Additive manufacturing is one of the most widely used techniques in domain manufacturing. Three-dimensional (3-D) printers are those systems that made additive easier. Fused-deposition-modeling-based 3-D provide cost-effective models. Like other mechanical systems, also face faults damage printing system. Hence, proper maintenance required. The data-driven-based approach diagnosis fault proposed this letter. Data collected for three scenarios—1) healthy condition, 2) bed failure, and 3) arm...

An accurate and reliable technique to predict rechargeable battery health proves helpful in battery-operated, low-resourced industrial IoT devices. The existing data-driven prediction techniques often require a comparatively large amount of computational power for predicting the State Health (SOH) Remaining Useful Life (RUL) due most methods being feature-heavy. Further, there are very limited works RUL nodes. To address this issue, paper presents unique IoT-based sensor node framework,...

An accurate and robust technique for sensor fault diagnosis proves useful an uninterrupted supply of correct monitoring data across the Internet-of-Things (IoT) network. The manual checking calibration thousands sensors deployed in IoT network is a challenging task. Furthermore, most techniques require additional hardware support calibration. To address these issues, unique IoT-based framework, SNRepair, has been proposed, which uses modified deep reinforcement learning (DRL) detecting...

The early detection of stator faults in three-phase induction motors is great importance for modern smart industries' safety, reliability, and performance. existing fault techniques are based on voltage-current parameters collected from the motor control system, making process invasive complex. To mitigate these drawbacks, a novel non-invasive data-driven-based technique using vibration signature has been proposed. statistical data analysis first used to optimize best accelerometer mounting...

An accurate technique for early detection of sensor faults proves useful in the uninterrupted supply correct monitoring data across Internet Things (IoT) network. Most existing AI-based fault diagnosis techniques have a high computational burden, and their "black-box" nature creates challenges generating adequate trust high-risk industrial applications. To address drawbacks, unique IoT-based method, i.e., XAI-LCS, has been proposed that uses eXtreme gradient boosting algorithm detecting...

The diagnosis of human knee abnormalities using the surface electromyography (sEMG) signal obtained from lower limb muscles with machine learning is a major problem due to noisy nature sEMG and imbalance in data corresponding healthy abnormal subjects. To address this challenge, combination wavelet decomposition (WD) ensemble empirical mode (EEMD) Synthetic Minority Oversampling Technique (S-WD-EEMD) proposed. In study, hybrid WD-EEMD considered for minimization noises produced during...

Understanding the State-of-Health (SoH) and Remaining Useful Life (RUL) of commonly used Lithiumion Batteries (LIBs) within emerging sector Electric Vehicles (EVs) is crucial for assuring their efficiency, stability, safety. However, predicting SoH precisely challenging due to intricacies electrochemical processes cells especially variability real-world operating circumstances an EV. As a result, numerous data-driven approaches estimating with resilient adaptable characteristics have been...

Abstract The performance of coal-fired boilers has a significant impact on the overall yield thermal power plants. Among various boiler faults, clinkering fault diagnosis is one most crucial and scarcely addressed topics in literature. Existing detection methods are boiler-specific require both healthy faulty data for training, which difficult to acquire. To overcome these drawbacks, generalized method early proposed that only requires normal operation training. A...

In modern industrial processes, monitoring the health of rotating machinery is an important task. Many machine learning (ML) and deep (DL)-based models have shown good fault detection diagnosis results. However, these need to provide explanations insights users experts in order increase adoption spread technologies. Another common problem lack labeled historical data, which makes it impossible use supervised models. Therefore, we propose a new approach for (FDD) systems overcome challenges....

An accurate and reliable technique for predicting Remaining Useful Life (RUL) battery cells proves helpful in battery-operated IoT devices, especially remotely operated sensor nodes. Data-driven methods have proved to be the most effective until now. These devices low computational capabilities save costs, but Data-Driven health techniques often require a comparatively large amount of power predict SOH RUL due being feature-heavy. This issue calls ways with least calculations memory. paper...

The utility of the deployment IoT sensor nodes in Smart Cities is mainly dependent on accurate data generated by sensors. Hence, automatic detection faults and self-calibration faulty sensors essential for uninterrupted operation Sustainable IoT. This paper presents sCalib technique this purpose, considering temperature a warehouse as an example. Suppose one becomes inaccurate due to drift or similar reasons; that case, can identify fault automatically calibrate using another non-faulty...

Real-time fault diagnosis in Unmanned Aerial Vehicles (UAVs) is a challenging task. Data-driven intelligent of faults ensures flight safety for UAVs. In this paper, realtime on small scale fixed-wing UAVs has been shown by data natural conditions with wrapped wing structure that breaks the geometric symmetry. AutoML based approach was taken multi-class classification. Two datasets were created from combination two days. The experimental results showed proposed Deep Learning model...

In Industry 4.0, the world is trying to move its dependency from non-renewable a renewable source of energy. Solar energy regarded as most reliable Over past few years, solar sector has seen tremendous growth due increased usage photovoltaic (PV) cells fulfill needs. The Internet Things (IoT) based solutions are now powered by PV become sustainable. To improve efficiency and reliability cell-powered IoT systems, detection prediction different defects in have critical. Several fault...

This is a demo abstract based on our previously published article [1]. One of the significant shortcomings in existing data-driven battery health prediction techniques include higher computational complexity requirement to predict State Health (SOH) and Remaining Useful Life (RUL). To mitigate this drawback, an accurate reliable solution has been proposed for heath battery-operated, low-resourced IoT devices. The voltage time-based features are extracted random learning algorithms with good...

Nowadays, sensors play a vital role in monitoring many appliances various sectors, including the medical field. The electrocardiogram (ECG) is test used to check heart's electric activity, using ECG sensors. These can get faulty and cause serious havoc. In this paper, real-time classification prognostics system proposed that could classify between normal signals even predict whether sensor will based on previous data. Drift bias fault are considered study. Four different machine learning...

The accurate estimation of the State Health (SoH) and Remaining Useful Life (RUL) Lithium-ion batteries are great significance for safety performance electric vehicles (EVs), However, existing SoH techniques involve non-dynamic feature extraction without considering time computation cost, leading to challenges in real-time implementation on with varying specifications. To address these issues, this paper proposes a novel hybrid model, DELiB, designed by integrating Convolution neural...

The safety of a wide range devices depends on the accurate long-term estimation State-of-health (SoH) Lithium-ion battery. However, existing techniques have several limitations in terms prediction accuracy, computational complexity and applicability. To mitigate these drawbacks, novel Deep learning-based method with low has been proposed that can predict battery's future performance upto 100 cycles, thereby providing an early warning battery failure. A unique feature called Interval time for...

Beyond the Automation Pyramid, industries are currently embracing intelligence. One of challenges in Industry 4.0 is to conduct Predictive maintenance (PdM) for Investment Casting Process, which one oldest metal-forming industrial processes. According existing works, PdM achieved by data-driven methods scheduling just-in-time maintenance. However, traditional Machine Learning (ML) techniques can make good predictions some extent but not guaranteed be accurate. This limitation served as...