Sustainable Energy Technologies and Assessments最新文献

{"title":"Driving sustainability through data: a systematic review of indicators shaping the European Union digital battery passport","authors":"Sara M. Pinto,&nbsp;Luís Oliveira,&nbsp;Emanuel J. Lourenço,&nbsp;Belmira Neto","doi":"10.1016/j.seta.2026.105449","DOIUrl":"10.1016/j.seta.2026.105449","url":null,"abstract":"<div><div>The European Battery Regulation (EU) 2023/1542 establishes the Digital Battery Passport (DBP) as a mandatory instrument for transparency, traceability, and regulatory compliance for lithium-ion batteries (LiBs). While the DBP marks a significant regulatory milestone, its current emphasis on carbon footprint reporting raises questions about whether it fully reflects the broader spectrum of environmental, economic, and social sustainability drivers documented in the scientific literature. This study addresses that gap through a systematic review of 48 life cycle sustainability studies on LiBs published between 2015 and 2025, comprising 31 Life Cycle Assessment, 9 Life Cycle Costing, and 8 Social Life Cycle Assessment studies, complemented by 5 Life Cycle Sustainability Assessment studies underpinning the cross-dimensional analysis. Recurrent environmental hotspots, cost drivers, and social risks were synthesised and mapped against current DBP requirements. Across environmental evidence, cathode material production and energy-intensive manufacturing consistently emerge as the dominant hotspots, while use phase impacts are governed by grid carbon intensity. Economic evidence identifies material cost volatility, manufacturing capital intensity, and total cost of ownership as recurrent drivers, none of which are addressed by the current regulation. Social risks concentrate in upstream cobalt, graphite, and nickel supply chains and are inadequately captured by country-level compliance reporting. Based on this evidence, a framework of eighteen DBP-ready indicators is proposed, distinguishing those already required by regulation, those partially addressed, and those currently absent. The findings carry direct implications across three fronts. For policy, they indicate which life-cycle indicators could strengthen future phases of the passport. For the industry, they clarify which sustainability data are most decision-relevant and where disclosure is most challenging. For research, they point to the need to consolidate economic and social life-cycle methods, so they match the maturity of environmental assessment.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"94 ","pages":"Article 105449"},"PeriodicalIF":7.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149161529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From forecast to dispatch: A physics-guided dual-stream diffusion framework for long-horizon probabilistic wind speed forecasting","authors":"Chang Chen,&nbsp;Jie Du,&nbsp;Youmin Mai,&nbsp;Wenbo Yu,&nbsp;Linlin Pan","doi":"10.1016/j.seta.2026.105447","DOIUrl":"10.1016/j.seta.2026.105447","url":null,"abstract":"<div><div>As wind power integration expands, grid scheduling urgently requires precise forecasting and fluctuation-risk quantification. However, deterministic models lack probabilistic boundaries, and generative models may diverge in long-sequence forecasting. Furthermore, existing models often treat interactions as instantaneous, ignoring propagation-induced time lags. To this end, we propose Phy-Diff, a robust physics-guided wind-speed forecasting system. First, to address these lags, we construct a dynamic directed graph using wind direction, spatial distance, and an estimated propagation delay to represent lagged dependencies between nodes. Second, to untangle coupled meteorological effects, a physics-guided dual-stream encoder separates large-scale advection-related information from local meteorological corrections. Third, to resolve forecasting instability, we design a gradient-anchored hybrid decoding strategy. A deterministic branch constrains low-frequency trends, guiding the diffusion branch to generate high-frequency details. Combined with explicit conditional injection, the model achieves non-autoregressive long-sequence generation. Experiments on 109 North China wind farms demonstrate that Phy-Diff outperforms 15 baselines in point and probabilistic accuracy. Its 72-hour Continuous Ranked Probability Score is 7.2% lower than that of the strongest baseline. Under the stated illustrative assumptions, its converted nominal 95% power interval is 5.38 percentage points narrower than that of the strongest probabilistic baseline.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"94 ","pages":"Article 105447"},"PeriodicalIF":7.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149161377","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing the biochemical methane potential of cattle manure through temperature–pressure co-variation during hydrothermal pretreatment: organic matter degradation, inhibitory substances generation, and energetic evaluations","authors":"Tianqi Kong,&nbsp;Aimin Li,&nbsp;SiQi Jiang","doi":"10.1016/j.seta.2026.105458","DOIUrl":"10.1016/j.seta.2026.105458","url":null,"abstract":"<div><div>This study characterized the properties and biochemical methane potential of cattle manure (CM) and its hydrothermal products (HTP) under hydrothermal treatment (HT) at 100–200 °C with 0–5  MPa N₂. HT elevated moisture, soluble chemical oxygen demand, total volatile fatty acids, and ammonia-N, while decreasing ash, total solids, volatile solids, and pH. Under non-pressurized HT, only hemicellulose degraded, and furan derivatives first increased then decreased with temperature. By contrast, pressurized-HT further degraded both hemicellulose and cellulose at identical temperatures. At 3 MPa, furan derivatives increased from 0.19 g/L at 100 °C to a maximum of 0.67 g/L at 150 °C, followed by a decrease to 0.12 g/L at 200 °C; at 150 °C, their concentration rose with pressure. The cumulative methane yield of HTP improved by 115–235%. Principal component analysis indicated that furan derivatives and pressure were negatively associated with methane production. Modified Haldane modeling showed furan derivatives, rather than ammonia-N, dominated the non-linear stimulation-inhibition behavior. Most HT-AD combinations achieved positive net energy gains, with maximum methane yield and energy gain at HT-CM-200 °C and 3 MPa ₂–100 °C, respectively. Pressurized-HT enhanced degradation efficiency and methane production at lower temperatures, reducing energy input and improving the economic feasibility of anaerobic digestion.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"94 ","pages":"Article 105458"},"PeriodicalIF":7.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149161467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Microwave-Assisted extraction of lipids from Chlorella sp. using green solvents for sustainable biodiesel production","authors":"Sapto Purnomo Putro,&nbsp;Fitra Ari Aditya,&nbsp;Rizki Sandhi Titisari,&nbsp;Tabina Shannaesha,&nbsp;Mohd Razif Harun,&nbsp;Natrah Fatin","doi":"10.1016/j.seta.2026.105364","DOIUrl":"10.1016/j.seta.2026.105364","url":null,"abstract":"<div><div>Microalgal lipids are a promising renewable feedstock for sustainable biodiesel production, but conventional extraction methods are limited by long processing times, high energy use, and fossil-based solvents. This study evaluated microwave-assisted extraction (MAE) with four green solvents (2-MeTHF, ethanol, D-limonene, and a deep eutectic solvent) for lipid recovery from <em>Chlorella</em> sp. across 50–80 °C and 5–30 min. The highest yield, 28.7 ± 0.9% DW using 2-MeTHF at 65 °C/15 min, exceeded the Bligh &amp; Dyer method (26.5 ± 0.8%) while cutting extraction time by over 95%. Kinetic modeling confirmed pseudo-first-order and Fickian diffusion behavior (R² &gt; 0.96), and Response Surface Methodology identified an optimal region at 68–72 °C and 14–18 min (predicted yield 27.9–29.1%). Energy consumption at optimal conditions (∼0.42 kWh/g biomass) was substantially lower than Soxhlet extraction (1.47 kWh/g). 2-MeTHF also showed the most favorable neutral lipid fraction and lowest residual solvent levels (&lt;5 ppm). These results identify MAE with green solvents, particularly 2-MeTHF, as an efficient and sustainable strategy for microalgal lipid extraction within a circular bioeconomy framework.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"93 ","pages":"Article 105364"},"PeriodicalIF":7.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148952631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"System-level maintenance strategy for offshore hybrid renewable energy systems based on cyclic non-homogeneous Markov Chain modelling","authors":"Panagiotis M. Psomas,&nbsp;Agapios N. Platis,&nbsp;Vasilis P. Koutras,&nbsp;Ioannis K. Dagkinis","doi":"10.1016/j.seta.2026.105369","DOIUrl":"10.1016/j.seta.2026.105369","url":null,"abstract":"<div><div>The reliability of offshore hybrid renewable energy systems (HRES) is critical due to seasonal environmental variability, and mechanical degradation. In this paper, a HRES that integrates wind, wave and solar power into an offshore platform is considered. The system is modeled using a Cyclic Non-Homogeneous Markov Chain (Cyclic NHMC), which unlike Homogeneous Continuous Time Markov Chains (HCTMC) with constant transition rates, the Cyclic NHMC incorporates season-dependent degradation, failure, maintenance, and environmental transitions. Building on this model, a Seasonal Energy Demand-Based Maintenance (SEDBM) strategy is introduced to allocate preventive maintenance according to seasonal demand criticality and environmental stress. The framework is validated through an experimental case study using in situ Aegean Sea environmental data, together with technical system data and expert-supported degradation and maintenance rates. System performance is assessed using Expected Energy Produced (EEP), Loss of Load Probability (LOLP), and Expected Energy Not Supplied (EENS). The results show that the Cyclic NHMC provides season-specific adequacy insights that are not captured by a homogeneous benchmark. Under SEDBM, EEP increases across all seasons, while LOLP and EENS are reduced during the critical seasonal periods. A modest increase in load loss in spring is observed, reflecting the deliberate reallocation of maintenance effort.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"93 ","pages":"Article 105369"},"PeriodicalIF":7.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148952665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Field thermal and energy performance of a closed-platform floating PV system on a tropical hydroelectric reservoir","authors":"Fabian Alonso Lara-Vargas,&nbsp;Idi Amin Isaac-Millan,&nbsp;Carlos Vargas-Salgado,&nbsp;Amarilis Castro-Morales","doi":"10.1016/j.seta.2026.105386","DOIUrl":"10.1016/j.seta.2026.105386","url":null,"abstract":"<div><div>FPV systems are expected to outperform ground-mounted PV (GPV) through water cooling, unconfirmed for closed-platform FPV in tropical reservoirs, where water temperature can exceed ambient air temperature. Platform type was isolated using a co-located, identical-hardware GPV reference, monitored for 12 months at hourly resolution in a Colombian hydroelectric reservoir. The GPV system generated 1407 kWh/kWp per year, compared to 1382 kWh/kWp for the FPV system, representing a shortfall of 25 kWh/kWp (1.75%), of which 12.4% is due to resistive losses in the FPV system’s longer wiring and the remaining 87.6% to thermal loss. The FPV modules operated at higher temperatures than the GPV modules during 70.5% of the monitored hours, at a mean differential of +1.17 °C. Water beneath the platform averaged 30.98 °C, above ambient throughout. Wind speed showed no cooling effect on the temperature difference between the FPV and GPV modules (Spearman: 0.090); airflow beneath the floating platform was not measured directly, and it is possible that the irregular bathymetry (depth of 6 to 38 m) limited renewal of the surface layer. Closed-platform FPV over warm water can act as a net heat source. The 25 kWh/kWp gap equals 567 USD/year (171 kWp) or 82,725 USD/25 years (1 MW), useful for prefeasibility assessments.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"93 ","pages":"Article 105386"},"PeriodicalIF":7.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148952668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Thermodynamic performance and multi-objective optimization of a low-grade heat-driven integrated desalination system combining absorption refrigeration, freeze crystallization and HDH process","authors":"Chan He,&nbsp;Sheng Yang,&nbsp;Nan Xie,&nbsp;Han Lei,&nbsp;Zhiqiang Liu","doi":"10.1016/j.seta.2026.105377","DOIUrl":"10.1016/j.seta.2026.105377","url":null,"abstract":"<div><div>To improve low-grade waste-heat utilization for high-salinity wastewater treatment in coal-fired power plants, a thermally driven integrated desalination system combining ammonia-water absorption refrigeration (AAR), freeze crystallization (FC), and humidification-dehumidification (HDH) is proposed. Flue-gas waste heat is converted into subzero cooling for freeze-based freshwater production, while residual low-grade heat is used for HDH polishing. A thermodynamic model is developed in Aspen Plus to evaluate heat and mass balances and component-level exergy destruction, and is further coupled with economic and life-cycle environmental assessments. Results show that the main irreversibilities arise from waste-heat transfer, ammonia-water solution regeneration, and coupled heat-mass transfer in the HDH unit, highlighting the importance of thermal matching and coordinated operation. A particle swarm optimization-backpropagation neural network (PSO-BP) surrogate model combined with the multi-objective grey wolf optimizer (MOGWO) is then used to optimize recovery ratio (RR), levelized cost of water (LCOW), and a 16-category life-cycle environmental impact index (EI16). The selected compromise solution achieves an RR of 49.29%, an LCOW of 3.97 $/t, and an EI16 of − 2.71 × 10⁷, providing a balanced trade-off between economic and environmental performance within the identified Pareto operating region. The results provide a decision-oriented basis for operating waste-heat-driven freeze-based ZLD systems.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"93 ","pages":"Article 105377"},"PeriodicalIF":7.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148952730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A DCEF-driven Stackelberg bi-level coordination framework for low-carbon grid-load interaction with multi-objective demand response of energy-intensive enterprises","authors":"Xiaoshun Zhang,&nbsp;Chundi Wang,&nbsp;Zhengxun Guo,&nbsp;Shuai Zhou","doi":"10.1016/j.seta.2026.105337","DOIUrl":"10.1016/j.seta.2026.105337","url":null,"abstract":"<div><div>To address the limitations of existing demand response, including insufficient coordination between system dispatch and the multi-objective decisions of enterprises, the compression of Pareto solution sets to a single solution, and oscillatory behavior during bi-level interactions, a Dynamic carbon emission factor (DCEF)-driven Stackelberg bi-level coordination framework is proposed for low-carbon dispatch. Within this framework, system power variance is minimized by the Distribution system operator (DSO) through Particle swarm optimization (PSO), while Gas turbine (GT) output is optimized and nodal DCEF are obtained through power-flow and carbon-flow tracing. Electricity cost and carbon emissions are optimized by energy-intensive enterprises using the Non-dominated sorting dung beetle optimizer (NSDBO). A multi-solution uploading mechanism is developed to retain multiple candidate solutions from the enterprise-side Pareto front for upper-level coordination, enabling the DSO to select the implemented load response only from the uploaded solutions. A progressive update mechanism is introduced to stabilize iterative coordination among enterprise loads, power-flow states, and nodal DCEF. Case studies on a modified IEEE 33-bus system validate the proposed framework. Under the 10% constraint, the proposed PU-DSM achieves the lowest mean system power variance among four modes, with reductions of 84.10% and 88.36% for workday and weekend scenarios, respectively. Under the 15% constraint, the reductions reach 85.37% and 90.61%. The IEEE 69-bus system further demonstrates the scalability of the proposed framework.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"93 ","pages":"Article 105337"},"PeriodicalIF":7.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148855080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From residual learning to explicit correlation revision: An interpretable machine learning framework for cross-fuel spray penetration prediction","authors":"Dexiang Xi,&nbsp;Wuqiang Long,&nbsp;Shaobin Chen,&nbsp;Yang Wang,&nbsp;Xilei Sun","doi":"10.1016/j.seta.2026.105402","DOIUrl":"10.1016/j.seta.2026.105402","url":null,"abstract":"<div><div>Predicting macroscopic spray characteristics is essential for optimizing fuel injection. To overcome the limited accuracy of conventional empirical correlations for liquid-phase penetration under high-temperature, high-pressure and cross-fuel conditions, this study proposes an interpretable machine-learning-assisted explicit correction method. A diesel/methanol spray database comprising 718 valid samples was established based on optical spray experiments conducted in this study. Using the Hiroyasu-Arai equation as the physical baseline, logarithmic relative residuals were defined and modeled using Light Gradient Boosting Machine (LightGBM). SHapley Additive exPlanations (SHAP) were then applied to identify dominant error sources and reconstruct nonlinear deviations into explicit correction terms. Results show that increasing ambient temperature suppresses liquid-phase penetration, with methanol exhibiting stronger thermal sensitivity. LightGBM outperformed eXtreme Gradient Boosting (XGBoost) and Categorical Boosting (CatBoost) in predictive accuracy and stability, while SHAP effectively revealed the origins of structural residuals. Compared with the original Hiroyasu-Arai correlation (R² = −1.3298), the corrected correlation achieved an R² of 0.9486, together with substantial reductions in Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The proposed closed-loop framework integrates experiments, baseline evaluation, residual learning, SHAP attribution, and explicit reconstruction, demonstrating an effective route for diagnosing and refining empirical spray correlations under complex conditions.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"93 ","pages":"Article 105402"},"PeriodicalIF":7.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148995850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Two stage framework for anomaly detection in cyber-physical system: A supervised deep learning approach","authors":"Akash Sharma,&nbsp;Rajive Tiwari","doi":"10.1016/j.seta.2026.105325","DOIUrl":"10.1016/j.seta.2026.105325","url":null,"abstract":"<div><div>The smart grid, a sophisticated cyber–physical system, integrates a physical power infrastructure with a cyber-communication network. It faces numerous anomalies that pose significant challenges to the accuracy and reliability of the system. This study presents a two-stage framework for detecting anomalies such as bad data, sudden load change (SLC), and false data injection attacks (FDIA). In the first stage, bad data is identified and removed using WLS state estimation. A feedforward neural network is employed in the second stage, to detect SLC and FDIA. To improve model performance, various techniques were implemented, including the categorical cross-entropy loss function for efficient training and regularization to reduce overfitting. Batch normalization was used to stabilize the learning process, while the ReLU activation function introduced non-linearity. An Adam optimizer ensured effective weight updates, and the softmax activation function was applied at the output layer for multi-class classification tasks. The methodology was validated on the IEEE 14-bus test system. Performance evaluation includes metrics such as accuracy, precision, recall, and F1 score, along with an analysis of the confusion matrix for detailed classification insights. A comparative analysis with other state-of-the-art algorithms revealed that the FFNN achieved an accuracy of 96.62%, surpassing other machine learning techniques.</div></div>","PeriodicalId":56019,"journal":{"name":"Sustainable Energy Technologies and Assessments","volume":"93 ","pages":"Article 105325"},"PeriodicalIF":7.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148855082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}