Inspection Maintenance Research Articles

Strategic asset management in LNG Plants: A holistic approach to long-term planning, rejuvenation, and sustainability

The liquefied natural gas (LNG) industry plays a vital role in the global energy landscape, balancing growing energy demands with the imperative of reducing environmental impacts. However, the sector faces significant challenges, including aging infrastructure, the need for operational efficiency, and adherence to sustainability goals. This paper explores the role of strategic asset management (SAM) as a holistic solution for optimizing LNG plant operations. It introduces a comprehensive framework that integrates organizational, financial, and technical strategies, emphasizing long-term planning, rejuvenation, and sustainability. Key strategies, such as predictive maintenance, digital twins, and risk-based inspection methods, are discussed for their effectiveness in extending asset lifecycles and minimizing downtime. Furthermore, the paper underscores the importance of environmental, social, and governance (ESG) factors, renewable energy integration, and resilience planning in addressing climate risks and market volatility. Concluding with actionable recommendations, this paper provides a roadmap for stakeholders to adopt SAM practices, ensuring sustainable, resilient, and competitive LNG operations. Keywords: Liquefied Natural Gas, Strategic Asset Management, Sustainability, Predictive Maintenance, Digital Twin Technology, Resilience Planning.

Human-centric lighting asset management for LED bulbs: a context-driven approach on prognostics and maintenance strategy development in public libraries

ABSTRACT Traditional asset management of lighting systems typically focuses on functionality, cost, and lifespan. In contrast, a human-centric approach prioritizes social sustainability and user well-being by ensuring lighting assets “provide the right light at the right time” for diverse activities. Light-emitting diode (LED) bulbs, known for energy efficiency and longevity, have become a preferred choice, yet public libraries often struggle to manage these assets sustainably, remaining in a reactive “fix/replace when it breaks” stage. Current predictive methods, such as artificial intelligence and machine learning, rely on laboratory data that often overlook real-world contexts, leading to performance gaps. This paper presents a context-driven, human-centric methodology for LED prognosis and maintenance strategies in public libraries, employing limited degradation data from LED testing. Advanced analytical techniques, including Markov Chain Monte Carlo (MCMC) and Deviance Information Criterion (DIC), support a shift from function-based to performance-based reliability assessment. By incorporating Mean Time of Exposure (MTOE) and Critical Integrated Levels (CILs), the approach defines optimal maintenance inspection intervals. This research enhances sustainable LED lighting management in public libraries, offering a framework adaptable to broader applications and aligned with human-centric goals..

Multi-Object Detection for Daily Road Maintenance Inspection With UAV Based on Improved YOLOv8

Multi-Object Detection for Daily Road Maintenance Inspection With UAV Based on Improved <i>YOLOv8</i>

Probabilistic risk assessment of civil aircraft associated failures under condition-based maintenance

Maintenance can improve an aircraft system's reliability over a long operation period or when a component has failed. However, inappropriate maintenance inspection intervals will cause latent failures to be covered or undetected, leading to a large number of unplanned flight disruptions for airlines. In this paper, we present a two-stage framework to assess the associated failure risk of civil aircraft under condition-based maintenance. In the first stage of the framework, the probability of primary functional failure across the lifecycles of the monitored component is determined by analyzing whether the current inspection interval prevents the component from progressing from latent failure to functional failure. In the second stage of the framework, the associated failure probability between components and related systems is formulated by the adjacency matrix. The structure and performance of the proposed model were tested on a case study by run-to-failure data associated with aircraft engines from a large airline. Focusing on the scenario of turbine disk cracking leading to fragment penetration of the fuel tank and causing fire as the consequential fault impact path, the results show that the risk of aircraft fire caused by turbine disk fragments falls within an acceptable range, necessitating the completion of inspections and subsequent monitoring within the stipulated timeframe. The method can be used to readjust the inspection interval, optimize the operation plan, improve the on-time performance of flights, and reduce the risk of aviation accidents.

Embracing drones and the Internet of drones systems in manufacturing – An exploration of obstacles

The manufacturing sector attributes the growing prominence of Drones and the Internet of Drones (IoD) systems to their multifaceted utility in delivery, process monitoring, infrastructure inspection, inventory management, predictive maintenance, and safety inspections. Despite their potential benefits, adopting these technologies faces significant obstacles that need systematic identification and resolution. The current literature inadequately addresses the barriers impeding the adoption of Drones and IoD systems in manufacturing, indicating a research gap. This study bridges this gap by providing comprehensive insights and facilitating the organisational transition towards embracing Drone and IoD technologies. This research identifies 20 critical barriers to deploying Drones and IoD in manufacturing. These barriers are validated through a global quantitative survey of 120 Drone experts and analyzed via Exploratory Factor Analysis (EFA). EFA categorises these challenges into six distinct dimensions. Utilizing the Analytical Hierarchy Process (AHP), these dimensions and individual barriers are ranked, incorporating feedback from five Drone specialists. The study highlights ‘Safety and Human Resource Barriers’ and ‘Payload Capacity and Battery Barriers’ as the most predominant obstacles. Key concerns include limited battery life, explosion risks, and potential damage to assets and individuals. This research significantly advances the existing literature by presenting a practical methodology for categorising and prioritising Drone and IoD adoption barriers. Employing EFA and AHP offers a globally relevant framework for stakeholders to strategically address these challenges, advancing the integration of drones and IoD systems in the manufacturing domain.

Analisis Prosedur Penggantian Arrester pada Jaringan Distribusi PT. PLN (PERSERO) UP3 Cikupa

The electricity distribution network is the backbone of energy supply from power plants to consumers. One important component of the distribution network is the arrester, which serves to protect equipment from sudden voltage surges, such as those caused by lightning strikes. Arresters are designed to withstand overvoltage and protect transformers and other sensitive components from damage. This research uses a case study method with a descriptive qualitative approach. Data were collected through interviews with technicians involved in network maintenance and field inspections of the condition of arresters in the distribution network of PT. PLN (Persero) UP3 Cikupa. The research findings indicate that several arresters in the distribution network of PT. PLN (Persero) UP3 Cikupa receive voltages that do not meet technical specifications, leading to voltage leakage. Based on the research results, it can be concluded that regular replacement of arresters is necessary to maintain the reliability of the electricity distribution network. Voltage leakage caused by discrepancies between incoming voltage and the technical specifications of the arresters has the potential to cause damage to network equipment, particularly transformers.

Advanced Image Stitching Method for Dual-Sensor Inspection.

Efficient image stitching plays a vital role in the Non-Destructive Evaluation (NDE) of infrastructures. An essential challenge in the NDE of infrastructures is precisely visualizing defects within large structures. The existing literature predominantly relies on high-resolution close-distance images to detect surface or subsurface defects. While the automatic detection of all defect types represents a significant advancement, understanding the location and continuity of defects is imperative. It is worth noting that some defects may be too small to capture from a considerable distance. Consequently, multiple image sequences are captured and processed using image stitching techniques. Additionally, visible and infrared data fusion strategies prove essential for acquiring comprehensive information to detect defects across vast structures. Hence, there is a need for an effective image stitching method appropriate for infrared and visible images of structures and industrial assets, facilitating enhanced visualization and automated inspection for structural maintenance. This paper proposes an advanced image stitching method appropriate for dual-sensor inspections. The proposed image stitching technique employs self-supervised feature detection to enhance the quality and quantity of feature detection. Subsequently, a graph neural network is employed for robust feature matching. Ultimately, the proposed method results in image stitching that effectively eliminates perspective distortion in both infrared and visible images, a prerequisite for subsequent multi-modal fusion strategies. Our results substantially enhance the visualization capabilities for infrastructure inspection. Comparative analysis with popular state-of-the-art methods confirms the effectiveness of the proposed approach.

High-detail and low-cost underwater inspection of large-scale hydropower dams

Abstract. The article presents a practical method that combines low-cost camera systems with remotely operated vehicles (ROVs) to accomplish a comprehensive but economically feasible underwater survey of large hydropower infrastructures. Typically, inspecting reservoirs entails draining them off to allow for visual inspections, which are time-intensive, pose risks to operators' safety and are associated with generation losses. In this regard, ROVs are a much safer and more efficient alternative to traditional methods. The study was conducted at the Pack reservoir in Austria, where a reference framework was set up using terrestrial laser scanning and checkerboard markings for the above-water components. A ROV equipped with a GoPro camera and lighting system for the underwater recordings has been employed. Via a close-range photogrammetric approach, it was possible to generate 3D point clouds of the submerged infrastructure with a survey-grade accuracy level. Various strategies were explored to perform bundle block adjustment (BBA), among these were strategies where ground control points (GCPs) were used, strategies without the use of GCPs but pre-calibrated initial camera parameters and strategies with a combination of using both GCPs and pre-calibrated camera parameters in the BBA. The deployment of an inspection technique using low-cost sensors that can generate highly detailed three-dimensional models of submerged infrastructure areas is presented and discussed, allowing easy detection and localization for maintenance inspection, all while being cost-effective. The paper strengthens the suggestion of best practices that optimize camera settings, considering the effect of electronic image stabilization, suggesting its avoidance, and using advanced calibration methods.

Minimizing Flight Hours Losses by Utilizing Deep-Q Learning Techniques in Scheduling Aircraft Maintenance

Currently, there are insufficiently good or efficient tools or procedures for long-term scheduling of aviation maintenance activities in the world, and Vietnam specifically. Airlines are impacted by a variety of elements, including the quantity of aircraft they operate, their capacity, their personnel resources, their maintenance resources, and unforeseen and urgent occurrences that cause schedule disruptions. There are no fast, automatic solutions to the above-described problems existing in aviation today. The reinforcement learning method as described here could potentially be one answer to these problems. The idea is to plan in years running across a specified period such that the aircraft is brought as close as possible to the inspection deadline. From there, the airworthiness of the aircraft increases while the maintenance inspection decreases, reducing the cost of maintenance. Application optimization of the scheduling plan is done using the Deep Q-learning method. The results achieved by the Q-learning and Deep Q-learning algorithms are better in terms of computation times as compared to the other current techniques. The research results of the checks showed reinforcement learning potential in dealing with this problem, where the fly hours loss of planned inspections was reduced by using data from Vietnam Airlines. Computational experiments show that our methods adapt for different purposes and settings of reality. After teaching the model with these simulated conditions, they show how well a reinforcement learning application quickly arrives at lean repair plans.

Development of Risk Informed Aging Management Program in Decommissioning

Abstract This paper proposes to solve the issues related to aging management program (AMP) during the decommissioning phase by using risk information. The AMP for the decommissioning plants has not been established while the number of permanently shut down and decommissioned nuclear power plants are increasing in the world. The first issue is the change of the functional importance of the safety systems at the beginning of decommissioning. We propose to evaluate the importance using the probabilistic risk assessment (PRA) model with the change of specification in the decommissioning plant and to determine new importance grade. The second issue is to make the maintenance based on the appropriate risk information. This risk information should be based on the progress of decommissioning, reflected and evaluated in the PRA model, as well as information obtained from maintenance inspections, including the ROP process, and failure experience during routine operation. To address these issues, the maintenance program in decommissioning that uses the evaluation of risk information according to the decommissioning phase is proposed in this paper.

Augmented dataset for multidimensional ballast segmentation and evaluation

Ballast plays an important role as a railway track supporting structure against repeated loading. Ballast degradation results in poor drainage, lateral instability, excessive settlement, service interruptions, and safety concerns. Therefore, utilizing efficient methods for evaluating ballast fouling is important to ensure safe railroad operations. Emerging computer vision approaches have been effectively incorporated into the ballast inspection processes to enhance their accuracy and robustness. Owing to the inherent data-driven nature of deep learning approaches, the efficacy of the developed model heavily depends on the quality of the training dataset, revealing an urgent need for a comprehensive and high-quality annotated ballast aggregate dataset. In this study, a multidimensional ballast aggregate dataset has been established which contains realistic data collected from field and laboratory setups, augmented by data produced by a synthetic ballast particle generator. In the dataset, the 2D data comprised ballast images with 2D annotations (masks) for localizing the particles in the images. The 3D data comprises height maps, point clouds, and 3D annotations for particle localization. The collected data included various environmental lighting levels and fouling conditions to ensure adequate coverage and diversity in the training dataset. A 2D ballast particle segmentation model previously developed by the authors was trained using an augmented dataset and demonstrated reliable and accurate results in field ballast inspections. The proposed dataset will be utilized in future studies, including 3D ballast particle segmentation and shape completion, to aid in the inspection and investigative methods for effective ballast maintenance.

Optimum maintenance units in multi‐indenture systems

AbstractThis paper deals with an optimization problem of determining the maintenance units and age‐based inspection schemes to minimize the expected maintenance cost rate and satisfy a target system availability for multi‐indenture systems that have a modular structure with redundant units. Failures of primary components directly lead to system failures and can therefore be identified and maintained immediately, while failures of redundant components need to be identified by inspections. And the designated maintenance units need to be replaced during maintenance. By constructing the dual structure of the fault tree, we present a failure path enumeration method, based on which the system availability and expected maintenance cost rate models considering maintenance time are derived. For redundant units, two types of inspection policies are considered in our models: one is the corrective maintenance (CM)‐based inspection policy, and the other one is the age‐based inspection (PM) policy, inspections are performed at failures or age h, depending on which occurred earlier. A heuristic algorithm is proposed to find the reasonable maintenance units and inspection intervals simultaneously, and its performance is also compared and analyzed. Several numerical examples are studied to investigate the effect of parameters on optimal solutions.

A Study on the Design and Implementation Technologies of EVA at the China Space Station

Extravehicular activity (EVA) is a key point and a difficult point for manned spaceflight tasks, as well as an inevitable trend in the development of the manned spaceflight industry. Equipment maintenance, load installation, and extravehicular routing inspection via EVA on the track are necessary to guarantee the safety and reliability of the long-term in-orbit operation of the China Space Station. In this paper, a comprehensive analysis was conducted on the features of multiple tasks, diverse working modes, and strong systematic coupling during the EVA of the China Space Station (CSS). On this basis, the design, implementation technologies’ development, and in-orbit performance evaluation during EVA were expounded. In the space station system, an extravehicular reliability verification and evaluation system suitable for the requirement for EVA under the conditions of China’s multi-mission, multi-module combination, and repairable spacecraft was constructed. Finally, the in-orbit EVA implementation of the China Space Station since the launch of the core module to the present was summarized, and the subsequent application of the extravehicular technologies in manned lunar landing projects and optical modules was anticipated.

Assessing and Improving Maintenance and Safety Inspection Policy for Suspension Bridges in Gyeonggi-do Through On-site Investigation

Assessing and Improving Maintenance and Safety Inspection Policy for Suspension Bridges in Gyeonggi-do Through On-site Investigation

3D Anisotropic Magnetoresistance sensor for steel health monitoring

Electromagnetic sensors are widely used for safety and maintenance inspection purposes of ferromagnetic materials used in aerospace, construction and transport industry. In this paper, a portable device based on an Anisotropic Magnetoresistance (AMR) sensor for steel health monitoring is presented. The principle of operation of this device is the detection of magnetic anomalies of ferromagnetic materials, such as steel, due to stresses, corrosion, or other kind of defections. For this purpose, a cost-effective, low-consumption and high-sensitivity device has been developed, consisting of a sensor, a magnetic yoke and a microcontroller. The AMR sensor, is placed above the material under test, giving the ability of contactless measurement of the magnetic flux density on 3 axes. Additionally, a yoke, consisting of two iron poles and a permanent magnet between them, is used in this implementation, in order to create a closed magnetic circuit with the steel sample, enhancing the measurements. Then, a microcontroller with the appropriate electronic components receives the acquired data, enabling their real-time analysis and visualization of the measurements through a personal computer, smartphone, etc. Moreover, a 3D-printed enclosure has been designed and constructed, which both protects the sensor and maintains the desired position and alignment with the samples. The device can be portable due to an integrated Li-Po battery and its Wi-Fi capabilities. It can be used to determine the location of the defective spots or zones of the material under inspection, facilitating its treatment through rehabilitation methods that can be applied on the affected area. Both single-point and scanning mode monitoring are possible, making it suitable for many testing scenarios. Furthermore, its ease of use and portability render it accessible to a wide range of users, and appropriate for on-site measurements. The initial measurements using the device on test samples, demonstrate its reliability and accuracy in steel health monitoring.

Methodology for Calculating the Damaged Surface and Its Relationship with Power Loss in Photovoltaic Modules by Electroluminescence Inspection for Corrective Maintenance.

Photovoltaic panels are exposed to various external factors that can cause damage, with the formation of cracks in the photovoltaic cells being one of the most recurrent issues affecting their production capacity. Electroluminescence (EL) tests are employed to detect these cracks. In this study, a methodology developed according to the IEC TS 60904-13 standard is presented, allowing for the calculation of the percentage of type C cracks in a PV panel and subsequently estimating the associated power loss. To validate the methodology, it was applied to a polycrystalline silicon module subjected to incremental damage through multiple impacts on its rear surface. After each impact, electroluminescence images and I-V curves were obtained and used to verify power loss estimates. More accurate estimates were achieved by assessing cracks at the PV cell level rather than by substring or considering the entire module. In this context, cell-level analysis becomes indispensable, as the most damaged cell significantly influences the performance of the photovoltaic model. Subsequently, the developed methodology was applied to evaluate the conditions of four photovoltaic panels that had been in operation, exemplifying its application in maintenance tasks. The results assisted in decision making regarding whether to replace or continue using the panels.

Investigation of fatigue crack closure effect on the evaluation of edge cracks with the fundamental mode of edge waves

Fatigue cracks often initiate and propagate from edges of structural components. Detection and evaluation of edge cracks could be very challenging, specifically, due to the crack closure phenomenon, which makes fatigue cracks to be partially closed when the applied loading is removed; this usually corresponds to maintenance and NDE inspection conditions. Despite that fatigue crack closure is well investigated, past experimental and theoretical studies related to guided wave-based NDEs largely ignored this phenomenon. In this article, the fundamental symmetric mode of edge waves (ES0) is used to evaluate crack closure effects on the evaluation of fatigue cracks. The experimental studies have demonstrated that the reflected and transmitted signals at different frequencies correlate very well with the length of the open region of fatigue cracks. However, an accurate evaluation of the total crack length can only be conducted under an applied loading, which fully separates the crack faces. Finally, a new FE model has been proposed to simulate the fatigue crack closure and its effects on propagation of ultrasonic bulk and guided waves. The outcomes of FE modelling and experimental study were found in a good agreement.

BoltVision: A Comparative Analysis of CNN, CCT, and ViT in Achieving High Accuracy for Missing Bolt Classification in Train Components

Maintenance and safety inspection of trains is a critical element of providing a safe and reliable train service. Checking for the presence of bolts is an essential part of train inspection, which is currently, typically carried out during visual inspections. There is an opportunity to automate bolt inspection using machine vision with edge devices. One particular challenge is the implementation of such inspection mechanisms on edge devices, which necessitates using lighter models to ensure efficiency. Traditional methods have often fallen short of the required object detection performance, thus demonstrating the need for a more advanced approach. To address this challenge, researchers have been exploring the use of deep learning algorithms and computer vision techniques to improve the accuracy and reliability of bolt detection on edge devices. High precision in identifying absent bolts in train components is essential to avoid potential mishaps and system malfunctions. This paper presents “BoltVision”, a comparative analysis of three cutting-edge machine learning models: convolutional neural networks (CNNs), vision transformers (ViTs), and compact convolutional transformers (CCTs). This study illustrates the superior assessment capabilities of these models and discusses their effectiveness in addressing the prevalent issue of edge devices. Results show that BoltVision, utilising a pre-trained ViT base, achieves a remarkable 93% accuracy in classifying missing bolts. These results underscore the potential of BoltVision in tackling specific safety inspection challenges for trains and highlight its effectiveness when deployed on edge devices characterised by constrained computational resources. This attests to the pivotal role of transformer-based architectures in revolutionising predictive maintenance and safety assurance within the rail transportation industry.

Multi-Criteria based Components Ranking for Predictive Maintenance: Application to Electro-Hydrogen Generator

In order to improve the reliability and efficiency of a manufacturing system, it is essential to prioritize its components based on their importance and determine the most critical components for predictive maintenance considering multiple criteria. However, this can be a challenging task as different criteria may have varying degrees of importance and correlation with each other. To address this issue, a methodology has been proposed in this paper, which utilizes a multi-criteria decision process. The method involves the identification of specific evaluation criteria, such as maintenance frequency, duration, training requirements, inspection difficulty, component lifespan, and cost, in collaboration with maintenance experts. The developed methodology also allows considering the importance of criteria and the correlation between them. The proposed methodology is then applied to a real Electro-Hydrogen Generator (GeH2) to identify the most important components for predictive maintenance purposes. The results obtained are in line with expert judgments and practical considerations.

Web platform for building roof maintenance inspection using UAS and artificial intelligence

PurposeThis work aims to develop a web platform for inspecting roof structures for technical assistance supported by drones and artificial intelligence. The tools used were HTML, CSS and JavaScript languages; Firebase software for infrastructure; and Custom Vision for image processing.Design/methodology/approachThis study adopted the design science research approach, and the main stages for the development of the web platform include (1) creation and validation of the roof inspection checklist, (2) validation of the use of Custom Vision as an image recognition tool, and (3) development of the web platform.FindingsThe results of automatic recognition showed a percentage of 77.08% accuracy in identifying pathologies in roof images obtained by drones for technical assistance.Originality/valueThis study contributed to developing a drone-integrated roof platform for visual data collection and artificial intelligence for automatic recognition of pathologies, enabling greater efficiency and agility in the collection, processing and analysis of results to guarantee the durability of the building.