China’s natural gas (NG) consumption is expected to grow in the short and medium term under the country’s carbon neutrality policy, but this growth is accompanied by increasing uncertainty driven by geopolitical developments and ongoing market liberalisation. This highlights the need for accurate demand forecasting. This study proposes a hybrid forecasting framework that integrates the Seasonal Autoregressive Integrated Moving Average with Exogenous Regressors (SARIMAX) model and the Long short‐term memory (LSTM) neural network through an artificial neural network (ANN). The analysis focuses on 1‐year‐ahead forecasting of China’s NG consumption at the national level using monthly data from April 2009 to July 2023. Eight key drivers reflecting market, economic, and infrastructure conditions are incorporated as model inputs. Forecasting performance is evaluated using out‐of‐sample R‐squared ( R 2 ), root mean square error (RMSE) and mean absolute error (MAE). The results demonstrate that the ANN‐optimised SARIMAX–LSTM hybrid model with a trend‐capturing function achieves the highest predictive accuracy, with an R 2 of 47.94%, RMSE of 5.23 and MAE of 1.70. This superior performance is further supported by the model’s lowest cumulative squared forecast error (CSFE), confirming its robustness and consistency in predictive performance. In addition, principal component analysis (PCA) confirms the validity of the selected input drivers and identifies NG‐related infrastructure variables as the most influential factors in near‐term forecasting. These findings provide robust evidence for the effectiveness of hybrid modelling in short‐term energy demand forecasting and offer practical implications for policy design, investment planning, and operational decision‐making in the context of China’s carbon‐neutral energy transition.
The optimization of multiobjective trade‐offs between thermal efficiency and pollutant emissions is recognized as a critical challenge in piloted premixed impinging combustion systems. In this study, this limitation is addressed by developing an integrated framework that combines thermodynamic exergy analysis, Kriging‐based surrogate modeling, and a genetic algorithm (GA). A verified computational fluid dynamics (CFD) database was established to evaluate four key performance targets of a ring‐piloted premixed impinging jet flame: thermal efficiency ( η T ), exergy efficiency ( η E ), nitrogen oxide emissions (EINO x ), and carbon monoxide emissions (EICO). The integration of exergy analysis enables precise, quantifiable measures of thermodynamic irreversibilities across the flame and impingement regions. High predictive accuracy was demonstrated by the optimized Kriging surrogate model, as evidenced by low external‐checkpoint average relative errors of 1.47% for η T , 0.06% for η E , 1.66% for EINO x , and 3.29% for EICO. Through multiobjective optimization via GA, the optimal operating envelope was identified, occurring at a lean central equivalence ratio ( φ C = 0.798), a rich coaxial pilot equivalence ratio ( φ P = 1.2), and a low coaxial fuel flow rate ( Q PF = 0.4 L/min). This rich‐lean flame structure retains 64.14% thermal efficiency and 70.41% exergy efficiency, while EINO x and EICO are suppressed to 4.328 × 10 −4 and 1.110 × 10 −4 , respectively. High thermodynamic performance is maintained, and efficiency penalties are restricted below 3.13%. Thus, a robust and computationally efficient methodology for the design of next‐generation low‐emission industrial burners is provided by this framework.
The decarbonization of the building sector is one of the central challenges to meet climate goals and mitigate climate change. For the retrofit of existing buildings, well‐informed retrofit decisions are crucial, and several modeling approaches exist to support these decisions. In this paper, we reviewed more than 200 approaches that support this decision‐making and the energy design of existing buildings. Our analysis covers both individual buildings and building stock perspectives, providing a comprehensive overview of the strengths and limitations of existing modeling approaches. Approaches can generally be categorized into scenario‐based, meta‐heuristic, and deterministic models. At the individual building level, mixed‐integer linear programming (MILP) and combinations of simulations and genetic algorithms (GA) are widely used. At the building stock level, scenario‐based approaches and metaheuristic approaches are the most common. Most of the analyzed approaches consider both the building envelope and the energy supply system (ESS), which has also been recommended in many works. While individual building works focus on applications of their methods in specific buildings, building stock approaches mostly use typical buildings. However, important challenges remain in modeling the decision‐making of modernizing building energy systems. Especially, improved representations of long‐term investment decisions, uncertainty in environmental developments, and limited generalization of results due to a strong focus on specific energy systems and building types. A common limitation of all methods is the limited integration of resource limitations and uncertainty assessments. Future research should therefore focus on integrating dynamic, long‐term planning approaches with uncertainty modeling. Hybrid approaches combining heuristics, machine learning (ML), and optimization can improve computational efficiency and scalability, especially for building stock modeling. In addition, incorporating resource allocation into models can improve their practical relevance. Finally, improving model validation through empirical case studies and benchmarking with real‐world data will be critical to ensure the reliability and practical implementation of all of the methods mentioned.
This article addresses the under‐representation of social aspects in the study of lithium batteries. Although lithium batteries are critical to global energy systems and to ongoing sustainability transitions, the literature has framed them predominantly as electrochemical devices, leaving their societal, economic and political dimensions in a peripheral position. From a science and technology studies (STS) perspective, this study asks how scientific literature constructs and represents the social dimensions of lithium battery technologies. Using a quantitative approach that combines bibliometric descriptive statistics with semantic network analysis on 5337 records retrieved from Scopus, we examine the disciplinary, geographical and conceptual patterns through which this social construction takes place. Three findings stand out. First, the natural and physical sciences, together with engineering and technology, dominate the production of knowledge that explicitly incorporates social terms. Second, although social vocabulary is present, it occupies peripheral positions in the semantic core, which is structured around materiality and electrochemistry. Third, and most strikingly, the major lithium‐producing countries of the Global South (Chile, Argentina and Bolivia) jointly contribute less than 3% of China’s output, evidencing a sharp asymmetry between extraction territories and centres of knowledge production. Building on these findings, we propose an ontological shift whereby lithium batteries should be understood not only as electrochemical devices but also as socio‐technological artefacts. This reframing motivates new research pathways comprising five lines of enquiry for the future social study of lithium batteries: practices, imaginaries, conflicts, policies and geopolitics.
In this paper, a methodology dedicated to improving the design of perovskite solar cells (PSCs), particularly those utilizing K 2 InSbBr 6 as the absorbing material, is presented. The methodology combines artificial intelligence’s potential and modern physics‐driven simulators’ capabilities. A detailed dataset created using the SCAPS‐1D simulator investigated the effect of various factors, such as the choice of electron transport‐layer (ETL) and hole transport‐layer (HTL) materials, absorber thickness, and material defects, on the solar cells’ efficiency. Based on this dataset, an artificial neural network (ANN) model was trained. This dataset was then exploited to predict the power conversion efficiency (PCE) of various solar cell configurations by training an ANN. The ANN model’s predictive results strongly correlate with the simulation results (Pearson coefficient = 0.801). Considering this ANN model’s interesting prediction capabilities, a solar cell is designed and optimized using such a model. This cell is characterized by a WS 2 (1 µm) as the ETL, K 2 InSbBr 6 (700 nm) as the absorber, and NiO (100 nm) as the HTL, and its simulation PCE is about 32.11%.
Solar water heaters are widely used to convert solar energy into thermal energy for residential, commercial, and industrial applications. This study investigates the thermal performance of a double‐pass solar water heater (DPSWH) using numerical and experimental approaches under different water mass flow rates. Three configurations were evaluated: a flat absorber plate, a V‐corrugated absorber plate, and a V‐corrugated absorber plate integrated with paraffin wax‐filled aluminum tubes for thermal energy storage (TES). Numerical simulations were conducted using ANSYS Fluent and validated through experimental testing under climatic conditions of Multan, Pakistan at mass flow rates of 0.003, 0.005, and 0.007 kg/s. Results showed that DPSWH with TES achieved the highest thermal performance. A maximum inlet–outlet water temperature difference of 39°C was obtained at 0.003 kg/s while a minimum difference of 35°C occurred at 0.007 kg/s. Solar fraction values ranged from 0.01 to 0.20. Furthermore, incorporation of paraffin wax enhanced thermal efficiency and prolonged heat delivery compared to systems without thermal storage. The findings demonstrate that lower mass flow rates improve utilization of stored thermal energy, resulting in enhanced system performance and extended hot‐water availability.
Phase change microcapsules (MCs) demonstrate significant application potential in thermal energy storage. However, conventional single‐component shell materials exhibit limitations, including low mechanical strength, poor thermal conductivity, and inadequate sealing performance. In this study, octadecane was used as the core material, and four kinds of MC systems were constructed by interfacial polymerization and in situ polymerization: melamine–urea–formaldehyde (MUF) single shell, polyurea (PUA) single shell with nano‐SiO 2 as Pickering emulsifier, MUF inner shell/PUA outer shell double shell, and PUA inner shell/MUF outer shell double shell. The packaging efficiency, thermal conductivity, leakage rate and heat‐transfer characteristics of different structures are systematically compared. The results show that the comprehensive performance of the PUA inner shell/MUF outer shell double‐shell structure is the best, the packaging efficiency is 76.2%, the thermal conductivity is 9.35% higher than that of pure PUA single shell, and the leakage rate is as low as 3.35% at 50°C for 20 h. The transient heat conduction simulation of ANSYS Fluent reveals the performance improvement mechanism: the heat conduction network formed by self‐assembly of nano‐SiO 2 at the oil–water interface accelerates the heat transfer, and the double‐shell synergistic effect of “flexible inner shell attached to the core material and rigid outer shell provided protection” significantly inhibits the core material leakage. The collaborative optimization strategy of shell structure design and nano‐Pickering emulsification proposed in this study provides a new technical path for realizing high sealing, high strength, and high thermal conductivity of phase change MCs at the same time and has application value in the fields of building energy saving and thermal management of electronic equipment.
Accurate short‐term predictions of building heating, ventilation, and air‐conditioning (HVAC) loads are central to model‐predictive control (MPC), demand response, and grid‐interactive efficient buildings (GEBs). Black‐box machine‐learning models routinely fail to generalize to unseen climates and out‐of‐distribution operating seasons, whereas pure white‐box reduced‐order resistance–capacitance (RC) models extrapolate well but cannot capture nonlinear HVAC behavior. We propose a physics‐informed residual network (PIRN) that adds a small multilayer perceptron (MLP) residual on top of a three‐parameter learnable RC envelope model—initialized at its exact nonnegative least‐squares (NNLS) optimum—and is trained end‐to‐end with a single mean‐squared‐error loss. We construct a benchmark spanning four generalization axes (in‐distribution accuracy, cross‐climate zero‐shot transfer, few‐shot fine‐tuning, and shoulder‐to‐extreme seasonal extrapolation), two capacity‐ and feature‐controlled ablations, and a sensor‐noise robustness probe from 8760 h annual EnergyPlus simulations of the U.S. Department of Energy small‐office prototype in five ASHRAE climate zones (2A, 3C, 4A, 5A, and 5B), with 10 random seeds for every primary experiment and Holm‐corrected paired statistics throughout. Three findings emerge. (i) At a matched parameter budget, PIRN and a plain MLP are practically interchangeable in‐distribution, in zero‐shot transfer, and in seasonal extrapolation (median |Δ R 2 | ≤ 0.02; every difference is either statistically indistinguishable or equivalent within a ± 0.03 R 2 margin by two one‐sided tests [TOSTs]); the often‐claimed physics‐informed‐versus‐black‐box gap disappears once the MLP receives the same physically meaningful features (under conservative few‐shot fine‐tuning, the MLP adapts modestly better, an effect our mitigation study traces to the fine‐tuning protocol). (ii) A capacity ablation shows the practical payoff of compactness: a 140‐parameter PIRN‐xs reproduces the in‐distribution accuracy ( R 2 ≈ 0.91) and the zero‐shot transfer accuracy ( R 2 ≈ 0.79) of its 9412‐parameter parent—a 67× reduction at equal accuracy, running in under 0.1 ms on microcontroller‐class hardware; a size‐matched small‐MLP control achieves the same, showing that this compactness reflects the low complexity of the well‐featurized task rather than the physics prior itself. (iii) The physics prior becomes decisive when sensing is sparse: with only three thermostat‐observable inputs, the MLP collapses on the colder climates while PIRN holds steady (median Δ R 2 = + 0.24, Holm‐adjusted p = 5 × 10 −7 ), an advantage a strict‐physics control traces to the physics channel’s use of cheaply available irradiance and occupancy signals; the same term also suppresses winter‐extrapolation failures, and fitted alone, it recovers stable, climate‐consistent envelope coefficients of the expected order for a single‐node model (UA within a factor 1.3–2.6 of the design envelope value). The combination—equal at scale, compact, and dominant under sparse sensing—makes PIRN a sensible default for embedded MPC controllers and retrofit settings with minimal instrumentation. We release the simulation pipeline, processed datasets, trained checkpoints, and benchmark code.
Hydrogen (H 2 ) has attracted considerable attention as a promising energy carrier due to its high energy content compared to conventional fossil fuels. The continuous growth of the global population is driving an increasing demand for energy, leading to higher consumption of fossil fuels and, consequently, increased greenhouse gas emissions and climate change. This has driven the research community toward renewable energy sources (RESs) such as solar, wind, and tidal energy. Among RES options, hydrogen stands out as one of the most viable alternatives. The significant research has focused on hydrogen production and its applications, comparatively less attention has been given to hydrogen storage. In this context, the hydrogen storage characteristics of alkali metal (AM) and alkaline earth metal (AEM) decorated and defective hexagonal boron nitride (h‐BN) nanosheets were systematically investigated using density functional theory (DFT) calculations. The AM atom (Li) and AEM atom (Be) are found to be uniformly dispersed on h‐BN surfaces containing boron vacancies and divacancy defects, effectively preventing atom clustering. Up to 19 H 2 molecules can be adsorbed on the surface of Be‐decorated Li‐doped h‐BN, exhibiting an average adsorption energy ranging from −0.23 to −0.42 eV per H 2 molecule. Electronic structure analysis reveals that the introduction of Li and Be atoms significantly alters the electronic properties of pristine h‐BN, as indicated by the modification of bands near the Fermi level, thereby enhancing conductivity and inducing semiconducting behavior. Furthermore, Be decoration Li doping collectively improves the hydrogen storage capacity (HSC), achieving a theoretical gravimetric density of up to 11.22 wt.%. The thermal stability of the H 2 ‐adsorbed system was further evaluated using ab initio molecular dynamics (AIMD) simulations and the van’t Hoff equation, confirming its robustness under thermal conditions.
Accurate productivity forecasting plays a vital role in ensuring the efficient and economical exploitation of gas hydrate reservoirs. This work proposes a modified inflow performance relationship (IPR) model tailored for the rapid prediction of Class I gas hydrate production behavior under vertical well and depressurization development strategies. The formulation of the revised IPR model is grounded in insights obtained from numerical simulations of production dynamics. Owing to its alignment with the typical IPR curve patterns and distinct production stages, the Fetkovich equation was selected as the core framework, with the late‐time slow decline stage omitted due to its marginal contribution to overall output. To enhance prediction accuracy, the conventional use of reservoir pressure was replaced by an effective average reservoir pressure, characterized through an energy coefficient denoted as B i . Sensitivity analyses revealed that this coefficient is primarily influenced by the initial formation pressure and temperature. By leveraging the phase equilibrium relationship between these two parameters, a fitted expression linking B i to temperature was derived based on orthogonal experimental results. Comparative evaluation against numerical simulation outputs indicates that the revised IPR model yields a mean relative error of 4.20%, significantly outperforming the original model’s 18.25% error. With its computational efficiency and practical applicability, the improved IPR formula provides a robust analytical alternative for productivity estimation in Class I gas hydrate systems developed via depressurization and vertical wells.
In this study, pore‐filling anion‐exchange membranes (PFAEMs) incorporating (vinylbenzyl)trimethylammonium chloride (VBTA) and 1,3,5‐triacryloylhexahydro‐1,3,5‐triazine (TATA) were investigated to enhance the initial durability and performance of AEM water electrolysis (AEMWE). The optimized VBTA‐TATA‐20 membrane exhibited superior OH − conductivity (99.5 mS·cm −1 at 60°C) and enhanced mechanical and dimensional stability compared to the commercial PiperION membrane. Building upon these intrinsic properties, AEMWE performance was systematically optimized under varying temperatures, flow rates, and flow configurations. While the transition from symmetric to asymmetric flow (dry cathode) resulted in a slight performance trade‐off, VBTA‐TATA‐20 demonstrated remarkable adaptability to practical operating conditions. Notably, although the ex situ areal specific resistance (ASR) of VBTA‐TATA‐20 was 129.7% of PiperION, the in situ ohmic resistance ( R ohmic ) gap derived from I-V narrowed to 108.9% under dry cathode conditions. This discrepancy suggests that the VBTA‐TATA‐20 membrane may facilitate more effective water retention and ionic transport under asymmetric dry cathode operation, thereby potentially mitigating membrane drying induced by Joule heating. In 100 h initial durability tests under asymmetric flow, VBTA‐TATA‐20 maintained a stable voltage plateau without the irreversible degradation observed in PiperION, despite exhibiting reversible voltage fluctuations attributed to dynamic gas transport in the cathode. These findings suggest that the triazine‐based cross‐linked architecture may contribute to enhanced mechanical robustness and improved water management, supporting stable operation under practical asymmetric flow conditions.
Accurate voltage prediction of proton exchange membrane fuel cells (PEMFCs) under variable operating conditions remains challenging because of nonlinear interactions among current density, temperature, pressure, humidity, and mechanical assembly conditions. This study develops a regime‐aware machine–learning framework for PEMFC voltage prediction using 9930 time‐resolved experimental observations collected from 11 distinct operating regimes. Random forest (RF), least‐squares boosting (LSBoost), Gaussian process regression (GPR), artificial neural networks (ANNs), and support vector regression (SVR) were evaluated using grouped five‐fold cross‐validation, in which all observations from the same temperature–humidity–pressure–torque combination were kept within a single fold. This strategy enabled assessment of model generalization to previously unseen operating regimes. RF showed the most balanced performance, with a mean R 2 of 0.9215 ± 0.0784, root mean squared error (RMSE) of 0.0421 ± 0.0236 V, and mean absolute error (MAE) of 0.0287 ± 0.0156 V. LSBoost achieved comparable average performance, whereas GPR produced highly accurate predictions in selected folds but showed greater fold‐to‐fold variability. ANN and SVR exhibited limited generalization under strict regime–level validation. A raw‐versus‐engineered‐feature ablation analysis demonstrated that current‐density transformations and selected interaction terms improved RF predictive performance. Complementary intrinsic and permutation importance analyses identified current‐density–related variables and inlet pressure as influential predictors. The results demonstrate the importance of grouped validation for realistic assessment of PEMFC voltage–prediction models and provide a basis for future optimization, monitoring, and control‐oriented applications.
This study presents a novel framework for enhancing the efficiency and reliability of distributed energy resources (DERs) in unbalanced low‐voltage (LV) distribution networks, with a particular focus on integrating battery energy storage systems (BESS) under photovoltaic (PV) generation uncertainties. In such networks, node voltages are highly sensitive to the collective behavior of DERs. To address this challenge, a Monte Carlo (MC)‐based predictive BESS control strategy is developed to optimize the tradeoff between BESS state‐of‐charge (SOC) tracking optimal references, control effort, and utility power exchange, while ensuring compliance with BESS operational limits, feeder voltage constraints, and demand requirements. The framework is validated on an unbalanced European LV test network under two operational scenarios. Simulation results demonstrate that the proposed strategy significantly improves voltage profiles across all buses and reduces utility power exchange under varying PV uncertainties. Benchmarked against a classical BESS model predictive control (MPC), the proposed controller demonstrates superior performance by robustly satisfying operational limits even under worst‐case uncertainty scenarios. Overall, the framework provides a robust and effective solution for the reliable operation of DERs in distribution networks.
Rapid electricity demand growth and high natural gas (NG) dependence challenge the integration of firm low‐carbon energy systems in emerging economies. This study introduces a hybrid uranium–thorium production index (UTPI) to quantify nuclear energy’s system‐level contribution relative to total final energy demand (TFED), using Türkiye as a representative gas‐dependent economy. Using 120 monthly observations (2016–2025), UTPI exhibits strong and statistically significant negative correlations with TFED ( r = −0.977), NG consumption (NGC; r = −0.911), and net gas imports ( r = −0.868), indicating a robust structural substitution effect. Explainable machine learning (ML) models capture complex nonlinear interactions between nuclear relevance and energy system variables, achieving high predictive performance (test R 2 ≈ 0.97; RMSE ≈ 0.20), while SHapley Additive exPlanation (SHAP) analysis identifies gas‐related variables as the dominant drivers. Scenario projections for 2026–2035 reveal differentiated transition pathways. Under a baseline with committed nuclear capacity, nuclear contribution remains limited (≈3.2% of TFED in 2035). A uranium‐based expansion increases this share to ≈10.2%, whereas an exploratory thorium‐oriented pathway, representing a long‐term technological potential rather than a commercially deployable near‐term option, enables a structural shift exceeding 21.2% of TFED. Corresponding fossil fuel import reductions reach 2.8–18.6 bcm of NG, generating annual economic savings between USD 3.5 and 22.9 billion. Overall, UTPI provides a demand‐normalized system‐level metric that links nuclear fuel‐cycle potential with broader energy‐system dynamics. The results should be interpreted in light of uncertainties associated with resource estimates, technology readiness, and future deployment conditions, particularly for advanced thorium fuel cycles. Nevertheless, the findings demonstrate that nuclear energy primarily substitutes gas‐fired baseload generation and that combined uranium–thorium transition strategies may strengthen long‐term energy security and decarbonization pathways in gas‐dependent systems.
Microgrids operating in islanded mode face significant frequency and voltage instability, particularly when dominated by inverter‐based renewable energy sources (RESs) with inherently low system inertia. This study presents a layered hierarchical control strategy that integrates three novel components: (1) an adaptive virtual synchronous generator (VSG) with coupled frequency–voltage co‐regulation, wherein both the inertia constant H ( t ) and the Q-V droop coefficient m Q ( t ) are dynamically tuned based on real‐time frequency deviation and battery state‐of‐charge (SoC); (2) a delay‐compensated consensus protocol employing second‐order Padé approximants to maintain distributed secondary control stability under communication delays up to 500 ms; and (3) a tri‐objective hybrid model predictive control (MPC)–genetic algorithm (GA) optimization framework for battery energy storage system (BESS) coordination, simultaneously minimizing frequency deviation, voltage deviation, and SoC degradation. Simulation results for a 500 kW microgrid with 70% renewable energy penetration demonstrate that the proposed framework significantly outperforms traditional fixed‐inertia VSG and droop control methods. Key outcomes include: frequency deviations limited to 0.05 Hz (vs. 0.38 Hz for fixed‐inertia VSG), voltage regulation maintained within ± 1.2% (vs. ± 4.2%), system stabilization time reduced by 30%, and communication load reduced by 50%. The BESS SoC remained within the safe 20%–80% operating range across all tested disturbance scenarios.
Small and medium‐sized enterprises (SMEs) account for a considerable share of industrial electricity consumption, yet continuous energy auditing and electrical load assessment remain limited by the cost and complexity of conventional energy management systems. This article presents an edge‐centric industrial energy auditing framework for three‐phase electrical loads that integrates synchronized measurement, automated auditing, operational intelligence, and real‐time fault awareness within a unified architecture. The framework constantly collects voltage, current, active power, energy usage, frequency, and power factor monitoring to evaluate the load behavior for balance of phases and use of power as well as abnormal conditions during operation. The validated measurements are automatically converted into time‐aligned audit logs, operational analytics for energy performance assessment, fault diagnostics, maintenance planning, and regulatory compliance. Conventional IoT monitoring systems primarily focus on visualization; however, the proposed framework integrates the functionality of measurement validation and assessment of energy, detect fault, and notify operator into a distributed edge‐computing context. Moreover, the system can detect voltage variations, load issues, and phase balancing under real‐time monitoring so as to mitigate energy losses, improve efficiency of electrical systems, and safeguard equipment from deterioration. Experimental validation was conducted through a 7‐day continuous deployment using representative three‐phase industrial loads under steady‐state and transient operating conditions. Results demonstrated reliable energy auditing, effective abnormal‐condition detection, and stable operation with low computational overhead (5%–8% CPU utilization and <150 MB RAM). The proposed system offers a scalable and cost‐effective solution for data‐driven energy assessments and operational reliability in SME industrial facilities.
Sustainable energy technologies play major role in mitigating global warming and climate change, yet its implementation ratio remains below the levels required to realize net‐zero emissions. The effectiveness and diffusion of sustainable energy technologies are inextricably linked to the design and implementation of critical success factors (CSFs). Existing studies often focus on context‐specific factors, limiting the exploration of broader aspects. This study dives deep into exploring a comprehensive set of critical factors, rank them, and evaluate their interdependencies in the Pakistani energy sector. The study employed interpretive structural modeling (ISM) matrix multiplication applied to classification (MICMAC) approach to validate the factors by gathering opinions from expert’s panel. The analysis identified 10 CSFs which were further clubbed under autonomous, dependent, independent, and linkage categories. The findings from the study highlight the “access to sufficient funding”, “support from authorities”, and “perceived benefits” as the top three ranked critical factors requiring immediate attention. Pakistan energy sector can accelerate the path towards sustainable energy technologies by integrating financial support and aligning policies with perceived benefits of sustainable energy technologies (SETs).
Biodiesel is an environmentally friendly and sustainable alternative fuel with several advantages over conventional diesel. This study explores the enhancement of safflower biodiesel (B20) with the addition of CeO 2 NPs (60 mg) in a modified piston‐top CI engine, operating at different CRs (CR19, CR18, CR17, and CR16). The modified piston‐top plays a major role by lowering the ignition duration, thereby increasing the HRR. The effect of antioxidant CeO 2 NPs as an additive in SAF B20 was analyzed for its catalytic action and ability to reduce NOx emission and its reduction of in‐cylinder temperatures, which persisted throughout the full combustion cycle. The results emphasize that the addition of CeO 2 led to BTE improvements by 6.67%, while also lowering BSFC by 17.64% compared to B20 due to its good thermal stability, advancing toward more complete combustion. The catalytic action of CeO 2 NPs facilitates the formation of oxygen‐containing radicals, such as O and OH, and accelerates chain reactions during combustion. Alongside, there were significant reductions in toxic emissions such as carbon monoxide (CO) (19.14%) and HC (24.01%) compared to B20, mainly due to the greater specific surface area of 25.8 m 2 /g, whereas NOx emissions were slightly higher by 7.54%. These findings suggest that safflower B20 with CeO 2 NPs offers promising potential for improved engine characteristics and reduced the environmental impact, placing it as a competitive green fuel in the near future for diesel engine applications.
The high cost of raw materials remains a critical barrier to the large‐scale deployment of hydrogen‐storage alloys. Feedstock blending ratios were developed using Excel Solver and a Python codebase to meet a specified alloy chemistry given the target composition and the prices and chemistries of low‐purity feedstocks. To reduce the Si content in the raw‐material composition obtained from the Solver to below 0.0128 wt %, a revised composition ratio was determined using a general‐purpose Python‐based heuristic procedure to balance the price–composition trade‐off. A 100 g alloy was produced by vacuum arc remelting and a 120 kg alloy was produced by vacuum induction melting, both according to the Solver‐derived recipe. In addition, a 25 g reagent‐grade vacuum arc melt was prepared as a control based on the target composition. The effects of industrial impurities such as Si and Hf on the H 2 sorption performance of hydrogen storage alloy were also analyzed. The results provide an experimental demonstration of recipe‐level cost‐performance trade‐offs, and offer economically viable feedstock‐blending strategies for hydrogen‐storage alloys while minimizing performance sacrifice.
The rapid electrification of maritime transport poses significant challenges for coastal electricity distribution networks due to the high‐power, intermittent charging demands of electric ferries (EFs). This research proposes a coordinated battery energy storage system (BESS) control framework based on a hybrid genetic algorithm–particle swarm optimization–bacterial foraging optimization (GA–PSO–BFO) algorithm to mitigate the adverse impacts of EF charging on a real‐world coastal distribution network located near Gladstone Marina, Queensland, Australia. A comprehensive quasi‐dynamic co‐simulation framework is developed by integrating DIgSILENT PowerFactory, MATLAB/Simulink‐based Corvus Orca ESS models and a Python‐based hybrid optimization controller to determine optimal BESS charge–discharge schedules under varying network operating conditions. The proposed framework evaluates coordinated and uncoordinated BESS operating strategies considering transformer loading, feeder loading, line current, bus voltage, power losses, power factor, and harmonic performance across multiple loading scenarios. Simulation results demonstrate that coordinated BESS operation reduces transformer loading by up to 4%, line loading by 3.5%, and voltage deviation by 1.5% compared with uncoordinated operation while simultaneously improving voltage regulation and reducing network losses. The proposed hybrid GA–PSO–BFO algorithm achieved the lowest average objective function value of 49.01, outperforming the individual GA, PSO and BFO algorithms in terms of solution quality, convergence characteristics, and optimization stability. Furthermore, harmonic analysis confirms that coordinated BESS operation maintains total harmonic distortion (THD) within the limits specified by IEEE 519 and IEC 61000 standards, whereas uncoordinated operation produces significantly higher harmonic distortion under heavily loaded conditions. These findings demonstrate the effectiveness of coordinated BESS scheduling for enhancing distribution network resilience and provide a practical framework for the reliable integration of high‐power EF charging infrastructure into coastal electricity networks.
