The recent development of highly automated machinery and intelligent industrial plants has increasingly enabled the continuous monitoring of their efficiency and condition, with the aim of maintaining high production efficiency and minimal malfunctions. Typical condition monitoring and fault detection applications are often achieved using acoustic and vibrational techniques, but the availability of distributed electrical measurements opens new opportunities for industrial fault detection with minimal impact on electrical systems. Even if artificial intelligence-based approaches can be used to model industrial equipment by means of measures made on electrical systems to which they are connected, machine learning algorithms have been demonstrated to be particularly adequate for this purpose due to the huge amount of data produced by interconnected sensors and devices. In this context, the aim of this work is to propose a new unsupervised analysis methodology for detecting anomalies in industrial machinery using electrical current values and other parameters measured on the power grid. The proposed framework is aimed at incorporating the advantages of Machine Learning algorithms and those of traditional analysis, optimizing their operation to improve performance and execution time; this also incorporates a methodology for analyzing the temporal dynamics of the anomaly based on Short-time Fourier transform to strengthen the performance of the detection. The results obtained showed excellent performance, both compared to the evaluations of a technical expert and to other methodologies used in the literature, with zero false positives detected in all datasets tested and a negligible number of undetected outlier events, less than 4% of the total in the datasets.

Battery state of charge estimation is not new, and the literature is rich in examples. Still, in most cases, state of charge algorithms are embedded in battery chargers and do not present the charging stage with a standard communication form, thus limiting the possibility of applying advanced smart charging algorithms. To overcome this limitation, this study proposes a method for the estimation of the state of charge of onboard batteries installed on light electric vehicles and the identification of their charging stage, if charged by the constant-current/voltage method, without the need for direct current measurements or communication from the light electric vehicle battery management system. The method is based on a preliminary off-line characterization of the full charge of a given battery/charger set by implementing a learning algorithm. The parameters learned during the off-line test are then used in on-line applications to estimate the light electric vehicle’s state of charge and identify its charging stage by only using the active power measured at the light electric vehicle charging socket. The method was tested on two different instances of the same battery/charger set family installed on two different light electric vehicles, providing a state of charge estimation with an overall accuracy of about 1.5%, computed as root mean square error over the whole state of charge range. The results showed that the proposed method could be successfully applied to estimate the real-time state of charge of light electric vehicles by only using low-cost electronic devices.

The Internet of Things (IoT) demands scalable, secure, and feeless distributed ledger technologies (DLTs) to enable seamless machine-to-machine transactions. The IOTA DLT was developed to fulfill this vision through its feeless Directed Acyclic Graph (DAG) named the Tangle, whose announced upgrade to IOTA 2.0 promised feeless microtransactions and coordinator-free (Coordicide) decentralization via a Nakamoto Consensus mechanism and a Mana anti-spam system. However, its delayed decentralization and scalability limitations hindered ecosystem growth and practical IoT adoption, leading to a new ledger architecture named IOTA Rebased. This paper critically analyzes this architectural pivot and its implications for IoT applications, contrasting the abandoned IOTA 2.0 protocol—a leaderless, feeless DAG designed for the IoT—with the adoption of a Move Virtual Machine-based, object-oriented ledger secured by a Delegated Proof-of-Stake consensus via the Mysticeti protocol in IOTA Rebased. We evaluate IOTA Rebased trade-offs: enhanced programmability and speed versus compromised IoT suitability due to fees, and explore mitigation strategies such as sponsored transactions, lightweight clients, and hierarchical tiered transaction architecture to align IOTA Rebased with IoT environments where microtransactions are prevalent. A use case analysis is provided for the integration of IOTA Rebased in IoT scenarios. This study underscores the tension between technological innovation and decentralization, offering insights for balancing scalability with the unique demands of the IoT.

The main goal of the paper is to fully explore the capability of the ‘No Motion No Integration’ (NMNI) technique for the optimization of Euler angles in orientation tracking. The gyroscope is the critical component in the Inertial Measurement Unit for angle detection in Industry 4.0. An upgraded NMNI model is introduced to remove the drift of the gyroscope significantly for the advanced measurement approach for roll, pitch, and yaw. A model of threshold update is implemented into the NMNI algorithm to compensate for the increased offset of temperature. This pre-processing method is applied to the Madgwick filter and Mahony filter to acquire the optimal performance. The experiments were carried out by using a lowcost platform equipped with Micro-Electro-Mechanical system sensors. A pan-tilt unit with high accurate positioning was used to move the sensors and obtain a reference angle during both static and dynamic experiments. A substantial improvement was clearly demonstrated after the optimization process. The measurements of Euler angles have minimized noise and tracks around the reference points properly. The results show strong competition from both fusion filters where the fused Mahony accomplishes more stable less variation in roll and pitch, but the fused Madgwick shows more precision in heading/yaw estimation.

Having become aware of how limited all natural resources are, the water leakage problem in piping systems has become a stimulating topic. This problem increased over the last few years even though innovative tools and techniques appeared in the literature and in the consumer market. Identifying water leaks at the nearest point, the household level, is still an unsolved problem because most water meters are mechanical and, therefore, cannot detect leaks. While the issue is not important for water service providers since consumption is charged to the user, the resolution is crucial due to the increasingly relevant concern of saving natural resources. The detection of small but continuous leaks of drinking water in domestic systems is addressed in this work. Machine learning approaches enabled image processing techniques also in uncontrolled environments, overcoming the classical methods but introducing new challenges like power consumption. Using a combination of Convolutional Neural Networks (CNNs) and Long Short–Term Memory Networks (LSTM) within an adaptive under–sampling strategy, it is possible to process the images captured from the mechanical water meter dial and identify the Period With Null Consumption (PWNC) or the consumption class. The presented solution can classify the water flow into four different classes, and, in the case of absence or small flow, its function becomes to detect leakages. Analyzing images from a mechanical water meter quadrant, it has been possible to identify PWNC and detect small water leakages in the domestic environment under common consumer flow profiles. In addition to the confusion matrices, the synthetic parameters of Sørensen–Dice coefficient (DSC) and Jaccard Index have been utilized and presented to quantify the performance of the proposed DNN. The conducted experiments on static and dynamic water flow demonstrated the applicability of this approach and the possibility of an increase in PWNC identification thanks to the adaptative increase in the sampling time. Moreover, the reduction of sampling time allows for the reduction of computational load and power consumption in embedded scenarios where limited energy is available.

Light and active mobility, as well as multimodal mobility, could significantly contribute to decarbonization. Air quality is a key parameter to monitor the environment in terms of health and leisure benefits. In a possible scenario, wearables and recharge stations could supply information about a distributed monitoring system of air quality. The availability of low-power, smart, low-cost, compact embedded systems, such as Arduino Nicla Sense ME, based on BME688 by Bosch, Reutlingen, Germany, and powered by suitable software tools, can provide the hardware to be easily integrated into wearables as well as in solar-powered EVSE (Electric Vehicle Supply Equipment) for scooters and e-bikes. In this way, each e-vehicle, bike, or EVSE can contribute to a distributed monitoring network providing real-time information about micro-climate and pollution. This work experimentally investigates the capability of the BME688 environmental sensor to provide useful and detailed information about air quality. Initial experimental results from measurements in non-controlled and controlled environments show that BME688 is suited to detect the human-perceived air quality. CO₂ readout can also be significant for other gas (e.g., CO), while IAQ (Index for Air Quality, from 0 to 500) is heavily affected by relative humidity, and its significance below 250 is quite low for an outdoor uncontrolled environment.