A method and system for intelligent aquaculture management and feeding optimization on offshore fish rafts

Abstract

This invention provides a method and system for intelligent aquaculture management and feeding optimization on offshore fish rafts, relating to the field of aquaculture management technology. The method includes the following steps: acquiring energy status information of energy equipment and biological characteristic information of the aquaculture species; acquiring multiple preset candidate feeding adjustment schemes, and calculating the estimated energy savings corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding equipment; calculating the aquaculture loss assessment value corresponding to each candidate feeding adjustment scheme based on the biological characteristic information, determining the target feeding scheme from the multiple candidate feeding adjustment schemes, and controlling the feeding equipment to execute the feeding operation. The method of this invention aims to solve the problem that existing intelligent aquaculture management systems for offshore fish rafts, when facing energy challenges, adjust feeding schemes solely based on power levels and equipment power consumption information, ignoring the physiological needs of the aquaculture species, leading to a decline in aquaculture efficiency. It achieves a smart balance between energy utilization efficiency and aquaculture benefits.

Description

Technical Field

[0001] This invention relates to the field of aquaculture management technology, and more specifically, to a method and system for intelligent aquaculture management and feeding optimization of offshore fish rafts. Background Technology

[0002] In offshore aquaculture environments far from the mainland, a stable power supply is fundamental for automated and intelligent management. To address the power supply issues of these aquaculture rafts and reduce operating costs, a power supply system consisting of wind turbines and photovoltaic panels is typically deployed. The electricity generated by this system is stored in battery banks to power various devices on the rafts. These devices include water quality monitoring sensors, such as probes for measuring water temperature, dissolved oxygen, and pH levels, as well as underwater cameras for observing fish activity. Furthermore, the most crucial piece of equipment is the automatic feeder, which precisely dispenses a fixed amount of feed at designated times according to instructions from the central control system. The central control system is the core of this intelligent management system. It collects information from all sensors, combines it with preset aquaculture models, calculates the most favorable feeding time and amount for fish growth, thus creating an optimized feeding plan, and instructs the automatic feeder to execute it.

[0003] Under ideal conditions of clear skies and moderate wind, the system operates very smoothly. Wind and sunlight generate ample electricity, not only meeting the daily operational needs of all equipment but also storing excess energy in batteries for emergencies. The central control system can execute optimal feeding plans without hesitation, for example, feeding multiple times during the fish's most active and feeding times to maximize feed utilization and promote healthy and rapid fish growth.

[0004] However, the weather at sea is unpredictable, and this ideal state cannot always be maintained. Problems begin to emerge when the fish rafts experience several consecutive days of rain, accompanied by calm or light winds. Under these extreme weather conditions, the power generation efficiency of the photovoltaic panels drops sharply, and the wind turbines almost shut down, resulting in the power input of the entire power supply system being far less than the normal consumption of the fish raft equipment. Initially, the system can still operate using the electricity stored in the battery banks, but as time goes on, the battery power continues to decline, quickly reaching a warning level. At this point, the energy management module of the central control system intervenes. To ensure the operation of the most basic life support systems (such as necessary oxygenation equipment) and the control system itself, it must implement energy-saving measures. Among the many electrical devices, the motor of the automatic feeder is the most energy-intensive component during startup and operation. Therefore, the primary energy-saving strategy of the energy management module is to adjust the original feeding schedule. It might simply reduce the planned four feedings per day to two, or directly cancel the energy-intensive nighttime feedings. This decision was based entirely on a simple logic: reduce the number of times the equipment runs to lower the total energy consumption, thereby extending the battery's lifespan and ensuring that the core equipment is always powered.

[0005] However, this approach hides a deeper problem. When deciding to reduce feeding frequency, the system relies solely on current power levels and equipment power consumption, failing to consider the crucial variable of the fish's physiological needs. The originally optimized feeding plan was a scientific conclusion drawn from considering various factors such as water temperature, fish age, and fish density, aiming to ensure the fish receive the optimal amount of nutrition at the most suitable time. Now, simply because of insufficient power, this scientific plan has been replaced by a crude energy-saving logic. This change has a direct negative impact on the farmed organisms. Irregular feeding or prolonged starvation can cause stress responses in the fish, affecting their normal digestion and absorption functions and reducing feed conversion efficiency. More seriously, if fish in their growth period do not receive a continuous and stable supply of nutrients, their growth rate will slow significantly, potentially even inducing disease, ultimately leading to a prolonged farming cycle and a decrease in yield. This creates a contradiction: an intelligent system designed to improve aquaculture efficiency through "optimization" adopts a "de-optimization" or even "anti-optimization" approach to protect itself when faced with energy challenges, and its behavioral logic is in fundamental conflict with the ultimate goal of improving aquaculture efficiency.

[0006] There is currently no effective technical solution to the above problems. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for intelligent aquaculture management and feeding optimization for offshore fish rafts. This invention aims to solve the problem that existing intelligent aquaculture management systems for offshore fish rafts adjust feeding plans based solely on power levels and equipment power consumption information when facing energy challenges, neglecting the physiological needs of the aquaculture species and resulting in decreased aquaculture efficiency. This invention achieves a smart balance between energy utilization efficiency and aquaculture efficiency.

[0008] In a first aspect, the present invention provides a method for intelligent aquaculture management and feeding optimization of offshore fish rafts, applied to an intelligent aquaculture management and feeding optimization system for offshore fish rafts. The intelligent aquaculture management and feeding optimization system for offshore fish rafts includes energy equipment and feeding equipment. The method for intelligent aquaculture management and feeding optimization of offshore fish rafts includes the following steps:

[0009] S1. Obtain the energy status information of the energy equipment and the biological characteristic information of the cultured object; the biological characteristic information includes the current growth stage of the cultured object;

[0010] S2. Obtain multiple preset candidate feeding adjustment schemes, and calculate the estimated energy saving value corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding device;

[0011] S3. Based on the biological characteristic information and combined with the preset physiological influence rules, calculate the aquaculture loss assessment value corresponding to each of the candidate feeding adjustment schemes; the aquaculture loss assessment value is obtained by weighting the preset estimated growth influence parameters with the weight coefficients corresponding to the growth stage;

[0012] S4. Taking the minimization of the aquaculture loss assessment value as the optimization objective, and under the condition that the energy status information meets the preset minimum operating constraints, determine the target feeding scheme from multiple candidate feeding adjustment schemes;

[0013] S5. Control the feeding equipment to perform feeding operations according to the target feeding plan.

[0014] The intelligent aquaculture management and feeding optimization method for offshore fish rafts provided by this invention can comprehensively consider the energy status and the biological characteristics of the aquaculture objects. Through optimization algorithms, it can achieve a balance between energy saving and aquaculture losses. Thus, even when there is insufficient power, it can still formulate a feeding plan that minimizes the impact on the aquaculture objects. This effectively solves the problem of declining aquaculture benefits due to simple energy saving in the prior art and realizes the optimization of intelligent aquaculture management.

[0015] Secondly, this invention provides an intelligent aquaculture management and feeding optimization system for offshore fish rafts, including energy equipment and feeding equipment, and further comprising:

[0016] The first acquisition module is used to acquire the energy status information of the energy equipment and the biological characteristic information of the aquaculture object; the biological characteristic information includes the current growth stage of the aquaculture object;

[0017] The second acquisition module is used to acquire multiple preset candidate feeding adjustment schemes and calculate the estimated energy saving value corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding device.

[0018] The calculation module is used to calculate the aquaculture loss assessment value corresponding to each of the candidate feeding adjustment schemes based on the biological characteristic information and in combination with the preset physiological influence rules; the aquaculture loss assessment value is obtained by weighting the preset estimated growth influence parameters with the weight coefficients corresponding to the growth stage;

[0019] The determination module is used to determine the target feeding scheme from multiple candidate feeding adjustment schemes, with the optimization objective of minimizing the aquaculture loss assessment value and under the condition that the energy state information meets the preset minimum operating constraints.

[0020] The control module is used to control the feeding device to perform feeding operations according to the target feeding scheme.

[0021] As can be seen from the above, the intelligent aquaculture management and feeding optimization method for offshore fish rafts provided by this invention overcomes the shortcomings of existing technologies, where the system adjusts the feeding plan solely based on power levels and equipment power consumption information when power is insufficient, neglecting the physiological needs of fish and leading to a decline in aquaculture efficiency. By comprehensively considering energy status and the biological characteristics of the aquaculture species, this application can achieve a balance between energy saving and aquaculture losses, ensuring that even under power constraints, a feeding plan with minimal impact on the aquaculture species can still be formulated. This effectively solves the problem of prolonged aquaculture cycles and reduced yields caused by simply saving energy in existing technologies, thereby improving the economic benefits and sustainability of intelligent aquaculture on offshore fish rafts.

[0022] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0023] Figure 1 This is a flowchart of an intelligent aquaculture management and feeding optimization method for offshore fish rafts provided in an embodiment of the present invention.

[0024] Figure 2 This is a schematic diagram of a marine fish raft intelligent aquaculture management and feeding optimization system provided in an embodiment of the present invention.

[0025] Label Explanation:

[0026] 100. First acquisition module; 200. Second acquisition module; 300. Calculation module; 400. Determination module; 500. Control module. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0028] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0029] In traditional offshore aquaculture systems, when energy supply is insufficient, the energy management module adjusts the feeding plan solely based on energy status and equipment power consumption information, without incorporating the physiological needs of the farmed organisms into the decision-making process. These physiological needs include specific nutritional requirements at different growth stages. This approach causes the feeding plan to deviate from the scientifically optimized plan based on aquaculture models, thereby affecting the normal growth of fish and feed conversion efficiency, specifically manifested as slower growth and reduced feed conversion efficiency.

[0030] Specifically, during several consecutive days of overcast and windless weather, the photovoltaic panel's power generation efficiency decreased, and the wind turbines stopped operating, resulting in insufficient electrical energy to maintain normal system operation. Energy status information showed that the battery pack's charge level continued to decline to a warning level. The energy management module simply reduced the daily feeding frequency from four times to twice, without considering that the fish were currently in a rapid growth phase and had high demands for the frequency and amount of nutrition. Consequently, the fish exhibited irregular feeding patterns, their digestive and absorptive functions were inhibited, and their growth rate slowed.

[0031] Furthermore, if the aforementioned problems are not addressed, the unmet physiological needs of farmed organisms will lead to prolonged growth cycles, decreased feed conversion efficiency, and increased susceptibility to disease. These effects will accumulate, ultimately resulting in reduced farmed output and impaired economic benefits, with the sustainability and economic efficiency of the farming system being significantly impacted.

[0032] For reference, see the appendix. Figure 1 This invention provides a method for intelligent aquaculture management and feeding optimization on offshore fish rafts, applied to an intelligent aquaculture management and feeding optimization system for offshore fish rafts. The system includes energy equipment and feeding equipment. The method comprises the following steps:

[0033] S1. Obtain energy status information from energy equipment and biological characteristic information from the cultured organisms; the biological characteristic information includes the current growth stage of the cultured organisms;

[0034] S2. Obtain multiple preset candidate feeding adjustment schemes, and calculate the estimated energy saving value corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding equipment;

[0035] S3. Based on biological characteristic information and combined with preset physiological influence rules, calculate the aquaculture loss assessment value corresponding to each candidate feeding adjustment plan; the aquaculture loss assessment value is obtained by weighting the preset estimated growth influence parameters with the weight coefficients corresponding to the growth stage;

[0036] S4. With the goal of minimizing the assessed value of aquaculture losses, and under the condition that the energy status information meets the preset minimum operating constraints, determine the target feeding scheme from multiple candidate feeding adjustment schemes;

[0037] S5. Control the feeding equipment to perform feeding operations according to the target feeding plan.

[0038] For ease of understanding, the following explains some key terms in this embodiment:

[0039] Intelligent Aquaculture Management and Feeding Optimization System for Offshore Fish Rafts: This system is a comprehensive platform integrating energy management, aquaculture monitoring, and feeding control functions. It aims to improve the efficiency and sustainability of offshore fish raft aquaculture through intelligent means. Its core function lies in its ability to dynamically adjust aquaculture strategies based on real-time data, especially in situations of energy constraints, balancing energy consumption and aquaculture benefits.

[0040] Energy equipment: This refers to all the equipment that provides electricity to offshore fish rafts, typically including wind turbines, photovoltaic panels, and battery packs for storing electrical energy. These devices work together to ensure the stable operation of various electrical appliances on the fish rafts.

[0041] Feeding equipment: This mainly refers to automatic feeders, whose function is to feed the aquaculture water at specified times and in specified quantities according to preset or system instructions. The power consumption characteristics of the feeding equipment are one of the key factors in evaluating energy savings in this solution.

[0042] Energy status information refers to real-time operating data of energy equipment, such as the current remaining charge of batteries, the real-time power generation of wind turbines, and the real-time power generation of photovoltaic panels. This information forms the basis for the system's energy management and feed optimization decisions.

[0043] Biological characteristics of farmed fish: This refers to physiological data on farmed fish, such as their current growth stage (e.g., juvenile stage, adult stage) and behavioral patterns (e.g., feeding activity level, schooling patterns). This information is crucial for assessing the impact of feeding programs on fish growth.

[0044] Candidate feeding adjustment schemes: These refer to the various feeding strategies preset by the system that can be selected when energy supply is tight. These schemes may include adjusting the feeding frequency, feeding amount, feeding time, etc., and each scheme corresponds to different energy consumption and impact on the farmed organisms.

[0045] Energy saving estimate: This refers to the amount of electricity saved compared to the original feeding plan after implementing a candidate feeding adjustment scheme. This value is obtained by accurately calculating the energy consumption data of the feeding equipment under different operating modes.

[0046] Physiological Influence Rules: A set of pre-established rules, based on extensive biological research and aquaculture practice, describing the expected effects of different feeding adjustments on fish growth rate, feed conversion efficiency, and health status.

[0047] Aquaculture Loss Assessment Value: A quantitative indicator obtained by multiplying the expected percentage of growth slowdown by the corresponding life stage importance weight. This indicator reflects the potential negative impact of implementing a feeding adjustment program on aquaculture profitability.

[0048] Minimum operating constraints: These refer to the minimum electrical threshold required to ensure the continuous operation of the most basic life support equipment (such as oxygen pumps) and control systems on the fish raft, for example, the battery charge should not be less than 20%.

[0049] Targeted feeding scheme: An emergency feeding strategy that minimizes total aquaculture losses while meeting minimum operational constraints.

[0050] This embodiment provides a method for intelligent aquaculture management and feeding optimization on offshore fish rafts. This method is applied to an intelligent aquaculture management and feeding optimization system for offshore fish rafts, which includes energy equipment and feeding equipment. The method aims to solve the problem in offshore fish raft aquaculture where, when energy supply is insufficient, the system adjusts the feeding plan solely based on energy-saving needs, ignoring the physiological needs of the farmed organisms, leading to a decline in aquaculture efficiency.

[0051] In step S1, it is necessary to acquire the energy status information of the energy equipment and the biological characteristic information of the cultured organisms. The biological characteristic information includes the current growth stage of the cultured organisms. Energy status information can be acquired in various ways. For example, voltage and current sensors can be installed on batteries to monitor their voltage and current in real time, thereby calculating the remaining power. Alternatively, power sensors can be installed on wind turbines and photovoltaic panels to acquire real-time power generation. This sensor data can be periodically uploaded to the central control system. For the biological characteristic information of the cultured organisms, such as their growth stage, the length and weight of the fish can be measured manually at regular intervals, and then the growth stage can be determined based on a preset growth curve model. Alternatively, images of the fish school can be captured by underwater cameras, and image recognition technology can be used to analyze the size and number of fish to estimate the growth stage.

[0052] In step S2, multiple preset candidate feeding adjustment schemes need to be obtained, and the energy saving estimate corresponding to each candidate feeding adjustment scheme is calculated based on the power consumption characteristics of the feeding equipment. The candidate feeding adjustment schemes can be pre-stored in the system's database. For example, Scheme A: feeding twice a day, with a 10% reduction in feeding amount each time; Scheme B: feeding three times a day, with the same feeding amount each time; Scheme C: feeding twice a day, with the same feeding amount each time, but a shorter feeding time. The power consumption characteristics of the feeding equipment can be obtained from the calibration data at the time of equipment leaving the factory. This data records in detail the actual energy consumption of the feeder under different feeding amounts, feeding times, and motor speeds. For example, if the original scheme feeds four times a day, running for 5 minutes each time, with a power of P watts, and the new scheme feeds three times a day, running for 5 minutes each time, with a power of P watts, then the energy saving can be calculated as (4 × P × 5 minutes) - (3 × P × 5 minutes).

[0053] In step S3, the aquaculture loss assessment value corresponding to each candidate feeding adjustment plan needs to be calculated based on biological characteristic information and preset physiological impact rules. The aquaculture loss assessment value is obtained by weighting preset estimated growth impact parameters with weight coefficients corresponding to the growth stages. The physiological impact rules are summarized based on a large amount of biological research and aquaculture practice experience. For example, the rules may include: "Reducing the feeding amount by 10% during the juvenile stage may lead to a 2% slowdown in growth rate"; "Reducing the feeding amount by 10% during the adult stage may lead to a 0.5% slowdown in growth rate". At the same time, the system will also apply a life stage importance weight table according to the current age and growth stage of the fish population. For example, the juvenile stage is a critical stage for fish growth and development, and any nutritional deficiency may have an irreversible impact on its future growth. Therefore, the slowdown in growth during the juvenile stage will be given a higher weight (e.g., a weight coefficient of 1.5), while the slowdown in growth during the adult stage will be relatively lower (e.g., a weight coefficient of 0.8). Finally, by multiplying the expected percentage of slowdown in growth by the corresponding life stage importance weight, a weighted aquaculture loss index is obtained. For example, if the scheme results in a 2% slowdown in the growth of juvenile fish, with a weight of 1.5, then the loss index is 2% × 1.5 = 3%.

[0054] In step S4, minimizing the assessed value of aquaculture loss is the optimization objective. Under the condition that the energy state information meets the preset minimum operating constraints, a target feeding scheme is determined from multiple candidate feeding adjustment schemes. This selection process can be understood as an optimization problem: minimizing the aquaculture loss index, while the constraint is that the sum of the current battery capacity and the expected energy saving is greater than or equal to the minimum operating energy threshold. For example, if the minimum operating energy threshold is 20% of the total battery capacity, the system will screen all schemes that meet this energy requirement and select the scheme with the smallest assessed value of aquaculture loss as the target feeding scheme.

[0055] In step S5, the feeding device is controlled to perform the feeding operation according to the target feeding plan. Once the optimal emergency feeding strategy is determined, the decision module will immediately send instructions to the control unit of the automatic feeder through the wireless communication module (e.g., a wireless transceiver based on the LoRaWAN protocol), so that it operates according to the new feeding time, frequency and feeding amount.

[0056] The core technological concept of this solution lies in establishing an "energy-biological state linkage decision-making module," which transcends the traditional limitation of treating energy management and aquaculture management as separate entities. When the power supply of offshore fish rafts is strained due to fluctuations in natural conditions, this module no longer simply reduces feeding based on remaining power. Instead, it proactively collects and comprehensively analyzes real-time power supply information (such as battery charge and power generation) and the life status information of the fish population (such as water quality parameters, behavioral patterns, and growth stages). Based on this, it evaluates multiple refined emergency feeding adjustment plans and quantifies the power savings of each plan and its weighted impact on fish growth and health. Ultimately, the module selects a strategy that minimizes total aquaculture losses while meeting the minimum operating power requirements of the fish rafts, thereby achieving a smart balance between energy efficiency and aquaculture benefits under extreme conditions of power constraints.

[0057] The following example will provide a more detailed explanation of the above technical solution:

[0058] Imagine an offshore aquaculture raft equipped with wind turbines, solar panels, and battery banks for energy, and an automatic feeder for feeding. Currently, the raft is experiencing several days of continuous rain and calm weather, leading to a shortage of energy supplies.

[0059] First, in step S1, the system acquires real-time energy status information of the energy equipment. For example, the battery currently has 30% remaining charge, and the real-time power generation of the wind turbine and photovoltaic panels is low. Simultaneously, the system acquires the biological characteristics of the aquaculture objects, and through historical data and image recognition, determines that the fish currently being farmed are in the juvenile stage.

[0060] Next, in step S2, the system will obtain multiple preset candidate feeding adjustment schemes. For example, the system has preset the following three schemes:

[0061] Option 1: Reduce the number of feedings per day from 4 to 3, while keeping the amount of food fed each time the same. Based on the power consumption characteristics of the feeding equipment, the estimated energy savings of this option are 10 kWh.

[0062] Option 2: Reduce the number of feedings per day from 4 to 2, while keeping the amount of food fed each time the same. Calculations show that this option can save an estimated 20 kWh of energy.

[0063] Option 3: Reduce the number of feedings per day from 4 to 3, and reduce the amount of food fed each time by 10%. Calculations show that this option can save an estimated 15 kWh of energy.

[0064] Then, in step S3, the system calculates the aquaculture loss assessment value corresponding to each candidate feeding adjustment scheme based on the biological characteristic information of the cultured fish (juvenile stage) and the preset physiological influence rules. Since the fish are in the juvenile stage, they are more sensitive to nutritional needs, so the slowdown in growth during the juvenile stage will be given a higher weight (e.g., a weighting coefficient of 1.5).

[0065] According to the rules of physiological influence:

[0066] Option 1: Reducing feeding by one session may slow the growth rate of juvenile fish by 1%, so the estimated loss for aquaculture is 1% × 1.5 = 1.5%.

[0067] Option 2: Reducing feeding by two times may slow the growth rate of juvenile fish by 3%, so the estimated loss for aquaculture is 3% × 1.5 = 4.5%.

[0068] Option 3: Reducing feeding by one time and decreasing the amount of feed by 10% may slow the growth rate of juvenile fish by 2%, so the estimated loss for aquaculture is 2% × 1.5 = 3%.

[0069] In step S4, the system optimizes by minimizing the assessed value of aquaculture losses and determines the target feeding scheme under the condition that the energy status information meets the preset minimum operating constraints. It is assumed that the preset minimum operating constraint is that the battery charge is not less than 20%.

[0070] The current battery level is 30%.

[0071] For Option 1: 30% + 10kWh (estimated savings) meets the minimum operating constraints, and the estimated loss from aquaculture is 1.5%.

[0072] For Option 2: 30% + 20kWh (estimated savings) meets the minimum operating constraints, and the estimated loss from aquaculture is 4.5%.

[0073] For Option 3: 30% + 15kWh (estimated savings) meets the minimum operating constraints, and the estimated loss from aquaculture is 3%.

[0074] By comparison, Option 1 has the lowest estimated aquaculture loss (1.5%). Therefore, the system determines Option 1 as the target feeding option.

[0075] Finally, in step S5, the system controls the feeding equipment to perform feeding operations according to the target feeding plan, that is, instructs the automatic feeder to feed 3 times a day, with the feeding amount remaining unchanged each time.

[0076] As can be seen from the above examples, the overall technical solution of this embodiment, when facing energy shortages, no longer simply reduces feeding based on remaining electricity. Instead, it proactively collects and comprehensively analyzes real-time power supply information and the fish's vital signs. It evaluates various refined emergency feeding adjustment schemes and quantifies the electricity saved by each scheme, as well as its weighted impact on fish growth and health. Ultimately, the module selects a strategy that minimizes total aquaculture losses while meeting the minimum operating electricity requirements of the fish raft, thus achieving a smart balance between energy efficiency and aquaculture benefits under extreme conditions of limited power. This dynamic adjustment capability is precisely designed to address the characteristics of large fluctuations in power supply and rapid environmental changes at sea.

[0077] The technical concept of this embodiment lies in establishing an "energy-biological state linkage decision-making module," which transcends the traditional limitation of treating energy management and aquaculture management independently. In the example above, when the power supply of the offshore fish raft is strained due to fluctuations in natural conditions, this module no longer simply reduces feeding based on remaining power. Instead, it proactively collects and comprehensively analyzes real-time power supply information (such as battery charge and power generation) and the life status information of the fish (such as growth stage). Based on this, it evaluates multiple refined emergency feeding adjustment schemes and quantifies the power savings of each scheme and its weighted impact on fish growth and health. Ultimately, the module selects a strategy that minimizes total aquaculture losses while meeting the minimum operating power requirements of the fish raft, thereby achieving a smart balance between energy utilization efficiency and aquaculture benefits under extreme conditions of power constraints.

[0078] Compared to existing technologies, when fish rafts encounter several consecutive days of rainy weather, accompanied by calm or light winds, the primary energy-saving strategy of the energy management module is simply to adjust the original feeding schedule. For example, reducing the planned four feedings per day to two, or even canceling energy-intensive nighttime feedings altogether. This decision is based entirely on a simple logic: reducing the number of times equipment operates lowers total energy consumption, thereby extending battery life and ensuring uninterrupted power to core equipment. However, this approach hides a deeper problem: when making the decision to reduce feeding frequency, the system relies solely on the current power level and equipment power consumption information, failing to consider the crucial variable of the fish's physiological needs. This change has a direct negative impact on the farmed organisms, leading to slower growth rates, potentially inducing disease, and ultimately resulting in a longer farming cycle and reduced yields.

[0079] This embodiment incorporates biological characteristic information and physiological influence rules of the farmed organisms, and uses minimizing the assessed value of farming losses as the optimization objective. This ensures maximum farming efficiency while meeting energy constraints. This decision-making mechanism, which comprehensively considers both energy and biological conditions, avoids the decline in farming efficiency caused by a single energy-saving logic in existing technologies. It achieves a smart balance between energy utilization efficiency and farming benefits, representing a significant technological advancement.

[0080] In some embodiments, step S2, which involves calculating the estimated energy savings for each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding device, includes the following specific steps:

[0081] S21. Based on the power consumption characteristics of the feeding equipment, calculate the initial energy saving estimate corresponding to each candidate feeding adjustment scheme;

[0082] S22. Obtain key information about the battery in the energy device; key information includes internal temperature, current voltage, and cumulative discharge cycle count;

[0083] S23. Based on the key information, adjust the battery's charge / discharge efficiency factor and effective usable capacity factor. Specific steps include:

[0084] S231. Adjust the charge / discharge efficiency factor of the battery according to the internal temperature of the battery;

[0085] S232. Adjust the effective available capacity factor of the battery based on the current voltage and cumulative discharge cycle count of the battery;

[0086] S24. Based on the adjusted charge / discharge efficiency factor and the adjusted effective available capacity factor, the initial energy saving estimate is corrected by multiplication to obtain the final energy saving estimate.

[0087] In the above scheme, firstly, based on the power consumption characteristics of the feeding equipment, the initial energy saving estimate corresponding to each candidate feeding adjustment scheme is calculated. The power consumption characteristics of the feeding equipment refer to the energy consumption pattern of the feeding equipment under different operating modes (e.g., different feeding amounts, feeding durations, or motor speeds). This power consumption data can be pre-calibrated experimentally and stored in the system, or obtained by real-time monitoring of the equipment's operating current and voltage. The initial energy saving estimate is calculated based on these power consumption characteristics by comparing the power consumption differences between different candidate feeding adjustment schemes and the current or benchmark feeding scheme. For example, it can be calculated by consulting a preset power consumption curve table or by using a simple power consumption model (such as average power multiplied by operating time).

[0088] Secondly, it's crucial to acquire key information about the batteries in energy devices. This key information comprises important parameters reflecting the battery's current state and health. Internal temperature can be acquired in real-time using built-in temperature sensors (e.g., thermistors or thermocouples), directly impacting the battery's chemical reaction rate, internal resistance, and charge / discharge efficiency. Current voltage can be monitored in real-time using high-precision voltage sensors, reflecting the battery's immediate charge level and load status. The cumulative number of discharge cycles can be recorded and statistically analyzed by the battery management system (BMS), serving as a vital indicator for assessing battery aging and lifespan. This information can be provided directly by the BMS or acquired in real-time and transmitted to the central control unit via a sensor network.

[0089] Secondly, based on the aforementioned key information, the battery's charge / discharge efficiency factor and effective usable capacity factor are adjusted. The charge / discharge efficiency factor refers to the energy conversion efficiency of the battery during charging and discharging, which is affected by factors such as temperature. The effective usable capacity factor refers to the proportion of actual charge the battery can provide in actual use to its nominal capacity, which is affected by factors such as aging and voltage. Adjusting these factors is to more accurately reflect the battery's actual performance under current conditions. For example, adjustments can be made based on preset lookup tables or empirical formulas. Specifically, the battery's charge / discharge efficiency factor is adjusted based on the battery's internal temperature. The battery's internal temperature has a significant impact on charge / discharge efficiency. In low-temperature environments, the battery's internal resistance increases, and charge / discharge efficiency decreases; in high-temperature environments, although efficiency may slightly improve, prolonged high temperatures will accelerate battery aging. Therefore, the system dynamically adjusts the charge / discharge efficiency factor based on the real-time monitored internal temperature by consulting preset temperature-efficiency curves or models to more accurately reflect energy loss. Simultaneously, the battery's effective usable capacity factor is adjusted based on the battery's current voltage and cumulative discharge cycle count. The current voltage of a battery directly reflects its immediate state of charge, while the cumulative number of discharge cycles reflects the battery's long-term aging. As the number of cycles increases, the battery's actual usable capacity gradually decreases. The system dynamically adjusts the effective usable capacity factor based on the current voltage and the cumulative number of discharge cycles, combined with a preset battery health model or capacity decay curve, to accurately assess the battery's actual usable capacity in its current state.

[0090] Finally, based on the adjusted charge / discharge efficiency factor and the adjusted effective available capacity factor, the initial energy saving estimate is corrected through multiplication to obtain the final energy saving estimate. The initial energy saving estimate only considered the power consumption of the equipment, without considering the battery's own losses and actual available capacity. By incorporating the adjusted charge / discharge efficiency factor and effective available capacity factor into the calculation, the initial estimate can be corrected. For example, the initial estimate can be multiplied by the charge / discharge efficiency factor and the effective available capacity factor to obtain a final energy saving estimate that is closer to reality and takes into account battery losses and capacity decay. This correction ensures that the energy saving assessment is more accurate and reliable.

[0091] This application's solution incorporates the real-time status and aging characteristics of the battery into the energy-saving estimation calculation, providing a more realistic energy efficiency assessment and thus offering more reliable energy data support for the formulation of emergency feeding strategies. Specifically, when calculating the energy-saving estimate corresponding to each candidate feeding adjustment scheme, the system first calculates the initial energy-saving estimate corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding equipment, providing a basic energy-saving estimate. However, this initial estimate may overlook the actual performance changes of the battery. To compensate for this deficiency, the system further acquires key information about the battery in the energy equipment, including internal temperature, current voltage, and cumulative discharge cycle count. These parameters directly capture the real-time status and aging effects of the battery. Subsequently, the system dynamically adjusts the battery's charge / discharge efficiency factor and effective usable capacity factor based on this key information. Internal temperature is used to adjust the charge / discharge efficiency factor because temperature fluctuations alter the battery's chemical reaction efficiency; current voltage and cumulative discharge cycle count are used to adjust the effective usable capacity factor because voltage reflects the immediate charge level, while the cycle count indicates capacity decay due to battery life. Finally, based on these adjusted charge / discharge efficiency factors and adjusted effective available capacity factors, the initial energy saving estimate is corrected through multiplication to obtain the final energy saving estimate. This correction process ensures that the energy saving estimate comprehensively considers the actual constraints of the battery, such as efficiency loss due to temperature or capacity reduction caused by aging, thereby improving the accuracy of decision-making. In this way, the system can more accurately assess the actual impact of different feeding adjustment schemes on the energy system, providing a solid data foundation for subsequently determining the target feeding scheme from multiple candidate feeding adjustment schemes under the condition that the energy state information meets the preset minimum operating constraints. This ensures that the final determined target feeding scheme not only considers minimizing aquaculture losses but also fully considers the actual availability and sustainability of energy.

[0092] As a specific implementation method, assume that the energy equipment of the offshore fish raft uses lithium iron phosphate battery packs and is equipped with a battery management system (BMS). When it is necessary to calculate the estimated energy savings of each candidate feeding adjustment scheme, the system first calculates an initial estimated energy savings based on the rated power of the feeding equipment and the expected operating time under different feeding schemes. For example, if a scheme reduces the number of daily feedings, thereby reducing the total operating time of the feeding equipment motor, the system will calculate a preliminary energy saving value based on the motor power and the reduced operating time. Simultaneously, the BMS monitors and provides key battery information in real time. For example, it obtains the battery's internal temperature through a built-in NTC thermistor, acquires the current voltage through a high-precision voltage sampling module, and records the cumulative discharge cycle count through an internal counter. After receiving this key information, the system will make the following adjustments: If the current battery internal temperature is 5°C, the system will consult a preset temperature-efficiency comparison table. This table may show that the charge / discharge efficiency factor of the lithium iron phosphate battery is 0.92 at 5°C and 0.95 at 25°C. The system will then adjust the charge / discharge efficiency factor to 0.92. If the current battery voltage is 3.2V (single cell) and the cumulative discharge cycle count has reached 1500, the system will adjust the effective usable capacity factor to 0.85 based on a preset battery health model. This model may indicate that after 1500 cycles, the battery's effective usable capacity has decreased to 85% of its nominal capacity, and the actual discharge capacity is further limited at 3.2V. Finally, the system will multiply the initial energy saving estimate by the adjusted charge / discharge efficiency factor and the effective usable capacity factor. For example, if the initial estimated energy saving is 1000Wh, the final energy saving estimate will be revised to 1000Wh × 0.92 × 0.85 = 782Wh. This revised value will serve as a more accurate energy saving estimate for subsequent feed plan optimization decisions.

[0093] Through the above technical solution, this application effectively solves the problem that traditional energy-saving estimates based on equipment power consumption cannot accurately reflect the actual range contribution of batteries. By acquiring key information such as the battery's internal temperature, current voltage, and cumulative discharge cycle count in real time, and dynamically adjusting the battery's charge / discharge efficiency factor and effective available capacity factor accordingly, this application can accurately correct the initial energy-saving estimate. This makes the final energy-saving estimate closer to the actual performance of the battery under the current environment and aging conditions, avoiding overestimation or underestimation of energy savings due to battery performance fluctuations. Therefore, in the subsequent feeding scheme optimization process, the system can make decisions based on more reliable energy data, ensuring that, under the premise of meeting minimum operating constraints, the determined target feeding scheme can minimize aquaculture losses and effectively manage energy consumption, avoiding unexpected power outages or unnecessary feeding reductions due to inaccurate energy estimates, thereby improving the robustness and decision-making accuracy of the intelligent aquaculture management and feeding optimization system for offshore fish rafts.

[0094] In some embodiments, the biometric information also includes the behavioral characteristics of the cultured object; these characteristics are acquired via pre-deployed optical sensors.

[0095] The specific steps in step S3 include:

[0096] S31. Evaluate the data quality of the corresponding portion of biometric information obtained from optical sensors, and obtain the data quality evaluation result;

[0097] S32. Based on the data quality assessment results, revise the biometric information and physiological influence rules to obtain the revised biometric information and adjusted physiological influence rules;

[0098] S33. Combining the revised biological characteristic information and the adjusted physiological impact rules, calculate the aquaculture loss assessment value corresponding to each candidate feeding adjustment scheme.

[0099] The behavioral characteristics of the farmed fish refer to dynamic physiological indicators such as their activity patterns, feeding status, and school density under specific environmental conditions. These behavioral characteristics can more comprehensively and in real-time reflect the physiological health status of the fish and their response to environmental changes, thus compensating for the limitations of relying solely on information from the growth stage. These behavioral characteristics can be acquired in various ways, such as through video analysis using underwater cameras, or through non-contact monitoring using optical devices such as infrared sensors and lidar.

[0100] The optical sensors are devices used to capture light signals and convert them into electrical signals, which are used in offshore fish farm environments to acquire data on the behavior of fish schools. These sensors may include underwater cameras to capture images of fish activity, or photoelectric sensors to detect the movement or distribution of fish schools.

[0101] In step S31, the data quality of the corresponding portion of the biometric information derived from the optical sensor is evaluated to obtain a data quality assessment result. This aims to identify and quantify potential errors, noise, or distortions in the sensor data. Data quality assessment can be based on various indicators, such as analyzing image sharpness, contrast, and signal-to-noise ratio to determine the quality of image data, or assessing the reliability of water quality sensor data by detecting the stability and abnormal fluctuations of sensor readings.

[0102] In step S32, based on the data quality assessment results, the biometric information and the physiological influence rules are corrected to obtain corrected biometric information and adjusted physiological influence rules. This is to adaptively adjust the information and rules input to the decision-making system when data quality is poor, thereby reducing the negative impact of low-quality data on decision-making. Correcting biometric information may include filtering, interpolation, or weighted averaging of the original data; adjusting physiological influence rules can be done by selecting a more conservative rule set based on the data quality level, or by dynamically adjusting the parameters in the rules.

[0103] In step S33, the revised biological characteristic information and the adjusted physiological impact rules are combined to calculate the aquaculture loss assessment value corresponding to each candidate feeding adjustment scheme, ensuring that the final assessment result is based on more reliable and accurate biological information and rules more adapted to the current data quality. This combination method can be to use the revised biological characteristic information as the input parameter of the adjusted physiological impact rules, or to fuse the two through weighted calculations or other methods to obtain a more accurate aquaculture loss assessment.

[0104] This application's solution, by introducing mechanisms for data quality assessment, biometric information correction, and physiological impact rule adjustment, can provide a more accurate assessment of aquaculture losses, more closely reflecting the true physiological state of fish populations. This provides more reliable biological data support for the formulation of emergency feeding strategies. Specifically, in the aforementioned intelligent aquaculture management and feeding optimization method for offshore fish farms, when the biometric information (such as the behavior of farmed organisms) acquired by optical sensors is affected by the deterioration of the aquatic environment, the system no longer blindly uses the raw data. Instead, it first assesses the quality of this data, identifying the degree of uncertainty or distortion. Subsequently, based on the assessment results, the system dynamically corrects the raw biometric information, for example, by filtering to remove noise or reducing its weight in decision-making when the data quality is extremely low. Simultaneously, the system adjusts the preset physiological impact rules to make them more conservative when data uncertainty is high; for example, by using broader or safer estimates when predicting the impact on fish population growth. Finally, by combining these corrected biometric information and adjusted physiological impact rules, the system can calculate a more accurate and reliable aquaculture loss assessment value. This mechanism enables the entire feeding optimization method to maintain the robustness and accuracy of decision-making when facing complex and ever-changing marine environments, avoiding suboptimal feeding decisions due to misjudgment of biological information, and thus maximizing aquaculture benefits even under energy-constrained conditions.

[0105] As a specific implementation method, the solution of this application can be implemented as follows: In the intelligent aquaculture management and feeding optimization system for offshore fish rafts, the central control unit (e.g., a high-performance industrial-grade microcontroller, such as an STM32H7 series chip based on an ARM Cortex-M7 core) periodically or under specific environmental conditions triggers data quality assessments of optical water quality sensors (e.g., fluorescence dissolved oxygen sensors) and underwater cameras. For optical water quality sensors, the system can monitor the attenuation of their internal optical path, for example, by analyzing the ratio between the driving current of the light-emitting diode and the intensity of the reference light signal received by the photodetector. If this ratio continuously deviates from the factory calibration value, it indicates that the optical window may be contaminated. In addition, the system can also activate the sensor-integrated micro-vibrator or local water flow flushing device for brief physical cleaning, and then compare the sensor readings before and after cleaning. If the reading after cleaning shows a significant improvement in a short period of time, it is determined that there was previous interference from deposits, and the difference between the stable reading after cleaning and the stable reading before cleaning is taken as a possible systematic deviation in the current data of the sensor. For underwater cameras, their image processing modules (e.g., edge computing modules equipped with NVIDIA Jetson Nano) continuously analyze captured images, evaluating their sharpness, contrast, and noise levels. This is achieved by analyzing the texture sharpness of preset static background areas in the image (such as edge density after local binarization or the contrast of the gray-level co-occurrence matrix). If these feature values ​​continue to decrease, it indicates that image sharpness is compromised. Simultaneously, the system also calculates the image's global noise level and contrast, such as the image's global average gray value and standard deviation, as well as the image's signal-to-noise ratio. When the water is turbid, the image's global contrast decreases, and noise increases. The system uses the image quality degradation index as a quantitative indicator of image data distortion.

[0106] Based on the self-diagnostic results of the optical sensors and the image quality assessment results of the underwater cameras, the central control unit dynamically assigns a real-time "confidence" score to each biological data source. For example, the confidence score of an optical sensor can be calculated based on the ratio of the quantified deviation value to the maximum permissible deviation of the sensor's range, while the confidence score of an underwater camera can be calculated based on the image quality degradation index. These confidence scores are updated at a frequency matching the data acquisition frequency and stored in the memory of the central control unit.

[0107] When the "Energy-Biological State Linked Decision Module" assesses the life status of fish, it no longer simply uses raw sensor data. Instead, it weights the data based on the real-time confidence level of each data source and corrects the parameters in the "Fish Physiological Impact Judgment Rules." For example, for key water quality parameters such as dissolved oxygen, if multiple sensors exist, the system will perform a weighted average of the different data sources based on their confidence levels to obtain a more reliable comprehensive water quality parameter. If only optical sensors are available, their readings will be corrected based on the confidence level. When analyzing fish activity and aggregation patterns, the image processing module adjusts the weight of its analysis results based on the confidence level of the underwater camera. When the confidence level is low, the system reduces its reliance on image analysis results and combines other non-optical data for auxiliary judgment. Furthermore, when applying the "Fish Physiological Impact Judgment Rules," the decision module introduces a "data uncertainty correction factor," which is inversely proportional to the average confidence level of all biological data sources. When the average confidence level is low, the correction factor increases, causing the "expected growth slowdown percentage" or "stress risk coefficient" predicted by the rules to be adjusted upwards, thus prompting the system to choose a more conservative feeding strategy. Through this weighted processing and rule correction, the system can effectively distinguish between the bias of the sensor itself and the actual physiological changes in the fish population.

[0108] Through the aforementioned technical solution, this application effectively addresses the problem of inaccurate biological characteristic information caused by the deterioration of optical sensor data quality in the complex environment of offshore fish rafts. By real-time evaluation of data quality, correction of biological characteristic information, and dynamic adjustment of physiological influence rules, the system can obtain more reliable biological data that more closely reflects the actual physiological state of fish populations. This significantly improves the accuracy and reliability of calculating aquaculture loss assessment values, thereby avoiding decision-making biases caused by misjudgments of biological information. Ultimately, under extreme conditions of limited energy, the system can determine the target feeding scheme that truly minimizes aquaculture losses based on a more accurate aquaculture loss assessment, improving the robustness and decision-making quality of intelligent aquaculture management and feeding optimization methods for offshore fish rafts, and ensuring the maximization of aquaculture benefits.

[0109] In some embodiments, the specific steps in step S32 include:

[0110] S321. Based on the data quality assessment results, determine the data quality level to which the data quality assessment results belong, based on multiple preset data quality thresholds; wherein, multiple data quality thresholds correspond to multiple data quality levels;

[0111] S322. Based on the data quality level, select the corresponding filtering method and filter the biometric information to obtain the corrected biometric information;

[0112] S323. Based on the data quality level, select the corresponding physiological impact rule from the preset set of physiological impact rules and use it as the adjusted physiological impact rule.

[0113] Determining the data quality level to which the data quality assessment result belongs refers to mapping the potentially continuous or multidimensional data quality assessment result obtained in step S31 to a discrete, manageable quality level by comparing or classifying it with multiple preset data quality thresholds. The data quality assessment result can be a comprehensive numerical score, such as a confidence score from 0 to 100, or a vector composed of multiple indicators (such as signal-to-noise ratio, sharpness, and bias quantification). The multiple preset data quality thresholds are pre-defined boundary points used to divide the data quality assessment result into different level intervals. These thresholds can be determined based on historical data analysis, expert experience, or system performance requirements. For example, 0-40 points can be classified as a "low" quality level, 41-70 points as a "medium" quality level, and 71-100 points as a "high" quality level. Through this grading process, complex data quality information can be simplified into discrete levels that are easier for subsequent decision-making.

[0114] Selecting a corresponding filtering method and applying it to the biometric information to obtain corrected biometric information refers to dynamically selecting and applying the most suitable data processing algorithm based on the determined data quality level. The filtering method aims to eliminate or reduce noise, outliers, or systematic biases in the biometric information, thereby improving the accuracy and reliability of the information. For example, when the data quality level is low, a stronger data smoothing algorithm, such as median filtering, Gaussian filtering, or Kalman filtering, can be selected to remove noise to the maximum extent; when the data quality level is high, a mild smoothing process, such as moving average filtering, or no filtering can be performed to preserve the details of the original data. Furthermore, filtering methods may also include data interpolation, outlier removal, or model-based data correction.

[0115] Selecting a corresponding physiological impact rule from a pre-defined set of physiological impact rules, and using it as the adjusted physiological impact rule, refers to choosing the rule that best reflects the reliability of the current data from a set containing multiple physiological impact rules, based on the data quality level. The pre-defined set of physiological impact rules is a pre-established set of mathematical models or logical judgments used to describe the impact of different feeding adjustment schemes on the physiological state of farmed animals (such as growth rate, feed conversion rate, stress response, etc.). These rules may differ in parameter settings, model complexity, or degree of conservatism. For example, when the data quality level is low, the system may tend to select a more conservative physiological impact rule, which may assess the negative impact of insufficient feeding more severely, or introduce a larger uncertainty correction factor to reduce decision-making risk; when the data quality level is high, the system may select a more refined and accurate physiological impact rule, which can more accurately predict the impact of feeding schemes on the physiological state of farmed animals.

[0116] This application's solution refines the data quality assessment results into different data quality levels and dynamically adjusts the filtering methods for biometric information and the selection of physiological influence rules based on these levels, thereby achieving refined correction of biometric information and physiological influence rules. Specifically, in the intelligent aquaculture management and feeding optimization method for offshore fish farms, when the data quality assessment result obtained in step S31 is uncertain, this solution first maps the assessment result to a preset data quality level. For example, if the image quality of the behavioral data of the aquaculture object acquired by the optical sensor decreases due to water turbidity, its data quality assessment result may be judged as "medium" or "low" quality level. Subsequently, the system selects an appropriate filtering method to process the biometric information based on this data quality level. For example, a stronger denoising algorithm is applied to the behavioral data of the "low" quality level to reduce misjudgments. At the same time, the system also selects a physiological influence rule from the preset set of physiological influence rules that best matches the reliability of the current data based on the data quality level. For example, for data of "low" quality, a more conservative physiological impact rule might be chosen, which considers greater uncertainty or potential risk when assessing the impact of feeding programs on aquaculture losses. Through this tiered, adaptive correction mechanism, this approach ensures that the biometric information and physiological impact rules used to calculate aquaculture loss assessments remain highly reliable and applicable despite fluctuations in data quality. Compared to simply correcting biometric information and physiological impact rules in the basic approach, this more accurately reflects the true physiological state of the farmed organisms and the potential impact of feeding programs, thus providing a more solid data foundation for subsequently determining target feeding programs.

[0117] The following is a concrete example to illustrate this. Suppose that the intelligent management system of an offshore fish farm acquires information on the behavior of the farmed organisms through underwater cameras, and the image processing module evaluates the data quality. When water transparency decreases due to the proliferation of plankton, the image quality assessment result (e.g., image sharpness score) will decrease accordingly. The system will classify this sharpness score into different data quality levels based on several preset data quality thresholds. For example, if the sharpness score is below 40 points (out of 100), it is determined to be of "low" quality; if it is between 40 and 70 points, it is of "medium" quality; and if it is above 70 points, it is of "high" quality. When it is determined to be of "low" quality, the system will choose a strong filtering method, such as applying the Kalman filter algorithm to the fish movement trajectory data, to smooth the data to the maximum extent and predict the true trajectory, while possibly discarding some image frames with extremely low confidence. Meanwhile, the system will select a physiological impact rule from a pre-set set of physiological impact rules for "low" data quality scenarios. This rule may include a large "data uncertainty correction factor," giving higher weight to any slowed growth or stress response that may result from reduced feeding when calculating the assessment value of aquaculture losses. Conversely, if the clarity score is high and it is determined to be of "high" quality, the system may choose to perform mild smoothing or no filtering, and select a more refined and sensitive physiological impact rule to more accurately capture the subtle effects of feeding programs on the physiological state of the aquaculture species.

[0118] Through the aforementioned technical solution, this application can adaptively select the most suitable biometric information filtering method and physiological influence rules based on the level of refinement of the data quality assessment results, thereby effectively solving the problem of traditional correction methods lacking refined processing when data quality is uncertain. This makes the corrected biometric information more accurate and the adjusted physiological influence rules more in line with the actual situation, thus significantly improving the reliability of aquaculture loss assessment values. Ultimately, in the intelligent aquaculture management and feeding optimization method for offshore fish rafts, this helps the system to more accurately weigh the relationship between energy saving and aquaculture benefits in complex environments with limited energy, determine the target feeding scheme that truly minimizes aquaculture losses, and improve the decision-making robustness and economic efficiency of the entire intelligent aquaculture management system.

[0119] In some embodiments, the energy status information includes the current remaining charge of the battery;

[0120] The specific steps in step S4 include:

[0121] S41. Calculate the fluctuation range of the aquaculture loss assessment value and the fluctuation range of the energy saving estimate value corresponding to each candidate feeding adjustment plan;

[0122] S42. For each candidate feeding adjustment plan, obtain the upper limit of the fluctuation range of the aquaculture loss assessment value as the aquaculture loss under the most unfavorable condition, and obtain the lower limit of the fluctuation range of the energy saving estimate value as the energy saving under the most conservative condition.

[0123] S43. Under the condition that the sum of the current remaining battery power and the energy saving under the most conservative condition meets the minimum operating constraints, select the candidate feeding adjustment scheme with the least breeding loss under the most unfavorable condition from multiple candidate feeding adjustment schemes as the target feeding scheme.

[0124] Energy status information is the foundational data for energy management and decision-making within the system. It can take various forms, such as battery voltage, current, and charging / discharging power. In this application, this information specifically refers to the current remaining battery charge, which can be directly read by the battery management system (BMS) or calculated by integrating the battery voltage and current. This information is a key indicator for assessing the power supply capacity of the fish raft and directly impacts the energy feasibility assessment of subsequent feeding plans.

[0125] The calculation of the fluctuation range of the assessed aquaculture loss value and the estimated energy saving value for each candidate feeding adjustment scheme aims to quantify the potential risks and benefits of each candidate feeding adjustment scheme under uncertain conditions. The fluctuation range of the assessed aquaculture loss value reflects the degree of uncertainty in the impact of the same feeding scheme on aquaculture benefits under different environmental or biological conditions. For example, it can be determined through historical data analysis combined with statistical methods (such as standard deviation and confidence intervals). The fluctuation range of the estimated energy saving value represents the possible deviation between the actual energy savings and the estimated value due to changes in equipment power consumption, environmental factors, etc., during actual operation. This can be obtained through statistical analysis of historical power consumption data of the feeding equipment under different operating conditions.

[0126] For each candidate feeding adjustment scheme, the upper limit of the fluctuation range of the assessed aquaculture loss is obtained as the aquaculture loss under the worst-case scenario, and the lower limit of the fluctuation range of the estimated energy savings is obtained as the energy savings under the most conservative scenario. This step is to introduce a risk aversion mechanism in the decision-making process. Obtaining the upper limit of the fluctuation range of the assessed aquaculture loss means that the system will consider the worst-case scenario when assessing potential losses, thereby ensuring that even under adverse conditions, the selected scheme can control the losses within an acceptable range. For example, if the fluctuation range is [L_min, L_max], then L_max is taken as the aquaculture loss under the worst-case scenario. Similarly, obtaining the lower limit of the fluctuation range of the estimated energy savings is to ensure the conservatism of energy constraints and avoid actual energy shortages due to overly optimistic estimates. For example, if the fluctuation range is [E_min, E_max], then E_min is taken as the energy savings under the most conservative scenario.

[0127] Under the condition that the sum of the current remaining battery power and the most conservative energy savings meets the minimum operating constraint, the system selects the candidate feeding adjustment scheme that minimizes the aquaculture loss under the worst-case scenario from among multiple candidate feeding adjustment schemes as the target feeding scheme. This step is the final decision-making and screening process. The minimum operating constraint can be a preset battery power threshold, for example, the battery power cannot be lower than 20% to ensure the operation of basic equipment. The system first adds the current remaining battery power to the most conservative energy savings achievable by each candidate scheme to determine whether the minimum operating constraint is met. Only schemes that meet this condition are further considered. Among all schemes that meet the energy constraint, the system selects the one with the smallest "aquaculture loss under the worst-case scenario" as the final target feeding scheme. This screening mechanism ensures that under extreme conditions of limited energy, the system can prioritize the basic operation of the fish rafts while minimizing the negative impact on aquaculture efficiency.

[0128] This application's solution addresses the issue of unsound decision-making in uncertain environments by incorporating consideration of the fluctuation range of the assessed value of aquaculture losses and the estimated value of energy savings. First, the system no longer relies solely on a single estimated value of energy savings and assessed value of aquaculture losses. Instead, it comprehensively captures the uncertainty of these assessed values ​​in actual operation by calculating the fluctuation range of both the assessed value of aquaculture losses and the estimated value of energy savings for each candidate feeding adjustment scheme. This calculation of fluctuation ranges allows the system to more realistically reflect potential risks and benefits. Second, to address the worst-case scenario, the system uses the upper limit of the fluctuation range of the assessed value of aquaculture losses for each candidate feeding adjustment scheme as the aquaculture loss under the worst-case scenario. This ensures that the potential negative impacts are always measured with the most conservative and cautious approach when evaluating aquaculture benefits. Simultaneously, the lower limit of the fluctuation range of the estimated value of energy savings is used as the energy savings under the most conservative scenario. This allows the system to avoid overly optimistic estimates when judging energy feasibility, thus ensuring a more reliable energy supply. Finally, in the decision-making stage, the system compares the current remaining battery power with the sum of the most conservative energy savings achievable by each candidate scheme to ensure that preset minimum operating constraints are met. Only those solutions that meet this energy constraint are considered. Based on this, the system selects the candidate feeding adjustment scheme that minimizes aquaculture losses under the most unfavorable conditions from these eligible schemes as the target feeding scheme. This decision-making logic enables the system to not only guarantee the minimum operating requirements of the fish rafts in complex and variable environments with limited energy, but also to minimize the negative impact on aquaculture efficiency under controllable risks, thereby achieving more robust and reliable intelligent aquaculture management and feeding optimization. In this way, the solution proposed in this application further improves the robustness and reliability of decision-making on the basis of basic feeding optimization methods, and is particularly suitable for application scenarios such as offshore fish rafts with variable environments and unstable energy supplies.

[0129] As a specific implementation method, suppose the intelligent management system of an offshore fish raft needs to determine the next feeding plan during a period of energy shortage (e.g., continuous rainy and windless weather). The system first obtains the current remaining battery power, for example, 30%. Simultaneously, the system has generated multiple candidate feeding adjustment plans, such as "Plan A: Feed twice a day, 5 minutes each time," and "Plan B: Feed three times a day, 4 minutes each time," etc. For each candidate plan, the system performs the following steps: First, it calculates the fluctuation range of the aquaculture loss assessment value and the fluctuation range of the energy saving estimate corresponding to each candidate feeding adjustment plan. For example, for Plan A, by analyzing historical data, the system may conclude that its aquaculture loss assessment value fluctuation range is [1.5%, 2.5%] (indicating a potential growth slowdown of 1.5% to 2.5%), and the energy saving estimate fluctuation range is [10kWh, 12kWh]. For Plan B, the aquaculture loss assessment value fluctuation range may be [1.0%, 2.0%], and the energy saving estimate fluctuation range is [8kWh, 10kWh]. Next, for each candidate feeding adjustment scheme, the system obtains the upper limit of the fluctuation range of the estimated breeding loss as the breeding loss under the worst-case scenario, and the lower limit of the fluctuation range of the estimated energy saving as the energy saving under the most conservative scenario. For example, for scheme A, the breeding loss under the worst-case scenario is 2.5%, and the energy saving under the most conservative scenario is 10 kWh. For scheme B, the breeding loss under the worst-case scenario is 2.0%, and the energy saving under the most conservative scenario is 8 kWh. Finally, under the condition that the sum of the current remaining battery power and the energy saving under the most conservative scenario meets the minimum operating constraint, the system selects the candidate feeding adjustment scheme with the smallest breeding loss under the worst-case scenario as the target feeding scheme from multiple candidate feeding adjustment schemes. Assume that the minimum operating constraint is that the battery power cannot be lower than 20%, and the current remaining power is 30%. For scheme A: current power 30% + saving 10 kWh. Assume 10 kWh corresponds to 5% of the battery capacity. Then the total power is 35%, meeting the minimum operating constraint of 20%. Its worst-case breeding loss is 2.5%. For Option B: Current power consumption is 30%+, saving 8kWh. Assuming 8kWh corresponds to 4% of the battery capacity, the total power consumption is 34%, meeting the minimum operating constraint of 20%. Its worst-case aquaculture loss is 2.0%. In this case, since Option B, under the premise of meeting the energy constraint, has a lesser worst-case aquaculture loss (2.0%) than Option A (2.5%), the system will select Option B as the target feeding option.

[0130] Through the aforementioned technical solution, this application effectively addresses the decision-making risks arising from energy supply uncertainty in its intelligent aquaculture management and feeding optimization method for offshore fish rafts. By specifying energy status information as the current remaining battery power and incorporating calculations of the fluctuation range of aquaculture loss assessments and energy saving estimates, the system no longer relies solely on a single estimate for decision-making but comprehensively considers various uncertainties. Obtaining the aquaculture loss under the worst-case scenario and the energy saving under the most conservative scenario allows the system to adopt more prudent and risk-averse strategies in decision-making, avoiding energy depletion or severe losses in aquaculture benefits due to optimistic estimations under extreme conditions. This optimization target determination method based on volatility analysis ensures that, in complex environments with limited energy, the determined target feeding scheme not only meets the minimum operating requirements of the fish rafts but also minimizes negative impacts on the growth and health of the farmed organisms, thereby significantly improving the robustness and reliability of the entire intelligent aquaculture management system.

[0131] In some embodiments, the specific steps in step S41 include:

[0132] S411. For each candidate feeding adjustment plan, obtain historical aquaculture loss data and historical energy saving data under similar environmental conditions and feeding strategies;

[0133] S412. Conduct statistical analysis on historical aquaculture loss data and historical energy saving data to obtain the distribution characteristics of aquaculture loss assessment values ​​and the distribution characteristics of energy saving estimates;

[0134] S413. Based on the corresponding distribution characteristics, determine the fluctuation range of the aquaculture loss assessment value and the fluctuation range of the energy saving estimate value.

[0135] First, for each candidate feeding adjustment scheme, historical aquaculture loss data and historical energy saving data corresponding to similar environmental conditions and feeding strategies are obtained. This step aims to provide a real and reliable data foundation for subsequent fluctuation range calculations, avoiding evaluations based on subjective assumptions or incomplete data, thereby improving the accuracy and objectivity of the evaluation. For example, a historical database can be established in the fish raft intelligent aquaculture management and feeding optimization system to continuously record actual aquaculture losses (such as slower growth rate, decreased feed conversion rate, disease incidence, etc.) and actual energy consumption data under different environmental conditions (such as water temperature, dissolved oxygen, weather conditions, etc.) and different feeding strategies (such as feeding amount, feeding frequency, feed type, etc.). When it is necessary to evaluate new candidate feeding adjustment schemes, the system can retrieve the historical records most similar to the current environmental conditions and candidate feeding strategies from this database. In addition, data interfaces can be established with external aquaculture data platforms or industry databases to obtain large-scale and diverse historical aquaculture data and energy consumption data. Through data cleaning, feature matching, and similarity calculation algorithms, historical datasets that highly match the current fish raft environment and candidate feeding schemes are selected.

[0136] Secondly, statistical analysis is performed on the acquired historical aquaculture loss data and historical energy saving data to obtain the distribution characteristics of the aquaculture loss assessment value and the energy saving estimate value. This step uses mathematical tools to quantify the uncertainty and variability of the data, making the extraction of distribution characteristics more scientific and objective, and reducing the interference of human factors. For example, descriptive statistical methods can be used, such as calculating the mean, median, standard deviation, variance, skewness, kurtosis, and other statistics of historical data, to characterize the central tendency and dispersion of the data. At the same time, histograms, kernel density estimation plots, etc., can be drawn to visually display the distribution pattern of the data. Furthermore, inferential statistical methods can also be used, such as probability distribution fitting, fitting the historical data to common statistical distribution models (such as normal distribution, gamma distribution, Weibull distribution, etc.), thereby obtaining the probability density function or cumulative distribution function of the aquaculture loss assessment value and the energy saving estimate value. These functions can accurately describe their distribution characteristics.

[0137] Finally, based on these obtained distribution characteristics, the fluctuation ranges of the assessed aquaculture losses and the estimated energy savings are determined. This step directly utilizes the results of statistical analysis to define the risk boundary, making subsequent decision-making more reliable and effectively reducing aquaculture risks that may result from estimation errors. For example, the fluctuation range can be determined based on the standard deviation or quantiles obtained from statistical analysis. The fluctuation range can be defined as the mean plus or minus two standard deviations (corresponding to approximately a 95% confidence interval), or a certain percentile of historical data (such as the 5th and 95th percentiles) can be used directly as the upper and lower limits of the fluctuation range. Alternatively, a prediction interval at a specific confidence level can be calculated based on a fitted probability distribution model. For example, for a normal distribution, a 90% or 95% prediction interval can be calculated, which can contain future aquaculture losses or energy savings with a certain probability. For an asymmetric distribution, its corresponding quantile interval can be calculated.

[0138] Through the above technical solution, this application effectively solves the problem of inaccurate assessment of aquaculture losses and energy savings caused by the lack of reliable historical data support in traditional methods. By acquiring historical data under similar environmental conditions and feeding strategies, and performing statistical analysis to obtain distribution characteristics, this application can more objectively and scientifically determine the fluctuation range of aquaculture loss assessment and energy savings estimate. This fluctuation range determination mechanism based on historical data analysis enables a more accurate assessment of aquaculture losses under the worst-case scenario and energy savings under the most conservative scenario for each candidate feeding adjustment scheme in the intelligent aquaculture management and feeding optimization method for offshore fish rafts, when energy status information includes the current remaining battery power. This significantly improves the reliability of target feeding scheme selection and reduces the uncertainty of aquaculture decisions caused by inaccurate fluctuation range estimation. Under extreme conditions of limited energy, the system can select a feeding scheme that meets the minimum operating constraints and minimizes aquaculture losses to the greatest extent based on more robust risk assessment results, thereby optimizing aquaculture efficiency while ensuring the basic operation of the fish rafts and avoiding irreversible negative impacts on aquaculture organisms caused by blindly saving energy.

[0139] Reference Appendix Figure 2 This invention provides an intelligent aquaculture management and feeding optimization system for offshore fish rafts (this system adopts the intelligent aquaculture management and feeding optimization method for offshore fish rafts described in the above embodiments; the specific process is described in the corresponding steps above), including energy equipment and feeding equipment, and further comprising:

[0140] The first acquisition module 100 is used to acquire energy status information of energy equipment and biological characteristic information of aquaculture objects; the biological characteristic information includes the current growth stage of the aquaculture objects;

[0141] The second acquisition module 200 is used to acquire multiple preset candidate feeding adjustment schemes and calculate the energy saving estimate corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding device.

[0142] The calculation module 300 is used to calculate the aquaculture loss assessment value corresponding to each candidate feeding adjustment plan based on biological characteristic information and in combination with preset physiological influence rules; the aquaculture loss assessment value is obtained by weighting the preset estimated growth influence parameters with the weight coefficients corresponding to the growth stage;

[0143] The determination module 400 is used to determine the target feeding scheme from multiple candidate feeding adjustment schemes with the optimization objective of minimizing the aquaculture loss assessment value and under the condition that the energy status information meets the preset minimum operating constraints.

[0144] The control module 500 is used to control the feeding equipment to perform feeding operations according to the target feeding plan.

[0145] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0146] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent aquaculture management and feeding optimization of offshore fish rafts, applied to an intelligent aquaculture management and feeding optimization system for offshore fish rafts, wherein the intelligent aquaculture management and feeding optimization system for offshore fish rafts includes energy equipment and feeding equipment, characterized in that, The intelligent aquaculture management and feeding optimization method for offshore fish rafts includes the following steps: S1. Obtain the energy status information of the energy equipment and the biological characteristic information of the cultured object; the biological characteristic information includes the current growth stage of the cultured object; S2. Obtain multiple preset candidate feeding adjustment schemes, and calculate the estimated energy saving value corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding device; S3. Based on the biological characteristic information and combined with the preset physiological influence rules, calculate the aquaculture loss assessment value corresponding to each of the candidate feeding adjustment schemes; the aquaculture loss assessment value is obtained by weighting the preset estimated growth influence parameters with the weight coefficients corresponding to the growth stage; S4. Taking the minimization of the aquaculture loss assessment value as the optimization objective, and under the condition that the energy status information meets the preset minimum operating constraints, determine the target feeding scheme from multiple candidate feeding adjustment schemes; S5. Control the feeding equipment to perform feeding operations according to the target feeding plan.

2. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 1, characterized in that, In step S2, the specific steps for calculating the estimated energy savings for each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding device include: S21. Based on the power consumption characteristics of the feeding device, calculate the initial energy saving estimate corresponding to each of the candidate feeding adjustment schemes; S22. Obtain key information about the battery in the energy device; S23. Based on the key information, adjust the charge / discharge efficiency factor and effective usable capacity factor of the battery; S24. Based on the adjusted charge / discharge efficiency factor and the adjusted effective available capacity factor, the initial energy saving estimate is revised to obtain the final energy saving estimate.

3. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 2, characterized in that, The key information includes internal temperature, current voltage, and cumulative discharge cycle count.

4. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 3, characterized in that, In step S23, the specific steps for adjusting the charge / discharge efficiency factor and effective usable capacity factor of the battery based on the key information include: S231. Adjust the charge / discharge efficiency factor of the battery according to the internal temperature of the battery; S232. Adjust the effective available capacity factor of the battery based on the current voltage and cumulative discharge cycle count of the battery.

5. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 2, characterized in that, The specific steps in step S24 include: Based on the adjusted charge / discharge efficiency factor and the adjusted effective available capacity factor, the initial energy saving estimate is corrected by multiplication to obtain the final energy saving estimate.

6. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 1, characterized in that, The biometric information also includes the behavioral characteristics of the cultured organisms; these behavioral characteristics are acquired through pre-deployed optical sensors. The specific steps in step S3 include: S31. Evaluate the data quality of the corresponding portion of the biometric information obtained from the optical sensor, and obtain the data quality evaluation result; S32. Based on the data quality assessment results, correct the biometric information and the physiological influence rules to obtain the corrected biometric information and the adjusted physiological influence rules; S33. Combining the corrected biological characteristic information and the adjusted physiological influence rules, calculate the aquaculture loss assessment value corresponding to each of the candidate feeding adjustment schemes.

7. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 6, characterized in that, The specific steps in step S32 include: S321. Based on the data quality assessment result, and using multiple preset data quality thresholds, determine the data quality level to which the data quality assessment result belongs; wherein, the multiple data quality thresholds correspond to multiple data quality levels; S322. Based on the data quality level, select the corresponding filtering method and filter the biometric information to obtain the corrected biometric information; S323. Based on the data quality level, select the corresponding physiological influence rule from the preset set of physiological influence rules, and use it as the adjusted physiological influence rule.

8. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 1, characterized in that, The energy status information includes the current remaining charge of the battery; The specific steps in step S4 include: S41. Calculate the fluctuation range of the aquaculture loss assessment value and the fluctuation range of the energy saving estimate value corresponding to each of the candidate feeding adjustment schemes; S42. For each of the candidate feeding adjustment schemes, obtain the upper limit of the fluctuation range of the aquaculture loss assessment value as the aquaculture loss under the most unfavorable condition, and obtain the lower limit of the fluctuation range of the energy saving estimate value as the energy saving under the most conservative condition. S43. When the sum of the current remaining power of the battery and the energy saving under the most conservative condition satisfies the minimum operating constraint, select the candidate feeding adjustment scheme with the least breeding loss under the most unfavorable condition from the multiple candidate feeding adjustment schemes as the target feeding scheme.

9. The intelligent aquaculture management and feeding optimization method for offshore fish rafts according to claim 8, characterized in that, The specific steps in step S41 include: S411. For each of the candidate feeding adjustment schemes, obtain historical aquaculture loss data and historical energy saving data under similar environmental conditions and feeding strategies; S412. Perform statistical analysis on the historical aquaculture loss data and the historical energy saving data to obtain the distribution characteristics of the aquaculture loss assessment value and the distribution characteristics of the energy saving estimate value; S413. Based on the corresponding distribution characteristics, determine the fluctuation range of the aquaculture loss assessment value and the fluctuation range of the energy saving estimate value.

10. A smart aquaculture management and feeding optimization system for offshore fish rafts, comprising energy equipment and feeding equipment, characterized in that, Also includes: The first acquisition module is used to acquire the energy status information of the energy equipment and the biological characteristic information of the aquaculture object; the biological characteristic information includes the current growth stage of the aquaculture object; The second acquisition module is used to acquire multiple preset candidate feeding adjustment schemes and calculate the estimated energy saving value corresponding to each candidate feeding adjustment scheme based on the power consumption characteristics of the feeding device. The calculation module is used to calculate the aquaculture loss assessment value corresponding to each of the candidate feeding adjustment schemes based on the biological characteristic information and in combination with the preset physiological influence rules; the aquaculture loss assessment value is obtained by weighting the preset estimated growth influence parameters with the weight coefficients corresponding to the growth stage; The determination module is used to determine the target feeding scheme from multiple candidate feeding adjustment schemes, with the optimization objective of minimizing the aquaculture loss assessment value and under the condition that the energy state information meets the preset minimum operating constraints. The control module is used to control the feeding device to perform feeding operations according to the target feeding scheme.

Images