historically, online condition monitoring systems are installed only on the most critical assets within a power plant, process plant, or manufacturing facility. Less critical equipment, while vital to operation of the plant, are only monitored or tested periodically using manual route based technologies. This historical practice leaves equipment specialists with a small amount of time for analysis of collected sensory data (vibration, temperature, oil, power, etc.) as they spend the vast majority of working hours collecting equipment sensory data. Fortunately, data acquisition technology has evolved, making it possible to transform standard and advanced machinery measurements from manual collections to online collections, increasing time for specialists to analyze, and yielding opportunities for automated diagnostics and prognostics. By taking advantage of automation, the ability of equipment owners and operators to lower life cycle costs and increase reliability of plant equipment is greatly improved.The transition from manual route based measurements to a fleetwide surveillance program touches many elements from sensors to networked data acquisition nodes to servers to historians and predictive technologies. Within power generation plants, installation costs, information technology strategies, and long term vision come together to create higher machine reliability at lower operational cost and new automation in performance monitoring, diagnostics, and advisory generation. With automation, comes increased sensory data from pumps and turbines that require new tools for data management, data mining, and data transformation into actionable information. A case study reviews the open and extensible data architecture of a fleetwide monitoring system deployed, the ongoing efforts, and current benefits delivered to the power generation industry participants.

Signal Processing Techniques to Improve an Acoustic Emissions Sensor


 
 
 Acoustic Emissions (AE) are stress waves produced by the sudden internal stress redistribution of material caused by changes in the internal structure of the material. Possible causes of these changes are crack initiation and growth, crack opening/closure, or pitting in monolithic materials (gear/ bearing material). Where as vibration can measure the effect of damage, AE is a direct measure of damage. Unfortunately, AE methodologies suffer from the need of high sample rates (4 to 10 Msps) and relatively immature algorithms for condition indictors (CI). This paper hypothesizes that the AE signature is the result of some forcing function (e.g. periodic motion in the case of rotating machinery). By using analog signal processing to demodulating the AE signature, one can reconstruct the information carried (e.g. modulation) by the AE signature and provide improved diagnostics. As most on-line condition monitoring systems are embedded system, analog signal processing techniques where used which reduce the effective sample rate needed to operate on the AE signal to those similarly found in traditional vibration processing systems. Further, by implementing another signal processing technique, time synchronous averaging, the AE signal is further enhanced. This allowed, for the first time, an AE signal to be used to identify a specific component within gearbox. This processing is tested on a split torque gearbox and results are presented.
 
 

Integrated online condition monitoring system for nuclear power plants

Abstract Online condition monitoring or online monitoring (OLM) uses data acquired while a nuclear power is operating to continuously assess the health of the plant's sensors, processes, and equipment; to measure the dynamic performance of the plant's process instrumentation; to verify in-situ the calibration of the process instrumentation; to detect blockages, voids, and leaks in the pressure sensing lines; to identify core flow anomalies; to extend the life of neutron detectors and other sensors; and to measure the vibration of reactor internals. Both the steady-state or DC output of plant sensors and their AC signal or noise output can be used to assess sensor health, depending on whether the application is monitoring a rapidly changing (e.g., core barrel motion) or a slowly changing (e.g., sensor calibration) process. The author has designed, developed, installed, and tested OLM systems (comprised of software/hardware-based data acquisition and data processing modules) that integrate low-frequency (1 mHz to 1 Hz) techniques such as RTD cross-calibration, pressure transmitter calibration monitoring, and equipment condition monitoring and high-frequency (1 Hz to 1 kHz) techniques such as the noise analysis technique. The author has demonstrated the noise analysis technique's effectiveness for measuring the dynamic response-time of pressure transmitters and their sensing lines; for performing predictive maintenance on reactor internals; for detecting core flow anomalies; and for extending neutron detector life. Integrated online condition monitoring systems can combine AC and DC data acquisition and signal processing techniques, can use data from existing process sensors during all modes of plant operation, including startup, normal operation, and shutdown; can be retrofitted into existing PWRs, BWRs, and other reactor types and will be integrated into next-generation plants.

A Novel Power Supply of Online Monitoring Systems for Power Transmission Lines

This paper presents a novel power supply of online condition monitoring systems for power transmission lines. The proposed power supply obtains energy from the magnetic field induced by transmission-line currents using a specially designed Rogowski coil. The design details of the power supply circuit including the unit for overvoltage and impulse line current protection are given in this paper. The impact to the power supply's reliability caused by abnormal high transmission-line currents is also addressed. Experimental tests, including the short-circuit and impulse-current tests, have been conducted on the power supply to verify its performance under different operating conditions. Test results show that the power supply is capable of providing stable outputs with no saturation and low heat generation for the entire range of line conducting currents and can sustain severe conditions such as abnormal impulse currents.

An Approach for Condition-Based Maintenance Optimization Applied to Wind Turbine Blades

A sound maintenance planning is of crucial importance for wind power systems. There is a large potential in cost savings by optimizing maintenance decisions to make the projects more cost-efficient. This paper presents an approach to optimize condition-based maintenance strategies for components which degradation can be classified according to the severity of the damage. The approach is applicable to inspection-based maintenance as well as maintenance based on online condition-monitoring systems. This enables us to compare the maintenance strategies within a unified framework. First decisions associated with the different maintenance strategies are optimized, then the expected life cycle maintenance costs of the strategies are compared. The model is applied to blades of wind turbine in order to compare three maintenance strategies: visual inspection, inspection with a condition-monitoring technique (e.g., ultrasonic or thermography), and online condition monitoring (e.g., fiber optic). It is shown that online condition monitoring is the optimal maintenance strategy for high failure rate and low crack time to failure. In other cases, inspection with condition-based maintenance is slightly more beneficial.

Online Degradation Assessment and Adaptive Fault Detection Using Modified Hidden Markov Model

Online condition monitoring and diagnosis systems play an important role in the modern manufacturing industry. This paper presents a novel method to diagnose the degradation processes of multiple failure modes using a modified hidden Markov model (MHMM) with variable state space. The proposed MHMM is combined with statistical process control to quickly detect the occurrence of an unknown fault. This method allows the state space of a hidden Markov model to be adjusted and updated with the identification of new states. Hence, the online degradation assessment and adaptive fault diagnosis can be simultaneously obtained. Experimental results in a turning process illustrate that the tool wear state can be successfully detected, and previously unknown tool wear processes can be identified at the early stages using the MHMM.

Automated Operating Mode Classification for Online Monitoring Systems

Transient operation of machinery can greatly complicate the task of vibration-based online condition monitoring. Because the operating mode of a machine affects the physical response and hence the diagnostic parameters, real-time information regarding the operating mode is likely to improve the performance of an online fault detection system. This paper proposes a method for automated operating mode classification to augment the performance of vibration-based online condition monitoring systems for applications such as gearboxes, motors, and their constituent components. Experimental work has been carried out on the swing machinery of an electromechanical excavator, which demonstrates how such a method might function on actual dynamic signals gathered from an operating machine. Several variations of the system have been tested and found to be successful.

Fault detection using transient machine signals

This paper describes the development and testing of a strategy for vibration-based online detection of faults in a particular class of machinery. This machinery is defined by two basic characteristics that preclude it from the application of standard online condition monitoring systems. The first characteristic is the absence of historical fault data. The second characteristic is that the machine is in a constant state of transient operation. An example of such a machine is the swing machinery of an electromechanical excavator. The monitoring strategy presented here employs an anomaly detection scheme together with various methods of signal processing and feature extraction. Experiments are carried out using a laboratory apparatus to show the how various configurations of the system are able to detect different types of faults. The results indicate that this approach is effective and merits further investigation.

The design of enhanced online condition monitoring systems for turning processes

The increasing requirements of high quality and low cost of products have created an urgent need to implement new technologies in current automated manufacturing environments. Condition monitoring systems of manufacturing processes have been recognised in recent years as one of the essential technologies that provide the competitive advantage in many manufacturing environments. This paper aims to developing an effective sensor fusion model for turning processes for the detection of tool wear. Multi-sensors combined with a novelty detection algorithm are used to detect tool wear and provide diagnostic and prognostic information. A novel approach using dynamic threshold is utilised to improve the accuracy of the novelty detection system. The results found indicate that the suggested approach provides a responsive and effective solution in monitoring tool wear in turning.