Aquaculture system, method and aquaculture equipment based on fish growth prediction model

Abstract

The application relates to the technical field of intelligent aquaculture, in particular to an aquaculture system and method based on a fish growth prediction model and an aquaculture device. The system comprises a water quality sensor array, a data acquisition module and an intelligent control module. The water quality sensor array is arranged in a breeding pond to collect water quality parameters such as dissolved oxygen, pH value, water temperature, ammonia nitrogen and nitrite in real time. The data acquisition module detects the water quality parameters according to a preset threshold value and generates an evaluation result. The intelligent control module generates a device control instruction based on a fish growth cycle prediction model, a feeding optimization model and an energy-saving control model of an oxygenation machine, so as to realize linkage control of the oxygenation machine, a water pump, a feeding machine and other devices. The application can realize all-weather intelligent monitoring and control of breeding water quality, the scientific growth cycle prediction can improve the survival rate of aquatic products, the energy-saving control system realizes 15% to 20% energy consumption reduction by adopting frequency conversion speed regulation technology, supports multi-pond cluster management, and can effectively reduce the breeding cost.

Description

Technical Field

[0001] This invention relates to the field of smart aquaculture technology, specifically to an aquaculture system, method, and aquaculture equipment based on a fish growth prediction model. Background Technology

[0002] Aquaculture, as an important component of my country's agricultural economy, has shown a trend towards large-scale and intensive development in recent years. However, traditional freshwater pond aquaculture still relies primarily on manual pond inspections and experience-based judgment as management methods, facing numerous technical bottlenecks.

[0003] In terms of water quality monitoring, aquaculture workers usually obtain water quality parameters by taking samples at regular intervals and conducting laboratory tests. However, this method has limitations such as low monitoring frequency and delayed response, making it difficult to keep track of water quality changes in real time. In terms of risk warning, due to the lack of continuous data accumulation and intelligent analysis capabilities, it is impossible to predict the trend of water quality deterioration. Measures are often only taken after fish have already shown symptoms such as oxygen deficiency and surfacing, and ammonia poisoning, thus missing the best time for regulation. In terms of daily management, the amount of feed is mainly determined by experience, and there is a lack of a dynamic adjustment mechanism based on the feeding status of the fish and the water environment factors, which can easily lead to feed waste and water pollution.

[0004] With the rapid development of IoT, cloud computing, and AI technologies, intelligent upgrading of aquaculture has become an industry trend. While some existing water quality monitoring devices can collect basic parameters such as dissolved oxygen and pH online, they still have shortcomings in areas such as multi-parameter collaborative monitoring, accuracy of early warning models, and the linkage capabilities of control systems. Specifically, existing water quality monitoring systems are mostly limited to single-parameter threshold alarms, failing to establish quantitative correlation models between water quality parameters and fish growth status. At the control level, the linkage control logic between the monitoring system and equipment such as aerators and feeders is relatively simple, typically using fixed threshold triggering methods, lacking adaptive adjustment capabilities based on dynamic environmental changes. This leads to problems such as frequent aerator start-stops or feeding strategies not matching actual needs. Furthermore, existing systems are mostly designed for single ponds, with weak capabilities in multi-pond cluster management, data traceability, and decision support, making it difficult to meet the needs of large-scale farms for full-process information management. Summary of the Invention

[0005] In view of the above situation and to overcome the shortcomings of the existing technology, the purpose of this invention is to provide a fish farming system, method and equipment based on a fish growth prediction model, which effectively solves the problems that existing aquaculture cannot be dynamically adjusted based on fish growth and that existing systems are not suitable for multi-pond cluster management.

[0006] The technical solution of this invention to solve the technical problem is an aquaculture system based on a fish growth prediction model, applied to fish aquaculture equipment, comprising: a water quality sensor array configured to be deployed in an aquaculture pond for real-time collection of water quality parameter data; and a data acquisition module communicatively connected to the water quality sensor array, configured to perform the following steps: acquiring water quality parameter data collected by the water quality sensor array when a preset acquisition cycle arrives; performing anomaly detection and level determination on the water quality parameter data based on the water quality parameter data, a preset water quality threshold range, and a preset water quality assessment model to obtain a water quality assessment result; generating corresponding equipment control commands based on the water quality assessment result, current aquaculture status information, and a preset intelligent control model; and sending the equipment control commands to the target aquaculture equipment to cause the target aquaculture equipment to perform corresponding control operations.

[0007] In some embodiments, the water quality sensor array includes: a dissolved oxygen sensor for collecting dissolved oxygen concentration data in the aquaculture water; a pH sensor for collecting acidity / alkalinity data in the aquaculture water; a water temperature sensor for collecting temperature data in the aquaculture water; an ammonia nitrogen sensor for collecting ammonia nitrogen concentration data in the aquaculture water; and a nitrite sensor for collecting nitrite concentration data in the aquaculture water.

[0008] In some embodiments, based on the water quality parameter data, a preset water quality threshold range, and a preset water quality assessment model, anomaly detection and level determination are performed on the water quality parameter data to obtain a water quality assessment result, including: determining whether each water quality parameter exceeds the corresponding safety threshold range; if any water quality parameter exceeds the safety threshold range, determining the risk level according to the extent of the exceedance, wherein the risk level includes low risk, medium risk, and high risk; and generating a comprehensive water quality assessment result based on the exceedance status of each water quality parameter.

[0009] In some embodiments, based on the water quality assessment results, current aquaculture status information, and a preset intelligent control model, corresponding equipment control commands are generated, including: generating an aerator start command when the water quality assessment results indicate that the dissolved oxygen concentration is lower than a first preset threshold; generating an aerator stop command when the water quality assessment results indicate that the dissolved oxygen concentration is higher than a second preset threshold; generating a water pump start command or a water change command when the water quality assessment results indicate that the ammonia nitrogen concentration is higher than a third preset threshold; and generating a pH adjustment device start command when the water quality assessment results indicate that the pH value deviates from the normal range.

[0010] In some embodiments, the intelligent control model includes a fish growth cycle prediction model, a feeding optimization model, and an aerator energy-saving control model.

[0011] In some embodiments, the fish growth cycle prediction model is established based on the following formula:

[0012] in, For a specific growth rate, The maximum specific growth rate under optimal conditions. Let the water temperature suitability function be... To measure the actual water temperature, The dissolved oxygen suitability function To measure the dissolved oxygen concentration, It is a pH fitness function. To measure the pH level, It is the non-ionic ammonia suppression function. To measure the concentration of non-ionic ammonia, The nitrite inhibition function, This is for the actual measurement of nitrite concentration.

[0013] In some embodiments, the water temperature suitability function is a Gaussian function:

[0014] in, For the optimal growth temperature, For temperature tolerance width, and This is the survival temperature limit.

[0015] In some embodiments, the dissolved oxygen suitability function is in the form of a modified Michaelis function:

[0016] in, To maintain the minimum dissolved oxygen for growth, It is the half-saturation constant.

[0017] In some embodiments, generating a feeding plan based on the feeding optimization model includes: calculating a basic feeding rate based on the total biomass of the fish population, the current water temperature, and the water quality stress coefficient; dynamically correcting the basic feeding rate based on the real-time feeding activity of the fish population; and determining the amount of feed per meal based on the corrected feeding rate.

[0018] In some embodiments, the basic feeding rate calculation formula of the feeding optimization model is:

[0019] in, Based on the basic daily feeding rate, The average weight of the fish. Water temperature , , These are empirical parameters related to fish species.

[0020] In some embodiments, the feeding activity index of the feeding optimization model is calculated using the following formula:

[0021] in, The feeding activity index, This represents the real-time feeding activity characteristic value. and These are the historical maximum and minimum characteristic values ​​recorded during the breeding cycle, respectively.

[0022] In some embodiments, generating aerator control instructions based on the aerator energy-saving control model includes: predicting the dissolved oxygen change trend in future periods according to the dissolved oxygen dynamic equation; calculating the target operating frequency of the aerator using a dead-zone incremental PI controller; and generating frequency conversion control instructions based on the target operating frequency.

[0023] In some embodiments, the dissolved oxygen dynamic equation is:

[0024] in, This represents the current dissolved oxygen concentration. This represents the saturated dissolved oxygen concentration. To match the frequency of the aerator The relevant oxygen mass transfer coefficient, The oxygen consumption rate of fish respiration. This refers to the oxygen consumption rate of the bottom sediment.

[0025] In some embodiments, the relationship between the frequency and power consumption of the aerator satisfies:

[0026] in, For frequency The corresponding power, Rated power, This is the rated frequency.

[0027] In some embodiments, the frequency increment calculation formula of the aerator energy-saving control algorithm is:

[0028] in, For the first Dissolved oxygen deviation at any given time Dead zone threshold, and These are the proportional coefficient and the integral coefficient, respectively.

[0029] In some embodiments, a data traceability module is also included, configured to: collect and store fish fry release records, historical water quality data, feeding records, medication records, and catch yield data; generate an electronic archive of the entire aquaculture process based on the stored data; and perform big data analysis based on the electronic archive to generate aquaculture optimization suggestions.

[0030] In some embodiments, an energy consumption analysis module is also included, configured to: collect statistics on the running time and energy consumption data of each aquaculture equipment; identify high energy consumption periods and high energy consumption equipment; generate energy consumption optimization strategies and push them to the user terminal.

[0031] A second aspect of this application provides a smart aquaculture management and control method, applied to the data acquisition module of the smart aquaculture management and control system described in the first aspect of this application. The management and control method includes: acquiring water quality parameter data collected by a water quality sensor array when a preset acquisition period arrives; performing anomaly detection and level determination on the water quality parameter data based on the water quality parameter data, a preset water quality threshold range, and a preset water quality assessment model to obtain a water quality assessment result; generating corresponding equipment control instructions based on the water quality assessment result, current aquaculture status information, and a preset intelligent control model; and sending the equipment control instructions to a target aquaculture equipment to cause the target aquaculture equipment to perform corresponding control operations.

[0032] In some embodiments, the control method further includes: predicting the fish's body weight growth trend based on a fish growth cycle prediction model and real-time water quality parameters; and generating a market launch reminder message when the prediction result reaches the preset market specifications.

[0033] In some embodiments, the control method further includes: dynamically adjusting the feeding strategy based on the feeding optimization model and the feeding activity of the fish school; and automatically stopping the feeding operation when a satiated signal of the fish school is detected.

[0034] In some embodiments, the control method further includes: controlling the operation of the aerator by means of frequency conversion speed regulation according to the aerator energy-saving control model; and slowly adjusting the operating frequency through a PI controller to reduce energy consumption when the dissolved oxygen concentration is in the target range.

[0035] The beneficial effects of this invention are as follows: This invention achieves all-weather perception and intelligent analysis of aquaculture water quality by using an array of water quality sensors deployed in aquaculture ponds to collect water quality parameter data in real time and generating equipment control commands based on an intelligent regulation model.

[0036] This invention can link aerators, water pumps, feeders, and medicine dispensers to form a closed-loop management system for the entire aquaculture process, from problem detection to problem resolution. This effectively solves the problems of low monitoring frequency, delayed response, and inaccurate early warning in traditional aquaculture models.

[0037] The fish growth cycle prediction model in this invention comprehensively considers the influence of multiple factors such as water temperature, dissolved oxygen, pH, ammonia nitrogen, and nitrite on fish growth, achieving scientific prediction of the growth cycle. The feeding optimization model dynamically adjusts the feeding amount based on the real-time feeding activity of the fish population, significantly improving feed conversion rate. Attached Figure Description

[0038] Figure 1 This is a logical framework diagram of the present invention; Figure 2 This is an overall layout diagram of the present invention; Figure 3 This is a flowchart of the process of this invention; Detailed Implementation

[0039] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0040] Refer to the instruction manual appendix Figure 1-3 A fish farming system based on a fish growth prediction model includes: A water quality sensor array is configured to be deployed in an aquaculture pond for real-time collection of water quality parameter data. A data acquisition module, communicatively connected to the water quality sensor array, is configured to perform the following steps: acquiring water quality parameter data collected by the water quality sensor array when a preset acquisition period arrives; performing anomaly detection and level determination on the water quality parameter data based on the water quality parameter data, a preset water quality threshold range, and a preset water quality assessment model to obtain a water quality assessment result; generating corresponding equipment control commands based on the water quality assessment result, current aquaculture status information, and a preset intelligent control model; and sending the equipment control commands to the target aquaculture equipment to enable the target aquaculture equipment to perform corresponding control operations.

[0041] In this embodiment, the water quality sensor array can employ probe-type industrial-grade corrosion-resistant sensors, deployed at different depths in the aquaculture pond, overcoming the limitations of traditional manual sampling monitoring, such as low frequency and delayed response. The water quality sensor array includes five core sensors: dissolved oxygen, pH, water temperature, ammonia nitrogen, and nitrite, enabling real-time monitoring of five key indicators of the aquaculture water. The data acquisition module can be a data acquisition terminal integrated into a protective box at the pond entrance. Specifically, it can communicate with the water quality sensor array via an RS485 bus to acquire water quality parameter data collected by the sensors in real time and transmit the data to a cloud-based big data platform for analysis and processing.

[0042] The dissolved oxygen sensor in the water quality sensor array is used to collect dissolved oxygen concentration data in the aquaculture water. Dissolved oxygen is a core indicator for fish survival, and its concentration directly affects fish respiration, metabolism, and survival rate. The pH sensor collects acidity / alkalinity data in the aquaculture water. Acidity / alkalinity is an important parameter affecting fish physiological functions and water quality stability. The water temperature sensor collects temperature data in the aquaculture water. Water temperature directly affects fish metabolic rate, feeding appetite, and growth rate. The ammonia nitrogen sensor collects ammonia nitrogen concentration data in the aquaculture water. Ammonia nitrogen is a toxic substance, mainly originating from fish excrement and feed residue; excessively high ammonia nitrogen concentrations can lead to ammonia poisoning in fish. The nitrite sensor collects nitrite concentration data in the aquaculture water. Nitrite is a secondary toxin produced after the oxidation of ammonia nitrogen and poses a serious threat to the fish's blood system.

[0043] The preset data collection cycle can be set according to actual aquaculture needs, such as automatically collecting water quality parameter data every 15 minutes. When the preset collection cycle is reached, the data acquisition module sends a data read command to each sensor via the RS485 bus to obtain the latest water quality parameter data collected by each sensor. The acquired water quality parameter data includes the current dissolved oxygen concentration (DO), pH value, water temperature (T), ammonia nitrogen concentration (NH3), and nitrite concentration (NO2).

[0044] Based on the water quality parameter data, the preset water quality threshold range, and the preset water quality assessment model, anomaly detection and level determination are performed on the water quality parameter data to obtain water quality assessment results. Specifically, this includes: determining whether each water quality parameter exceeds the corresponding safety threshold range; if any water quality parameter exceeds the safety threshold range, determining the risk level based on the extent of the exceedance, where the risk level includes low risk, medium risk, and high risk; and generating a comprehensive water quality assessment result based on the exceedance status of each water quality parameter.

[0045] In this embodiment, the safety threshold range can be determined based on the optimal growth environment of different fish species and industry aquaculture standards. For example, for freshwater carp farming, the safety threshold range for dissolved oxygen can be set to 3 mg / L to 8 mg / L, the safety threshold range for pH can be set to 6.5 to 8.5, the safety threshold range for water temperature can be set to 18℃ to 30℃, the safety threshold range for ammonia nitrogen can be set to 0 to 0.5 mg / L, and the safety threshold range for nitrite can be set to 0 to 0.1 mg / L. When any water quality parameter exceeds the corresponding safety threshold range, the system determines the risk level based on the extent of the exceedance: an exceedance of less than or equal to 20% is considered low risk, an exceedance of greater than 20% but less than or equal to 50% is considered medium risk, and an exceedance of greater than 50% is considered high risk. The comprehensive water quality assessment results take into account the exceedance status and risk level of each water quality parameter, generating a comprehensive assessment result that includes a description of the water quality status, risk level, and control recommendations.

[0046] Based on the water quality assessment results, current aquaculture status information, and a preset intelligent control model, corresponding equipment control commands are generated, including: when the water quality assessment results indicate that the dissolved oxygen concentration is lower than a first preset threshold, an aerator start command is generated; when the water quality assessment results indicate that the dissolved oxygen concentration is higher than a second preset threshold, an aerator shutdown command is generated; when the water quality assessment results indicate that the ammonia nitrogen concentration is higher than a third preset threshold, a water pump start command or a water change command is generated; when the water quality assessment results indicate that the pH value deviates from the normal range, a pH adjustment device start command is generated.

[0047] In this embodiment, the first preset threshold can be 3 mg / L, and the second preset threshold can be 5 mg / L. When the dissolved oxygen concentration is below 3 mg / L, the system generates an aerator start command and sends the command to the aerator's control terminal via the 4G network to start the aerator for aeration. When the dissolved oxygen concentration is above 5 mg / L, the system generates an aerator shutdown command to stop the aerator from operating, thus avoiding energy waste. The third preset threshold can be 0.5 mg / L. When the ammonia nitrogen concentration is above 0.5 mg / L, the system generates a water pump start command or a water exchange command to start the circulating water equipment or the water exchange pump for water purification. When the pH value is below 6.5 or above 8.5, the system generates a pH adjustment device start command, such as adding quicklime or acetic acid to maintain the acid-base balance of the water.

[0048] The intelligent control model includes a fish growth cycle prediction model, a feeding optimization model, and an aerator energy-saving control model. The fish growth cycle prediction model is based on the following formula:

[0049] in, For a specific growth rate, The maximum specific growth rate under optimal conditions. Let T be the water temperature suitability function, and T be the measured water temperature. Let DO be the dissolved oxygen suitability function, and DO be the measured dissolved oxygen concentration. This is a pH suitability function, where pH is the measured acidity or alkalinity. It is the non-ionic ammonia suppression function. To measure the concentration of non-ionic ammonia, The nitrite inhibition function, This is for the actual measurement of nitrite concentration.

[0050] In this embodiment, the specific growth rate (SGR) is an important indicator for measuring the growth rate of fish, representing the relative growth rate of fish body weight per unit time. Determined by fish species and genetic characteristics, it needs to be calibrated based on historical aquaculture data. Various environmental suitability functions. The value of SGR ranges from 0 to 1 and is used to describe the degree of inhibition of fish growth when various environmental factors deviate from optimal conditions. When all environmental factors are at their optimal conditions, each fitness function takes the value of 1, at which point SGR equals 1. This indicates that the fish are in their optimal growth state; when any environmental factor deviates from the optimal conditions, the corresponding fitness function value is less than 1, resulting in a decrease in SGR and a slowdown in growth rate.

[0051] The water temperature suitability function is a Gaussian function.

[0052] in, For the optimal growth temperature, For temperature tolerance width, and This is the survival temperature limit.

[0053] In this embodiment, the Gaussian function can well describe the influence of water temperature on fish growth: the closer the water temperature is to the optimal growth temperature... The higher the suitability, the more the water temperature deviates. The farther away, the lower the suitability. Temperature tolerance width. The fitness curve is determined in The steepness of the surrounding terrain, and the different fish species The values ​​differ. When the water temperature is below the lower limit of the survival temperature... Or above the upper limit of survival temperature When the fitness level is 0, it means the fish cannot survive. For example, for freshwater carp, It can be set to 25℃. It can be set to 5℃. It can be set to 10℃. It can be set to 35℃.

[0054] The dissolved oxygen suitability function is in the modified Michaelis function form:

[0055] in, To maintain the minimum dissolved oxygen for growth, It is the half-saturation constant.

[0056] In this embodiment, the modified Michaelis function can describe the saturation effect of dissolved oxygen on fish growth: when the dissolved oxygen concentration exceeds the minimum dissolved oxygen level... Afterwards, fitness increases with increasing dissolved oxygen concentration; when the dissolved oxygen concentration exceeds... add At this point, the fitness level is close to 0.5, and further increasing the dissolved oxygen concentration has a weaker effect on promoting growth. and The value depends on the fish species; for freshwater carp, It can be set to 2mg / L. It can be set to 1.5 mg / L. When the dissolved oxygen concentration is lower than... At this time, fish cannot grow normally, and the fitness level is 0.

[0057] The pH suitability function is a parabolic function:

[0058] in, For the optimal pH value, and This is the range within which the plant can grow.

[0059] In this embodiment, the parabolic function can describe the unimodal effect of pH on fish growth: the closer the pH value is to the optimum pH value... The higher the suitability, the more likely it is to deviate. The farther away, the lower the suitability. When the pH value is below the lower limit... or above the upper limit When the suitability is 0, it means that fish cannot grow. For freshwater carp, It can be set to 7.0. It can be set to 6.0. It can be set to 9.0.

[0060] The non-ionic ammonia suppression function adopts a linear decay form:

[0061] in, For safe concentration thresholds, For lethal or completely inhibiting concentrations.

[0062] In this embodiment, the linear decay function can describe the inhibitory effect of ammonia nitrogen on fish growth: when the concentration of non-ionic ammonia is below the safe concentration... When the concentration of non-ionic ammonia is between [value missing], it has no inhibitory effect on growth, and the fitness level is 1; when the concentration of non-ionic ammonia is between [value missing], it has no inhibitory effect on growth. and Between these concentrations, the degree of inhibition increases linearly with increasing concentration; when the non-ionic ammonia concentration reaches or exceeds [a certain value], [the degree of inhibition increases linearly with increasing concentration]. At that time, growth was completely inhibited. It can be set to 0.02 mg / L. It can be set to 0.5 mg / L.

[0063] The nitrite suppression function adopts a Sigmoid type function:

[0064] in, To achieve a 50% inhibition concentration, This is the curve steepness coefficient.

[0065] In this embodiment, the Sigmoid function can describe the S-shaped inhibition curve of nitrite on fish growth: when the nitrite concentration is below... When the concentration of nitrite is close to the threshold, the inhibitory effect is weak; when the concentration of nitrite is close to the threshold, the inhibitory effect is weak. When the nitrite concentration exceeds a certain level, the inhibitory effect increases dramatically; when the nitrite concentration exceeds a certain level... At that time, the inhibitory effect tends to stabilize. It can be set to 0.5 mg / L. It can be set to 2.0.

[0066] The feeding plan is generated based on the feeding optimization model, including: calculating the basic feeding rate based on the total biomass of the fish population, the current water temperature, and the water quality stress coefficient; dynamically correcting the basic feeding rate based on the real-time feeding activity of the fish population; and determining the amount of feed per meal based on the corrected feeding rate.

[0067] In this embodiment, the goal of the feeding optimization model is to dynamically adjust the feeding rate and total amount during each feeding, combining water quality parameters and the real-time feeding activity of the fish population, to avoid feed waste and improve feed conversion rate. The total biomass B of the fish population can be calculated by multiplying the fish population N by the current average weight. The calculation yielded: The water quality stress coefficient is calculated by substituting current water quality parameters into a fish growth cycle prediction model, and is used to reflect the impact of the water quality environment on the feeding intentions of fish.

[0068] The basic feeding rate calculation formula for the feeding optimization model is as follows:

[0069] in, Based on the basic daily feeding rate, , where T is the average weight of the fish population, T is the water temperature, and a, b, and c are empirical parameters related to the fish species.

[0070] In this embodiment, the basal feeding rate formula comprehensively considers two main influencing factors: the average weight of the fish population and water temperature. The empirical parameters a, b, and c need to be calibrated based on the farming data of different fish species. For freshwater carp, a can be set to 0.05, b to -0.2, and c to 0.05. The basal feeding rate also needs to be corrected by incorporating the water quality stress coefficient.

[0071] The calculation formulas for each water quality suitability function are the same as those in the fish growth cycle prediction model, but the parameter values ​​focus on reflecting the impact on feeding behavior. The planned feeding amount per meal can be determined by the number of feedings per day. Distribute: .

[0072] The formula for calculating the feeding activity index of the feeding optimization model is as follows:

[0073] in, The feeding activity index, This represents the real-time feeding activity characteristic value. and These are the historical maximum and minimum characteristic values ​​recorded during the breeding cycle, respectively.

[0074] In this embodiment, the feeding activity index is used to quantify the intensity of competition for food among fish during the feeding process. Real-time feeding activity characteristic value. The feeding activity can be extracted using machine vision technology, specifically by: installing an underwater camera above the feeding area to capture video of water surface ripples as fish compete for food; calculating the mean vector magnitude of pixel motion between video frames using optical flow to serve as a feeding activity feature value; or identifying the rate of change in the area of ​​water splashes on the water surface through image processing to serve as a feeding activity feature value. and These are the historical maximum and minimum characteristic values ​​recorded during the breeding cycle, used to normalize the real-time characteristic values ​​so that the feeding activity index ranges from 0 to 1.

[0075] The dynamic feeding control strategy adopts a feedback correction method, and the formula for calculating the instantaneous feeding rate is:

[0076] in, As the baseline feeding rate, The threshold for the satiety signal.

[0077] In this embodiment, the instantaneous feeding rate is positively correlated with the feeding activity index: when the feeding activity index is high, the feeding rate is increased; when the feeding activity index is low, the feeding rate is decreased. When the feeding activity index is below the satiety signal threshold... When the value remains at 0.15 for a certain period of time (e.g., 30 seconds), the system considers the fish to be satiated and automatically stops feeding. The execution steps of the single-meal feeding algorithm are as follows: First, read the current water quality parameters and calculate the planned feeding amount. Then start feeding and set the baseline feeding rate. ; every time interval Get real-time feeding activity index in seconds And update the instantaneous feeding rate When the cumulative feeding amount reaches or Below After 30 seconds, turn off the feeder; finally, record the actual amount fed, which will be used for the next correction and feed conversion ratio (FCR) calculation.

[0078] The aerator control command is generated based on the energy-saving control model of the aerator, including: predicting the trend of dissolved oxygen change in the future period according to the dissolved oxygen dynamic equation; calculating the target operating frequency of the aerator using a dead-zone incremental PI controller; and generating frequency conversion control command according to the target operating frequency.

[0079] In this embodiment, the goal of the aerator energy-saving control model is to maintain the dissolved oxygen concentration within the target range while minimizing the aerator's total energy consumption. Compared to traditional on / off control, variable frequency speed control can precisely adjust the aerator's operating power according to the actual dissolved oxygen demand, achieving energy savings of 15% to 20%.

[0080] The dynamic equation for dissolved oxygen is:

[0081] Where DO is the current dissolved oxygen concentration. This represents the saturated dissolved oxygen concentration. The oxygen mass transfer coefficient is related to the aerator frequency f. The oxygen consumption rate of fish respiration. This refers to the oxygen consumption rate of the bottom sediment.

[0082] In this embodiment, the dissolved oxygen dynamic equation describes the changing pattern of dissolved oxygen concentration in the aquaculture pond: the aerator supplements oxygen to the water body through aeration, causing the dissolved oxygen concentration to tend towards the saturation dissolved oxygen concentration. Fish respiration and the consumption of oxygen in the bottom sediment consume dissolved oxygen, thus reducing the dissolved oxygen concentration. It is related to the current water temperature and salinity, and can be looked up in the saturated dissolved oxygen table. It is related to the operating frequency f of the aerator; the higher the frequency, the better. The larger the size, the stronger the oxygen-generating capacity. It is related to the total biomass of the fish population, water temperature, and dissolved oxygen concentration. The calculation formula is: Where V is the volume of the aquaculture water. , d, and q are empirical constants. The oxygen consumption rate of sediment, plankton, and other organisms can be considered as a slowly time-varying parameter.

[0083] The relationship between the frequency and power consumption of the aerator satisfies:

[0084] in, The power corresponding to frequency f. Rated power, This is the rated frequency.

[0085] In this embodiment, the power of the aerator is proportional to the cube of the operating frequency, which is determined by the operating characteristics of the asynchronous motor. As the frequency decreases, the power decreases cubically, resulting in significant energy savings. For example, when the frequency decreases from 50Hz to 25Hz, the power decreases to 12.5% ​​of the rated power.

[0086] The relationship between the oxygen mass transfer coefficient and frequency can be expressed as:

[0087] in, The oxygen mass transfer coefficient at the rated frequency. This is the frequency index, typically ranging from 1.5 to 2.1.

[0088] In this embodiment, the oxygen mass transfer coefficient The efficiency increases with increasing frequency, but the rate of increase is lower than the rate of increase in power consumption. Therefore, there exists an optimal frequency range that achieves the best balance between oxygenation efficiency and energy consumption.

[0089] The formula for calculating the frequency increment of the energy-saving control algorithm for the oxygenator is as follows:

[0090] in, Let the dissolved oxygen deviation be at time k. Dead zone threshold, and These are the proportional coefficient and the integral coefficient, respectively.

[0091] In this embodiment, the working principle of the dead-zone incremental PI controller is as follows: A target dissolved oxygen value is set. Calculate the dissolved oxygen deviation using the median value of the target range, for example, 4 mg / L. When the absolute value of the deviation is less than the dead zone threshold (For example, at 0.3 mg / L), no integral calculation is performed, and the frequency remains constant to avoid frequent adjustments to the controller when there are small fluctuations; when the absolute value of the deviation is greater than or equal to At that time, the frequency increment is calculated according to the incremental PI formula. The controller output needs to be limited:

[0092] in This is the minimum permissible frequency. Auxiliary rules include: when dissolved oxygen is below the lower limit of 0.5 mg / L, the forced aerator operates at the rated frequency; when dissolved oxygen is above the upper limit of 1.0 mg / L for 5 minutes, the forced aerator operates at the lowest frequency or shuts down; when dissolved oxygen is within the normal range, the PI controller slowly adjusts the frequency to keep the dissolved oxygen level at the setpoint.

[0093] In some embodiments, a data traceability module is also included, configured to: collect and store fish fry release records, historical water quality data, feeding records, medication records, and catch yield data; generate an electronic archive of the entire aquaculture process based on the stored data; and perform big data analysis based on the electronic archive to generate aquaculture optimization suggestions.

[0094] In this embodiment, the data traceability module records data throughout the entire aquaculture process, forming an unalterable traceability archive. Fish fry release records include release time, fish species, quantity, size, and origin; historical water quality data includes water quality parameter values ​​and assessment results at each collection point; feeding records include feeding time, amount, feed type, and feeding strategy; medication records include medication time, drug name, dosage, and withdrawal period; and catch yield data includes catch time, yield, and size grading. This data is stored and analyzed through a cloud-based big data platform, supporting one-click report export, providing a basis for scientific decision-making, and reducing reliance on aquaculture experience.

[0095] In some embodiments, an energy consumption analysis module is also included, configured to: collect statistics on the running time and energy consumption data of each aquaculture equipment; identify high energy consumption periods and high energy consumption equipment; generate energy consumption optimization strategies and push them to the user terminal.

[0096] In this embodiment, the energy consumption analysis module collects real-time power consumption data from various aquaculture equipment, including aerators, water pumps, and feeders, through the electricity metering module. The system calculates the monthly energy consumption of each device, identifies high-energy-consuming components and periods, such as anomalies like aerators running idling for 30% of the time at night. Energy consumption optimization strategies are generated and pushed to the user terminal, such as suggestions to adjust start-up and shutdown times based on dissolved oxygen trends and optimize feeding strategies to reduce water pollution, thus helping farmers reduce water and electricity costs.

[0097] This system comprises a water quality sensor array, a data acquisition module, a cloud-based big data platform, and terminal devices. The water quality sensor array is deployed in the aquaculture ponds, specifically at the four corners and the center, at depths of 0.5 meters, 1.0 meter, and 1.5 meters respectively, to acquire water quality parameter data at different depths. The data acquisition module is installed inside a protective box at the pond entrance and connects to the water quality sensor array via an RS485 bus and to the cloud-based big data platform via a 4G network. The cloud-based big data platform is deployed on a cloud computing server, supporting multi-pond, multi-farm cluster management and possessing data storage, cleaning, analysis, and model calculation capabilities. Terminal devices include a mobile app, a computer client, and a monitoring screen, used to display real-time water quality data, early warning information, and aquaculture recommendations to users.

[0098] Figure 3 This is a flowchart of a smart aquaculture management and control method provided in an embodiment of this application, applied to the data acquisition module of the smart aquaculture management and control system of the first aspect of this application embodiment, from... Figure 3 As can be seen from this, the control method steps include: Step S1: When the preset acquisition period arrives, acquire the water quality parameter data collected by the water quality sensor array.

[0099] In this embodiment, the preset acquisition period can be set to 15 minutes. The data acquisition module sends data reading commands to each sensor via RS485 bus to obtain water quality parameters such as dissolved oxygen concentration, pH value, water temperature, ammonia nitrogen concentration, and nitrite concentration. For the raw data output by the sensors, data verification and outlier removal processing are performed to eliminate erroneous data caused by sensor failure or signal interference.

[0100] Step S2: Based on the water quality parameter data, the preset water quality threshold range, and the preset water quality assessment model, perform anomaly detection and level determination on the water quality parameter data to obtain the water quality assessment result.

[0101] In this embodiment, the data acquisition module compares the acquired water quality parameter data with preset safety threshold ranges to determine whether each parameter exceeds the standard. If any parameter exceeds the standard, the risk level is calculated based on the extent of the exceedance, and a comprehensive water quality assessment result is generated, including a description of the water quality status, the risk level, and control recommendations. The assessment result is uploaded to a cloud-based big data platform via a 4G network, simultaneously triggering local audible and visual alarms and a push notification to a mobile app.

[0102] Step S3: Based on the water quality assessment results, current aquaculture status information, and preset intelligent control model, generate corresponding equipment control commands.

[0103] In this embodiment, the intelligent control model generates targeted equipment control commands based on water quality assessment results and current aquaculture status information. Aquaculture status information includes fish population size, average weight, growth stage, and historical feeding records. The cloud-based big data platform calculates equipment control commands based on the intelligent control model, including aerator start-up, aerator shutdown, water pump start-up, water change commands, pH adjustment device activation, and feeding plan adjustments.

[0104] Step S4: Send the equipment control command to the target aquaculture equipment so that the target aquaculture equipment performs the corresponding control operation.

[0105] In this embodiment, equipment control commands are sent to the control terminal of the target aquaculture equipment via a 4G network. The control terminal parses the commands and executes the corresponding operations, such as starting the aerator, turning off the water pump, or adjusting the feeder speed. After execution, the control terminal feeds back the results to the cloud-based big data platform and records the operation log.

[0106] In some embodiments, the control method further includes: predicting the fish's body weight growth trend based on a fish growth cycle prediction model and real-time water quality parameters; and generating a market launch reminder message when the prediction result reaches the preset market specifications.

[0107] In this embodiment, fish farmers can set a target market weight, such as 1.5 kg / fish. The system predicts the average weight of the fish on future dates based on a fish growth cycle prediction model and real-time water quality parameters (water temperature, dissolved oxygen, pH, ammonia nitrogen, and nitrite). When the predicted weight reaches the target market size, the system generates a market alert and pushes it to the fish farmers to help them determine the optimal time to sell the fish.

[0108] In some embodiments, the control method further includes: dynamically adjusting the feeding strategy based on the feeding optimization model and the feeding activity of the fish school; and automatically stopping the feeding operation when a satiated signal of the fish school is detected.

[0109] In this embodiment, the feeding optimization model calculates the basic feeding rate based on the total biomass of the fish population and the water quality stress coefficient, and then dynamically corrects the feeding rate based on the real-time feeding activity index. The system collects videos of fish competing for food using an underwater camera, extracts feeding activity feature values, and calculates the feeding activity index. When the feeding activity index is lower than the satiety signal threshold and remains below it for a certain period of time, the system automatically stops feeding to avoid feed waste and water pollution caused by overfeeding.

[0110] In some embodiments, the control method further includes: controlling the operation of the aerator by means of frequency conversion speed regulation according to the aerator energy-saving control model; and slowly adjusting the operating frequency through a PI controller to reduce energy consumption when the dissolved oxygen concentration is in the target range.

[0111] In this embodiment, the energy-saving control model predicts the future trend of dissolved oxygen changes based on the dissolved oxygen dynamic equation and adjusts the aerator frequency in advance to avoid large fluctuations in dissolved oxygen. When the dissolved oxygen concentration is within the target range (e.g., 3 mg / L to 5 mg / L), a dead-zone incremental PI controller is used to slowly adjust the frequency to stabilize the dissolved oxygen near the set point while minimizing energy consumption. When the dissolved oxygen is below the lower limit, the aerator is forced to operate at the rated frequency to ensure that the fish do not suffer from oxygen deficiency; when the dissolved oxygen is above the upper limit and remains above it for a certain period of time, the aerator is forced to operate at the lowest frequency or shut down to avoid energy waste.

[0112] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the intelligent aquaculture management and control method as described in this application.

[0113] This application also provides a smart aquaculture equipment, including the smart aquaculture management and control system of the first aspect of this application.

[0114] In this embodiment, the smart aquaculture equipment can be a smart aquaculture pond terminal integration system, including a water quality sensor array, a data acquisition module, a control terminal, and a communication module. The system collects water quality parameters in real time through the water quality sensor array, generates equipment control commands based on an intelligent control model, and links equipment such as aerators, water pumps, feeders, and dosing machines to achieve all-weather perception and intelligent analysis of aquaculture water quality, forming a closed-loop management of the entire aquaculture process from problem detection to problem resolution.

[0115] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0116] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods and apparatus according to embodiments of the present invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0117] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0118] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0119] The present invention provides a detailed description of an IoT-based smart aquaculture management and control system, method, storage medium, and aquaculture equipment. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A fish farming system based on a fish growth prediction model, applied to fish farming equipment, characterized in that, include: A water quality sensor array is configured to be deployed in aquaculture ponds for real-time collection of water quality parameter data; The data acquisition module, which is communicatively connected to the water quality sensor array, is configured to perform the following steps: when a preset acquisition cycle arrives, acquire water quality parameter data collected by the water quality sensor array; based on the water quality parameter data, a preset water quality threshold range, and a preset water quality assessment model, perform anomaly detection and level determination on the water quality parameter data to obtain a water quality assessment result; based on the water quality assessment result, current aquaculture status information, and a preset intelligent control model, generate corresponding equipment control commands. The device control command is sent to the target aquaculture equipment so that the target aquaculture equipment performs the corresponding control operation.

2. The aquaculture system according to claim 1, characterized in that, Based on the water quality parameter data, the preset water quality threshold range, and the preset water quality assessment model, anomaly detection and level determination are performed on the water quality parameter data to obtain water quality assessment results, including: determining whether each water quality parameter exceeds the corresponding safety threshold range; if any water quality parameter exceeds the safety threshold range, the risk level is determined according to the extent of the exceedance, and the risk level includes low risk, medium risk, and high risk; and generating a comprehensive water quality assessment result based on the exceedance status of each water quality parameter.

3. The aquaculture system according to claim 1 or 2, characterized in that, Based on the water quality assessment results, current aquaculture status information, and a preset intelligent control model, corresponding equipment control commands are generated, including: when the water quality assessment results indicate that the dissolved oxygen concentration is lower than a first preset threshold, an aerator start command is generated; when the water quality assessment results indicate that the dissolved oxygen concentration is higher than a second preset threshold, an aerator shutdown command is generated; when the water quality assessment results indicate that the ammonia nitrogen concentration is higher than a third preset threshold, a water pump start command or a water change command is generated; when the water quality assessment results indicate that the pH value deviates from the normal range, a pH adjustment device start command is generated.

4. The aquaculture system according to claim 3, characterized in that, The intelligent control model includes a fish growth cycle prediction model, a feeding optimization model, and an aerator energy-saving control model; the fish growth cycle prediction model is established based on the following formula: in, For a specific growth rate, The maximum specific growth rate under optimal conditions. Let T be the water temperature suitability function, and T be the measured water temperature. Let DO be the dissolved oxygen suitability function, and DO be the measured dissolved oxygen concentration. This is a pH suitability function, where pH is the measured acidity or alkalinity. It is the non-ionic ammonia suppression function. To measure the concentration of non-ionic ammonia, The nitrite inhibition function, This is for the actual measurement of nitrite concentration.

5. The aquaculture system according to claim 4, characterized in that, The water temperature suitability function is a Gaussian function. in, For the optimal growth temperature, For temperature tolerance width, and This is the survival temperature limit.

6. The aquaculture system according to claim 4, characterized in that, The dissolved oxygen suitability function is in the modified Michaelis function form: in, To maintain the minimum dissolved oxygen for growth, It is the half-saturation constant.

7. The aquaculture system according to claim 4, characterized in that, The feeding plan is generated based on the feeding optimization model, including: calculating the basic feeding rate based on the total biomass of the fish population, the current water temperature, and the water quality stress coefficient; dynamically adjusting the basic feeding rate based on the real-time feeding activity of the fish population; and determining the amount of feed per meal based on the adjusted feeding rate. The formula for calculating the basic feeding rate of the feeding optimization model is as follows: in, Based on the basic daily feeding rate, , where T is the average weight of the fish population, T is the water temperature, and a, b, and c are empirical parameters related to the fish species.

8. A fish farming method based on a fish growth prediction model, characterized in that, A data acquisition module applied to the fish growth prediction model aquaculture system according to any one of claims 1 to 7, the control method comprising: step S1, acquiring water quality parameter data collected by a water quality sensor array when a preset acquisition cycle is reached; step S2, performing anomaly detection and level determination on the water quality parameter data based on the water quality parameter data, a preset water quality threshold range, and a preset water quality assessment model, to obtain a water quality assessment result; step S3, generating corresponding equipment control commands based on the water quality assessment result, current aquaculture status information, and a preset intelligent control model; and step S4, sending the equipment control commands to the target aquaculture equipment to enable the target aquaculture equipment to perform corresponding control operations.

9. A fish farming equipment, characterized in that, Including the aquaculture system based on the fish growth prediction model as described in any one of claims 1 to 7.

Images