Abstract:In order to accelerate the healthy and sustainable development of deep-sea aquaculture in China and promote the spatial expansion of mariculture and the transformation and upgrading of marine fishery industry, the article expounds the current status of deep-sea aquaculture facilities, and equipment as well as technology in China and lists the latest technical progress and typical facilities of the main types of deep-sea aquaculture including gravity cages, truss-type cages and aquaculture vessels etc. Then in response to the technical requirements for the construction of deep-sea aquaculture system, the article clarifies the progress and development trend of science and technology innovation at home and abroad in the areas of breed cultivation, fry production, aquaculture facilities, and production management equipment, etc. The article further dissects the main problems existing at present, such as lack of good breeds, insufficient integration of disciplines, technical immaturity of facilities and equipment, as well as lack of standardization in design and construction based on the analysis on how to realize the sustainable and healthy development of deep-sea aquaculture in China. Last but not least, the key technical breakthroughs in species, facilities and equipment, and industrial chain are proposed to achieve safety, quantity and quality breeding, as well as deeper and farther deep-sea aquaculture in China, with the aim to provide reference for accelerating the sustainable development of deep-sea aquaculture in China.

Abstract:The industrialized recirculating aquaculture system （RAS） has emerged as a promising approach to achieving standardized culturing and intelligent management in aquaculture. This system conforms to the principles of resource conservation and environmentally friendly green development, making it a leading direction for the future of the industry. The aquaculture tank, being a vital infrastructure component of RAS, plays a crucial role in optimizing space utilization, water-injection energy efficiency, waste collection and discharge effectiveness, and ensuring uniform distribution of dissolved oxygen to support the healthy growth of aquatic organisms. This article provides a comprehensive review of both domestic and international research on the hydrodynamic characteristics of RAS. By summarizing the current research hotspots and trends in this field, it aims to analyze how various factors such as tank structure, water circulation driving devices, and waste discharge devices influence hydrodynamics within RAS. The research results presented in this article lay a solid foundation for the development of an "energy-saving, emission-reducing, ecological, and efficient" aquaculture system.

Abstract:In response to the current situation that most of the bait feeders in pond culture rely on manpower, which is labor-intensive, and it is difficult to feed the baiting boats on steep sloping banks, an energy-saving feeding truck without power discharging has been designed. The truck uses a rotating discharging valve for feeding, and a scissor lift platform for adjusting the height of feeding. Through the design of the spherical limit rotating connector located at the discharge port of the rotating discharging valve to adjust the inclination angle of the feeding tube to realize the non-powered discharging,a new type of telescopic feeding tube is designed based on cylindrical cam. The feeding truck can realize directional, fixed-distance and fixed-point feed addition through the movement of the truck body, the lifting and lowering of the scissor lift platform, and the adjustment of the inclination angle and length of the feeding tube. The prototype of the feeding truck was made, and the feeding test was carried out with the baiting boat and stationary baiting machine as the feeding target, and the test results show that the feeding truck has the universality for different specifications of feeds, and the telescopic length of 1.2 m and the lifting height of 1.5 m can satisfy the feeding needs of the baiting boat and stationary baiting machine, and the adjusting is flexible, and the rate of feeding is 6.2 kg/min. The designed feeding truck is flexible in use, and it can replace the manual feeding to the baiting boat and stationary baiting machine. It can replace manual feeding for baiting boat and fixed baiting machine, reduce manual labor intensity, improve the mechanization level of breeding process, and lay the foundation for further automation and intelligence.

Abstract:Aquatic weed management is an important part of crab aquaculture. In order to reduce the labor intensity of aquatic weed collection equipment control and decrease the use of personnel, the fan driving system of the aquatic plants comb-collect boat has been developed. The driving system is mainly composed of fan blades, protective devices, brackets, motors, reduction drives and waterproof sealing covers. Through CFD fluid simulation to calculate and verify the resistance of the practical boat, the fan was selected, the rated power is 1 100 W, the working voltage was 48 V, the maximum width of the blade was 100 mm, the inclination angle of the blade root was 28°,and the steering system with self-locking function was designed. The tests of using the drive system in the crab breeding pond showed that the continuous working time was 1.5 hours under the 48 V/24 Ah power battery, and the unloaded and fully loaded boat speeds were 1.14 and 0.43 m/s respectively in the absence of anhydrous weeds. The maximum speed of unloaded and fully loaded boats was 1.06 and 0.26 m/s respectively under the condition of mild aquatic weeds （the top of aquatic plants was >3 cm and <8 cm from the water surface）. The maximum boat speeds of unloaded and fully loaded boats under severe aquatic plants （the top of aquatic plants was ≤3 cm from the water surface） were 0.71 and 0.12 m/s, respectively, and they could maintain a forward posture and continue to collect aquatic plants. The average turning radius was 1.2 meters. The on-site application showed that the driving system meets the dynamical requirements of the combing boat in the actual aquatic weed environment of the crab pond, and the weight of the aquatic weed increased by 94.3% per unit time compared with the manual cleaning of the aquatic weed, laying the foundation for the realization of automatic aquatic weed management.

Abstract:The aquatic plants comb-collect boat has gradually become an important equipment for the management of aquatic plants in river crab breeding ponds. In order to realize the automation of the aquatic plants comb-collect boat, an automatic navigation control system for the boat with a fan as the driving device was designed and developed. Firstly, a set of the aquatic plants comb-collect boat software and hardware control system was designed； Then, the dynamic model of the aquatic plants comb-collect boat was established and verified by experiments； Then, the motion control and path tracking of the aquatic plants comb-collect boat were realized by using the heading estimation PID control method and the LOS circle of sight method； Finally, the speed test, path tracking and waterweed harvesting experiments were carried out in the farm. The results show that the maximum speed of the aquatic plants comb-collect boat can reach 1.14m/s, the PID heading control method and the LOS circle of sight method can accurately realize the path point tracking of the aquatic plants comb-collect boat. In good weather conditions, the average error of path tracking is less than 0.96 m. compared with the manual method, the cleaning weight per unit time is increased by 94.3%, which effectively reduces the labor intensity of harvesting aquatic plants and improves the automation level of aquatic plant management in crab ponds.

Abstract:In order to achieve accurate and reliable prediction of the dissolved oxygen in ponds and mitigate aquaculture risks, we propose a predictive model based on Feature Construction （FC） pretreatment and Temporal Convolutional Network （TCN） coupled with Bidirectional Gate Recurrent Unit （BiGRU）. By constructing statistical features, environmental factor features and seasonal features from the samples, deep-level correlations between variables are explored. Then, the structural feature sequences are processed using multiple layers of convolution and dimensionality reduction through TCN, while preserving the global temporal characteristics and removing redundant information. By integrating BiGRU to model the reduced features, accurate prediction of dissolved oxygen levels is achieved. Furthermore, the Sand Cat Swarm Optimization （SCSO） algorithm is employed to optimize the non-parametric Kernel Density Estimation （KDE） for estimating the distribution range of dissolved oxygen prediction errors. The experimental results indicate that the proposed model achieves superior performance compared to other comparative models, with respective values of 0.027 5 for Mean Squared Error （MSE）, 0.143 2 for Mean Absolute Error （MAE）, 0.165 8 for Root Mean Squared Error （RMSE）, and 0.94 for the Coefficient of Determination （R²）. Meanwhile, the interval estimation effectively covers the fluctuation range of dissolved oxygen, thereby quantifying the uncertainty in the prediction process. In the short-term prediction of dissolved oxygen levels in ponds, this model demonstrates notable accuracy and robustness. It is instructive for water quality monitoring in aquaculture and the enhancement of farming efficiency.

Abstract:Aiming at the current situation of low endurance and high cost of mobile monitoring equipment for aquaculture, which limits the widespread application of such equipment, a small portable AUV for mobile and fixed-point monitoring of fishery aquaculture is proposed and developed. A seal-like operation mode is proposed, replacing expensive underwater inertial navigation systems with conventional GNSS. The three main hulls are arranged in a pin pattern with a modular structure to increase smoothness. The unpowered diving and floating are controlled by the motor-controlled piston altering the volume of the external oil bladder. With the low-power STM32F407 chip as the control core, FreeRTOS embedded real-time operating system and LabVIEW development environment are employed to develop the AUV water surface autonomous cruise and dynamic obstacle avoidance system, data image acquisition and communication system, and shore-based intelligent fisheries big data interaction interface. The results of energy consumption analysis and simulation experiments show that the AUV has low energy consumption, and the endurance time is about 77 h when carrying a lithium battery of 5.184 kWh, which can perform autonomous cruising. Tests show that the prototype can effectively communicate with the shore-based host computer, collect water quality information and underwater video images. The AUV for aquaculture monitoring has flexibility in carrying sensors, long endurance, low power consumption, low cost, easy maintenance and upgrading, and is able to meet the demand for three-dimensional movement and long-term fixed-point monitoring of various aquaculture.

Abstract:Baiting is an important part of shrimp pond culture, and obtaining the weight of the remaining bait is the key to achieving accurate baiting. In order to solve the problem of inaccurate weighing of the remaining bait caused by the environment, the system itself and other factors, an adaptive strong-tracking unscented Kalman filtering algorithm （SHFL-ASTUKF） for the dynamic weighing of the bait on baiting boats is proposed. Firstly, a second-order model of the weighing system was established and the measurement results were filtered. Then, based on the fast introduction of the asymptotic cancellation factor, the singular value decomposition was used to replace the Cholesky decomposition to deal with the error covariance problem. Meanwhile, the fuzzy control algorithm and Sage-Husa adaptive filtering were combined to adaptively update the measurement noise covariance and the system noise covariance, so as to suppress the divergence in the filtering process. Example data simulation and experimental validation of SHFL-ASTUKF show that, compared with the strong tracking unscented Kalman filter, the RMSE is improved by 14.9% in the example data simulation, and the RMSE is improved by an average of 15.15% in the experiments of the bait weighing of the remaining bait in the baiting boat and the MAE is improved by 17.27%. The proposed algorithm has higher dynamic measurement accuracy and better noise reduction effect.

Abstract:To enhance the effectiveness of pollution collection and self-purification within a circulating water aquaculture system, a regular octagonal aquaculture tank structure with a deflector was developed. A three-dimensional non-constant flow field numerical calculation model was established. Based on the validation of the calculation method, this study investigated the impact of geometric parameters and the positioning of flow-guide discs on hydrodynamic traits, including flow velocity distribution, vortex strength, wall shear stress, and water body mixing uniformity. The findings indicate that, while keeping the ratio of deflector diameter to pond width constant, elevating the deflector height yields a pattern of diminishing and then augmenting vortex columns at the center of the flow field. Additionally, the number of vortex rings displays a similar tendency, transitioning from decrease to increase and gradually evolving into an annular configuration from its irregular shape. Similarly, at the same height, raising the ratio of deflector diameter to pond width leads to an initial growth and subsequent reduction in the vortex column, while the count of vortex rings experiences a growth followed by a relatively stable phase. This initial growth in vortex rings is followed by a largely consistent state. By situating the deflector 20- -40 mm above the pool bottom and maintaining a pool width-to-diameter ratio of 0.05- -0.08, an amplified velocity gradient close to the pool's base can be achieved. Furthermore, this installation reduces collisions between the pool wall surface and the water body, enhancing vortex strength and promoting hydraulic mixing uniformity. This configuration fosters the necessary hydrodynamic conditions for the aggregation and expulsion of particulate matter.

Abstract:For the microgrid system under the "fish-light complementarity", a multi-objective optimization energy scheduling method is proposed to balance the "impossible triangle" of energy. Firstly, based on the background of "fish-light complementarity", a light-storage-load microgrid system model is established, and multidimensional dynamic objective functions that include economic, environmental, and comprehensive aspects are proposed. Secondly, considering the shortcomings of the sparrow search algorithm, such as insufficient diversity and easy trapping in local optima, an improved sparrow search algorithm （SA-ISSA） based on simulated annealing algorithm and hybrid strategy is proposed for microgrid energy optimization scheduling. Finally, a case study was conducted on the "fish-light complementarity" factory breeding base in Puyue Town, Chongming, Shanghai, where the SA-ISSA algorithm is applied to solve the microgrid model and obtain the optimal energy scheduling scheme. The experimental results demonstrate that the SA-ISSA algorithm performs better and has higher application value in solving the microgrid energy optimization scheduling problem, providing reference for energy optimization scheduling strategies in microgrid systems under the "fish-light complementarity".

Abstract:In this paper, an autonomous mobile feeding robot system for eel farming is developed, which mainly includes AGV handling trolley, XYZ CNC electric synchronous belt sliding table linear guide module, feeding peristaltic pump, motion controller, etc. The AGV handling trolley detects the guiding magnetic stripe in real time to determine the working path through the sensing device. The motion controller outputs pulse signals to the stepping motor of the XYZ three-dimensional CNC electric synchronous belt slide table linear guide module to achieve the spatial positioning of the robot feeding position. By controlling the rotational speed of the peristaltic pump drive motor, the precise feeding of bait and the stability of the feeding process are realised. Nine prototype performance tests were carried out for the three-layer three-dimensional eel pond with a target feeding volume of 9.5 g. The results showed that the average error of the robot spatial positioning accuracy was （2.1, 1.8, 2.1） mm, and the standard deviation was （2.6, 2.3, 2.5） mm, and the average feeding rate of the robot was 1.85 g/s, and the average feeding volume was 9.3 g, with an error of less than 5%. To assess the stability and reliability of the system, the robot was operated continuously for 50 hours at a travelling speed of 0.5 km/h without technical failures or system interruptions. In conclusion, the autonomous mobile feeding robot is stable and reliable, with high feeding accuracy and good feeding effect, which has strong practical and popularisation value.

Abstract:Based on binocular vision technology,a target detection and ranging technology suitable for unmanned feeding equipment is studied.A high-precision target detection and recognition and binocular ranging method was designed,and related simulation experiments were carried out.The experimental results show that the average accuracy of the improved recognition and positioning algorithm is 99.53%.When the binocular camera is 40 cm to 90 cm away from the feeding device,the relative error of ranging is between 0.59% and 1.26%,which can effectively assist the short-distance positioning feeding device.It better meets the docking function requirements of the feeding device and the silo,and has positive significance for enhancing the automation level of the feeding device.

Abstract:YOLO-based mussel （Mytilus edulis） recognition and detection technology is the key to achieving mechanization and intelligence in the grading, seedling separation, and other operational links of mussels. However, the unclear and indefinite external features of mussels present a challenge to improving recognition accuracy. This paper proposes a mussel object detection model based on the improved YOLOv5 algorithm （CST-YOLO）, which integrates the CoordAttention attention mechanism to enhance feature expression ability, uses the SIoU boundary box regression Loss function to reduce the boundary box regression loss and improve the detection speed of the model, and designs an improved decoupled head TSCODE to enhance the detection accuracy. The algorithm testing was performed on a self-constructed dataset of mussels, and the experimental results demonstrated that the CST-YOLO algorithm exhibited an improvement in accuracy （P） by 0.428% compared to the YOLOv5 algorithm, and mAP （mean Average Precision） at IoU thresholds of 0.5 and 0.95 reached 92.221%, indicating a significant increase of 1.583%. These findings highlight the effective enhancement of mussel object detection accuracy while ensuring optimal detection speed achieved by the CST-YOLO algorithm. The present study contributes to the advancement of machine vision technology in facilitating the automation and intelligent production and processing within the mussel farming industry.

Abstract:In order to improve the ability of cavitation water jet to clean the surface of the net, an Inner wall-Central body cavitation nozzle is designed. The numerical method of cavitation nozzle cavitation flow CFD was used to compare and analyze the simulation and theoretical solution, and the reliability of the two-dimensional axisymmetric cavitation flow numerical model based on the two-equation turbulence model was verified. The numerical calculation of the cavitation flow of the Inner wall-Central body cavitation nozzle in the environment of submerged water jet was carried out, and the influence of the Inner wall cavitator and the Central body on the cavitation effect of the nozzle was obtained, and the influence of water depth on the cavitation effect was calculated and discussed. The results show that： The water jet passes through the nozzle throat, creating cavitations on the surface of the Inner wall cavitator and the Central body, respectively； The Central body cavitation effect is better than that of the Inner wall cavitation, but the effect on jet blockage is stronger； The Inner wall cavitator increases the nozzle throat pressure and further improves the cavitation effect of the Central body； Both the Inner wall cavitator and the Center body can reduce the sensitivity of the nozzle cavitation effect to water depth. This study provides a basis for the research of cavitation flow dynamics mechanism for nozzle design of net cleaning in aquaculture cages, and provides design ideas with low cost and high efficiency.

Abstract:Due to the long distance from the coastline and mild environmental conditions, the sea near islands has become a key area for the development of aquaculture. It is found that violent wave reflections near islands can lead to a relative concentration of wave energy, which easily causes structural damage to aquaculture facilities. Most of the studies on the dynamic response characteristics of aquaculture facilities have been focused on the flat seabed areas, while there are fewer studies on the sea near islands. To solve this problem, the dynamic response of a hinged multi-body floating aquaculture platform deployed in the sea near islands in the survival condition are studied in this paper by means of physical model experiment. The results show that the presence of island increases the surge motion of the platform and the maximum value of power spectrum density increases by 61.78% compared with that of the flat seabed. The significant value of the cable force on the windward side of the platform increases by 2.03 times compared with that of the flat seabed in the survival condition. At the same time, the wave reflection caused by the island topography has transferred the position of the platform with the largest heave motion from the cage on the windward side to the middle cage. The pitch motion of the platform has been decreased, but the relative rotation between the middle cage and other cages has been increased. The results provide a theoretical reference for the design of aquaculture facilities and mooring system deployed in the sea near islands.

Abstract:With the in-depth development of marine ranching construction, dynamic monitoring and assessment of marine ecological environment and fishery resources have become particularly important, and mobile autonomous monitoring is one of the feasible solutions. Aiming at the problems of low efficiency, poor flexibility, high cost and difficult transportation and delivery of existing monitoring methods, a combined unmanned monitoring ship equipped with multiple types of monitoring equipment was proposed. The hull structure design and monitoring instrument installation mechanism design were carried out, and the counterweight mechanism design was carried out considering the hull draft load and stability. Based on the computational fluid dynamics method, the hull resistance performance was simulated and calculated, and the underwater thruster selection was completed, and the static strength finite element analysis was carried out on the unmanned ship connecting bridge structure in high waves, and the strength of the hull structure could resist the 4th level wave. The test results show that the unmanned monitoring vessel has stable operation without interference, the design draft is 0.25 m, the speed is 2.5 m/s, and it can be competent for marine ranching work sea conditions. The monitoring equipment works stably after being installed, and can monitor underwater fishery resources, explore underwater terrain, monitor water pH, temperature and dissolved oxygen in real time. This study provides technical support for ocean dynamic, real-time monitoring and digitalization, and provides a low-cost and reliable implementation scheme for the design and research of marine ranching combined unmanned monitoring vessels.

Abstract:Ultra-large aquaculture vessel is a new direction for the development of deep-sea aquaculture industry. With a new anchoring system, it can be used to reduce environmental pollution, improve the quality of aquatic products, avoid the attack of natural disasters such as typhoons, and solve the problem of unsustainable development of traditional aquaculture industry. The mooring system of aquaculture vessel is not only different from the temporary mooring system of transport ship, but also different from the positioning mooring system of ocean engineering, and is closely related to the operation safety and cost control of aquaculture vessel. Therefore, the design and optimization analysis of the mooring system of ultra-large aquaculture vessel is very necessary and important. Based on the project of 300 000 t large aquaculture vessel, the design and optimization of single point anchoring system were carried out. WRF model and 3DVAR system were used to simulate the wind field, and the wind field models were used to simulate the annual maximum wind speed of the anchorage area in a 30-year series. Wind speed extremes in different recurrence periods were calculated based on Weibull type distribution, and wind field models were used to forecast waves for 30 years. FVCOM model was used to analyze the current. The time-domain analysis of hull and mooring system was carried out with the help of Orcaflex software, and the mooring system analysis of aquaculture vessel was carried out with the input of wave simulation results. The appropriate anchor chain parameters were obtained by theoretical calculation. The results show that： For the 300 000 t aquaculture vessel, the grade-3 anchor chains with 122 mm diameter and 440 m length meet requirements of mooring under level 6 sea conditions, and the safety factor of anchor chain tension is greater than 1.67, which is safe and feasible. It provides a reference for the design of the single mooring system of ultra-large aquaculture vessel.

Abstract:Aquaculture vessels are one of the new ways to expand the space for offshore farming, and if particles such as undigested feed residues and excreta cannot be discharged in a timely manner during the shipboard tank farming process, it will have a serious negative impact on the health of fish and their growing environment. The article uses FLOW-3D software to carry out numerical simulation calculations to study the effect of the number of outlets in the aquaculture tank of the 300 000 t-class aquaculture vessel on the flow field characteristics under the stationary condition, and on this basis to explore the removal effect of particles with different particle sizes and densities in the tanks. The results showed that： when the total inlet and outlet flow rate and the total area of the outlet are constant, as the number of outlets increased from 1 to 4, V_99% decreased from 0.244 m/s to 0.232 m/s and V_95% decreased from 0.215 m/s to 0.203 m/s. As the number of outlets increases, the flow velocity in the aquaculture tank decreases. Measured by the time it takes for 95% of the total amount of particles to be discharged from the aquaculture tank, when the particle density was 1 050 kg/m³, discharge of particles with a particle size of 1 mm took about 24 minutes, discharge of particles with a particle size of 0.5 mm took about 45 minutes, and discharge of particles with a particle size of 0.1 mm took about 78 minutes. So at the same density of particles, as the particle size of particles increases, their retention time in the aquaculture tanks decreased and their removal efficiency increased significantly. When the particle size was 1 mm, discharge of particles with a density of 1 100 kg/m³ took about 15 minutes, and discharge of particles with a density of 1 150 kg/m³ took about 11.6 minutes. When the particle size of the particles was the same, the higher the density of the particles, the shorter their retention time in the flow field of the aquaculture tank, and the greater their removal efficiency. In this study, an integral analysis method combining CFD numerical simulation and mathematical statistics was formed to provide research means and theoretical basis for the complex working conditions with different transverse rolling periods and larger transverse rolling amplitudes.

Abstract:Automatic and accurate collection of fishery catch data is an important part of the electronic observer system. However, automated tuna longline catch estimation remains challenging to deploy in practice due to the complexity of the working environment and the instability of tracking. In this study, a lightweight counting network was designed to automate the processing of real-time video data from fishing vessels in order to enable real-time tracking and counting of tuna catches on board fishing vessels. YOLOv5s was selected as the benchmark network in this study. The channel pruning algorithm was first used to prune the backbone network of YOLOv5s, and the results showed that the detection accuracy of the pruned model mAP_0.5~0.95 reached 68.8%, and the detection speed was 16.5 frames per second （FPS） under CPU, which was basically unchanged compared with the original model. The number of parameters, model size and computation of the model were reduced by 67.2%, 66.4% and 42.5% respectively, and the detection speed was increased by 33.1%. Secondly, the ByteTrack algorithm was used to achieve real-time tracking of multiple targets, optimize the shape of the counting area and solve the problem of counting deviation caused by the jump in the identity （ID） of the tuna being tracked. The test results of 10 videos showed that the average counting accuracy of the method was 80% and the video processing speed was 50.7 FPS, which meets the requirements of industrial-grade real-time detection. In summary, the model had the advantages of light weight, high accuracy and real-time, which could complete the real-time monitoring of longline catch in complex working environments and provide a solution to realizing fisheries automation.

Abstract:The deep-sea unmanned vessel faces the problems of insufficient range and slow convergence and low accuracy of path planning algorithm when exploiting fishery resources.For the purpose of minimizing the environmental impact and optimizing navigation routes of unmanned fishing vessels during the actual mission execution of the unmanned vessel in fisheries, a path planning algorithm with the objective of minimizing multiple parameters such as path length, smooth and current energy was designed under the premise of ensuring its safe navigation.The multi-objective computational model of unmanned vessel under time-varying current disturbance was established by analyzing the ocean environment and mission objectives of the unmanned ship during navigation, and an improved adaptive gray wolf optimization algorithm was employed to settle the problem, which is integrated optimization by introducing multiple strategies. Eventually, the feasibility of the algorithm in the field of multi-objective optimization of unmanned surface vehicles, as well as the usefulness of the improved strategy can be confirmed by simulation experiments. The optimization rate of total objective value was improved by 9.2%, 1.7% and 11.9% for multi-objective compared to three single-objective simulation results. The simulation paths under different current conditions show that the distance-optimal algorithm can save more cost than the traditional distance-optimal algorithm and effectively improve the performance of global trajectory planning.