Machinery Downtime Research Articles

Enhancing robotic manipulator fault detection with advanced machine learning techniques

The optimization of rotating machinery processes is crucial for enhanced industrial productivity. Automatic machine health monitoring systems play a vital role in ensuring smooth operations. This study introduces a novel approach for fault diagnosis in robotic manipulators through motor sound analysis to enhance industrial efficiency and prevent machinery downtime. A unique dataset is generated using a custom robotic manipulator to examine the effectiveness of both deep learning and traditional machine learning in identifying motor anomalies. The investigation includes a two-stage analysis, initially leveraging 2D spectrogram features with neural network architectures, followed by an evaluation of 1D MFCC features using various conventional machine learning algorithms. The results reveal that the proposed custom CNN and 1D-CNN models significantly surpass traditional methods, achieving an F1-score exceeding 92%, highlighting the potential of sound analysis for automated fault detection in robotic systems. Additional experiments were carried out to investigate 1D MFCC features with various machine learning algorithms, including KNN, DT, LR, RF, SVM, MLP, and 1D-CNN. Augmented with additional data collected from the locally designed manipulator, our experimental setup significantly enhances model performance. Particularly, the 1D-CNN stands out as the top-performing model on the augmented dataset.

Smart Quantum Tunneling Composite Sensors to Monitor FKM and FFKM Seals.

Operators of industrial machinery relentlessly pursue improving safety, increasing productivity, and minimizing unplanned downtime. Elastomer seals are ubiquitous components of this machinery. In general, static seals are designed to be compressed at a fixed level of compression, taking gland geometry, loading condition, temperature range of operation, fluid media exposure, and other factors into account to ensure the safe operation of equipment. Over time, seals experience compression set, chemical-induced swelling, erosion, and other phenomena which can compromise the compressive force generated by the seal and cause leaking. This is particularly important in critical applications, where high pressure, high temperature, and aggressive media are present, and fluorinated elastomers are common materials for seals. Further, changes in operating conditions at manufacturing plants, either intentional or through regular process variation, create unknown operating conditions for seals. This unknown and variable application environment makes seal performance hard to predict. Therefore, machinery utilizing seals is, at best, serviced preventatively at certain intervals, where seals are removed, and the remaining useful life of the seal is unknown. This leads to unnecessary machinery downtime and increases consumable costs for manufacturers. In the worst case, the seal is run to failure, creating machinery and plant safety concerns. Both scenarios are undesirable for manufacturers using industrial machinery. This paper reports on the development of "smart" intrinsic self-sensing seals, which enable performance monitoring of the compression behavior of seals while in use. In addition, this paper examines quantum tunneling elastomeric composites (QTC) to demonstrate a method of component performance monitoring by modifying the underlying elastomeric material itself. This paper studies QTC sensor-based fluorinated (FKM) and per-fluorinated (FFKM) compositions, which are modified to incorporate varying levels of carbon nanostructure (CNS) material. The resulting seal's resistive properties are shown to be a function of the level of compression, the first time this phenomenon has been demonstrated in high-performing FKM and FFKM seal materials.

Diagnosis of Multiple Faults in Rotating Machinery Using Ensemble Learning.

Fault diagnosis of rotating machines is an important task to prevent machinery downtime, and provide verifiable support for condition-based maintenance (CBM) decision-making. Deep learning-enabled fault diagnosis operations have become increasingly popular because features are extracted and selected automatically. However, it is challenging for these models to give superior results with rotating machine components of different scales, single and multiple faults across different rotating components, diverse operating speeds, and diverse load conditions. To address these challenges, this paper proposes a comprehensive learning approach with optimized signal processing transforms for single as well as multiple faults diagnosis across dissimilar rotating machine components: gearbox, bearing, and shaft. The optimized bicoherence, spectral kurtosis and cyclic spectral coherence feature spaces, and deep blending ensemble learning are explored for multiple faults diagnosis of these components. The performance analysis of the proposed approach has been demonstrated through a single joint training of the entire framework on a compound dataset containing multiple faults derived from three public repositories. A comparison with the state-of-the-art approaches that used these datasets, shows that our method gives improved results with different components and faults with nominal retraining.

IMPROVING THE QUALITY OF MACHINERY AND EQUIPMENT BASED ON A PROPRIETARY TECHNICAL SERVICE

The work is aimed at solving the problems of import dependence and import substitution in the technical sphere on the basis of improving the quality, reliability and efficiency of Russian-made machines and equipment, ensuring their operability during use for their intended purpose. (Research purpose) The research purpose is improving the quality, reliability and efficiency of Russian-made machinery and equipment based on the development and implementation of a proprietary method of technical service of machinery and equipment. (Materials and methods) The main research methods of the work: comparative assessments, analysis and synthesis of the results of the survey of the operational reliability of agricultural machinery and equipment in real operation. In the course of the study, the reliability and efficiency indicators of technological machines and equipment of domestic and imported production for the agro-industrial complex were studied and analyzed, their comparative assessments were carried out, the causes of failures and the duration of downtime were established when eliminating their consequences. (Results and discussion) It was noted that the main results of organizational and technological research are aimed at fully satisfying the needs of machine-technological industries in high-quality and reliable equipment. We have identified the main problem in this direction - increasing the level of reliability of domestic machinery and equipment and, first of all, reliability, which is significantly lower than the best world analogues. (Conclusions) Providing branded engineering and technical support of machines and equipment throughout the entire life cycle of machines is a prerequisite for solving the problem of improving their quality and reliability, ensuring the priority of consumer rights in the fight for the consumer. The expected efficiency from the implementation of the research results at the enterprises of the agro-industrial complex of the Russian Federation can be obtained by reducing downtime of machinery and equipment for technical reasons.

A Hybrid Attention-Based Paralleled Deep Learning model for tool wear prediction

In modern manufacturing process, tool condition significantly affects work efficiency, machinery downtime and operating profit. Convolutional neural network (CNN), recurrent neural network (RNN) or other deep learning models are widely adopted to learn sensitive features individually or sequentially from enormous samples for tool status monitoring. However, these models only learn partial features due to their inherent structures. And features extraction performance of the model with simple sequential combination is also restricted by inner mutual block interference. In this paper, a novel deep learning network named Hybrid Attention-Based Parallel Deep Learning (HABPDL) model is proposed to address these problems. Specifically, ResNet and BiLSTM blocks individually learn features. Their corresponding attention layers, namely convolutional block attention module (CBAM) and general attention unit in BiLSTM, are stacked in sequence to highlight extracted features. And global average pooling (GAP) is applied to reduce superfluous spatial features and increase model interpretability after CBAM layer. Finally, these features maps from CNN and RNN parts are concatenated to predict tool wear value more accurately. Life cycle milling experiments are conducted, and vibration signals are acquired for model training and validation. After model hyperparameters optimization, comparison experiment results validate that the proposed model can learn more complete features without any inner interference, and own brilliant prediction performance due to well-designed parallel structure and block-attention units. Proposed HABPDL model achieves the best prediction results, and MAPE, MAE, RMSE and R2 reach 10.8%, 6.072, 7.955, and 0.933, respectively. The model also outperforms other models even under noisy environment.

The implementation of discrete-event simulation and demand forecasting using Temporal Fusion Transformers to validate spare parts inventory policy for petrochemicals industry

One of the important strategic factors of petrochemicals plant maintenance is the spare parts inventory policy, It is significant for the efficiency, reliability, and productivity of the petrochemicals industry. An unsuitable spare part inventory policy will lead to a loss in long engineering machinery downtime due to a shortage of spare parts. To implement the spare parts inventory policy which is able to fulfill the future demand for spare parts, calculation by various statistical theories and working processes is used to custom the spare parts inventory policy. However, to validate whether or not the customized spare parts inventory policy is suitable, the discrete-event simulation library SimPy is used to mimic the actual spare parts inventory system. It must be involved in the performance evaluation process of the customized spare parts inventory policy. The inventory simulation model consists of many events depending on the supply chain system. The crucial event which is the most complex for the simulation of spare parts inventory is the demand event. This work applies the demand forecasting technique to the simulation by using deep learning with a prebuilt architecture model called Temporal Fusion Transformers (TFT). The averaged MAE of the point predictions from a global model is 0.4874+/- 6.7744 on the validation dataset and 0.6424+/-3.4963 on the test dataset. Our method predicts a quantile forecast of the future demand which is able to handle the stochastic nature of the spare parts demand in the petrochemicals industry and the result from the simulation outcome is more accurate and close to the outcome from an actual inventory system. The information from the analysis of the simulation outcome is used by the inventory management team to make decisions about the custom inventory policy before deploying it to the actual system.

The implementation of discrete-event simulation and demand forecasting using Temporal Fusion Transformers to validate spare parts inventory policy for petrochemicals industry

One of the important strategic factors of petrochemicals plant maintenance is the spare parts inventory policy, It is significant for the efficiency, reliability, and productivity of the petrochemicals industry. An unsuitable spare part inventory policy will lead to a loss in long engineering machinery downtime due to a shortage of spare parts. To implement the spare parts inventory policy which is able to fulfill the future demand for spare parts, calculation by various statistical theories and working processes is used to custom the spare parts inventory policy. However, to validate whether or not the customized spare parts inventory policy is suitable, the discrete-event simulation library SimPy is used to mimic the actual spare parts inventory system. It must be involved in the performance evaluation process of the customized spare parts inventory policy. The inventory simulation model consists of many events depending on the supply chain system. The crucial event which is the most complex for the simulation of spare parts inventory is the demand event. This work applies the demand forecasting technique to the simulation by using deep learning with a prebuilt architecture model called Temporal Fusion Transformers (TFT). The averaged MAE of the point predictions from a global model is 0.4874+/- 6.7744 on the validation dataset and 0.6424+/-3.4963 on the test dataset. Our method predicts a quantile forecast of the future demand which is able to handle the stochastic nature of the spare parts demand in the petrochemicals industry and the result from the simulation outcome is more accurate and close to the outcome from an actual inventory system. The information from the analysis of the simulation outcome is used by the inventory management team to make decisions about the custom inventory policy before deploying it to the actual system.

Online Unbalance Detection and Diagnosis on Large Flexible Rotors by SVR and ANN trained by Dynamic Multibody Simulations

Multiple-stage steam turbine generators, like those found in nuclear power plants, pose special challenges with regards to mechanical unbalance diagnosis. Several factors contribute to a complex vibrational response, which can lead to incorrect assessments if traditional condition monitoring strategies are used without considering the mechanical system as a whole. This, in turn, can lead to prolonged machinery downtime. Several machine learning techniques can be used to integrally correlate mechanical unbalance along the shaft with transducer signals from rotor bearings. Unfortunately, this type of machinery has scarce data regarding faulty behavior. However, a variety of fault conditions can be simulated in order to generate these data using computational models to simulate the dynamic response of individual machines. In the present work, a multibody model of a 640 MW steam turbine flexible rotor is employed to simulate mechanical unbalance in several positions along the shaft. Synchronous components of the resulting vibration signals at each bearing are obtained and utilized as training data for two regression models designed for mechanical unbalance diagnosis. The first approach uses an artificial neural network and the second one utilizes a support vector regression algorithm. In order to test their performance, the stiffness of each bearing in the multibody simulation was altered between 50% and 150% of the training model values, random noise was added to the signal and several dynamic unbalance conditions were simulated. Results show that both approaches can reliably diagnose dynamic rotor unbalance even when there is a typical degree of uncertainty in bearing stiffness values.

Increasing life of pit dump truck motor-wheel gearboxes by introducing the monitoring of the actual technical condition

At present, Kuzbass coal strip mines pay great attention to improving quality performance of mining equipment operation, including reliability and durability of components and units. One of the ways of the performance improvement is decreasing number of unforeseen failures. To achieve this purpose a mine dump truck part diagnostics should be introduced into a maintenance service procedure. At the same time the process of diagnostics should not increase the machinery downtime, but effectively reveal a condition of motor-wheel gearboxes in the course of dump truck operation. The aim of the research is to increase the reliability and service life of motor-wheel gearboxes of large BelAZ dump trucks. Failure of a motor-wheel gearbox is a rare phenomenon, but the cost of a new gearbox can vary from 3.5 to 10 million rubles. That is why it is important to implement such methods of diagnostics, which allow revealing the condition of gearboxes in the shortest possible time and without disassembling corresponding units. Determination of the actual technical condition of motor-wheel gearboxes is possible by different methods: vibroacoustic; acoustic; thermal; physical and chemical analysis of spent operating materials. The studies showed that none of these methods can be used as a universal one. When justifying and selecting the most suitable method, different factors should be considered, including technological, or a combination of methods should be applied, which will reduce risks, but at the same time increase costs. It is necessary to develop a better diagnostic method based on the use of several methods simultaneously.

Age-based preventive maintenance with multiple printing options

In today's economic context, production systems must be readily available and machinery downtime kept to a minimum. Maintenance and spare parts inventory management play a vital role in achieving these goals, and preventive maintenance has increasingly been considered in maintenance policies. Additive manufacturing (AM) has recently been combined with preventive maintenance, and thus represents an emerging research direction. However, few studies have as yet been conducted in this research stream, and we intend to fill this gap. Our study makes three main contributions. First, we address the main limitations of two current models (i.e., assuming that no failure occurs during the replenishment lead time of the spare parts). Second, we propose a new maintenance policy that considers two printing options with different levels of reliability and unitary purchase costs. Third, we develop a decision support system (DSS) to assist managers in deciding whether to implement a preventive maintenance policy that includes AM or conventional manufacturing (CM) parts. We take an interdisciplinary approach to conducting a parametrical analysis where we consider real data on the reliability of CM and AM parts, in addition to the impact of post-processing operations and optimization routines. We find that AM-based preventive maintenance policies are favored when the MTTF and the backorder costs are low and when the failure and maintenance costs are high. These findings have been incorporated into the DSS, which provides thresholds for every parameter to guide practitioners in choosing between AM and CM parts for preventive maintenance, without requiring time-expensive calculations.

Multifunctional Safety System as a Framework for the Digital Transformation of the Coal Mining Industry

Accidents related with operation of mining equipment is a serious problem in the extractive industries. The accidents can result in damage and downtime of machinery, or even can lead to injuries and fatalities. These problems seem to become more acute in the modern conditions of Industry 4.0 implementation. This paper focuses on applicability of a multifunctional safety system in opencast coal mining, with regard to features of this method and using digital technologies. The authors emphasize essentiality of a business-model of opencast coal mining to reach the wanted results.The first section in your paper.

Hankel Matrix-Based Condition Monitoring of Rolling Element Bearings: An Enhanced Framework for Time-Series Analysis

Robust bearing fault detection is significant to reduce the machinery down-time and to prevent catastrophic failure. Many algorithms are proposed for the faults feature extraction, but it remains challenging to monitors the condition of the mechanical systems from the overwhelming interference noise contained signal in a short response time. To address this problem, as an extension of our recent work, this article introduces an enhanced framework using acquired time-series signals. Specifically, an improved Hankel matrix-based method is proposed for the identification of the state from the sampled vibration signal for each spindle turn, where matrix similarity is employed for the mechanical operation state monitoring. The experimental results indicate that the proposed method performs considerably well in fault identification (100% identification accuracy in three tests) even with a few data samples and phase shift. This work, therefore, would have more hopeful prospects in a variety of engineering fault detection applications.

Research on Health Management Technology and Application of Lifting Machinery

Lifting machinery is a key facility for the development of modern industry. It’s not noly an important support for the construction of major national projects but also one of the core competitiveness of my country’s equipment manufacturing industry. With the development of large-scale, high-parameterization and intelligentization of lifting appliances, technologies such as safety monitoring and health monitoring have been gradually applied to industrial fields such as ports and metallurgy. In order to further develop the data-driven health management technology and engineering application of cranes, on the basis of elaborating the health management needs of cranes, the current status and development trends at home and abroad, we explored and proposed a technical framework for health management of cranes, and analyzed the lifting Key technologies such as machinery safety monitoring, health monitoring, health status diagnosis, failure trend prediction, and predictive maintenance decisions. Finally combined with a 200t casting crane in a steel plant, carried out safety monitoring and data-driven diagnosis, prediction and system visualization management. The results show that: the hoisting machinery health management system can monitor and monitor in real time and carry out early warning and forecasting to avoid the occurrence of unexpected accidents; based on predictive maintenance decisions, it helps enterprises to reasonably arrange maintenance cycles and maintenance methods to reduce the maintenance cost of the enterprise; reduce hoisting machinery downtime Time to enhance the effective value of assets; and enhance the level of enterprise information management.

Mitigating cavitation erosion using biomimetic gas-entrapping microtextured surfaces (GEMS).

Cavitation refers to the formation and collapse of vapor bubbles near solid boundaries in high-speed flows, such as ship propellers and pumps. During this process, cavitation bubbles focus fluid energy on the solid surface by forming high-speed jets, leading to damage and downtime of machinery. In response, numerous surface treatments to counteract this effect have been explored, including perfluorinated coatings and surface hardening, but they all succumb to cavitation erosion eventually. Here, we report on biomimetic gas-entrapping microtextured surfaces (GEMS) that robustly entrap air when immersed in water regardless of the wetting nature of the substrate. Crucially, the entrapment of air inside the cavities repels cavitation bubbles away from the surface, thereby preventing cavitation damage. We provide mechanistic insights by treating the system as a potential flow problem of a multi-bubble system. Our findings present a possible avenue for mitigating cavitation erosion through the application of inexpensive and environmentally friendly materials.

Development of method to form age structure of Rogun Dam construction machinery fleet

Dedicated to the commencement of operation of the second unit of the Rogun DamThe construction of the Rogun Dam (HPP) has drawn attention of the government of the Republic of Tajikistan. Due to the energy shortage for the population and industry in the winter period, the government resumed the construction of the Rogun HPP. The insufficient capacity of the construction machinery (CM) fleet at the Rogun Dam can be explained by both the unsatisfactory number of construction machines and the significant wear of the available units, resulting in machinery downtime due to failures, which accounts for up to 30% of the working time fund.

A one-class SVM based approach for condition-based maintenance of a naval propulsion plant with limited labeled data

Maintenance of the main propulsion system is an important part of ship operation, whose costs account for the most of the costs of ship machinery maintenance. Related literature has shown that condition-based maintenance can minimize maintenance costs and machinery downtime. In this respect, we propose a one-class support vector machine based approach for machinery decay status assessment which helps to achieve condition-based maintenance. The novelty of our work lies in the using of decision values returned by the trained models to estimate the decay degree and the main decay direction. In particular, only the normal data and a small amount of labeled decayed data is necessary in model training process which greatly reduces the requirement for labeled data. Consequently, it can be used as a decision support tool for condition-based maintenance not only on this plant, but also on many other machinery.

An Experimental Investigation of Hydrodynamic Journal Bearing with Different Oil Grades

Journal bearings are known to be intrinsic components in different marine applications. They are regarded as the most instrumental means by which large loads could be transmitted at mean speed of rotation. Hydrodynamic journal bearings, based on hydrodynamic lubrication, have undoubtedly proved to be among the most effective types of journal bearings designs, commonly used in marine applications. In hydrodynamic lubrication, metal-to-metal contact could be prevented via a separation between the load carrying surfaces of the bearing, that could be attained by means of a relatively thick film of lubricant. Extending bearing life in marine propulsion systems, auxiliary equipment and diesel engines, reducing friction energy losses and wear, minimizing maintenance expenses and downtime of machinery due to frequent bearing failure are the most important objectives being born in mind on launching the bearing design procedures.In the study at hand, an attempt has been made to perform a study of pressure distribution within hydrodynamic journal bearing on experimental bases. The experimental study has entailed the use of versatile lubrication oils, among of which comprised SEA 20W50, SEA 10W40 and SEA 5W30, for the sake of identifying their individual role in determining the condition of lubrication “hydrodynamic or hydrostatic”. Besides, the study has been extended to cover the pressure behaviour of different lubricants within the hydrodynamic journal bearing, at different speeds ranging from 50 to 400 RPM at constant loads. For accurately testing the 105 mm internal diameter, 58 mm bearing length and “l/d = 0.5” made of white metal, a use has been made of the universal journal bearing test rig “UJBTR”. Pressure distribution is further circumferentially investigated under constant loading, with operating at different journal rotational speeds.

ОПРЕДЕЛЕНИЕ КОМПЛЕКСНОГО ПОКАЗАТЕЛЯ ТЕХНИЧЕСКОГО СОСТОЯНИЯ ОБЪЕКТА НА ПРИМЕРЕ БЛОКА СТАБИЛИЗАЦИИ УСТАНОВКИ КАТАЛИТИЧЕСКОГО РИФОРМИНГА

Fail-safety taken as a factor of completeness of using the resources of the production facility is important for characterizing the technological process. In most modern automated process control systems the above factor is implemented by means of the emergency protection system which allows to avoid the progressing emergency. However, such measures significantly increase the downtime of process equipment, since the possibility of developing pre-emergency and emergency situations when combining several permissible and close to emergency values of parameters is not taken into account. To solve the problem of reducing machinery downtime, a method for determining the object state indicator has been proposed. It takes into account all variants of the influence of technological parameters on each other as well as their combinations leading to the emergency progress. As well, the proposed solution helps to set the proximity of the operating mode of the production equipment to the threshold emergency, which greatly reduces the risk of emergency in situ, thereby reducing equipment downtime and economic losses. The method of determining the indicator of the state of the object is considered on the example of the stabilization block of hydrogenate catalytic reforming unit.

Machinery Health Prognosis - Data Driven Approach Using Threshold Regression

Machinery health data is the backbone of prognostics. Effective prognostic, from the machinery data, leads towards operational reliability, reduced machinery downtime, cost savings, secondary/catastrophic failures etc. Various methodologies have been adopted by the researchers in an effort to precisely forecast/predict machinery health. In this study, Threshold Regression Methodology has been applied to a machinery vibration data to estimate future health state of machinery. The results show that the proposed method is an effective and reliable approach for data driven prognostics.

Simulation modelling and analysis of a production line

Production lines modelling has many problems that are difficult to be solved using analytical solutions, due to uncertainty and/or variability in variables and parameters. This paper is targeting unreliable production lines with finite buffers for the objective of evaluating and analysing the current situation and identifying bottlenecks. We use discrete event simulation to model a real production line based on one-year historical data about the breakdowns of all the production line's machinery. This data is used to find appropriate probability distributions to represent each machinery downtime and uptime. The simulation model is built using AnyLogic software to abstract the actual production line, and the model is validated using the actual data and information about the production line. Improvement scenarios are proposed to resolve the observed bottlenecks, and therefore give managerial insights to increase the throughput according to the increasing demand. Finally, production plans are set for different demands along the year.