Publikacje i wystąpienia

Photovoltaic (PV) installations increasingly experience performance losses caused by surface contamination. Conventional cleaning approaches rely on predefined maintenance schedules, which may increase water consumption, costs, or response delays. This paper presents the design and experimental evaluation of a smart self-cleaning system for sustainable photovoltaic energy harvesting. The system integrates a YOLO-based visual contamination detection module, embedded control, and a rail-guided cleaning mechanism with an automotive windscreen wiper and water spraying system. A laboratory-scale prototype based on a Raspberry Pi controller was developed to validate a condition-based maintenance strategy. Experimental investigations confirmed detection of representative contaminants and autonomous initiation of cleaning without operator intervention. The cleaning subsystem completed all cycles with low energy demand. Measurements showed that a complete cleaning cycle required approximately 0.042 Wh. The results confirm the feasibility of combining machine vision, autonomous decision-making, and automated cleaning in a low-cost prototype for photovoltaic maintenance.

Accurate and reliable rainfall measurements are essential for sustainable environmental monitoring, hydrology, agriculture, and climate studies. However, environmental contamination, including leaves, dust, insects, and other debris, can obstruct rain gauge funnels and introduce significant measurement errors. This paper presents an autonomous self-cleaning rain gauge designed to improve the reliability and sustainability of rainfall monitoring systems. The proposed solution integrates an autonomous cleaning module with embedded control and remote monitoring capabilities. Controlled laboratory experiments were conducted using a tipping-bucket rain gauge under different contamination scenarios. A baseline calibration established a reference rainfall value of 5.23 mm, which was used to evaluate performance under contaminated and cleaned conditions. The results showed that severe blockage produced measurement errors of up to 35%. After activation of the autonomous cleaning mechanism, the mean measurement error decreased from −0.40 mm to −0.09 mm, corresponding to an average error reduction of 73.8%. The findings indicate that autonomous cleaning significantly improves measurement accuracy, repeatability, and long-term operational reliability. By reducing maintenance requirements and extending autonomous operation, the proposed approach supports the sustainability of distributed environmental sensing infrastructures.

Excessive speed remains one of the main factors contributing to road‑traffic fatalities, while current Intelligent Speed Assistance (ISA) and map‑ or vision‑based systems often fail because of environmental limitations, outdated data, or high infrastructure costs. This paper proposes a cloud‑based Unified Driver Speed‑Assistance System (UDSAS), inspired by the European Train Control System (ETCS). The aim of the concept is to create a centralised and authoritative digital source of speed‑limit information and to enable real‑time computation of vehicle‑specific braking curves. The system integrates existing in‑vehicle components, including navigation, GNSS positioning, and speed‑limiter interfaces, to ensure low implementation cost and high scalability. The driver remains in the loop, while the system provides continuously updated speed guidance along the planned route. Optional extensions include offline operation, integration with vision‑based recognition, digital signage, Vehicle‑to‑vehicle (V2V) communication, and support for autonomous vehicles. The proposed solution addresses the current lack of a unified and trustworthy speed‑limit database and provides a simple, cost‑effective, and evolvable framework that can enhance road‑traffic safety, environmental performance, and driver comfort.

High-impedance ion-selective electrodes are susceptible to electromagnetic interference (EMI), which degrades potentiometric sodium and pH monitoring in automation. This study presented a disturbance-aware reconstruction pipeline for dual-channel recordings sampled at 1 kHz under laboratory conditions. A windowed sinusoidal regression was estimated and the 60 Hz interference component was subtracted using overlapping 0.5 s windows. Disturbances were detected per channel using robust amplitude and derivative scores with hysteresis, minimum duration rules, and mask dilation. Masked samples were replaced by an interpolated quiet-state baseline estimated from locally weighted scatterplot smoothing (LOWESS) using unmasked data only. Disturbed datasets contained events that masked up to 33% of the sodium record, which were reconstructed without changing the time axis. Two outputs were generated, including a drift-preserving trace and a drift-removed trace anchored to an initial reference level. Six smoothing kernels and a zero-phase second-order sections (SOS) Butterworth low-pass filter were benchmarked using quiet-only metrics. The SOS low-pass filter outperformed six alternative kernels in 22 of 24 experimental test cases, including one tie. Under the best quiet segment, robust variability decreased by approximately 50 times after reconstruction and low-pass filtering overall. Power spectral density values near -200 dB reflected a numerical reporting floor set by the safeguard, not floating-point quantisation.

Reliable ion-selective electrode (ISE) monitoring in industrial matrices is often compromised by electromagnetic interference (EMI). Traditional uncertainty models assume independent Gaussian noise; however, ISE data sampled at 1 kHz reveals temporal instabilities and cross-channel dependencies that violate these frameworks. This study presents a two-phase metrological evaluation of ISE responses (EH-01 and ENa-01). First, statistical characterization reveals extreme non-Gaussian behavior, with excess kurtosis reaching κ=20, negative skewness (γ = -3.40), and a Pearson correlation shift from r ≈ 0.26 to r ≈ 0.69 during impulsive activity. In the second phase, we evaluate uncertainty propagation using the Guide to the Expression of Uncertainty in Measurement (GUM) and validate results via Monte Carlo simulation (MCs). Our findings demonstrate that GUM underestimates measurement risk for EH-01 by 15.0% due to heavy-tailed noise, while overestimating ENa-01 uncertainty by 22.0% due to extreme central peak density. These results invalidate the second-moment-only GUM approximation in high-interference settings, necessitating multivariate, non-Gaussian uncertainty evaluation. These findings motivate the development of ML-based filtering approaches to mitigate EMI-induced transients and potentially restore GUM validity in real-time industrial process control.

Pomiary jakości powietrza wykonywane z użyciem bezzałogowych statków powietrznych stwarzają dla badaczy możliwości do tej pory trudne do osiągnięcia. W literaturze pojawia się coraz więcej publikacji na ten temat. Popularność niedrogich czujników wiąże się jednak z ich niską dokładnością i wrażliwością na zmianę warunków otoczenia. W tej pracy wykazano, że czujniki CO2 dość dobrze wskazują faktyczne stężenia tego gazu w powietrzu pod warunkiem uwzględnienia wpływu temperatury.

Due to additional parameters and similar structure with PID, fraction-order (FO) PID controllers offer tuning flexibility for better performance. More degree of freedom poses challenges in applying simple tuning rules for FO-PID controllers. The dilemma of practical realization of the FO controllers further limits their benefits. This work proposes simple tuning rules for the parameters of the FO-PI controllers by following the frequency-domain methods. We apply three different approximation approaches to realize the controller implementation and validate its performance on a simulation example. Comparison with a FO-PID controller evidences the effectiveness of the proposed approach

Electromyography (EMG) signals, crucial for neuromuscular assessment, are frequently corrupted by noise, impairing signal fidelity and subsequent analysis across diverse applications. Conventional filters often inadequately address non-stationary noise or introduce signal distortion. This paper introduces an advanced deep learning framework for EMG denoising, centred on a U-Net-inspired convolutional autoencoder with integrated residual blocks and skip connections. Training utilised synthetic EMG data, closely emulating physiological frequency bands and burst dynamics, subsequently corrupted by a comprehensive noise model encompassing electrode, crosstalk, electronic, drift, and contact artefacts. Training was guided by a custom loss function that combined weighted mean squared error (MSE) with signal-to-noise ratio (SNR). The proposed autoencoder achieved substantial improvements, SNR increased from -0.95 dB (noisy) to 14.64 dB (denoised), and MSE was drastically reduced from 0.001493 V2 to 0.000041 V2 on the test dataset. Qualitative analysis confirmed effective noise suppression while retaining crucial EMG burst characteristics. This advanced framework offers a promising solution for robust restoration of EMG signals in practical settings.

Recent review articles highlight an exponential rise in publications on Digital Twin (DT) technology. Despite its recognised potential, DTs have yet to achieve widespread practical use. Following a presentation of the state of the art and an original conceptual diagram illustrating the technology, this work presents the key factors contributing to the gap between conception and implementation. Using experimental data from a laboratory water system and a reliability-based perspective, this analysis examines the practical limitations of DT application. The findings indicate that broader use of DTs depends on the technological maturity and reliability of all system components, which still require further development.

This study presents the application of an Internet of Things (IoT)-based system for environmental data acquisition in a scientific research setting. The system comprises a network of 12 sensor nodes and a central server. Each node is built around a D1 Mini module, which collects data from an attached sensor and transmits the measurements to the server via web requests. A server-side script processes these requests and stores the data in structured text files. The collected data can be analysed either in real time during the experiment or retrospectively. To ensure durability and reliability in outdoor conditions, all sensor nodes are enclosed in protective housings. This work highlights the practicality, cost-effectiveness, and efficiency of a custom-designed, application-specific IoT measurement system, demonstrating its suitability for rapid deployment in environmental monitoring applications.

Train on-board system calculates its speed based on values provided by several sensors. This work deals with a hypothetical case where the speed is estimated throughout 60 seconds during braking. Information from five sensors characterized by different error models is fused in three ways: mean, median, and the weighted method. Although the weighted method is the most complex, it exhibits the best fit to the reference value.

Understanding dataset characterisation is fundamental to achieving accurate statis- tical modelling, particularly in the context of Global Navigation Satellite System (GNSS) data analysis. GNSS data exhibit heavy-tailed and skewed distributions, prompting this study to eval- uate non-Gaussian models (Cauchy, Student’s t, lognormal, skew-normal) in modelling GNSS data collected from a stationary receiver. This study uses maximum likelihood estimation for parameter estimation with a confidence i nterval. I t evaluates t he m odel’s p erformance using log-likelihood analysis, the Akaike Information Criterion, the Bayesian Information Criterion, and the Root Mean Squared Error. The comparative assessment of these models highlights that lognormal and skew-normal outperform in capturing extreme deviations and provide a better fit than the normal distribution. These findings underscore the importance of selecting appropriate statistical models to enhance uncertainty quantification in GNSS-based measurements.

The paper deals with the problem of ion activity determination for a mixture by means of ion-selective electrodes. Mathematical model of the analysed phenomenon is described by the Nicolsky-Eisenman equation, which relates activities of ions and ion-selective electrode potentials. The equation is strongly nonlinear and, especially in the case of multi-compound assays, the calculation of ion activities becomes a complex task. Application of multilayer perceptron artificial neural networks, which are known as universal approximators, can help to solve this problem. A new proposition of such network has been presented in the paper. The main difference in comparison with the previously proposed networks consists in the input set, which includes not only electrode potentials but also electrode parameters. The good network performance obtained during training has been confirmed by additional tests using measurement results and finally compared with the original as well as the simplified analytical model.

Determination of pH using a typical glass electrode requires prior calibration in order to determine the electrode parameters. Knowledge about uncertainties of the parameters is insufficient to calculate the uncertainty of measured pH because of existing correlation. In the paper, an example illustrating the problem is presented. Two ways of proper uncertainty assessment are suggested: (1) analytical with removing the correlated variables and (2) numerical using Monte Carlo simulations. The second one seems to be much less time-consuming and allows easier investigations of the uncertainty properties.