Abstract: This study provides a comprehensive overview of research and advancements on carbon materials with regard to practical targets for hydrogen storage in terms of gravimetric and volumetric capacities. For the sake of clarity, only the most relevant references on hydrogen storage by adsorption are presented, although the study was conducted in the same exhaustive manner as the one initially carried out by Anne C. Dillon and Michael J. Heben [Appl. Phys. A 2001, 72, 133–142] with a particular emphasis on emerging technologies and potential applications in various sectors, and focusing on the importance of carbon-based materials with high specific surface areas and porous structures optimised to maximise adsorption — including at high pressure —, while primarily limiting references herein to experimentally validated results. It therefore offers insights into the porous materials as well as the methodologies — including a fully comprehensive and so far proven highly transferable intermolecular hydrogen model combining van-der-Waals's and Coulomb's forces — used to improve hydrogen solid storage efficiency.

Abstract: Understanding CO₂ transport in fractal porous media requires models capable of capturing multi-scale structural variability and temporal correlations inherent to complex geological formations. In this work, we develop a mechanistic stochastic framework based on wavelet-assisted damped fractional Brownian motion (WA-DFBM) to describe CO₂ migration and diffusion across fractal pore structures. The method integrates multi-resolution wavelet decomposition with the long-range dependence and damping characteristics of fractional Brownian motion, enabling simultaneous representation of microscopic heterogeneity, temporal memory, and dissipative effects. The resulting WA-DFBM framework reproduces key transport signatures observed in porous media, including anomalous diffusion, non-stationary fluctuations, and scale-dependent variance evolution. Comparison with conventional Brownian-based models demonstrates that WA-DFBM provides enhanced capability for representing multi-scale pore heterogeneity and dynamic variability. This approach offers improved mechanistic insight into CO₂ transport behavior in fractal porous media and establishes a generalized modeling framework applicable to a wide range of subsurface flow and transport problems.

Abstract: Growing cooling demand and environmental concerns surrounding mechanical vapour compression systems motivate research into alternative technologies capable of converting low-grade heat into useful cooling. Silica-gel/water single-stage dual-bed adsorption chillers (ADCs) are promising candidates, however, their design must balance conflicting performance targets. This study proposes a regression-assisted multi-objective optimisation framework for low-grade-heat ADC, combining statistically validated surrogate models with the Ant Lion Optimiser and its multi-objective variant. Three co-equal objectives, coefficient of performance (COP), cooling capacity (Q_cc) and waste-heat recovery efficiency (η_e) are jointly maximised to map an operational envelope for sustainable cooling. Two-dimensional Pareto-optimal solutions exhibit a one-dimensional ridge in which η_e declines, and COP and Q_cc increase simultaneously. Within the explored bounds, non-dominated ranges span COP=0.675–0.717, Q_cc=18.3–27.5 kW and η_e=0.118–0.127, with a practical compromise near COP ≈ 0.695, Q_cc ≈ 24 kW and η_(e )≈ 0.122–0.123. While mass flow rate decisions increase Q_cc at the expense of η_e, a one-at-a-time sensitivity analysis with re-optimisation identifies the hot- and chilled-water inlet temperatures and exchanger conductance as the dominant decision variables and maps diminishing-return regions. The proposed framework can effectively use low-grade heat in future low-carbon buildings and processes and supports the configuration of ADC systems.

Abstract: Rain erosion testing of wind turbine blade coatings is still based almost entirely on kinetic test parameters while ignoring the temperature–salinity domains that control field damage. This short communication quantifies how far current rain erosion test conditions diverge from offshore environmental temperature–salinity envelopes. Using seas surface climatology for four offshore regions (North Sea, US Atlantic shelf, Taiwan Strait–South China Sea, Bass Strait) and temperature and water composition parameters from 11 water-based erosion studies, we show the environmental sea surface temperature (SST) spans 5 – 17 °C in the North Sea and 8 – 24 °C in the US Atlantic, rising to 20 – 32 °C in the Asia–Pacific and Bass Strait, with sea surface salinity (SSS) typically 31 – 35.5 PSU. In contrast, all reported droplet erosion tests were run between 18 and 29 °C; three of 11 used only freshwater (~ 0 PSU) and the remainder a single seawater-like level (3.0 – 3.5% NaCl, ≈30 – 35 PSU). No study combined marine salinity with cold (<10 °C) or tropical (≥28 – 30 °C) temperatures, despite evidence of markedly higher damage in saline media and up to an order of magnitude increase in polyurethane rates near the glass transition region.

Abstract: Ethanol, a key renewable biofuel, is vital for decarbonizing transportation and industry. Accurate simulation of its combustion is crucial for optimizing engines and burners, especially in diffusion flames common in diesel engines, turbines, and furnaces. While detailed kinetic models offer precise predictions, they are computationally expensive, prompting the creation of reduced and global mechanisms.However, these vital modeling tools remain scattered across the literature, and no systematic overview exists for models validated specifically for ethanol diffusion flames. Researchers currently lack a consolidated guide on which mechanisms are available, what reduction methods were used, and against which specific non-premixed experimental targets they have been validated.This scoping review addresses this gap by systematically identifying, mapping, and categorizing the existing literature on both detailed and reduced kinetic mechanisms for ethanol diffusion flames. We will collate mechanisms from numerical and experimental studies, classifying them by their size, reduction methodology, and the specific diffusion flame configurations and properties used for their validation.This review will facilitate appropriate model selection for diffusion flame simulations, highlight key progress, identify unaddressed research gaps, and ultimately accelerate the development and deployment of cleaner, more efficient ethanol combustion technologies.

Abstract: Accurate forecasting of wind power is essential for maintaining the stability and efficiency of power networks as renewable energy sources become more integrated. This study proposes a multilevel spatial-temporal graph convolution network (MLAGCN) for wind power forecasting. The framework combines a multilevel adaptive graph convolution (MLAGC) and a lightweight temporal transformer (LWTT) to jointly model complex spatial-temporal relationships in wind power data. MLAGC is constructed using three adaptive graphs: a local-aware graph, a global-aware graph, and a structure-aware graph. These components form a flexible graph structure that effectively represents dynamic spatial interactions while LWTT learns short- and long-term sequential patterns. Experiments on real wind farm datasets demonstrated that the proposed model outperforms existing baselines. The model achieved an improved prediction accuracy and generalization, as indicated by a lower score of 43.44, mean absolute error (38.83), root mean square error (48.05) and a forecast loss of 0.22. These results demonstrates the effectiveness of temporal modeling and multilevel attention-based adaptive graph learning for high-resolution wind power forecasting.

Abstract: Thorium has been recognized as a potentially viable alternative fuel for Molten Salt Reactors, Advanced Heavy Water Reactors, and High Temperature Reactors, and it is also being considered for Accelerator Driven Systems. Modular thorium reactor designs are expected to generate electrical outputs ranging from 100 to 3000 MW, underscoring their adaptability to national energy strategies. The role of thorium in nuclear reactors and fuel cycles, with a particular focus on advanced pressurized water reactors (PWRs), is reviewed. Currently, most Generation II nuclear reactors still depend on uranium within once-through fuel cycles. In nuclear-producing countries, growing interest in thorium reflects a wider effort to diversify energy generation, strengthen national nuclear programs, and align technological development with environmental goals. Key priorities include preserving limited uranium reserves, addressing rising global energy demand, and facilitating rural electrification. Domestic thorium-based power plants offer the potential to support sustainable development by curbing carbon dioxide and nitrogen oxide emissions. Furthermore, their resilience to seasonal variability enhances electricity supply stability, a crucial consideration for agriculture and irrigation. This study aims to explore current and future R&D priorities for innovative modular thorium reactors in nuclear-producing countries while presenting preliminary design concepts and simulation results that illustrate their prospective role in advanced nuclear energy systems. A preliminary electricity production cost estimate of US$20 per MWh was informed by Copenhagen Atomics' work on thorium molten salt reactors.

Abstract: This study investigates the influence of array height, irradiance, and wind speed on temperature difference and thermal gradients in photovoltaic (PV) arrays operating in hot, arid conditions. A field experiment was conducted in Mesa, Arizona (latitude 33° N), using two fixed-tilt PV module arrays installed at different elevations—one at 1 m and the other at 2 m above ground level. Each array comprised seven monocrystalline PV modules arranged in a single row with an 18° tilt angle optimized for summer performance. Data were collected between June and September 2025 and the analysis was restricted to 10:00–13:00 h to avoid shading and ensure uniform irradiance exposure on both arrays. Measurements included module backsheet temperatures at the center and edge modules, ambient temperature, plane-of-array (POA) irradiance, and wind speed. By maintaining identical orientation, tilt, and exposure conditions, the evaluation isolated the effect of height on module operating temperature and intra-array thermal gradients. Results indicate that the 2 m array consistently operated 1–3°C cooler than the 1 m array, confirming the positive impact of elevation on convective cooling. This reduction corresponds to a 0.4–0.9 % improvement in module efficiency or power based on standard temperature coefficients of crystalline silicon modules. The 1 m array exhibited a mean edge–center temperature gradience of −1.54°C, while the 2 m array showed −2.47°C, indicating stronger edge cooling in the elevated configuration. The 1 m array displayed a broader temperature range (−7 °C to +3°C) compared to the 2 m array (−5°C to +2°C), reflecting greater variability and weaker convective uniformity near ground level. The temperature gradience became more negative as irradiance increased, signifying intensified edge cooling under higher solar loading. Conversely, wind speed inversely affected ΔT, mitigating thermal gradients at higher airflow velocities. Overall, elevating PV arrays enhances convective heat transfer, reduces module temperature, and improves reliability and power output. These findings highlight the importance of array height, array length, irradiance, and wind conditions in optimizing PV system thermal and electrical performance.

Abstract:

Reduce fuel costs, improve waste utilization, and enhance energy efficiency by steaming mushroom substrate cubes using a mixed-fuel burner and furnace system that uses crude glycerol and used vegetable oil as alternative low-cost energy sources. This was the objective of this study. The experimental method measured boiler performance, exhaust-gas composition, temperature profiles, steam generation, and combustion-gas distribution inside the furnace. It was supported by analytical modeling of pressure, temperature, and combustion-gas distribution. Five fuel mixtures were prepared and tested, including 100% used vegetable oil, 100% glycerol, and 50/50, 25/75, and 10/90 blends. The tests were conducted in accordance with DIN EN 203-1. Blending used vegetable oil with glycerol improves flame stability, increases peak temperatures, and reduces the formation of incomplete combustion products compared to pure glycerol. The results also show that the mixture achieves high combustion efficiency (≈90-99%) and boiler thermal efficiency (≈72-73%). The optimal blend for stability, efficiency, and cost savings was 25/75 glycerol and vegetable oil. By cutting yearly fuel expenses by almost half, reducing steaming time by 2 hours per batch, and achieving a quicker payback period (3.26 months), the mixed-fuel system proved to be economically more advantageous than LPG, making it evidently practicable for agricultural producers. This study's conclusions suggest that, to maximize the use of renewable waste fuels and improve long-term sustainability, the following actions should be taken: further optimizing the air-fuel mixing process to improve combustion of higher-glycerol blends; scaling the system for larger mushroom farms; and expanding testing to other agricultural heating applications.

Abstract:

Oil spills pose severe ecological and economic threats, making rapid detection and severity assessment essential for effective environmental response and mitigation. Traditional remote-sensing approaches rely heavily on manual interpretation or rule-based algorithms, both of which are limited by variability in weather, illumination, and sea conditions. With the growing availability of satellite imagery and advancements in artificial intelligence, deep learning techniques offer powerful alternatives for automated oil spill identification. This study develops and evaluates a two-stage deep learning pipeline designed to (1) detect and segment oil spill regions in satellite images using semantic segmentation, and (2) classify the severity of identified spills using a supervised image-level classifier. The project utilizes the publicly available altunian/oil_spills dataset, consisting of 1,040 paired satellite images and color-encoded segmentation masks representing four classes: Background, Water, Oil, and Others. Stage 1 of the pipeline employs a U-Net architecture with a ResNet-18 encoder pretrained on ImageNet. The model performs pixel-level segmentation to isolate oil regions from surrounding ocean and environmental structures. Stage 2 uses a modified ResNet-18 classifier that accepts four-channel one-hot encoded segmentation outputs and predicts one of three spill severity levels derived from the proportional area of oil pixels: No Oil (<5%), Minor (5–15%), and Major (>15%). The pipeline was trained using the PyTorch framework with separate training cycles for each stage, enabling modular evaluation and interpretability. A systematic experimental setup including an 80/10/10 training–validation–test split, cross-entropy loss functions, Adam optimization, and 20-epoch training windows was used to assess model performance. Results show that the U-Net segmentation model achieves a mean Intersection-over-Union (IoU) of 0.8156 on the test set, with particularly strong performance on the Background (0.9123) and Water (0.8567) classes and lower, but still effective, performance on the Oil class (0.7234). These findings reflect the inherent class imbalance in satellite imagery, where oil occupies a small proportion of total pixels. The ResNet classifier achieved an overall accuracy of 88.76%, with F1-scores of 0.90 for No Oil, 0.85 for Minor, and 0.90 for Major severity levels. Classification errors were concentrated around the Minor category, consistent with threshold-based class definitions and segmentation uncertainty. The combined results demonstrate that a two-stage deep learning approach offers substantial improvements in both accuracy and interpretability over single-stage or heuristic-based systems. Segmentation masks provide visual justification for classification outputs, enabling a more transparent workflow for environmental monitoring agencies. Despite strong performance, limitations include dataset size, imbalance across severity classes, and dependency of classification accuracy on segmentation quality. Future work may incorporate data augmentation, advanced architectures such as U-Net++ or DeepLabv3+, temporal satellite imagery, or uncertainty quantification models for risk-aware operational deployment. Overall, our two-stage pipeline provides a robust, interpretable, and scalable framework for real-time oil spill detection and severity assessment in satellite imagery.

Abstract: Ensuring the structural integrity of high-energy piping systems is a critical requirement for the safe operation of nuclear power plants. This paper presents the design, imple-mentation, and three-year operational validation of a novel three-dimensional dis-placement monitoring system installed on the Steam Generator Blowdown pipeline of the Krško Nuclear Power Plant. The system was developed to confirm plant operating procedures will not cause excess dynamic displacements during operation. The measurement system configuration utilizes three non-collinear inductive dis-placement transducers (HBM WA/500 mm-L), mounted via miniature universal joints to a reference plate and to a defined observation point on the pipeline. The arrangement enables real-time monitoring of X, Y, and Z displacements within a spherical meas-urement volume of approximately 0.5 m. Data are continuously acquired by an HBM QuantumX MX840B amplifier and processed using CATMAN Easy-AP software through a fiber-optic communication link between the containment and control areas. The system has operated continuously for more than three years under elevated tem-perature and radiation conditions, confirming its reliability and robustness. The corre-lation of measured displacements with process parameters such as flow rate, pressure, and temperature provides valuable insight into transient events and contributes to predictive maintenance strategies. The presented methodology demonstrates a practical and radiation-tolerant approach for continuous structural monitoring of nuclear plant piping systems.

Abstract:

This work investigates the performance of a solar air heater (SAH) equipped with ten baffles whose angles can be adjusted in real time by a PLC. Many SAH systems operate passively, which makes their outlet temperature sensitive to daily variations in solar radiation. This study aims to show that an actively controlled SAH can maintain stable and efficient operation under practical outdoor conditions. Experiments were carried out at two set-point temperatures commonly used in drying applications, 54 °C and 60 °C, and the system was assessed through energy, exergy, and sustainability indicators. Greater baffle inclination increased turbulence and heat transfer, yielding thermal efficiencies up to 76.8%. The friction factor followed the Reynolds number closely, indicating that overall flow resistance depends mainly on the airflow rate. Exergy efficiency remained between 1.24% and 2.69%, while the Sustainability Index stayed near unity due to fan power related losses. A regression model was also developed to estimate the airflow needed to keep the outlet temperature at the desired level. Long-term projections show that the system can supply 20–22 MWh of heat and avoid nearly 9 tons of CO₂ emissions over 20 years. These findings highlight that combining PLC-based control with adjustable baffles offers a practical and environmentally meaningful improvement for solar air heating systems.

Abstract: Ammonia, as a carbon-free energy carrier, is gaining prominence for hydrogen storage and power generation applications due to its high energy density and ease of transport. However, the practical adoption of ammonia in combustion systems faces major stability challenges—chiefly its low reactivity, slow laminar burning velocity, narrow flammability envelope, and high ignition temperature. These attributes increase the risks of flame instability, misfire, and incomplete combustion, which, in turn, can elevate levels of unburned ammonia and greenhouse gas emissions such as NOx—posing significant health and climate concerns. Stable ammonia combustion demands optimization of several interrelated factors: the air–fuel equivalence ratio, flame temperature, flow regime, and combustor design are critical for maintaining reliable operation. Particularly pivotal is the control of the air–fuel equivalence ratio; excessively lean conditions can trigger flameout. Modern systems utilize real-time monitoring of flame and exhaust properties to diagnose and prevent instabilities. Advanced combustion strategies, such as transitioning to diffusion or flameless (MILD) regimes, substantially expand the stable operating window, especially under lean conditions. Overall, sustaining stable ammonia combustion is essential for maximizing efficiency and emission control, and integrating aftertreatment (deNOx) technologies is crucial for sustainable, clean-energy implementation.

Abstract: This study presents a techno-economic assessment of a modular solar-assisted me-thane pyrolysis pilot plant designed for sustainable hydrogen production in Nigeria using Concentrated Solar Power (CSP). Driven by the need to convert flare gas into value and reduce emissions, the work evaluates a hypothetical 100 kg/day hydrogen system by integrating a methane pyrolysis reactor with a solar heliostat–receiver field. Process modelling was carried out in DWSIM, while solar concentration behavior was represented using Tonatiuh. Mass and energy balance results show a hydrogen output of 3.95 kg/h accompanied by 12.30 kg/h of carbon black, with the reactor demanding roughly 44 kW of high-temperature heat at 900 °C. The total capital cost of the ≈50 kW pilot plant is approximately $1.5 million, with heliostat and receiver technologies forming the bulk of the investment. Annual operating costs are estimated at $69,580, along-side feedstock expenses of $43,566. Using annualized cost and discounted cash flow approaches, the resulting levelized cost of hydrogen (LCOH) is $5.87/kg, competitive with off-grid electrolysis in the region, though still above blue and gray hydrogen benchmarks. Financial indicators reveal a positive NPV, a 13% IRR, and a 13-year dis-counted payback period, highlighting the promise of solar-assisted methane pyrolysis as a transitional hydrogen pathway for Nigeria.

Abstract: This study introduces an integrated Life Cycle Assessment–Multi-Criteria Decision Analysis–Nash Equilibrium (LCA–MCDA–NE) framework to assess the feasibility of hydrogen energy storage (HES) in Philippine island grids. It starts with a cradle-to-gate LCA of hydrogen production across various electricity mix scenarios, from die-sel-dominated Small Power Utilities Group (SPUG) systems to high-renewable config-urations, quantifying greenhouse gas emissions. These impacts are normalized and in-tegrated into an MCDA framework that considers four stakeholder perspectives: Regu-latory (PRF), Developer (DF), Scientific (SF), and Local Social (LSF). Attribute utilities for Maintainability, Energy Efficiency, Geographic–Climatic Suitability, and Regulatory Compliance inform a 2×2 strategic game where net utility gain (Δ) and switching costs (C₁, C₂) influence adoption behavior. The findings indicate that the baseline Nash Equilibrium favors non-adoption due to limited utility gains and high switching barriers. However, enhancements in Main-tainability and reduced costs can shift this equilibrium toward adoption. The LCA results show meaningful decarbonization occurs only when low-carbon generation exceeds 60% of the electricity mix. This integrated framework highlights that successful HES de-ployment in remote grids relies on stakeholder coordination, reduced risks, and access to low-carbon electricity, offering a replicable model for emerging economies.

Abstract: To investigate the influence of different tip clearances on the hydraulic performance and cavitation characteristics of a cryogenic inducer, this study builds upon previous research by employing the SST k-ω turbulence model and modifying the empirical coefficients for evaporation and condensation in the Zwart cavitation model. Numerical simulations of the full flow field within an LNG cryogenic inducer were conducted. The results yielded cavitation performance curves, pressure distributions at incipient cavitation, vapor volume fraction contours, and leakage flow streamlines for various tip clearances. The impact of tip clearance on the overall hydraulic performance and cavitation behavior of the LNG inducer was systematically examined, with particular attention given to the microscopic evolution of the Tip Leakage Vortex (TLV) during the initial stages of cavitation. Experimental findings indicate that as the tip clearance increases, the tip leakage flow intensifies, leading to greater energy losses within the inducer and a consequent slight reduction in pump head and efficiency. A critical clearance value, δ, exists within the range of 0.4 mm to 0.6 mm, which governs the development pattern of the TLV. When the clearance is smaller than δ, the TLV forms more rapidly, and cavitation development is significantly more sensitive to increases in tip clearance. Conversely, when the clearance exceeds δ, the formation of the TLV is delayed, and cavitation progression becomes less responsive to further increases in tip clearance.

Abstract: This paper investigates the comprehensive energy profile and renewable energy solutions for a rural village comprising 30 houses. The study begins by analyzing the load demand distribution across day and night periods, with maximum daytime consumption recorded at 57.860 kWh and total daily usage of 509.040 kWh, while nighttime consumption peaks at 11.060 kWh with a total of 80.460 kWh. Detailed hourly consumption patterns are presented for various village components, including residential, water pumping, street lighting, medical facilities, and a supermarket, offering a granular view of energy use. To address the village’s energy requirements sustainably, a photovoltaic (PV) system with a capacity of 60 kW is proposed, supplemented by a solar thermal water heating system designed to meet hot water demands efficiently. The paper outlines the design and simulation of the solar water heating system, including calculations for water tube diameters, thermal resistance, and necessary tube length to transfer absorbed solar energy. MATLAB (V.22b) simulations further illustrate the performance of the integrated system, modeling energy production, battery charging/discharging cycles, and temperature fluctuations in water systems over a 24-hour period. Comparative analyses between standalone PV, PV/T hybrid, and combined PV plus solar thermal solutions reveal that the most cost-effective and maintenance-efficient strategy involves separate PV and thermal installations. The study highlights significant cost savings and environmental benefits over traditional diesel-based systems, positioning solar technologies as a reliable, sustainable, and economically viable solution for rural electrification and domestic hot water supply. Additional technical analysis confirms system sustainability and economic efficiency in real practice.

Abstract: This work examines the behaviour of a spark-ignition engine using oxy-fuel combustion, coupled with an oxygen production cycle based on a mixed ionic-electronic ceramic membrane. Through 1D-0D simulations, two compression ratios are studied: the original ratio of 9.6 and the optimised CR of 20, under various load levels and altitude conditions. The results show that operational limits exist at part-load conditions, where reducing the load without implementing additional control strategies may compromise system performance. It is observed that at low loads, the intake pressure can fall below atmospheric pressure, encouraging the presence of N2 in the combustion process. Additionally, the engine can operate efficiently up to an altitude of 4,000 m, although increasing boosting is required to maintain proper membrane conditions. These findings emphasise the importance of load control and the potential need for energy assistance under certain circumstances.

Abstract: A computational fluid dynamics analysis of the anode side of a proton exchange membrane (PEM) electrolyzer cell has been conducted. The geometry is symmetrical and allows for the investigation of a single feed channel and a single exhaust channel in an interdigitated flow field. The model utilizes the Eulerian approach and thus solves a full set of conservation equations for both gas and liquid phase. Moreover, it is non-isothermal and it includes phase change of water. The operating stoichiometric flow ratio results in segregated flow in the horizontal flow channels. At a current density of 1.0 A/cm², a local hot spot with a temperature increase of 7 °C is predicted. A reduction in the operating pressure below atmospheric pressure results in a more favorable concentration ratio of water vapor to oxygen at the PTL/CL interface, and the temperature distribution is more even. However, when the outlet pressure is too low, the outlet temperature is below the inlet temperature which makes this operation mode unfeasible. Adjustment of the back pressure can generally be used to control the temperature of the electrolyzer.

Abstract: Accurate binarization of phase-detection probe signals (gas vs. liquid) is necessary for the estimation of local void fraction, interfacial velocity, and bubble statistics in gas–liquid flows. Classical threshold based methods—single or double level— performs well on clean laboratory signals but degrades under realistic industrial conditions where noise, baseline drift, and clustered (slug-like) events challenge fixed rules. This work investigates whether deep learning (DL) models trained exclusively on synthetic data can deliver robust, generalizable binarization on real probe measurements. We (i) build a parametric generator of realistic time series from bubbly pulse templates, extended to clusters/slug patterns and perturbed with controlled noise, drift, and oscillatory baselines; (ii) train four lightweight DL architectures—one-dimensional U-Net (UNET-1D), Temporal Convolutional Network (TCN), a minimal one-dimensional Convolutional Neural Network (CNN-1D), and a Bidirectional Long-Short Memory network (BiLSTM)—only on synthetic signals; and (iii) evaluate them against classical threshold methods using event-level and sample-level metrics. On synthetic signal evaluation, UNET-1D and TCN achieve near-perfect event detection and sub-millisecond onset errors. On real bubbly and slug flow sensor data, classical threshold based methods remain highly competitive on clean sensor signals, while DL models retain advantages under non-stationary baselines and clustered events, yielding accurate void and timing with no hand-tuned assumptions. Results support DL as a practical, data-driven complement to fixed algorithms, particularly in noisy or drift-dominated measuring conditions.