Application of Artificial Intelligence for the Implementation of Mismatch Negativity Potential Algorithms in Industrial Automated Predictive Maintenance Systems

PurposeThe purpose of this paper is to propose a predictive maintenance (PdM) system for hybrid degradation processes with continuous degradation and sudden damage to improve maintenance effectiveness.Design/methodology/approachThe PdM system updates the degradation model using partial condition monitoring information based on degradation type judgment. In addition, an extended multi-step-ahead updating stopping condition is adopted for performance enhancement of the PdM system.FindingsAn extensive numerical investigation compares the performance of the PdM system with the corresponding preventive maintenance (PM) policy. By carefully choosing the updating stopping condition, the PdM policy performs better than the corresponding PM policy.Research limitations/implicationsThe proposed PdM system is applicable to single-unit systems. And the continuous degradation process should be well modeled by the stochastic linear degradation model (Gebraeel et al., 2009).Originality/valueIn literature, there are abundant studies on PdM policies for continuous degradation processes. However, research on hybrid degradation processes still focuses on condition-based maintenance policy and a PdM policy for a hybrid degradation process is still unreported. In this paper, a PdM system for hybrid degradation processes with continuous degradation and sudden damage is proposed. The PdM system decides PM schedules by fully utilizing the condition monitoring data of each specific product, and can hopefully improve maintenance effectiveness.

In this paper we present a model-based approach for designing efficient control strategies with the aim of increasing the performance of Heating, Ventilation and Air- Conditioning (HVAC) systems with ice Cold Thermal Energy Storage (ice CTES). The use of TES systems ensures reduced energy costs and energy consumption, increased flexibility of operation, reduced equipment size and pollutant emissions. A simulation environment based on Matlab/Simulink® is developed, where the thermal behaviour of the plant is analysed by a lumped formulation of the conservation equations. In particular, the ice CTES is modelled as a hybrid system, where the water phase transitions (solid-melting-liquid, liquidfreezing- solid) are described by combining continuous and discrete dynamics, thus considering both latent and sensible heat. Three standard control strategies and a model predictive control approach are developed and compared. Extensive simulations confirm that the MPC provides the best control in terms of energy efficiency and cooling load demand satisfaction with respect to standard control strategies.

The universal predictive maintenance(PdM) system framework integrated several common and important technologies is established in the paper. It is very important to choose and design the relative technologies for the PdM system in the real application at different factories. Here the reasons and method of the system establishment are explained. The functions and structures of the system are analyzed. The establishment of the database of the system is proposed. The PdM system can be realized flexibly based on the real conditions in the real industry application.

Predictive Maintenance (PdM) systems use process and equipment state information to predict when a tool or a particular component in a tool might need maintenance. PdM systems can be realized cost-effectively by leveraging Advanced Process Control (APC) technologies and infrastructure. APC data collection infrastructure can provide the state information necessary for prediction. APC fault detection systems contain necessary algorithms to identify features important to prediction, including tool health. In leveraging APC systems in a reusable and reconfigurable way, cost-effective PdM systems can be realized as part of existing fab infrastructure, leading to lower unscheduled downtimes, reduced mean-time-to-repair, reduced scrap, and increased life of components and consumables.

Predictive maintenance (PdM) is quickly revolutionizing the industrial processes, which is moving from time-based reactive technology to data-driven proactive technology in a very short period. Bearings in electric motors are important components used in a wide variety of industrial operations. Such components are prone to premature failure, loss of revenue because of downtime, and repair. The Remaining Useful Life (RUL) estimation of the bearings in electric motor is important because such a prediction assists in reducing equipment downtime, extending equipment life, and minimizing maintenance cost. Conventional methods of PdM employ machine learning (ML) models which are trained from historical data and, being black boxes, lack the physical intelligibility and robustness for the prediction of RUL, under varying operating conditions. This limitation is because only data-driven models are used and it is hard to extract the physical mechanisms of bearing damage. In this work, a systematic study on hybrid predictive maintenance frameworks for electric motor bearings is presented, which incorporates both ML and physics-based models looking to address the mentioned challenges and improve the reliability of RUL predictions. A key aspect to improve the reliability and effectiveness of predictive maintenance strategies is the precise estimation of the Remaining Useful Life (RUL) for industrial bearings. (Jiang et al., 2022; Hu et al., 2023) mentioned the NASA bearing dataset they used in the study, and a number of advanced pre-processing techniques are used, such as the statistical features (Root-Mean-Square (RMS), Peak Frequency, Kurtosis, Crest Factor), and Archard wear model as a physics-guided degradation metric [1, 2]. A Bidirectional LSTM model is trained with optimal sequence lengths for learning long-term dependencies on sensor signals, and the model is evaluated using Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared (R²) score. Comparisons between purely data-driven models and physics-guided counterpart demonstrate better accuracy and robustness. The results indicated that the inclusion of the wear-related degradation characteristics based on Archard improved remarkably the predictive model, and the R² values are greater than 0.67, the lowest MAE is 3.96. (Wang et al., 2017; Xue et al., 2022) went on to add, trends-based features as well as correlation matrices were used to confirm that the selected inputs were reliable, strengthening the interpretability of the models and reducing those overfitting risks [3, 4]. This paper makes a novel contribution to the study of integrated applications of physics-based AI models in predictive maintenance systems in challenging environments such as industrial automation in Zimbabwe,(Liu and Zhang, 2020) proving that the integration of physics and AI can yield scalable, explainable and high-performing predictive maintenance systems [5].

With the continuous improvement of intelligent level and measurement and control level in mines, higher requirements are put forward for the effective maintenance of mining equipments. In this paper, a study has been made of the condition monitoring, fault prediction and predictive maintenance scheme of coal mining equipments. Taking the shearer as the research object, this paper analyzes the common faults and characteristics of the hydraulic system of the shearer, formulates the construction scheme of the digital twin body of the hydraulic system of the shearer, clarifies the mechanism and implementation process of the predictive maintenance system, puts forward the condition monitoring and fault prediction method based on the hydraulic state signal, with the digital twin body of the hydraulic system of the shearer constructed, thus the predictive maintenance system is designed based on MR (mixed reality), which ensures the safe and stable operation of the shearer. The system has been experimentally verified for the functions of condition monitoring, fault prediction and predictive maintenance, and the results show that all modules achieve the expected functions.

A boiler plays a vital role in the process of generating electricity, hence power plants need to invest in effective maintenance strategies that ensure the boiler equipment and components are well maintained and function with minimal failures. As the technology evolves, the design of boiler equipment has also become more complex which has resulted in power plants favoring the predictive and preventive maintenance strategies; however, the power plant needs to ensure that the systems supporting the predictive maintenance strategies are properly implemented. The main function of the boiler predictive maintenance system is to predict the health status of the boiler, through tube wear rate monitoring to prevent unforeseen failure leading to a boiler emergency shutdown. At power plant X, the major cause of the boiler shutdown is boiler tube failure incidents. Such incidents still occur despite the implementation of the predictive maintenance system, thus the need for this study. In this study the effectiveness of the boiler's predictive maintenance systems is evaluated, and it is identified that the data collected during inspections for system prediction is not sufficient to predict the health status of the boiler, as the system is not configured to monitor the entire tube length and does not consider all failure mechanisms, hence production loss time of 9487 hours for 12 years.

Predictive maintenance (PdM), leveraging Artificial Intelligence (AI), is transforming the telecommunications industry by enabling the prediction and prevention of network failures. This proactive strategy reduces network outages and maintenance costs while enhancing overall system performance. By employing AI technologies such as machine learning algorithms, big data analytics, and sensor data analysis, telecom operators can identify patterns and anomalies indicative of potential component failures. AI-driven models continuously monitor network health, facilitating highly accurate failure predictions and enabling timely interventions. This article examines the application of AI for PdM within the telecom sector, focusing on its impact on operational efficiency, resource optimization, and service stability. The findings highlight significant cost reductions and operational improvements achievable with PdM systems. Furthermore, the paper discusses implementation challenges and key considerations for transitioning to these systems. The future outlook for telecom PdM suggests a continued evolution towards more automated, seamless network management and an improved customer experience.

The quality of the processed products in CNC machining centers is a critical factor in manufacturing equipment. The anomaly detection and predictive maintenance functions are essential for improving efficiency and reducing time and costs. This study aims to strengthen service competitiveness by reducing quality assurance costs and implementing AI-based predictive maintenance services, as well as establishing a predictive maintenance system for CNC manufacturing equipment. The proposed system integrates preventive maintenance, time-based maintenance, and condition-based maintenance strategies. Using continuous learning based on long short-term memory (LSTM), the system enables anomaly detection, failure prediction, cause analysis, root cause identification, remaining useful life (RUL) prediction, and optimal maintenance timing decisions. In addition, this study focuses on roller-cutting devices that are essential in packaging processes, such as food, pharmaceutical, and cosmetic production. When rolling pins are machining with CNC equipment, a sensor system is installed to collect acoustic data, analyze failure patterns, and apply RUL prediction algorithms. The AI-based predictive maintenance system developed ensures the reliability and operational efficiency of CNC equipment, while also laying the foundation for a smart factory monitoring platform, thus enhancing competitiveness in intelligent manufacturing environments.

Industry 4.0 is the latest industrial revolution primarily merging automation with advanced manufacturing to reduce direct human effort and resources. Predictive maintenance (PdM) is an industry 4.0 solution, which facilitates predicting faults in a component or a system powered by state-of-the- art machine learning (ML) algorithms (especially deep learning algorithms) and the Internet-of-Things (IoT) sensors. However, IoT sensors and deep learning (DL) algorithms, both are known for their vulnerabilities to cyber-attacks. In the context of PdM systems, such attacks can have catastrophic consequences as they are hard to detect due to the nature of the attack. To date, the majority of the published literature focuses on the accuracy of DL enabled PdM systems and often ignores the effect of such attacks. In this paper, we demonstrate the effect of IoT sensor attacks (in the form of false data injection attack) on a PdM system. At first, we use three state-of-the-art DL algorithms, specifically, Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Convolutional Neural Network (CNN) for predicting the Remaining Useful Life (RUL) of a turbofan engine using NASA's C-MAPSS dataset. The obtained results show that the GRU-based PdM model outperforms some of the recent literature on RUL prediction using the C-MAPSS dataset. Afterward, we model and apply two different types of false data injection attacks (FDIA), specifically, continuous and interim FDIAs on turbofan engine sensor data and evaluate their impact on CNN, LSTM, and GRU-based PdM systems. The obtained results demonstrate that FDI attacks on even a few IoT sensors can strongly defect the RUL prediction in all cases. However, the GRU-based PdM model performs better in terms of accuracy and resiliency to FDIA. Lastly, we perform a study on the GRU-based PdM model using four different GRU networks with different sequence lengths. Our experiments reveal an interesting relationship between the accuracy, resiliency and sequence length for the GRU-based PdM models.

Maintenance logs serve as the backbone of data-driven Predictive Maintenance (PdM) systems by providing information that can be used to create and label datasets for training survival analysis and machine learning (ML) models. However, due to personnel manually entering information into maintenance logs and the various levels of flexibility that maintenance tracking systems allow, service records often contain errors. Currently, the cleaning of equipment maintenance records is performed manually by experts such as data scientists or reliability engineers. Nevertheless, this task is time-consuming and often does not entirely eliminate noise from the data. In this paper, we propose using large language model (LLM)-based agents to automate the cleaning of maintenance logs. We provide an implementation that allows the agents to perform data cleaning as well as metrics to assess agents' performance. Finally, we compare the performance of several LLMs on this task. Our empirical results indicate that LLM-based agents are a promising solution for improving the quality of the datasets used in PdM systems and ultimately developing predictive maintenance models that are more reliable and useful.

Traditional preventive maintenance operations are being abandoned and electric utilities are becoming more failure driven due to the financial constraints being placed on them. When some distribution equipment begins to deteriorate, intermittent incipient faults persist in the system from as little as several days to several months. The failure of equipment in power distribution systems can have a direct or indirect impact on the reliable delivery of quality power. Also, certain failures can result in loss of service. There is great interest in the utility industry for low-cost, automated, real-time approaches which can detect distribution incipient faults and locate their source. This paper discusses an expert system based incipient failure detection and predictive maintenance (FDPM) system being developed for application in distribution systems. The FDPM system includes an expert system engine, a knowledge base, mathematical and neural network models of aging of distribution equipment, historical measurements databases, a distribution state estimator, a fault and disturbance event locator and a distribution system interconnection map. The FDPM system detects incipient disturbances, classifies the type of disturbance, and locates the source of the incipient behavior. If the source is one of the components under observation by the FDPM system, it assesses the integrity of the distribution system component and predicts maintenance needs.

Rapid developments in technologies such as Robotics, Digital Automation, Internet of Things and AI have heralded the Fourth Industrial Revolution, commonly referred to as Industry 4.0 (i4.0). Industrial operations and products have since become more competitive and hence more demanding. Systems have also become more complex and inter-disciplinary in nature. Diligent surveillance of operating conditions of such systems and initiation of appropriate actions based on monitored conditions have become indispensable for sustainability of businesses. Significant amount of research is being undertaken world over to meet this requirement of the day. In line with the ongoing research, this paper highlights the need for identifying the needs of condition monitoring preparedness of process plants located in remote places, especially in a logistic sense. Issues related to assessment of the need for the new paradigm in condition monitoring, challenges faced by such plants in the transition from legacy systems to a new system and customisation and optimisation of Predictive Maintenance under Industry 4.0 (PdM 4.0) have been discussed. A Case Study pertaining to remote monitoring of a gas compressor system of a petroleum refinery in North Eastern India and a Case Discussion on Basic Technical Requirements for the implementation of Industrial internet of Things (IIOT) based predictive maintenance system are presented to highlight the benefits and issues associated with the radical shift in paradigm from legacy systems to Industry 4.0 based predictive maintenance (PdM 4.0) system. Frameworks for PdM 4.0 system decision making and development are also suggested for supporting future work in this area.

This article presents an industrial predictive maintenance (PdM) system used in soybean processing based on artificial intelligence (AI) and Industrial Internet of Things (IIoT) technologies. The PdM system allows for the continuous monitoring of relevant production equipment/motor parameters, such as vibration, sound/noise, temperature, and current/voltage. It is designed to identify abnormalities and potentially break down situations to prevent damage, reduce maintenance costs and increase productivity. Condition monitoring is combined with AI-based methods and edge processing to identify the parameter changes and unusual patterns that occur before a failure and predict impending failure modes well before they occur. The PdM demonstrator currently under evaluation is planned to integrate intelligent IIoT-based sensors to measure parameters, convolutional neural network and Wi-Fi, LoRaWAN, Bluetooth low energy (BLE) technologies for intelligent communication.

The Frameworks of EcoPFM Predictive Maintenance (PM) System presents a novel approach to maintenance optimization within eco-friendly power facilities, addressing the critical need for sustainable, efficient asset management. This paper introduces an integrated framework leveraging advanced predictive analytics, machine learning algorithms, and Internet of Things (IoT) technology to enable proactive maintenance interventions based on real-time data insights. Focusing on the context of the United States it highlights the significance of implementing such a system in the realm of eco-friendly energy infrastructure. The automotive and heavy-duty truck industries in the United States grapple with the challenge of optimizing maintenance strategies to ensure vehicle reliability, safety, and environmental sustainability. Traditional maintenance approaches, primarily reactive or scheduled maintenance, fall short in addressing the complexities of modern vehicle operations. The U.S. Department of Transportation reports that heavy-duty trucks transport approximately 70% of the nation's freight by weight, underscoring the sector's critical role in the economy. However, inefficiencies in maintenance strategies contribute to significant economic and operational setbacks. According to the American Transportation Research Institute, unscheduled truck maintenance and repairs are leading operational costs for fleets, with an average expense of 16.7 cents per mile in 2020, highlighting the financial strain of current maintenance practices. In the United States, the demand for eco-friendly power solutions is rapidly increasing, driven by a growing awareness of environmental sustainability and the imperative to reduce carbon emissions. As the nation transitions towards renewable energy sources and eco-friendly power facilities, the effective management of these assets becomes paramount to ensuring reliability, performance, and longevity. The EcoPFM PM System integrates diverse data sets sourced from eco-friendly power facilities across the USA, encompassing historical operational data, sensor readings, and environmental parameters. Through predictive analytics, the system identifies patterns and trends within these data sets to forecast equipment failures and performance degradation accurately. By prioritizing maintenance tasks based on risk assessment models and condition monitoring, the system enables organizations to optimize resource allocation, minimize downtime, and extend asset lifespan. Embracing the Frameworks of EcoPFM Predictive Maintenance System holds immense promise for organizations operating eco-friendly power facilities in the United States. By harnessing data-driven insights and proactive maintenance strategies, this system offers a pathway towards enhanced operational efficiency, cost reduction, and sustainability, ultimately contributing to the advancement of eco-friendly energy infrastructure in the nation.
 Keywords: Predictive Maintenance, System, ECOPFM, Technology.