Aquacultural Engineering最新文献_第6页

RC-UMT: Multimodal fish feeding intensity classification network with enhanced feature extraction capability in aquaculture RC-UMT：水产养殖中具有增强特征提取能力的多模式鱼类摄食强度分类网络

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-09-20 DOI: 10.1016/j.aquaeng.2025.102639

Shantan Li , Xinting Yang , Zhitao Huang , Tingting Fu , Pingchuan Ma , Dejun Feng , Chao Zhou

{"title":"RC-UMT: Multimodal fish feeding intensity classification network with enhanced feature extraction capability in aquaculture","authors":"Shantan Li , Xinting Yang , Zhitao Huang , Tingting Fu , Pingchuan Ma , Dejun Feng , Chao Zhou","doi":"10.1016/j.aquaeng.2025.102639","DOIUrl":"10.1016/j.aquaeng.2025.102639","url":null,"abstract":"<div><div>Quantifying fish feeding intensity in real-time is crucial for devising scientific feeding strategies and improving production efficiency. Unimodal fish feeding intensity recognition tasks, such as audio or video, often fail to fully reflect the global characteristics, and lead to unreliable results. To address the limitations, a video and audio fusion model Reinforced Cross-modal Uni-Modal Teacher (RC-UMT) is introduced to evaluate fish feeding intensity classification, which can categorise feeding intensity into four different categories. The specific implementation is as follows: Firstly, the backbone in the original Uni-Modal Teacher (UMT) model is replaced with a Res2Net network, by weighting spatial coordinates of input feature maps, the Coordinate Attention (CA) mechanism in the Res2Net network enhances the ability to capture key multi-frequency features in audio signals. Secondly, RepViT, a lightweight convolutional network inspired by Vision Transformer principles, is introduced to the UMT, which can extract both global and local visual features simultaneously, thereby improving multi-level semantic representation. Finally, an affine transformation matrix with manifold preservation is introduced to enhance the fusion process of the original UMT, which preserves the manifold structure of original modal features, thereby achieving more efficient cross-modal fusion. Results from the experimentation indicate that the RC-UMT model achieves a classification accuracy of 93 %, outperforming the original UMT model by 7 % while reducing parameters by 25.41 %. Compared to audio-only and video-only modalities, the accuracy improves by 7 % and 6.5 %, respectively. Therefore, the proposed video-audio multimodal model enables real-time, high-precision feeding intensity classification, providing technical support for improving smart feeding equipment.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102639"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145119287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exposed vessel operations in aquaculture: Experimental validation of numerical models 水产养殖中暴露的船只操作：数值模型的实验验证

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-09-12 DOI: 10.1016/j.aquaeng.2025.102617

Robert Indergård, Simen B. Lysthaug, Eivind Lona, Henning Braaten

{"title":"Exposed vessel operations in aquaculture: Experimental validation of numerical models","authors":"Robert Indergård, Simen B. Lysthaug, Eivind Lona, Henning Braaten","doi":"10.1016/j.aquaeng.2025.102617","DOIUrl":"10.1016/j.aquaeng.2025.102617","url":null,"abstract":"<div><div>As more of the aquaculture industry is expected to move production towards exposed waters, accurate numerical models are essential for ensuring safe and efficient operations at sea.</div><div>In this paper, a numerical model of a vessel moored to a fish cage was developed and validated using model test experiments. The tests were conducted in irregular waves, wind, and current, with the vessel positioned in both head sea and beam sea conditions. The same environmental conditions were applied in the numerical analysis. The vessel motions in all degrees of freedom were considered, along with the forces in anchor lines, vessel mooring lines and bridle lines. The numerical model was created using SIMA.</div><div>Various floater ring bending stiffness values were tested in the numerical simulations. The system showed minimal sensitivity to these variations. The numerical model generally over-predicted peak loads and motions due to lack of friction between the vessel and the floating ring. However, it accurately recreated the vessel response and line force response, with mean values and phases matching the model test results. Hence, SIMA can predict motions and forces in typical exposed vessel operation in aquaculture.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102617"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145217588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DPAF-YOLO: YOLO-based dual-path attention fusion for glass eel detection and counting DPAF-YOLO：基于yolo的双路注意融合玻璃鳗检测与计数

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-11-05 DOI: 10.1016/j.aquaeng.2025.102661

Qi Chen , Weifeng Zhou

{"title":"DPAF-YOLO: YOLO-based dual-path attention fusion for glass eel detection and counting","authors":"Qi Chen , Weifeng Zhou","doi":"10.1016/j.aquaeng.2025.102661","DOIUrl":"10.1016/j.aquaeng.2025.102661","url":null,"abstract":"<div><div>The larvae of <em>Anguilla japonica</em> in the so-called \"glass eel\" stage are characterized by their small body size, transparent appearance, and prominent eye structure. Their transparency makes it difficult for conventional models to accurately detect and count individuals. To address the challenges of detecting and calculating glass eels from images, particularly given the morphological phenotype of these elvers, this study proposes a dual-channel attention fusion network model based on YOLOv10, named DPAF-YOLO (Adaptive Feature Fusion Module-YOLO). Firstly, the images containing high-density and overlapping glass eels were collected in situ from the fishing site. Then it takes the processing of sharpening to enhance the eye features of glass eel and to construct the necessary dataset. Based on the YOLOv10-m model, this study expands the detection heads to four by introducing an additional one, adopts an attention mechanism to achieve feature fusion of two different channels and Simplified SPPF module, and improves the loss function with the Focal DIoU to enhance detection accuracy and robustness. The effectiveness of the improvements was verified through ablation and comparative experiments. The result shows that the DPAF-YOLO model achieved an accuracy rate of 94.6 % and a recall rate of 95.2 %, with mAP50 and mAP50–95 reaching 96.8 % and 42.8 %, respectively. Compared with other YOLO models, this model demonstrates a significant improvement in accuracy for glass eel’s detection. Furthermore, we employ Eigen-CAM to generate heatmaps that highlight the image regions most critical to the model's classification decisions enhancing model interpretability, which indicates that the proposed algorithm performs well in both dense and sparse scenarios, with a particular focus on the eye region of glass eels, providing a reliable solution for glass eel detection and fry counting in the fishery domain.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102661"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145525329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Feasibility study of Field Asymmetric Ion Mobility Spectrometry (FAIMS) for rapid off-flavor detection in recirculating aquaculture system cultured Atlantic salmon 现场不对称离子迁移谱法（FAIMS）快速检测大西洋鲑鱼循环水养殖系统异味的可行性研究

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-09-01 DOI: 10.1016/j.aquaeng.2025.102622

Rakesh Ranjan , Gajanan S. Kothawade , John Davidson , Lav R. Khot

{"title":"Feasibility study of Field Asymmetric Ion Mobility Spectrometry (FAIMS) for rapid off-flavor detection in recirculating aquaculture system cultured Atlantic salmon","authors":"Rakesh Ranjan , Gajanan S. Kothawade , John Davidson , Lav R. Khot","doi":"10.1016/j.aquaeng.2025.102622","DOIUrl":"10.1016/j.aquaeng.2025.102622","url":null,"abstract":"<div><div>This study evaluates the feasibility of a portable Field Asymmetric Ion Mobility Spectrometry (FAIMS) system for detecting geosmin-related off-flavors in Atlantic salmon (<em>Salmo salar</em>) fillets from fish cultured in a recirculating aquaculture system. Salmon were exposed to a concentrated geosmin solution to spike off-flavor concentration in the fish tissues, followed by purging to create varying geosmin levels. Baseline geosmin and 2-methylisoborneol (MIB) in water and fillet samples were quantified using solid-phase microextraction coupled with gas chromatography-mass spectrometry (GC-MS). Volatile organic compound (VOC) fingerprints from geosmin-treated cold and heated fillet samples were analyzed via FAIMS, and the results were compared against the untreated control (UTC). For cold samples, the VOC signatures in FAIMS spectra of the UTC and treatment groups with spiked off-flavor were nearly identical. Heated samples, however, displayed distinct ion current profiles within the compensation voltage range of −2–2 V across the full dispersion field range. Principal component and statistical analysis also demonstrated evident separation between UTC and treatment groups. While FAIMS showed promise in effectively differentiating the VOC fingerprints between UTC and geosmin-treated fish fillets, the system could not consistently distinguish concentration differences within treatment groups, regardless of thermal treatment. Overall, the sampling and pre-processing protocols developed in this study can help fish farmers quickly identify geosmin-contaminated batches during grow-out stages and prior to processing, enabling timely corrective actions. Future studies could focus on training FAIMS for the quantitative analysis of off-flavor compounds, utilizing key biomarkers identified by GC-MS.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102622"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144997563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Effect of non-uniform biofouling distribution on the drag coefficient of aquaculture nets 生物污垢不均匀分布对水产养殖网拖曳系数的影响

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-10-25 DOI: 10.1016/j.aquaeng.2025.102653

Si Thu Paing, Louise Kregting, Suzy Black

{"title":"Effect of non-uniform biofouling distribution on the drag coefficient of aquaculture nets","authors":"Si Thu Paing, Louise Kregting, Suzy Black","doi":"10.1016/j.aquaeng.2025.102653","DOIUrl":"10.1016/j.aquaeng.2025.102653","url":null,"abstract":"<div><div>Biofouling on aquaculture nettings poses several issues, especially increased drag on net panels or aquaculture structures to which the net is attached, potentially affecting the structural integrity of the infrastructure. Despite biofouling growth on nets not being uniformly distributed, only a few studies have explicitly investigated the influence of naturally fouled nets on drag. In this study, the effects of uniform and non-uniform (half- and quarter-concentrated) hard fouling distributions on the drag coefficient of knotless square net were numerically investigated using a Computational Fluid Dynamics (CFD) approach. Based on the experimental data obtained from the literature, a CFD model was developed, validated, and analyses were undertaken on the effects of fouling distribution on associated drag coefficients of netting panels. At low net solidities less than 0.6, nettings with half- and quarter-concentrated fouling had significantly lower drag coefficients, up to 40 % and 35 % respectively, than a netting with uniform fouling that had the same solidity. However, the difference in drag coefficient on nettings due to fouling distribution attenuated with a further increase in fouling (net solidity). The correlations of drag coefficient with net solidity for netting panels with half- and quarter-concentrated fouling were proposed, which can provide a conservative drag coefficient estimate for the evaluation of cage safety based on the fouling distribution. This study also highlights the need to take uniformity of fouling distribution on aquaculture nets into consideration when estimating drag force, especially at solidities less than 0.6, as it significantly differs from that of uniform fouling distribution.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102653"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145412547","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Discrete element method for non-spherical feed particle transport in offshore truss cage aquaculture: Modeling and applications 海上网箱养殖非球形饲料颗粒输运的离散元法：建模与应用

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-10-07 DOI: 10.1016/j.aquaeng.2025.102644

Haibo Liu , Yunpeng Zhao , Chao Ma , Guangchen Jia , Zhan Zhang , Dong An

{"title":"Discrete element method for non-spherical feed particle transport in offshore truss cage aquaculture: Modeling and applications","authors":"Haibo Liu , Yunpeng Zhao , Chao Ma , Guangchen Jia , Zhan Zhang , Dong An","doi":"10.1016/j.aquaeng.2025.102644","DOIUrl":"10.1016/j.aquaeng.2025.102644","url":null,"abstract":"<div><div>Optimizing feed utilization is a critical challenge in offshore aquaculture, where low efficiency and high costs persist due to difficulties in predicting feed dispersal. This study addresses this issue by employing a Computational Fluid Dynamics–Discrete Element Method (CFD-DEM) framework, featuring a non-spherical drag model, to analyze the transport and dispersion of feed pellets in a truss net cage. One-way coupling was utilized to simulate the trajectories of dilute concentrations of both short cylindrical (<em>H/D</em> = 0.2, 0.4, 0.6) and slender cylindrical (<em>H/D</em> = 2.24, 2.47, 2.83) pellets. The results reveal that under cross-flow conditions, the particle Stokes number (<em>S</em><sub><em>t</em></sub><em>)</em> is significantly less than unity (<em>S</em><sub><em>t</em></sub>≪1), indicating that trajectories are primarily governed by the local flow field. The pellet aspect ratio was identified as a key determinant of rotational dynamics, slender pellets exhibited more vigorous rotation, which enhanced their spatial dispersion. Particle shape significantly influenced residence time. While slender pellets settled faster in still water, short cylindrical pellets had shorter residence times in the presence of a current. By accurately modeling small-scale particle dynamics within a large-scale environment, this research provides valuable insights for optimizing feed pellet geometry to enhance aquaculture efficiency.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102644"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145263200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A review of generative AI in aquaculture: Applications, case studies and challenges for smart and sustainable farming 水产养殖中的生成式人工智能综述：智能和可持续农业的应用、案例研究和挑战

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-09-14 DOI: 10.1016/j.aquaeng.2025.102637

Waseem Akram, Muhayy Ud Din, Lyes Saad Saoud, Irfan Hussain

{"title":"A review of generative AI in aquaculture: Applications, case studies and challenges for smart and sustainable farming","authors":"Waseem Akram, Muhayy Ud Din, Lyes Saad Saoud, Irfan Hussain","doi":"10.1016/j.aquaeng.2025.102637","DOIUrl":"10.1016/j.aquaeng.2025.102637","url":null,"abstract":"<div><div>Generative Artificial Intelligence (GAI) is revolutionizing aquaculture by providing practical and scalable solutions to longstanding industry challenges, including limited data availability, labor-intensive underwater inspections, disease outbreaks, and inefficiencies in resource management. As the sector evolves toward the Aquaculture 4.0 vision of intelligent, interconnected, and sustainable systems, GAI offers transformative capabilities across perception, planning, optimization, and communication. GAI enhances automation, decision support, and situational awareness across the aquaculture value chain through the intelligent synthesis of multimodal data ranging from sensor logs and underwater imagery to textual records and simulations. This review presents the first comprehensive synthesis of GAI in aquaculture, covering foundational models (e.g., diffusion models, transformers, and GANs), domain-specific applications, and emerging deployment scenarios. We demonstrate how GAI drives industry innovation in areas such as ROV-based infrastructure inspection, digital twins for farm design, synthetic data generation for fish health diagnostics, multimodal sensor fusion, and personalized advisory systems. Importantly, we map GAI models to specific aquaculture tasks, highlighting their suitability and advantages. We also offer a critical assessment of their operational readiness, including trust, performance, and environmental impact issues. In addition, we provide a systematic classification of applications, case studies, and future directions to guide the responsible and scalable integration of GAI in aquaculture. This review highlights GAI as a powerful tool and a foundational enabler of innovative, resilient, and ecologically aligned aquaculture systems, accelerating the industry’s transition toward more efficient, transparent, and adaptive practices.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102637"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145107186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Habitat-enhanced recirculating aquaculture systems improve post-transport recovery and survival of Scylla serrata (Forsskål, 1775) 生境增殖型循环水养殖系统可提高锡拉（Scylla serrata）运输后的恢复和存活（forsskamatl, 1775）

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-10-03 DOI: 10.1016/j.aquaeng.2025.102642

Naveen Nivas S , Dinesh Kaippilly , Geeji MT , Jowin T James , Subi S , Ayyaru Gopalakrishnan , Anisha Shafni John , Keerthika Ganesan , Arundathy A

{"title":"Habitat-enhanced recirculating aquaculture systems improve post-transport recovery and survival of Scylla serrata (Forsskål, 1775)","authors":"Naveen Nivas S , Dinesh Kaippilly , Geeji MT , Jowin T James , Subi S , Ayyaru Gopalakrishnan , Anisha Shafni John , Keerthika Ganesan , Arundathy A","doi":"10.1016/j.aquaeng.2025.102642","DOIUrl":"10.1016/j.aquaeng.2025.102642","url":null,"abstract":"<div><div>Mud crab aquaculture is expanding rapidly, but its potential is constrained by high mortality rates linked to post-transport stress and suboptimal rearing systems. The industry’s reliance on prolonged transport and clear-water recirculating aquaculture systems (RAS) is fundamentally misaligned with the physiology of <em>Scylla serrata</em>, the Indo-Pacific’s most widely farmed crab species. This study evaluated a habitat-enhanced RAS designed to mimic estuarine conditions through the use of mud substrate, turbid water and shelters to address these challenges. Crabs subjected to transport durations of 0 h, 24 h and 48 h were reared in either conventional clear-water or naturalised RAS. Extended transport significantly increased <em>Vibrio</em> spp. loads in haemolymph and tissues, induced gill damage and suppressed immune function. In contrast, naturalised RAS substantially improved outcomes, achieving 80 % survival even after 48 h of transport. These systems also enhanced key health indicators, including immune parameters (total haemocyte count, phenoloxidase) and antioxidant capacity (superoxide dismutase, glutathione peroxidase), while simultaneously increasing moulting frequency and growth rates compared with clear-water controls. Kaplan-Meier analysis further confirmed a significantly lower mortality risk in naturalised systems. Collectively, these findings demonstrate that adopting naturalised, species-specific RAS designs is a viable and critical strategy for mitigating transport stress, improving animal welfare and ensuring the long-term sustainability of mud crab aquaculture.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102642"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145217599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Smart aquaculture revolution: Triple-hybrid AI framework for ultra-precise aquaculture energy forecasting using adaptive markov imputation, neutrosophic intelligence, and shrimp-inspired multi-scale optimization 智能水产养殖革命：三重混合AI框架，用于超精确水产养殖能量预测，使用自适应马尔可夫imputation，嗜中性智能和虾类多尺度优化

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-10-24 DOI: 10.1016/j.aquaeng.2025.102651

Hanaa Salem Marie , Mostafa Elbaz , Farahat S. Moghanm , Hosam O. Elansary , Yasser Alginahi , Wael Said , Abdallah Elshawadfy Elwakeel , Ahmed Z. Dewidar

{"title":"Smart aquaculture revolution: Triple-hybrid AI framework for ultra-precise aquaculture energy forecasting using adaptive markov imputation, neutrosophic intelligence, and shrimp-inspired multi-scale optimization","authors":"Hanaa Salem Marie , Mostafa Elbaz , Farahat S. Moghanm , Hosam O. Elansary , Yasser Alginahi , Wael Said , Abdallah Elshawadfy Elwakeel , Ahmed Z. Dewidar","doi":"10.1016/j.aquaeng.2025.102651","DOIUrl":"10.1016/j.aquaeng.2025.102651","url":null,"abstract":"<div><div>Accurate energy prediction in aquaculture systems is critical for sustainable operations, yet existing methods fail under high missing data rates and provide no uncertainty quantification. This study presents a novel triple-hybrid AI framework integrating three algorithmic innovations: (1) Adaptive Markov-based Missing Value Imputation (AM-MVI) with dynamic transition probabilities for handling missing sensor data up to 35 %, (2) Shrimp Attitude Multi-Scale Optimization (SAMSO) algorithm inspired by marine organism behaviors for multi-dimensional feature selection across temporal, spatial, and operational scales, and (3) Adaptive Neutrosophic Energy Prediction (ANEP) providing explicit uncertainty quantification through truth, indeterminacy, and falsity membership functions. Experimental validation utilized 8760 hourly measurements across 24 months from five international aquaculture facilities. The integrated framework achieved superior performance compared to twelve state-of-the-art methods: 23.7 % RMSE improvement (29.83–22.76 kWh), 24.3 % MAE reduction (23.14–17.52 kWh), and enhanced coefficient of determination from 0.894 to 0.932. Uncertainty quantification achieved 94.8 % prediction interval coverage probability at 90 % confidence. Statistical analysis confirmed extremely strong significance (p < 0.001) with large effect sizes (Cohen's d: 1.11–1.82). Cross-validation demonstrated consistent performance across seasonal variations and geographic contexts. The framework enables real-time processing with 18.6 ms inference latency, supporting practical aquaculture energy management applications with potential for 5–8 % operational cost reductions.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102651"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145412467","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Safety evaluation of physical plasma for water treatment in recirculating aquaculture systems 循环水养殖系统水处理用物理等离子体的安全性评价

IF 4.3 2区农林科学

Aquacultural Engineering Pub Date : 2026-01-15 Epub Date: 2025-10-09 DOI: 10.1016/j.aquaeng.2025.102646

Martina Balazinski , Veronika Hahn , Robert Wagner , Mirko Basen , Klaus-Dieter Weltmann , Juergen F. Kolb

{"title":"Safety evaluation of physical plasma for water treatment in recirculating aquaculture systems","authors":"Martina Balazinski , Veronika Hahn , Robert Wagner , Mirko Basen , Klaus-Dieter Weltmann , Juergen F. Kolb","doi":"10.1016/j.aquaeng.2025.102646","DOIUrl":"10.1016/j.aquaeng.2025.102646","url":null,"abstract":"<div><div>Cold atmospheric plasma (CAP) is a promising technology for wastewater remediation, particularly for recirculating aquaculture systems (RAS) with persistent microbial residues. In our previous studies, CAP already demonstrated a high inactivation efficacy against fish-pathogenic microorganisms (up to 5 orders of magnitude) in model aquaculture water, which was composed according to the cultivation of Baltic sturgeon at the Research Centre for Agriculture and Fisheries (Born, Germany). However, in addition to the antimicrobial efficacy fish welfare and product safety, as in particular described by toxicity and mutagenicity, are crucial requirements for the future implementation of this technology in RAS. Therefore, internationally accepted toxicity tests were conducted in the present studies. The potential toxicity of a plasma treatment was analyzed using a fish cell line (RTgill-W1) following OECD Test Guideline No. 249. A possible water toxicity was evaluated in accordance to the DIN EN ISO 11348–1 by a bioluminescence assay using <em>Aliivibrio fischeri</em> as a water borne microorganism resulting in a IC<sub>50</sub> of 21 %. No poisoning of fish cells could be identified for CAP admixtures of up to 30 %. Additionally, a mutagenicity assay with <em>Escherichia coli</em> WP2 <em>uvrA</em> revealed no changes. Moreover, analyses by ion-chromatography of CAP-treated water showed neither a formation nor a degradation of the toxic agent bromate in RAS water. In conclusion, CAP is a promising, adjustable and safe technology for the treatment of water in aquaculture systems although further research towards optimization, scaling, and associated economic considerations for an implementation of the technology is needed.</div></div>","PeriodicalId":8120,"journal":{"name":"Aquacultural Engineering","volume":"112 ","pages":"Article 102646"},"PeriodicalIF":4.3,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145320678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0