Predictive Maintenance Activities Research Articles

Channel Coordination under Predictive and Corrective Maintenance Outsourcing

With respect to Industry 4.0, predictive maintenance plays an important role in production systems. A better predictive maintenance strategy is absolutely essential to an efficient, reliable and safe production process. This study considers a supply chain that consists of a manufacturer and an external contractor wherein the manufacturer is willing to outsource the predictive maintenance activities provided by the contractor. In order to achieve channel coordination, we introduce three incentive contracts: 1) maintenance cost subsidisation contract; 2) uptime target and bonus contract; 3) uptime reward contract, which enhance the cooperation willingness of the contractor and increases profit for both parties, thereby leading to a win-win situation. We also discuss the feasibility of the contracts given improvements in the failure rate and maintenance capability. Theoretical results show that the uptime target and bonus and uptime reward contracts can achieve channel coordination under specific conditions given increases in the failures rate and maintenance capacity. Numerical results are further presented to reveal the managerial insights.

A Novel Method for Deposit Accumulation Assessment in Dry Etching Chamber

Particle contamination in the dry etching chamber is one of the major issues in the semiconductor manufacturing. Particles on the wafer surface may cause wafer defects and yield loss. Therefore, particle monitoring is important to support contamination control and predictive maintenance activities. To achieve this goal, this paper proposes a virtual metrology (VM) method to estimate the deposit accumulation on the chamber wall, which can be a major root cause of the particle contamination. In the proposed method, the piecewise linear approximation (PLA) is employed and modified to derive a general segmentation template for the trace signals. Based on the trace segmentation template, important features that relate to the deposit accumulation are better extracted and modeled. To justify the effectiveness of the proposed method, we validate our method on the etching data from three different maintenance cycles. In the results, the proposed method with trace segmentation has improved performance compared to the VM model that does not segment the data, as well as the VM model that partitions the trace data by step number. The deposit accumulation indicator given by the proposed method demonstrates a slow trend over etching cycles and it matches well with the off-line metrology measurements.

Risk-based optimal scheduling of maintenance activities in a railway network

In a railway system, maintenance activities need to be continuously performed to ensure safety and continued rail operations. In this framework, while on one hand unplanned corrective maintenance activities performed when a fault is occurred are expensive and would cause low service quality, on the other hand preventive maintenance that does not consider the actual asset condition is often unnecessary and turns out to generate avoidable costs. To deal with this issue, in this paper, a risk-based decision support system to schedule the predictive maintenance activities is proposed. In such a framework, the interventions are planned by taking into account the forecast degradation state of railway assets and performed when a given threshold is reached, thus minimizing the probability of both sudden and unnecessary operations. With the end of finding the optimal scheduling of predictive maintenance, in this paper also the space-distributed aspect of railway infrastructure is considered, defining the best path and the activities assignment for each maintenance team. The scheduling model is formulated as a Mixed Integer Linear Programming (MILP) problem aiming at based on the risk minimization, according to the ISO 55000 guidelines. A matheuristic solution approach is proposed and applied to a real rail network. The relevant results show how the proposed scheduling model can use the outputs of predictive tools and degradation models, based on data from field, to mitigate the sudden failure risk by means of a cost-effective maintenance plan at a network level.

A new swarm intelligence optimized multiclass multi-kernel relevant vector machine: An experimental analysis in failure diagnostics of diesel engines

This work attempts to introduce a new intelligent method for condition monitoring of diesel engines. Diesel engine is one of the most important power providers for various industrial applications, including automobiles, ships, agricultures, construction, and electrical machinery. Due to harsh working environment, diesel engines are vulnerable to failures. This article addresses a significant need to improve predictive maintenance activities in diesel engines. A new failure diagnostics approach was proposed based on the manifold learning and swarm intelligence optimized multiclass multi-kernel relevant vector machine. Three manifold learning algorithms were first respectively used to fuse the features that extracted from the original vibration data of the diesel engines into a new nonlinear space. The fused features contain the most distinct health information of the engine by discarding redundant features. Then, the swarm intelligence optimized multiclass multi-kernel relevant vector machine was proposed to identify the failures using the fused features. The contribution of this research is that the dragonfly algorithm is employed to optimize the weights of the multi-kernel functions in the multiclass relevant vector machine. It was also applied to establishing a weighted-sum model by combining the outputs of swarm intelligence optimized multiclass multi-kernel relevant vector machine models with different manifold learning algorithms. Robust failure detection of diesel engines is achieved owing to combined strengths of multiple kernel functions and weighted-sum strategy. The effectiveness of the proposed method is demonstrated by experimental vibration data collected from a commercial diesel engine. The failure detection capability of the proposed manifold learning and swarm intelligence optimized multiclass multi-kernel relevant vector machine method for diesel engines will potentially benefit the machine condition monitoring industry by improving budgeting/forecasting and/or enabling just-in-time maintenance.

Operational Maintenance Philosophy for Tritium Processing Systems

The high contamination potential of the release of radioactive tritium facilitates the demand for and development of a stringent and comprehensive approach to operational maintenance of tritium systems. Prompt and efficient maintenance is necessary to ensure the accepted operational safety basis is adhered to and a continued safe state of operation is achieved. This will help to mitigate and avoid potential hazards that result from a tritium release to the public and facility personnel. Because of the hazards associated with a release of tritium contamination the process systems are in large kept within a series of inerted glovebox environments that must be maintained to keep structural integrity. The nature of a tritium release from a glovebox could have significant consequences for the general public as well as for personnel. As such, the maintenance philosophy is developed to help facilitate operations in the adherence to the facility’s safety code of conduct.To effectively facilitate the safe operation goals mentioned a well-defined maintenance philosophy has been developed that encompasses routine and non-routine maintenance activities. Examples of routine activities include preventative maintenance such as line-break inspections, helium leak tests to ensure components are leak tight, weld inspections and overall surveillance testing of essential components and infrastructure. Predictive maintenance also falls into this category. Predictive maintenance activities are developed over time in response to non-routine maintenance work. Non-routine maintenance or corrective maintenance activities are performed in response to a specific failure or to resolve a particular inadequacy in performance of tritium systems. When corrective maintenance is performed trends are often studied and more predictive maintenance can be scheduled to compensate for more routine failures.This technical note will identify key operational maintenance considerations which when applied, will ensure that tritium handling systems are operated safely.

Insights towards Condition Monitoring of Fixed Railway Crossings

Evaluation methods and field trial results in terms of geometry and loading of fixed railway crossings are proposed in this paper. Based on these experiences the fundamentals towards condition monitoring are created to allow predictive maintenance actions. For this, a fixed crossing has been instrumented with strain gauges and installed in track with mixed traffic. The strain gauge signals were recorded in irregular intervals to establish a database for the development of model based condition monitoring. A signal processing chain was developed that shall serve as base for continuous monitoring of crossing nose via strain measurements. To this end a data processing work flow is proposed that enables to monitor changes in the combined system of crossing geometry/ wear and bedding state.To quantify the geometry development of the crossing nose (due wear and plastic deformation) a laser based non-contacting measurement of the crossing geometry has been set up. 2D profiles recorded at predefined positions of the crossing enabled a 3D reconstruction of the crossing geometry. During 30 months in service the geometry of the crossing nose and adjacent regions has been recorded in 3 to 6 months intervals. The superposition of the corresponding 2D profiles and the 3D reconstruction allowed a quantitative measurement of the geometry changes during service.

Using a fuzzy logic approach for the predictive maintenance of textile machines

The state of the textile machines can be evaluated by monitoring different parameters, such as vibration and temperature, so their maintenance can be performed a short time before the failure. However, the evolution of each monitored parameter can conduct to different decisions regarding the maintenance actions. Within this framework, fuzzy logic was proposed for planning the predictive maintenance activities of the textile machines. A fuzzy decision making system was developed and its effectiveness was illustrated for the planning of the predictive maintenance of a sewing machine needle. The results of the case study show fuzzy logic as a proactive approach to perform when needed maintenance activities of textile machines.

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.

Parametric Analysis of a Two-Shaft Aeroderivate Gas Turbine of 11.86 MW

The aeroderivate gas turbines are widely used for power generation in the oil and gas industry. In offshore marine platforms, the aeroderivative gas turbines provide the energy required to drive mechanically compressors, pumps and electric generators. Therefore, the study of the performance of aeroderivate gas turbines based on a parametric analysis is relevant to carry out a diagnostic of the engine, which can lead to operational as well as predictive and/or corrective maintenance actions. This work presents a methodology based on the exergetic analysis to estimate the irrevesibilities and exergetic efficiencies of the main components of a two-shaft aeroderivate gas turbine. The studied engine is the Solar Turbine Mars 100, which is rated to provide 11.86 MW. In this engine, the air is compressed in an axial compressor achieving a pressure ratio of 17.7 relative to ambient conditions and a high pressure turbine inlet temperature of 1220 °C. Even if the thermal efficiency associated to the pressure ratio of 17.7 is 1% lower than the maximum thermal efficiency, the irreversibilities related to this pressure ratio decrease approximately 1 GW with respect to irreversibilities of the optimal pressure ratio for the thermal efficiency. In addition, this paper contributes to develop a mathematical model to estimate the high turbine inlet temperature as well as the pressure ratio of the low and high pressure turbines.

Condition-based maintenance of wind power generation systems considering different turbine types and lead times

Condition monitoring measurements, such as vibration data, acoustic emission data, oil analysis data, power voltage and current data, etc., can be obtained from wind turbine components and be utilised to evaluate and predict the health conditions of the components and the turbines. The objective of condition-based maintenance (CBM) is to optimise the predictive maintenance activities based on the condition monitoring and prediction information to minimise the overall costs of wind power generation systems. In existing work, all the wind turbines are assumed to be of the same type and the lead times of different components are assumed to be constant. This is not the case in many practical applications. In this paper, we develop a CBM approach for wind turbine systems considering different types of wind turbines in a wind farm and different lead times for different turbine components, which lead to more accurate modelling of CBM activities in actual wind farms. In the proposed CBM approach, we present a new CBM policy involving two design variables for each turbine type, a method for turbine failure probability evaluation considering different lead times and a CBM cost evaluation method. Numerical examples are provided to demonstrate the proposed CBM approach.

A Technical Framework and Roadmap of Embedded Diagnostics and Prognostics for Complex Mechanical Systems in Prognostics and Health Management Systems

Prognostics and Health Management (PHM) technologies have emerged as a key enabler to provide early indications of system faults and perform predictive maintenance actions. Implementation of a PHM system depends on accurately acquiring in real time the present and estimated future health states of a system. For electronic systems, built-in-test (BIT) makes it not difficult to achieve these goals. However, reliable prognostics capability is still a bottle-neck problem for mechanical systems due to a lack of proper on-line sensors. Recent advancements in sensors and micro- electronics technologies have brought about a novel way out for complex mechanical systems, which is called embedded diagnostics and prognostics (ED/EP). ED/EP can provide real-time present condition information and future health states by integrating micro-sensors into mechanical structures when designing and manufacturing, so ED/EP has a revolutionary progress compared to traditional mechanical fault diagnostic and prognostic ways. But how to study ED/EP for complex mechanical systems has not been focused so far. This paper explores the challenges and needs of efforts to implement ED/EP technologies. In particular, this paper presents a technical framework and roadmap of ED/EP for complex mechanical systems. The framework is based on the methodology of system integration and parallel design, which includes six key elements (embedded sensors, embedded sensing design, embedded sensors placement, embedded signals transmission, ED/EP algorithms, and embedded self-power). Relationships among these key elements are outlined, and they should be considered simultaneously when designing a complex mechanical system. Technical challenges of each key element are emphasized, and the corresponding existed or potential solutions are summarized in detail. Then a suggested roadmap of ED/EP for complex mechanical systems is brought forward according to potential advancements in related areas, which can be divided into three different stages: embedded diagnostics, embedded prognostics, and system integration. In the end, the presented framework is exemplified with a gearbox.

Opportune maintenance and predictive maintenance decision support

Conventional maintenance strategies on a single component are being phased out in favour of more predictive maintenance actions. These new kinds of actions are performed in order to control the global performances of the whole industrial system. They are anticipative in nature, which allows a maintenance expert to consider non-already-planned maintenance actions. Two questions naturally emerge: when to perform a predictive maintenance action; how a maintenance expert can take advantage of a given predictive maintenance action that will have to be performed. These questions concern maintenance decision making and lead to the notion of opportune maintenance actions. This new concept extends the investigation of predictive maintenance actions from one single component to several ones.

Opportunistic maintenance based on fuzzy modelling of component proximity

Conventional maintenance strategies on a single component are being phased out in favor of more predictive maintenance actions. These new forms of maintenance are emerging today for better controlling the global performances of the whole system all along its life cycle. They are anticipative in nature allowing developing non-already-planned maintenance actions or to move earlier or later a planned maintenance action. It leads to investigate opportunistic interventions by considering, for example, the direct environment of a given component to be maintained in order to run additional maintenance actions on other components considered enough closed (proximity principle). Thus, this paper presents a new fuzzy methodology to assess component proximity in the design phase to impact design out maintenance. The methodology is applied on an industrial case study.

Determining the maximum periodic inspection interval for medium voltage motors using a Markov model

Maintenance issues have been receiving ever greater attention in the industrial sector, because equipment ageing and their performance are of major concern both for capital loss and safety. Medium voltage motors are common drivers for most applications in the process sector. Predictive maintenance activities are of paramount importance for assuring effective machine performance in a specified time horizon. The identification of the periodic test interval plays a major role in achieving a safety integrity target. This paper presents a mathematical model, based on discrete step-wise Markov processes, predicting the probability of the machine failure depending on the periodic test interval. The ‘health state’ of the machine is depicted by a bulk index, called integrity level (IL), currently used by several companies. The model has been built and tested on a large maintenance database involving 111 motors from four facilities.

OVERVIEW ON OPPORTUNISTIC MAINTENANCE

Conventional maintenance strategies on a single component are being phased out in favour of more predictive maintenance actions. These new forms of maintenance are emerging today to help controlling the global performances of the whole system. They are anticipative in nature, which allows considering non-already-planned maintenance actions. Thus they lead to investigate potential opportunistic interventions. Opportunistic maintenance, the way it is understood or practiced today, its tools and criteria are questioned in the present work. An overview and some illustrations of this concept are also provided.

Real‐time degradation monitoring and failure prediction of pulse energy conversion systems

PurposeThe paper aims to present a new approach for safe operation, and maintenance cost reduction, regarding electrical and electromechanical systems of power production, power conversion and power transmission, primarily in industrial units.Design/methodology/approachThe paper adapts a theoretical approach to real‐time monitoring of pulse energy conversion systems (PECSs), and prediction of abnormal dynamics incipient and developing failure. The approach utilizes the preliminary bifurcation analysis results and the geometrical interpretation of the fractal regularities in PECS dynamics, to reveal degradation development.FindingsIt turns out that this new approach enables one to fill the joint requirements of real‐time failure prediction of the high frequency power control devices, and of the relating failure symptoms to cause parameters. Discussions are led on the fundamental outcomes of numerical and experimental investigations of a DC‐DC buck voltage converter with pulse‐width‐modulation (PWM) control.Practical implicationsThe real‐time monitoring of incipient abnormal dynamics in key nonlinear devices of electrical and electromechanical systems constitutes a mean to predict and prevent failures. It provides invaluable information for deciding and planning predictive maintenance actions, from the insurance of optimal operating conditions to abnormal operating prevention, either by means of modification of controlled parameters and control laws or, in the worst case, by change of the power components' structure.Originality/valueAbout “failure prediction”, this paper proposed to pay attention, not to identification of the dynamics evolution specific reason, but real‐time monitoring of this reason consequence – incipient abnormal dynamics in the electrical and electromechanical systems – that can lead to failure. The advantage of this approach consists in the possibility of taking into account PECS operating conditions of models ambiguity of both disturbing parameter changes and PECS behavior.