A method for predicting feed ration of fish based on a dynamic growth and energy requirement model thereof

Abstract

The application discloses a method for predicting the feed feeding amount of fish based on a dynamic growth and energy demand model, and comprises the following steps: screening an optimal dynamic growth model, calculating a body weight prediction value of a measured fish at a current time in an actual growth cycle of the fish through the optimal dynamic growth model, and obtaining growth and metabolism demand of the fish at the current time, water temperature of water, dissolved oxygen, ammonia nitrogen and nitrite concentration in the water; obtaining the energy demand of the measured fish at the current time through a constructed fish energy demand model, and obtaining the feed feeding amount at the current time in combination with the energy content of unit feed. The application can accurately calculate the energy demand of the fish at different growth time points, so that the feed use is optimized, and the breeding efficiency is improved.

Description

Technical Field

[0001] This invention relates to the fields of bioenergetics and aquaculture, and more specifically, to a method for predicting feed intake based on a dynamic growth and energy demand model of fish. Background Technology

[0002] In modern aquaculture, effectively managing fish growth and feed consumption is a crucial research topic. Traditional aquaculture methods typically rely on the farmer's experience and manual record-keeping, which inherently involve subjectivity and uncertainty, making precise control difficult. Furthermore, the aquatic environment (such as temperature) significantly impacts fish growth and energy requirements; neglecting these factors can lead to feed waste or stunted fish growth.

[0003] As an important farmed fish, the growth and energy requirements of largemouth bass are closely related to a variety of environmental parameters. Currently, there is a lack of a comprehensive model that considers multiple growth parameters to accurately predict the growth and energy requirements of largemouth bass under different conditions.

[0004] Existing growth models, such as Specific Growth Rate (SGR), Daily Growth Rate (DGC), Average Daily Weight Gain (ADG), Temperature-Growth (TGC), and polynomial fitting models, each have their own advantages and disadvantages. For example, the Specific Growth Rate (SGR) model can reflect the relative growth rate of fish over a certain period of time, but its predictive accuracy may be low under different temperatures and feeding conditions. The Daily Growth Rate (DGC) and Average Daily Weight Gain (ADG) models focus on growth at different time points, but may not fully reflect long-term growth trends. The Temperature-Growth (TGC) model considers the influence of temperature on growth, but the adaptability of a single model is limited in complex environments. Polynomial fitting models can fit growth curves more flexibly, but are prone to overfitting, leading to poor predictive performance in practical applications.

[0005] Furthermore, existing technologies have significant limitations in calculating fish energy requirements. Most methods fail to incorporate dynamic growth models, resulting in insufficient accuracy and practicality in energy requirement prediction. These shortcomings affect the formulation of feed feeding strategies, thereby impacting the economic benefits and resource utilization efficiency of aquaculture production.

[0006] To address the aforementioned problems, this invention proposes a method for predicting feed intake based on a dynamic growth and energy demand model for fish. By collecting fish growth data and water parameters, specific growth rate models, daily growth rate models, daily average weight gain models, temperature growth coefficient models, and polynomial fitting models are constructed. The optimal model is selected for growth prediction, and the energy demand of the fish is further calculated to guide feed intake and optimize aquaculture management. Summary of the Invention

[0007] To address the problems existing in current technologies, this invention provides a fish growth model and energy requirement calculation method based on bioenergetics. The aim is to optimize feed use and improve aquaculture efficiency by accurately calculating the energy requirements of fish at different growth stages. This method consists of two main parts: a multi-model selection mechanism and the calculation of fish feeding energy requirements.

[0008] The technical solution adopted in this invention is as follows:

[0009] A method for predicting feed intake based on a dynamic growth and energy requirement model of fish includes the following steps:

[0010] 1) Collect growth data and water data of the fish at each moment in the current growth cycle, and combine multiple dynamic growth models to obtain the predicted weight value under the corresponding model.

[0011] 2) Based on the coefficient of determination R 2 As an evaluation index for the fitting effect, the R-value for each model is obtained by fitting multiple predicted weight values ​​to the measured weight values ​​at each time step in the existing growth data. 2 Value, in R 2 The dynamic growth model with the highest value is selected as the optimal dynamic growth model.

[0012] 3) Energy demand model construction: The predicted weight of the fish at any time during the growth cycle is obtained through the optimal dynamic growth model. Based on the predicted weight, the growth and metabolic needs of the fish are considered, and the energy demand model of the fish at any time is constructed by taking into account the water temperature, dissolved oxygen, ammonia nitrogen and nitrite concentration in the water.

[0013] 4) Calculate the current feed amount; calculate the predicted weight of the fish at the current moment using the optimal dynamic growth model, obtain the energy requirement of the fish at the current moment based on the fish energy requirement model, and obtain the current feed amount by combining the energy content of the feed per unit.

[0014] Furthermore, in step 1, the existing growth data and water data include: the number of days the fish were raised, the measured weight of the fish at each moment, and the corresponding water temperature.

[0015] Furthermore, in step 1), the multiple dynamic growth models include: a specific growth rate model (SGR), a daily growth rate model (DGC), a daily average weight gain model (ADG), a temperature growth coefficient model (TGC), and a polynomial fitting model.

[0016] Furthermore, in step 2), the coefficient of determination R... 2The value ranges from [0,1] to characterize the ability of the dynamic growth model to explain data variation. The closer the value is to 1, the better the surface model fits the growth pattern of the tested fish.

[0017] Furthermore, in step 3), the expression for the fish energy demand model is:

[0018]

[0019] In the formula, E d W(t) represents the energy requirement at any time t; W(t) represents the predicted weight at any time t; E g E is the baseline energy required per unit of body weight gain. m The baseline energy required to maintain metabolism per unit body weight; T(t) is the water temperature at any given time t; T opt The suitable water temperature for the growth of the tested fish; DO(t) is the dissolved oxygen concentration in the water at any time t; DO opt NH3(t) represents the suitable dissolved oxygen concentration for the fish being tested; NH3(t) represents the ammonia nitrogen concentration in the water at any given time t; NH 3,max This represents the maximum tolerance value for ammonia nitrogen concentration in the tested fish; NO 2(t) NO is the concentration of nitrite in the water at any given time t; 2,max α represents the maximum tolerance value for nitrite concentration in the tested fish; α, β, γ, and δ are the sensitivity coefficients of environmental factors.

[0020] Furthermore, in step 4), the current time t a Feeding amount F(t) a )for:

[0021]

[0022] In the formula, E d (t a (t) represents the current time. a Energy demand at time, C f This refers to the energy content per unit of feed.

[0023] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the methods described above for predicting feed intake based on a dynamic growth and energy requirement model of fish.

[0024] A system for predicting feed intake based on a dynamic growth and energy requirement model of fish, the system comprising:

[0025] One or more processors;

[0026] Memory, used to store one or more programs;

[0027] When the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described above for predicting feed intake based on a dynamic growth and energy demand model of fish.

[0028] The beneficial effects of this invention are:

[0029] By constructing an energy demand calculation model, this invention can comprehensively and accurately assess the energy requirements of fish at different growth stages and under various environmental conditions. The model not only considers the growth and metabolic needs of the fish but also integrates the influence of environmental factors such as water temperature, dissolved oxygen, ammonia nitrogen, and nitrite, thereby formulating more scientific and efficient feeding strategies. This will effectively optimize feed use, reduce resource waste, and improve aquaculture efficiency and economic benefits, demonstrating significant application value and promising prospects for wider adoption. Attached Figure Description

[0030] Figure 1 This is a schematic diagram illustrating the process of establishing and selecting a fish growth model according to an embodiment of the present invention. Detailed Implementation

[0031] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Example:

[0033] This embodiment uses California bass as the test fish;

[0034] I. Establishment and selection of a growth model for California bass:

[0035] Based on the principles of bioenergetics, this study proposes a multi-model dynamic optimization strategy for the growth curve of largemouth bass. By integrating four classic growth prediction models and polynomial fitting (as shown in Table 1), the optimal solution is selected through a model optimization mechanism. According to different aquaculture scenarios, the current optimal growth prediction model is selected, thus solving the adaptability limitations of traditional empirical models under dynamic environmental fluctuations.

[0036] Table 1 Mathematical expressions for fish growth models

[0037]

[0038] The specific steps are as follows:

[0039] 1. Data Collection:

[0040] Time: Record the total growth cycle d of the California bass and the time t. In this embodiment, t is taken as the number of days in Table 2.

[0041] Body weight: At each time t, the measured body weight W of the largemouth bass was measured and recorded. obs,t .

[0042] Water temperature: Record the water temperature T(t) at each time t during the growth period of the largemouth bass.

[0043] 2. Model Establishment: The five models in Table 1 were used to analyze the growth data (final body weight W) of California bass. d Initial body weight W0, number of days of rearing (d), rearing temperature T(t)

[0044] To perform the fitting, specifically, model t in Table 1 is set to d, meaning that data from the last day of the breeding period is selected for calculation to obtain SGR, DGC, AD, TGC, and α. i Then, based on the obtained SGR, DGC, ADG, TGC, and α i By combining the input time t, water temperature T(t), and initial weight W0, the system can predict the weight at any time t and obtain the predicted weight value W. pred,t ;

[0045] 3. Model Comparison and Selection:

[0046] (1) Evaluation of model growth prediction performance:

[0047] To quantify the predictive performance of the dynamic growth model, this study uses the coefficient of determination (R²). 2 R is used as the core evaluation index for model fitting performance. 2 R is a classic parameter in statistics used to measure the ability of a regression model to explain the variation in observed data. In the field of fish growth modeling, R... 2 It has been widely used in model validation, and its expression is:

[0048]

[0049] In the formula, W obs，t W represents the measured body weight at time t. pred，t The model prediction value at time t. R represents the actual average weight of the sample. 2 The value represents the model's ability to explain data variation, and its range is [0, 1]. The closer the value is to 1, the better the model fits the growth pattern.

[0050] (2) Multi-model selection mechanism:

[0051] Based on R 2As an evaluation index for fit performance, four growth prediction models (SGR, DGC, ADG, and TGC) were used to fit existing data using polynomial fitting. The R-values ​​of each model were compared. 2 The value is used to select the optimal growth model. Specifically, R... 2 The model with the highest value can capture the nonlinear relationship between fish growth dynamics and environmental factors to the greatest extent, thus providing a high-precision mathematical basis for subsequent predictions.

[0052] 4. Model Application:

[0053] Once the optimal model is determined, it can be used to predict the weight of largemouth bass at different growth stages. By inputting new time and water temperature data, the predicted weight of largemouth bass is calculated using the optimal model. Table 2 shows the weight changes of largemouth bass under different water temperature conditions.

[0054]

[0055] Table 3. Fitting results of each model under different temperature conditions

[0056]

[0057] II. Establishment of the energy requirement model for California bass:

[0058] 2.1 Data Collection and Input:

[0059] (1) Body weight: The predicted body weight of California bass at different time points was obtained from the optimal growth model in Part 1.

[0060] (2) Water temperature: Real-time recording and input of water temperature data in the aquaculture environment.

[0061] (3) Dissolved oxygen time: Record and input dissolved oxygen time data.

[0062] (4) Ammonia nitrogen concentration: Record and input the ammonia nitrogen concentration in the water body.

[0063] (5) Nitrite concentration: Record and input the nitrite concentration in the water.

[0064] (6) Feed energy value: Enter the energy content of the feed used, usually in kJ / kg

[0065] express.

[0066] 2.2 Energy Demand Calculation:

[0067] After determining the optimal growth model, the second part of this invention aims to accurately calculate the energy requirements of largemouth bass at different growth stages based on the selected growth model and in conjunction with bioenergetics principles. The calculation of energy requirements considers not only the growth needs of the fish but also the influence of environmental factors on metabolic rate, thereby formulating a scientific feeding strategy.

[0068] The specific steps are as follows:

[0069] 2.2.1 Energy Demand Model Construction

[0070] Energy demand E d The calculation of (t) is based on the following comprehensive formula:

[0071]

[0072] The variables are defined as follows:

[0073] ·E d (t): Energy requirement at time t (kilojoules, kJ).

[0074] W(t): Predicted body weight (grams, g) at time t.

[0075] · Weight gain rate, which is the derivative of weight with respect to time (grams per day, g / day).

[0076] ·E g The baseline energy required per unit of body weight gain (kJ / g) is set at 25.37, and will be adjusted according to actual breeding conditions.

[0077] ·E m The baseline energy required per unit body weight to maintain metabolism (kJ / g) is set at 4.73, and will be adjusted according to actual breeding conditions.

[0078] • T(t): Water temperature at time t (degrees Celsius, °C).

[0079] ·T opt The suitable water temperature for the growth of California bass (°C) is approximately (26±0.5)°C.

[0080] •DO(t): Dissolved oxygen concentration at time t (mg / L).

[0081] ·DO opt The optimal dissolved oxygen concentration (mg / L) for California bass is approximately (6.8 ± 1).

[0082] 0.3) mg / L.

[0083] ·NH3(t): The concentration of ammonia nitrogen in the water at time t (mg / L).

[0084] ·NH 3,max The maximum tolerable concentration of ammonia nitrogen in California bass (mg / L) is approximately 0.3 mg / L.

[0085] NO 2(t) : Nitrite concentration in the water at time t (mg / L).

[0086] NO 2,max The maximum tolerable concentration of nitrite in California bass (mg / L) is approximately 0.3 mg / L.

[0087] • α,β,γ,γ: Environmental factor sensitivity coefficients, reflecting the degree of influence of each environmental parameter on energy demand.

[0088] 2.2.2 Energy Demand Calculation Process

[0089] 1. Data Input:

[0090] The collected data (predicted weight, temperature, dissolved oxygen, ammonia nitrogen, and nitrite concentration) are input into the model to ensure the accuracy and real-time nature of the data.

[0091] 2. Growth rate calculation:

[0092] Calculating the weight gain rate using the optimal growth model

[0093] 3. Calculation of basic energy requirements:

[0094] Calculate basal energy requirements based on weight gain rate.

[0095] 4. Adjustment of environmental factors:

[0096] Calculate the impact of various environmental factors on energy demand, and adjust the baseline energy demand to obtain the final energy demand E. d (t).

[0097] 5. Energy demand output:

[0098] Energy demand E at output time t d (t) is used to guide the formulation of feed feeding amounts.

[0099] 2.2.3 Determining the Feed Amount

[0100] Based on the calculated energy demand E d (t), combined with the energy content of the feed (C) f Given kJ / kg, determine the feed amount F(t) at the current time t:

[0101]

[0102] This formula allows the calculated energy requirements to be converted into specific feed amounts, ensuring the rational use of feed, reducing waste, and improving breeding efficiency.

[0103] Furthermore, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for predicting feed intake based on a dynamic growth and energy demand model of fish as described in any one of the claims.

[0104] This invention also provides a system for predicting feed intake based on a dynamic growth and energy requirement model of fish, the system comprising:

[0105] One or more processors;

[0106] Memory, used to store one or more programs;

[0107] When the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described above for predicting feed intake based on a dynamic growth and energy requirement model of fish.

[0108] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] The method for predicting feed intake based on the dynamic growth and energy demand model of fish, provided by this invention, has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and core ideas of this invention. It should be noted that those skilled in the art can make various improvements and modifications to this invention without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this invention.

Claims

1. A method for predicting feed intake based on a dynamic growth and energy requirement model of fish, characterized in that, Includes the following steps: 1) Collect growth data and water data of the fish at each moment in the current growth cycle, and combine multiple dynamic growth models to obtain the predicted weight value under the corresponding model. 2) Based on the coefficient of determination R 2 As an evaluation index for the fitting effect, the R-value for each model is obtained by fitting multiple predicted weight values ​​to the measured weight values ​​at each time point in the growth data. 2 Value, in R 2 The dynamic growth model with the highest value is selected as the optimal dynamic growth model. 3) Energy demand model construction: The predicted weight of the fish at any time during the growth cycle is obtained through the optimal dynamic growth model. Based on the predicted weight, the growth and metabolic needs of the fish are considered, and the energy demand model of the fish at any time is constructed by taking into account the water temperature, dissolved oxygen, ammonia nitrogen and nitrite concentration in the water. 4) Calculate the feed amount at the current moment; calculate the predicted weight of the fish at the current moment using the optimal dynamic growth model, obtain the energy requirement of the fish at the current moment based on the fish energy requirement model, and obtain the feed amount at the current moment by combining the energy content of the feed per unit. In step 1, the growth data and water data include: the number of days the fish were raised, the measured weight of the fish at each moment, and the corresponding water temperature. The multiple dynamic growth models include: Specific Growth Rate (SGR) model, Daily Growth Rate (DGC) model, Average Daily Weight Gain (ADG) model, Temperature Growth Coefficient (TGC) model, and polynomial fitting model. In step 2), the coefficient of determination R 2 The value range is [0,1]. The closer the value is to 1, the better the surface model fits the growth pattern of the tested fish. In step 3), the expression for the fish energy demand model is: In the formula, The energy requirement at any given time t; The predicted weight at any given time t; The baseline energy required per unit of body weight gain; The baseline energy required to maintain metabolism per unit body weight; Let t be the water temperature at any given time. The optimal water temperature for the growth of the fish being tested; Let be the dissolved oxygen concentration in the water at any given time t; The optimal dissolved oxygen concentration for the fish being tested; Let be the concentration of ammonia nitrogen in the water at any given time t; This represents the maximum tolerance value for ammonia nitrogen concentration in the tested fish. Let be the concentration of nitrite in the water at any given time t; This represents the maximum tolerance value for nitrite concentration in the tested fish. This represents the sensitivity coefficient to environmental factors.

2. The method for predicting feed intake based on a dynamic growth and energy demand model of fish according to claim 1, characterized in that, In step 4), the current time Feeding amount F(t) a )for: ； In the formula, For the current moment Energy requirements at that time This refers to the energy content per unit of feed.

3. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the method for predicting feed intake based on a dynamic growth and energy demand model of fish, as described in any one of claims 1-2.

4. A system for predicting feed intake based on a dynamic growth and energy requirement model of fish, characterized in that, The system includes: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method for predicting feed intake based on a dynamic growth and energy demand model of fish as described in any one of claims 1-2.

Images