A sow nutrition feed formula screening method based on an optimization algorithm

Abstract

The present application belongs to the technical field of feed formula research, and more particularly relates to a sow nutrition feed formula screening method based on an optimization algorithm. By analyzing population initialization distribution, a Bernoulli mapping sequence is introduced to initialize the population, so that the population distribution is more uniform, and the search ability of the algorithm in the early stage is improved. An improved adaptive nonlinear convergence factor is introduced to balance and enhance the global search ability and local development ability of the algorithm. A neighborhood disturbance optimization mechanism is added to generate new neighborhood solutions in the iteration process to prevent the algorithm from easily falling into a local optimal solution in the iteration, especially in the later stage. Finally, by comparing the optimization of sow feed formula before and after the improvement of ChOA, the effectiveness of the BANChOA algorithm is further verified. Through the analysis of the improvement strategy of the present application, it can be seen that the improved algorithm requires a large number of parameters and the optimization performance of the algorithm is greatly affected by the neighborhood disturbance optimization. There is still room for improvement in population initialization and convergence factor.

Description

Technical Field

[0001] This invention belongs to the field of feed formulation research technology, and more specifically, relates to a method for screening sow nutritional feed formulations based on optimization algorithms. Background Technology

[0002] In pig farming, most large-scale farms adopt a self-breeding and self-raising model. From a cost perspective, feed costs account for the highest proportion of pig farming costs. Precise feed formulation not only meets the needs of breeding sows for normal reproduction and piglets for future development, but also reduces the cost of raising sows. Traditional feed formulation methods struggle to handle complex multi-ingredient optimization problems, easily leading to increased feed costs. Therefore, research on precise and rational sow feed formulation using intelligent optimization algorithms is of great significance to today's large-scale self-breeding and self-raising farming model.

[0003] Domestic and international scholars have widely applied intelligent optimization algorithms to feed formulation optimization. Foreign scholars TOAliu et al. established a linear programming model to calculate and minimize the cost of feed formulations while meeting nutritional requirements. They optimized feed for pigs at different stages, including pregnant and lactating sows. Experimental results showed that optimization using the linear programming model significantly reduced pig feed costs while maintaining nutritional quality. Dipti Singh et al. were the first to combine linear programming (LP) with Self-Organizing Transfer Genetic Algorithm (SOMGA) for feed formulation optimization. This combination utilizes the rapid preliminary solution capability of linear programming and the fine-grained optimization capability of SOMGA. Compared to LP, SOMGA not only significantly improves computational efficiency and solution quality but also finds lower-cost feed formulations. Adem AlpaslanAltun and Mehmet Akif... Two researchers proposed using the Particle Swarm Optimization (PSO) algorithm to solve the cost optimization problem of mixed feed. They compared the results of this method with those of linear programming and real-valued genetic algorithms, finding that the PSO algorithm performed better. Domestic scholars, such as Li Hao et al., used PSO to optimize pig feed formulation, overcoming the difficulty of traditional methods in handling complex multi-constraint or multi-objective problems. Simulation results showed that PSO outperformed the traditional linear programming (LP) method in solving pig feed formulation problems. Zhang Xuecong et al. used an improved tabu search algorithm (ITS) to optimize the cost of pig feed formulation. By introducing new neighborhood movement rules and shortening the tabu list, they optimized the traditional Tabu Search (TS) algorithm. Compared with traditional algorithms such as LP and genetic algorithms (GA), they proved that ITS is superior in solving livestock feed formulation problems. Yang Shasha et al. used an improved genetic algorithm to optimize pig feed formulation. They adopted real-valued encoding, randomly generated the initial population, used an improved proportional selection strategy and an elite preservation strategy to select superior individuals for the next generation, used partial matching crossover for adaptive crossover, and used Gaussian mutation for adaptive mutation, reducing the cost of pig farming. Liu Qing et al. first applied the Artificial Fish Swarm Algorithm (AFSA) to feed formulation optimization. To meet the convergence accuracy requirements of feed formulation optimization, they adopted an artificial fish swarm algorithm operating framework based on a symbiotic system, which significantly improved the convergence accuracy and speed of the original algorithm. The calculation results were compared with some existing intelligent algorithms, and the results showed that it was superior to other existing intelligent algorithms.

[0004] While existing literature has used intelligent algorithms to optimize sow feed formulations, research on applying novel algorithms to this field remains relatively limited. ChOA, as a novel intelligent algorithm, has two main characteristics: first, it divides the population into individual entities, effectively improving the algorithm's exploration capabilities; second, it introduces a chaos factor, which helps improve the convergence speed and accuracy of the development process. Summary of the Invention

[0005] This invention improves upon the ChOA (Cheaper of the Oxygen Optimizer) algorithm, proposing a method for screening sow nutrient feed formulations based on an optimization algorithm. Addressing the shortcomings of ChOA, the population initialization of the chimpanzee optimization algorithm is improved by replacing random initialization with a Bernoulli mapping sequence and introducing an adaptive nonlinear convergence factor to better adapt to the actual search conditions. Furthermore, a neighborhood perturbation mechanism based on greedy selection is added. Experimental comparisons verify the feasibility and superiority of the improved chimpanzee optimization algorithm in cost optimization of sow feed formulations with multiple constraints and various raw material combinations.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for screening sow nutrient feed formulations based on an optimization algorithm is disclosed, where m represents the number of raw materials in the sow feed formulation and n represents the number of nutrients. The steps of the sow nutrient feed formulation screening method are as follows:

[0008] Step 1: Initialize the algorithm-related parameters, including the dimension of the raw solution, the maximum number of iterations, the upper and lower bounds of the search space, and the objective function;

[0009] Step 2: Define a fitness function to evaluate the merits of each formulation scheme. The fitness function has two evaluation indicators: total cost and whether the nutritional components of each raw material are within the constraints.

[0010] Step 3: Generate a population using the Bernoulli mapping sequence, where each individual in the population represents the formula content of a set of m raw materials;

[0011] Step 4: Enter the main loop, calculate the dynamic coefficients of f and CR, calculate the fitness of each search agent position, and evaluate the quality of the formula based on its nutritional components and cost.

[0012] The parameter f is the convergence factor. In the improved algorithm, the linearly decaying convergence factor is changed to a non-linearly decaying convergence factor to balance the algorithm's global and local search capabilities. A neighborhood perturbation strategy is added to the improved algorithm, and CR is a parameter used to determine whether to perform neighborhood perturbation. As the algorithm progresses into later stages, the probability of performing neighborhood perturbation increases with the increase of CR.

[0013] Step 5: Update the search agent's location based on the positions of attackers, obstacle users, chasers, and exterminators to explore better feed formulations;

[0014] Step 6: Determine if a neighborhood operation can be performed. If so, use the neighborhood operation to further optimize the position of each search agent; otherwise, proceed to Step 7.

[0015] Step 7: If the maximum number of iterations is reached, record the current optimal solution and output the result; otherwise, return to Step 4.

[0016] Further optimization of this technical solution, the objective function minF(X) model is as follows:

[0017]

[0018] In the formula, c i —The price of each raw material, x i —The proportion of different raw materials in 1kg of feed formulation.

[0019] Further optimization of this technical solution involves the following constraints on the upper and lower limits of raw material usage, nutritional components, and the total proportion of raw materials in the sow feed:

[0020]

[0021] x1+x2+…+x m =1 (18)

[0022] In the formula: a ij —The percentage of the j-th nutrient contained in the i-th raw material; d i —Minimum standards for various nutritional indicators that the formula should meet; m—Number of ingredients in the formula; n—Number of nutrients.

[0023] In a further optimization of this technical solution, the fitness function is used to evaluate the merits of each formulation scheme. The fitness function has two evaluation indicators: total cost and whether the nutritional components of each raw material are within the constraints. The fitness function is as follows:

[0024] f(x)=minF(x)+pen(x) (19)

[0025] Where minF(x) is the objective function of equation (16); pen(x) is a penalty term used to penalize solutions that do not meet the constraints, and its expression is as follows:

[0026] pen(x)=p1(x)+p2(x)+p3(x) (20)

[0027]

[0028] In the formula: p1(x) is the penalty for the component ratio constraint, which is used when the raw material does not meet the upper and lower limits; p2(x) is the penalty for the total ratio constraint, which is used when the formula (18) is not met; p3(x) is the penalty for the nutritional requirement constraint, which is used when the formula (19) is not met; where in the formula (21), li and ui are the lower and upper limits of the i-th raw material, respectively.

[0029] Unlike existing technologies, the above technical solution has the following beneficial effects:

[0030] By analyzing the initial population distribution, a Bernoulli mapping sequence is introduced to initialize the population, resulting in a more uniform population distribution and improving the search capability in the early stages of the algorithm. An improved adaptive nonlinear convergence factor is introduced to balance and enhance the algorithm's global search capability and local exploitation capability. A neighborhood perturbation optimization mechanism is added to prevent the algorithm from easily getting trapped in local optima, especially in the later stages of iteration, by generating new neighborhood solutions during the iteration process. The superiority of the improved algorithm is verified through optimization tests on two basic test functions. Finally, by comparing the optimization of sow feed formulation before and after the ChOA improvement, the effectiveness of the BANChOA algorithm (an improvement on the basic chimpanzee optimization algorithm that introduces a Bernoulli mapping sequence to initialize the population; introduces an improved adaptive nonlinear convergence factor; and adds a neighborhood perturbation optimization mechanism, named BANChOA) is further verified. Analysis of the improvement strategy of this invention shows that the improved algorithm requires a large number of parameters, and its optimization performance is greatly affected by neighborhood perturbation optimization. There is still room for improvement in population initialization and convergence factor. In future work, we will continue to research improved optimization strategies to enhance the operability and applicability of ChOA and apply it to more complex optimization problems. Attached Figure Description

[0031] Figure 1 Here is the flowchart for the BANChOA algorithm;

[0032] Figure 2 A comparison graph of F1 for unimodal functions;

[0033] Figure 3 A comparison chart of F2 for nonlinear multimodal functions;

[0034] Figure 4 A comparison chart of price trends for feed formulations of ChOA and BANChOA in early pregnancy (<90 days);

[0035] Figure 5 A comparison chart of price trends for feed formulations of ChOA and BANChOA in late pregnancy (>90 days);

[0036] Figure 6 This is a comparison chart of the price trends of ChOA and BANCHOA in lactation feed formulations. Detailed Implementation

[0037] To explain in detail the technical content, structural features, objectives, and effects of the technical solution, the following description is provided in conjunction with specific embodiments and accompanying drawings.

[0038] The data on sow feed and required nutrient content for this invention's experiments were primarily derived from the "Chinese Feed Composition and Nutritional Value Table" (30th edition, 2019) and "Nutritional Requirements of Pigs." The feed ingredients were formulated according to their proportions per kilogram, including 11 ingredients such as corn, wheat bran, fish meal, sodium-supplementing salt, and additives containing minerals and vitamins, as well as 16 nutrients, including digestible energy. Because sows go through different stages in their breeding process, namely gestation and lactation, and gestation is further divided into early gestation (<90 days) and late gestation (>90 days), there are corresponding ranges for ingredient usage and different nutrient requirements at each stage. Specific feed ingredient and nutrient composition data are shown in Table 1-1 below, and ingredient prices are shown in Table 1-2.

[0039] Table 1-1 Feed ingredient data and nutrient ratios

[0040]

[0041]

[0042] Table 1-2 Raw Material Prices

[0043]

[0044] The required dosage ranges for feed ingredients vary depending on the stage of the sow's growth. Specific dosage ranges are shown in Table 1-3.

[0045] Table 1-3 Range of Feed Ingredient Usage for Sows at Different Stages

[0046]

[0047] Salt is used to supplement the sow's sodium levels, while additives are used to supplement the sow's required vitamins and trace elements, with the dosage kept constant.

[0048] The nutritional needs of sows vary at different stages. In early gestation, protein requirements are moderate, supporting protein synthesis in the sow and early embryonic development. In late gestation, protein requirements increase to support fetal muscle and tissue formation, while calcium and phosphorus supplementation is needed to support fetal bone development and farrowing. During lactation, protein requirements are highest, used for protein synthesis in milk, while large amounts of calcium and phosphorus are required to support milk production and sow health. Specific data are shown in Table 1-4.

[0049] Table 1-4 Minimum Nutrient Requirements for Sows at Different Stages

[0050]

[0051] The proposed multi-strategy neighborhood perturbation chimpanzee algorithm (BANChOA) is based on ChOA and incorporates the three improved strategies mentioned above. The input to the BANChOA algorithm is the Bernoulli mapping sequence to initialize the population and related parameters; the output is the position of the optimal chimpanzee individual. The specific implementation steps and process are as follows: Figure 1 As shown.

[0052] Step 1: Initialize the population size, maximum number of iterations Max_ter, search range lb, ub, and set the corresponding parameters, where lb is the lower bound of the search range and ub is the upper bound of the search range.

[0053] Step 2: Initialize population X using the Bernoulli mapping sequence. i , i = 1, 2, ..., N.

[0054] Step 3: Calculate the fitness value of each individual, and identify the top four values, which are denoted as four types of chimpanzee individuals, and labeled as X. Attacker X Barrier X Chaser and X Driver .

[0055] Step 4: Update the values ​​of coefficients a, c, and convergence factor f according to the formula.

[0056] Step 5: Use the probability factor CR to determine whether to further update the positions of the four types of chimpanzees and the positions of other chimpanzees.

[0057] Step 6: Determine whether the target satisfactory value has been reached. If it has, output the optimal value; otherwise, proceed to Step 7.

[0058] Step 7: Determine if the maximum number of iterations has been reached. If not, go back to Step 3; otherwise, end the algorithm and output the location of the optimal chimpanzee.

[0059] Experimental environment and parameter settings

[0060] The hardware environment for the simulation experiments of this invention is based on an Intel(R) Core(TM) i5-4590 CPU, 3.30GHz clock speed, 16.0GB memory, and a Windows 10 (64-bit) operating system. The programming software is Matlab R2022a. Benchmark test functions were selected to perform performance tests on the improved algorithms. Table 2-1 shows the basic information of the unimodal function F1 and the nonlinear multimodal function F2. Table 2-2 shows the parameter settings for each algorithm.

[0061] Table 2-1 Benchmark Functions

[0062]

[0063]

[0064] Table 2-2 Algorithm Parameter Settings

[0065]

[0066] Algorithm Model Simulation Result Analysis and Evaluation

[0067] Comparison results of Bernoulli mapping sequence initialization population

[0068] The ChOA using the Bernoulli mapping sequence is called BChOA (Chimp Optimization Algorithm for Bernoulli). Bernoulli (the fundamental quantity of the mapping sequence is λ = 0.7, initial z...) k Let the random values ​​be between (0,1). Assume the population size N = 50 and the maximum number of iterations Max_ter = 500. The comparison of the optimization of ChOA and BChOA in the unimodal function F1 and the nonlinear multimodal function F2 is shown in Table 2-3.

[0069] Table 2-3 Comparison of optimization results between ChOA and BChOA

[0070]

[0071] Table 2-3 shows that in the optimization tests for unimodal functions F1 and nonlinear multimodal functions F2, BChOA's ability to find the optimal value is superior to that of ChOA. The standard deviation comparison shows that BChOA has a smaller standard deviation for both test functions, indicating better stability and robustness. In conclusion, the improvement of the Bernoulli mapping sequence initialization population has an impact on the performance of the chimpanzee optimization algorithm.

[0072] Comparison results with the addition of an adaptive nonlinear convergence factor

[0073] The BChOA with the addition of an adaptive nonlinear convergence factor f is called BAChOA (Chimp Optimization Algorithm for Bernoulli and Adaptive nonlinear convergence factor). The comparison between BChOA and BAChOA in the optimization of unimodal function F1 and nonlinear multimodal function F2 is shown in Table 2-4.

[0074] Table 2-4 Comparison of optimization results between BChOA and BAChOA

[0075]

[0076] Table 2-4 shows that in the optimization tests for unimodal functions F1 and nonlinear multimodal functions F2, BAChOA's ability to find the optimal value is superior to BChOA's. The standard deviation comparison shows that BAChOA has a lower standard deviation, indicating better stability and robustness. In conclusion, adding an adaptive nonlinear convergence factor f improves the performance of the Chimpanzee optimization algorithm.

[0077] Comparison results of optimization with added neighborhood perturbation

[0078] The BAChOA with added neighborhood perturbation function is called BANChOA. Table 2-5 compares the optimization of BAChOA and BANChOA for unimodal functions F1 and nonlinear multimodal functions F2.

[0079] Table 2-5 Comparison of Optimization Results between BAChOA and BANChOA

[0080]

[0081] Table 2-5 shows that in the optimization tests for both the unimodal function F1 and the nonlinear multimodal function F2, BANChOA outperforms BAChOA in finding the optimal value, and BANChOA even reaches the theoretical value when optimizing the test function F2. BANChOA also exhibits a lower standard deviation in the optimization tests for both functions, indicating better stability and robustness. In conclusion, the addition of neighborhood perturbation optimization has an impact on the performance of the Chimpanzee optimization algorithm.

[0082] Comprehensive comparison results of different improvement strategies

[0083] Based on the above ablation experiments, a comprehensive comparison of the average convergence curves of ChOA, BChOA, BAChOA, and BANChOA for the unimodal function F1 and the nonlinear multimodal function F2 is presented. The comparison results are as follows: Figure 2 and Figure 3 As shown, each addition of an improvement strategy enhances the algorithm's optimization. Compared to ChOA, the improved BANChOA exhibits significantly stronger average convergence performance, achieving the expected results.

[0084] Comparison experiment with other algorithms

[0085] To further observe the effectiveness of the improved chimpanzee optimization algorithm, BANChOA and ChOA were compared with the classic Particle Swarm Optimization (PSO) algorithm and the Marine Predators Algorithm (MPA) proposed by Afshin Faramarzi et al. in 2020. Two benchmark functions were used: a unimodal function F1 and a nonlinear multimodal function F2, with a population size of 50 and a maximum number of iterations of 500. The comparison results are shown in Table 2-6.

[0086] Table 2-6 Comparison of Results of Different Optimization Algorithms

[0087]

[0088]

[0089] As shown in the table above, BANChOA outperforms other intelligent optimization algorithms in finding the optimal value, and its standard deviation is also lower and more stable compared to other algorithms, further demonstrating that the improved chimpanzee optimization algorithm BANChOA has stronger optimization performance and robustness.

[0090] Optimization of sow feed formulation based on BANChOA

[0091] Constructing the fitness function

[0092] BANChOA was used to optimize the cost of sow feed formulation. The optimization problem of sow feed formulation is a combinatorial optimization problem with multiple objectives and constraints.

[0093] The experimental design uses the minimum economic cost as the objective function, with upper and lower limits for raw material usage at different stages, and nutritional indicators for sows at different weights and stages as constraints. The objective function minF(X) model is as follows:

[0094]

[0095] In the formula, c i —The price of each raw material, x i —The proportion of different raw materials in 1kg of feed formulation.

[0096] The upper and lower limits of raw material usage in sow feed are shown in Table 1-3, and the constraints on nutrient composition and total raw material proportions are as follows:

[0097]

[0098] x1+x2+…+x m =1 (18)

[0099] In the formula: a ij —The percentage of the j-th nutrient contained in the i-th raw material; d i —The minimum standards for various nutritional indicators that the formula should meet; m—the number of raw materials in the formula; n—the number of nutrients. As shown in Table 1-1, there are a total of 11 raw materials and 16 nutrients in sow feed, so the value of m is 11 and the value of n is 16.

[0100] A fitness function is defined to evaluate the merits of each formulation. The fitness function has two evaluation metrics: total cost and whether the nutritional composition of each ingredient is within the constraints. The fitness function is as follows:

[0101] f(x)=minF(x)+pen(x) (19)

[0102] Where minF(x) is the objective function of equation (16); pen(x) is a penalty term used to penalize solutions that do not meet the constraints, and its expression is as follows:

[0103] pen(x)=p1(x)+p2(x)+p3(x) (20)

[0104]

[0105] In the formula: p1(x) is the penalty for the component ratio constraint, which is used when the raw material does not meet the upper and lower limits; p2(x) is the penalty for the total ratio constraint, which is used when the formula (18) is not met; p3(x) is the penalty for the nutritional requirement constraint, which is used when the formula (19) is not met; where in the formula (21), li and ui are the lower and upper limits of the i-th raw material, respectively.

[0106] Optimization process

[0107] As shown in Table 1-1, there are a total of 11 raw materials and 16 nutrients in sow feed, so m is 11 and n is 16. The specific implementation steps of BANChOA in the research on sow feed formulation are as follows.

[0108] Step 1: Initialize algorithm-related parameters, such as the dimension of the raw material solution, the maximum number of iterations, the upper and lower bounds of the search space, and the objective function.

[0109] Step 2: Define a fitness function to evaluate the merits of each formulation. The fitness function has two evaluation metrics: total cost and whether the nutritional components of each ingredient are within the constraints.

[0110] Step 3: Use the Bernoulli mapping sequence to generate a population, with each individual in the population representing the formula content of a set of 11 raw materials.

[0111] Step 4: Enter the main loop, calculate the dynamic coefficients f and CR, calculate the fitness of each search agent position, and evaluate the quality of the formula based on its nutritional components and cost. The parameter f is the convergence factor; in the improved algorithm, the linearly decaying convergence factor is changed to a non-linearly decaying one. A neighborhood perturbation strategy is added to the improved algorithm, and CR is a parameter used to determine whether to perform neighborhood perturbation. As the algorithm progresses into later stages, the probability of performing neighborhood perturbation increases with the increase of CR.

[0112] Step 5: Update the search agent's location based on the positions of attackers, obstacle users, chasers, and exterminators to explore better feed formulations.

[0113] Step 6: Determine if a neighborhood operation can be performed. If so, use the neighborhood operation to further optimize the position of each search agent; otherwise, proceed to Step 7.

[0114] Step 7: If the maximum number of iterations is reached, record the current optimal solution and output the result; otherwise, return to Step 4.

[0115] Formula optimization results analysis

[0116] The above experimental design was used to compare BANChOA and ChOA in sow feed formulation to verify the effectiveness of the improved algorithm in solving practical sow feed formulation problems. The following is a comparison of feed formulations for sows in early gestation (<90 days), late gestation (>90 days), and lactation periods.

[0117] Table 2-7 Comparison of ChOA and BANChOA in feed formulations for early pregnancy (<90 days)

[0118]

[0119] Table 2-8 Comparison of ChOA and BANChOA in feed formulations during late pregnancy (>90 days)

[0120]

[0121]

[0122] Table 2-9 Comparison of ChOA and BANChOA in lactating feed formulation

[0123]

[0124]

[0125] Tables 2-7, 2-8, and 2-9 show the proportions and costs of different ingredients in the feed formulations for sows at different stages obtained using ChOA and BANChOA. Figure 4 , Figure 5 and Figure 6 These are comparison charts of the corresponding total cost trends. The experimental results show that, while meeting the nutritional requirements of sows, the improved ChOA reduces costs by 8.5% per kilogram in early gestation, 14.4% per kilogram in late gestation, and 18.2% per kilogram in lactation compared to standard ChOA. The price of BANChOA-optimized sow feed is significantly reduced.

[0126] It should be noted that in this invention, relational terms such as "first" and "second" are used merely 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..." or "including..." does not exclude the presence of additional elements in the process, method, article, or terminal device that includes said element. Additionally, in this invention, "greater than," "less than," "exceeding," etc., are understood to exclude the stated number; "above," "below," "within," etc., are understood to include the stated number.

[0127] Although the above embodiments have been described, those skilled in the art, once they understand the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the above descriptions are merely embodiments of the present invention and do not limit the scope of patent protection of the present invention. Any equivalent structural or procedural transformations made using the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for screening sow nutrient feed formulations based on optimization algorithms, characterized in that, Sow feed The number of ingredients in the formula. The steps for screening sow nutrient feed formulations based on the number of nutrients are as follows: Step 1: Initialize algorithm-related parameters, including the dimension of the raw material solution, the maximum number of iterations, the upper and lower bounds of the search space, and the objective function; Step 2: Define a fitness function to evaluate the merits of each formulation scheme. The fitness function has two evaluation indicators: total cost and whether the nutritional components of each raw material are within the constraints. Step 3: Use the Bernoulli mapping sequence to generate a population, where each individual in the population represents the formula content of a set of m raw materials; Step 4: Enter the main loop and calculate the convergence factor. The fitness of each search agent location is calculated using the neighborhood perturbation CR dynamic coefficient, and the quality of the formula is evaluated based on its nutritional composition and cost. Convergence factor CR is the convergence factor for nonlinear decay. CR is a parameter used to determine whether to perform neighborhood perturbation. As CR increases, the probability of performing neighborhood perturbation increases. Step 5: Update the search agent's location based on the positions of attackers, obstacle users, chasers, and exterminators to explore better feed formulations; Step 6: Determine if a neighborhood operation can be performed. If so, use the neighborhood operation to further optimize the position of each search agent; otherwise, proceed to Step 7. Step 7: If the maximum number of iterations is reached, record the current optimal solution and output the result; otherwise, return to Step 4.

2. The method for screening sow nutrient feed formulations based on optimization algorithms as described in claim 1, characterized in that, The objective function The model is as follows: (16) In the formula, —The price of each raw material, —The proportion of different raw materials in 1kg of feed formulation.

3. The method for screening sow nutrient feed formulations based on optimization algorithms as described in claim 2, characterized in that... The constraints on the upper and lower limits of raw material usage, nutritional composition, and total proportion of raw materials in the sow feed are as follows: (17) (18) In the formula: —No. The first type of raw material contains Percentage of each nutrient component; —Minimum standards that the formula should meet for each nutritional indicator; —Number of ingredients in the formula; —Number of nutrients.

4. The method for screening sow nutrient feed formulations based on optimization algorithms as described in claim 3, characterized in that, The fitness function is used to evaluate the merits of each formulation scheme. The fitness function has two evaluation metrics: total cost and whether the nutritional components of each raw material are within the constraints. The fitness function is as follows: (19) in, It is the objective function of equation (16); This is a penalty term used to penalize solutions that do not meet the constraints. Its expression is as follows: (20) (21) (22) (23) In the formula: It is a penalty for component ratio constraints, used when raw materials do not meet the upper and lower limits. It is a penalty for the overall proportion constraint, used when equation (18) is not satisfied. It is a penalty for nutritional requirement constraints, used to address non-compliance with equation (19); where in equation (21), and They are the first The lower and upper limits of each raw material.

Images