Predictive Condition Monitoring Research Articles

Quality analysis for condition monitoring of a manufacturing process integrating acoustic processing based on three distinct ML algorithms

Abstract Over the years, it has been proven that all machines that run continuously or even for a short period of time experiences breakdowns, reduced efficiency and damage to the parts. Considering one such manufacturing process is welding process where strip welding is studied for condition monitoring and prediction of online quality of the weld. The proposed embodiment discusses a predictive condition monitoring of a welding process, using acoustic sensors and processing their data. Converting the processed data to a predictive model, comparing different Machine Learning (ML) algorithms like Decision Tree (DT), Artificial Neural Networks (ANN) and Convolutional Neural Networks (CNN) based on the validation and training results, selecting the suitable algorithm for online or real-time weld quality prediction is the objective of the proposed embodiment. According to the research, effectively executing the proposed solution can improve its life, efficiency, and utilization. The training accuracy is observed to be 97 % for DT, 50 % for ANN and 67 % for CNN. DT, on the other hand, has a validation accuracy of 94.7 %, ANN has 47.65 % and CNN has 65.8 %. DT produced the best results in the study, initially DT prediction model is used in offline testing of the new data-set for individual weld class and the system produced 97 % accuracy. Furthermore the DT prediction model is implemented in online weld quality prediction in three classes: Bad weld, OK weld and Good weld where the system produced 90 % accuracy in real-time prediction.

Context-informed conditional anomaly detection approach for wave power plants: The case of air turbines

The reliability and energy production of wave power plants (WPPs) depend on sea-state conditions, operation efficiency and degradation of its constituent assets. Air turbines are key assets for the efficient and reliable operation of WPPs and ensuring their correct operation leads to enhance the efficiency of WPPs. However, the lack of run-to-failure data and scarce fault records hampers the development of predictive condition monitoring solutions. In this context, focusing on unsupervised health monitoring methods, this paper presents an air turbine conditional anomaly detection (CAD) approach with a practical case study tested and validated on the Mutriku wave power plant. In contrast to anomaly detection models, which model the health-state without taking into account the influence of the operating context, the proposed CAD approach learns the expected air turbine operation conditioned on specific sea-states information modelled through wave energy flux concepts. This is achieved through an ensemble of Gaussian Mixture models and the expectation–maximization algorithm. Results show that, the integration of sea-states in the anomaly detection learning process improves the discrimination capability of the CAD model compared with the anomaly detection model without sea-state information, reducing false positive events and improving the accuracy of the CAD model.

A Review paper on Vibration-Based Fault Diagnosis of Rolling Element Bearings

Abstract: One of the common crucial elements that decide the condition of machinery including its residual lifespan in the energy industry, automotive, hydroelectric dams, etc., is the rolling element bearing. Throughout the process, rigorous predictive condition monitoring tools like vibration control, and analysis are valuable methods that are demanded to ensure the safe condition of the rolling element bearings. The expected faults that offer ample cycle times for preventive management are suggested by a Predictive Health Forecasting instrument. Instead of simply identifying the faults, the Predictive bearing health Monitoring platform seeks to map the degradation, like wear progression. From the point of view of system condition control using vibration analysis, this article aims to summarize the fundamental science of rolling element bearings through these varied applications. Numerous vibration analysis approaches are employed to detect defects in REB. The aforementioned article explained different techniques for vibration analysis for REB defect investigation. Keywords: Frequency domain, REB, Fault Investigation, Vibration Analysis, Fault Investigation.

Transient Behavior in Variable Geometry Industrial Gas Turbines: A Comprehensive Overview of Pertinent Modeling Techniques.

Generally, industrial gas turbines (IGT) face transient behavior during start-up, load change, shutdown and variations in ambient conditions. These transient conditions shift engine thermal equilibrium from one steady state to another steady state. In turn, various aero-thermal and mechanical stresses are developed that are adverse for engine’s reliability, availability, and overall health. The transient behavior needs to be accurately predicted since it is highly related to low cycle fatigue and early failures, especially in the hot regions of the gas turbine. In the present paper, several critical aspects related to transient behavior and its modeling are reviewed and studied from the point of view of identifying potential research gaps within the context of fault detection and diagnostics (FDD) under dynamic conditions. Among the considered topics are, (i) general transient regimes and pertinent model formulation techniques, (ii) control mechanism for part-load operation, (iii) developing a database of variable geometry inlet guide vanes (VIGVs) and variable bleed valves (VBVs) schedules along with selection framework, and (iv) data compilation of shaft’s polar moment of inertia for different types of engine’s configurations. This comprehensive literature document, considering all the aspects of transient behavior and its associated modeling techniques will serve as an anchor point for the future researchers, gas turbine operators and design engineers for effective prognostics, FDD and predictive condition monitoring for variable geometry IGT.

A taxonomy approach to failure mode analysis for use in predictive condition monitoring

An extensive knowledge of a system’s failures is crucial for identifying areas where the reliability of the system can benefit from improvements, as well as informing the design of new systems. Moreover, relationships between faults and failures can be used to enhance the maintenance of the system.In this paper we present a taxonomy of failure modes of the Joint European Torus (JET) Remote Handling System (RHS). This system is used during maintenance and enhancements of in-vessel systems, and consists of two transporters (articulated booms) and a two-armed manipulator, along with a number of supporting systems. In this work we first present a failure taxonomy suitable for our specific system, and then we use a clustering approach to introduce example failure modes into the taxonomy. The presented failures have been collected during commissioning and operations over a period of over 5 years. Cataloged failures are extracted from the logs produced by the control system and from the daily log books recorded by the system operator.

Gaussian Process Operational Curves for Wind Turbine Condition Monitoring

Due to the presence of an abundant resource, wind energy is one of the most promising renewable energy resources for power generation globally, and there is constant need to reduce operation and maintenance costs to make the wind industry more profitable. Unexpected failures of turbine components make operation and maintenance (O&M) expensive, and because of transport and availability issues, the O&M cost is much higher in offshore wind farms (typically 30% of the levelized cost). To overcome this, supervisory control and data acquisition (SCADA) based predictive condition monitoring can be applied to remotely identify early failures and limit downtime, boost production and decrease the cost of energy (COE). A Gaussian Process is a nonlinear, nonparametric machine learning approach which is widely used in modelling complex nonlinear systems. In this paper, a Gaussian Process algorithm is proposed to estimate operational curves based on key turbine critical variables which can be used as a reference model in order to identify critical wind turbine failures and improve power performance. Three operational curves, namely, the power curve, rotor speed curve and blade pitch angle curve, are constructed using the Gaussian Process approach for continuous monitoring of the performance of a wind turbine. These developed GP operational curves can be useful for recognizing failures that force the turbines to underperform and result in downtime. Historical 10-min SCADA data are used for the model training and validation.

An unscented Kalman filter-based rolling radius estimation methodology for railway vehicles with traction

Monitoring the conditions of railway vehicle systems plays an important role in the maintenance of safety and performance of railway vehicles. Rolling radius is one of the properties that should be monitored continuously for the predictive maintenance of a railway vehicle since it changes with time due to wheel wear. In this study, a model-based condition monitoring methodology, which is based on an unscented Kalman filter, is proposed. The model includes the torsional dynamics of an independently rotating tram wheel with a traction motor and a contact model. The rolling radius is estimated by considering the traction effort of the motor and the angular velocity measurements. The proposed methodology is tested on a tram wheel test stand (roller rig), which has a wheel on roller configuration. First, a mathematical model is validated by the measurements taken from the test stand. Second, the unscented Kalman filter is applied as a parameter estimator. The results demonstrate that the proposed scheme is a promising option to be used in the predictive condition monitoring of the wheel profile for traction vehicles.

Modelling of microbial fuel degradation in liquid fuels for a gas turbine engine application

Microorganisms are important in fuel deterioration and consequently in engine performance. To assess the impact of fuel deterioration on gas turbines, a bio-mathematical model, Bio-fAEG, was developed to simulate microbial fuel degradation and estimate hydrocarbon loss. Simulation of four conventional diesel fuels and a biodiesel fuel estimated hydrocarbon loss between 0.37 mg/L per day (diesel-type fuel) and 1.48 mg/L per day (biodiesel-type fuel), with nearly complete degradation of the bioavailable fraction of the fuels within 20–60 days. Assuming biodegradation of these fuels is not impeded and conditions for growth remained constant, the Bio-fAEG model predicted that significant degradation begins in four months for a less degradable fuel and two months for a more degradable fuel. Analysis carried out with this model provides insight into cumulative biodegradation of engine fuels and the importance of microbial characteristics, initial microbial population and residence time. The integration of this model in engine performance and emission tools could provide further information on the effect of fuel degradation on engine systems. This is a first step in quantifiable assessment of microbial fuel degradation towards predictive condition monitoring in gas turbine fuels and fuel systems.

Application of embedded condition monitoring systems in pallet industry

The goal of the predictive condition monitoring of planetary power transmissions in pellet mills is to indicate deterioration in the condition of the power transmissions before the actual failure occurs. In many cases, the defect does not cause an immediate interruption of the process. If this is the case, the defective part can be replaced or repaired during normal, scheduled maintenance periods, provided that the defect has been found in a sufficiently early stage. The most common causes of failure of rotary machines are faults in bearings, the stator and the rotor. There are many methods for the predictive condition monitoring of rotary machines. The analysis can be based on different measured quantities. These include, for example, temperature, current, magnetic flux density and vibration. The basic design idea is to create a measurement and data collection system for condition monitoring in which the data analyses and decision-making are based on fuzzy logic programming. In this paper, a low-cost optimal micro configuration for measurement and condition monitoring of data collection system of pellet mills power transmission is presented. The system is based on PIC (Programmable Interface Controller) microcontrollers and represents the complete solution for condition monitoring regarding vibration, temperature and rpm measurement. The microcontroller based system also has an integrated function with control application based on fuzzy logic.

The Development of a Model for the Assessment of Biofouling in Gas Turbine System

A significant problem encountered in the gas turbine industry with fuel products is the degradation of fuel and fuel systems by micro-organisms, which are largely bacteria, embedded in biofilms. These micro-organisms cause system fouling and other degradatory effects, extending often to sudden failure of components with cost implications. Current methods of assessment are only postimpact evaluation and do not necessarily quantify the effects of fuel degradation on engine performance and emission. Therefore, effective models that allow predictive condition monitoring are required for engine's fuel system reliability, especially with readily biodegradable biofuels. The aim of this paper is to introduce the concept of biofouling in gas turbines and the development of a biomathematical model with potentials to predict the extent and assess the effects of microbial growth in fuel systems. The tool takes into account mass balance stoichiometry equations of major biological processes in fuel biofouling. Further development, optimization, and integration with existing Cranfield in-house simulation tools will be carried out to assess the overall engine performance and emission characteristics. This new tool is important for engineering design decision, optimization processes, and analysis of microbial fuel degradation in gas turbine fuels and fuel systems.

Predictive Condition Monitoring and Fault Diagnosis Techniques for Offshore Wind Turbines

Wind energy conversion is the fastest-growing source of new electric generation in the world. The unforeseen damages of a component in a wind turbine have significant impact on the wind turbines, especially offshore ones. There has been a trend towards adoption of various forms of condition monitoring and fault diagnosis to know the deteriorating condition for both mechanical and electrial components, well in advance of a breakdown. This paper review the most recent advances in these areas with a special focus on offshore wind turbines. Different type of faults on offshore wind turbines are reviewed, including gearbox and bearing failures, generators failures, rotor failures and electrical failures. And then two widely-used methods(vibration analysis and oil monitoring) for predictive condition monitoring and fault diagnosis in offshore wind turbines are reviewed.

Condition monitoring of laminated composite sliding bearings using embedded capacitors

A new online monitoring tool for predictive condition monitoring of laminated composite sliding bearings is proposed. This tool consists of embedded capacitors between the layers of a laminated composite bearing. Preliminary experiments were performed on a glass–epoxy plain bearing equipped with two capacitors. The capacitors were made of three copper foils attached to a paper ribbon. Besides the experiments, a simplified two-dimensional finite element model was built up to simulate the variation of the capacitors in relation to the applied load. The results of the experiments and the simulations show that the variation of the capacitors corresponds to the stress distribution in the bearing.

Non‐destructive subsurface damage monitoring in bearings failure mode using fractal dimension analysis

PurposeBearings in field applications with high dynamic loading, e.g. wind energy plants, suffer from sudden failure initiated by subsurface material transformation, known as white etching cracks in a typical scale of μm, preferably around the maximum Hertzian stress zone. Despite many investigations in this field no precise knowledge about the root cause of those failures is available, due to the fact that failure under real service conditions of wind energy plants differs from what is known from test rig results in terms of contact loading, lubrication or dynamics. The purpose of this paper is to apply Barkhausen noise measurement to a full bearing test ring running under conditions of elastohydrodynamic lubrication (EHL) with high radial preload.Design/methodology/approachFull bearing tests are carried out by use of DGBB (Deep Grove Ball Bearings) with 6206 specification, material set constant as 100Cr6, martensitic hardening, 10‐12 percent maximum retained austenite and radial preload of 3500 MPa. Speed is set 9000 rpm, temperature is self setting at 80°C by test conditions. For tests, synthetic hydrocarbon base oil (Poly‐α‐Olefine) with a 1 percent amount of molydenum‐dithiophosphate (organic chain given as 2‐ethylhexyl) was used.FindingsNon‐destructive fractal dimension analyses by use of Barkhausen noise measurements is of versatile value in terms of recording bearing manufacturing processes, but can also be part of non‐destructive condition monitoring of bearings in field applications, where predictive reactive maintenance is crucial for availability of the plant.Research limitations/implicationsBarkhausen noise signal recording may also be valuable for case studies related to microstructure changes of steel under operation conditions. Bearings are exposed in plenty of conditions to phenomena such as straying currents, subsequently straying magnetic fields. Hardly anything is known about how microstructure of bearing steel is susceptible to such conditions. This will be part of further studies.Originality/valueResults given in the paper show that sudden bearing failure, according to formation of subsurface material property changes might be driven by activities of dislocations. Since those activities start with sequences of stress field‐induced formation of domains, later by formation of low‐angle subgrains, and at least phase transformation, recording of the Barkhausen signal would lead to real predictive condition monitoring in applications where a highly dynamic loading of the contact, even with low nominal contact pressure leads to sudden failure induced by white etching.

Probabilistic Planning for Predictive Condition Monitoring and Adaptation Within the Self-Optimizing Energy Management of an Autonomous Railway Vehicle

Self-optimizing mechatronic systems are a new class of technical systems. On the one hand, new challenges regarding dependability arise from their additional complexity and adaptivity. On the other hand, their abilities enable new concepts and methods to improve the dependability of mechatronic systems. This paper introduces a multi-level dependability concept for selfoptimizing mechatronic systems and shows how probabilistic planning can be used to improve the availability and reliability of systems in the operating phase. The general idea to improve the availability of autonomous systems by applying probabilistic planning methods to avoid energy shortages is exemplified on the example of an innovative railway vehicle.

Vibration analysis of rotating machine shop test using Kalman filter

Factory approval test (FAT) or shop run test is a unique opportunity to evaluate rotating machine performance before delivery to site. This paper presents a new signal processing method to evaluate vibration measurement during FAT or shop run test based on recorded noisy data. It is also a necessary step for machine reliability, fault diagnosis, future predictive maintenance and condition monitoring.

P22 Kenyan children with cancer: Controlling Burkitt's lymphoma

Microorganisms are important in fuel deterioration and consequently in engine performance. To assess the impact of fuel deterioration on gas turbines, a bio-mathematical model, Bio-fAEG, was developed to simulate microbial fuel degradation and estimate hydrocarbon loss. Simulation of four conventional diesel fuels and a biodiesel fuel estimated hydrocarbon loss between 0.37 mg/L per day (diesel-type fuel) and 1.48 mg/L per day (biodiesel-type fuel), with nearly complete degradation of the bioavailable fraction of the fuels within 20–60 days. Assuming biodegradation of these fuels is not impeded and conditions for growth remained constant, the Bio-fAEG model predicted that significant degradation begins in four months for a less degradable fuel and two months for a more degradable fuel. Analysis carried out with this model provides insight into cumulative biodegradation of engine fuels and the importance of microbial characteristics, initial microbial population and residence time. The integration of this model in engine performance and emission tools could provide further information on the effect of fuel degradation on engine systems. This is a first step in quantifiable assessment of microbial fuel degradation towards predictive condition monitoring in gas turbine fuels and fuel systems.

Numerical simulation and experimental study of a two-stage reciprocating compressor for condition monitoring

A numerical simulation of a two-stage reciprocating compressor has replicated the operations of the compressor under various conditions for the development of diagnostic features for predictive condition monitoring. The simulation involves the development of a mathematical model of five different physical processes: speed–torque characteristics of an induction motor, cylinder pressure variation, crankshaft rotational motion, flow characteristics through valves and vibration of the valve plates. Modelling both valve leakage and valve spring deterioration has also been achieved. The simulation was implemented in a MATLAB environment for an efficient numerical solution and ease of result presentation. For normal operating conditions, the simulated results are in good agreement with the test results for cylinder pressure waveforms and crankshaft instantaneous angular speed (IAS). It has been found that both the IAS fluctuation and pressure waveform are sensitive detection features for compressor faults such as valve leakage and valve spring deterioration. However, IAS is preferred because of its non-intrusive measurement nature. Further studies using the model and experiments are being undertaken in order to develop fault detection features for compressor driving motors and transmission systems.

Corn Processor Employs ITT Technology for Predictive Condition Monitoring

Corn Processor Employs ITT Technology for Predictive Condition Monitoring