Live pig low-protein low-soybean-meal diversified daily ration prediction method established based on demand quantity subdivision method

Abstract

The invention belongs to the field of agricultural information technology and animal nutrition cross technology, and discloses a live pig low-protein low-soybean-meal diversified daily ration prediction method based on demand quantity subdivision, which adopts a mathematical model to obtain the growth characteristics of growing-finishing pigs of the variety. On the basis, the protein and amino acid demand amounts of growing-finishing pigs with different growth performances are obtained by adopting a demand amount subdivision method model, and the protein and amino acid demand amounts of the growing-finishing pigs with different growth performances are obtained according to comprehensive indexes such as low-protein daily ration models in different stages of the pigs, soybean meal substitute raw material shadow prices and growth performance requirements. And constructing the low-protein low-soybean-meal diversified daily ration for the growing-finishing pigs by using a linear programming method. Aiming at the growth difference of different varieties of growing-finishing pigs, adopting the screened core production indexes as input variables to establish a growth model, and adopting a demand quantity subdivision method to obtain corresponding protein and amino acid demand quantities under the growth model; the accuracy and the reliability of the low-protein low-soybean-meal diversified daily ration for the growing-finishing pigs are improved.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of agricultural information technology and animal nutrition, specifically relating to a method for predicting diversified diets of low-protein and low-soybean meal for growing-finishing pigs based on a requirement breakdown method. Background Technology

[0002] Soybean meal is widely used in pig farming in my country due to its high protein content and amino acid composition that closely matches animal needs. However, protein feed ingredients, represented by soybeans, face risks such as high import dependence and concentration of import sources, posing a significant risk to my country's food security. Furthermore, my country's pig farming has long followed the US model, relying primarily on high-protein soybean meal diets. This has not only fostered a mindset of judging feed quality based on protein or soybean meal content but has also led to substantial waste of feed resources and nitrogen pollution. Moreover, the increasingly tight global supply and rising prices of protein raw materials have resulted in continuously increasing feed costs, placing enormous economic pressure on the pig industry. Therefore, precise formulation technology for low-protein, low-soybean meal diets has become an urgent industry need, crucial for cost savings for farmers, feed production savings for feed producers, increased efficiency across the industry, and national food security. Current technologies mainly employ low-protein diets, which reduce the crude protein level in the diet by 2-4% from the NRC (1998) recommended level while rationally adding synthetic amino acids to meet the needs of livestock and poultry.

[0003] However, the existing methods have the following main shortcomings:

[0004] 1. Current low-protein diet programs mainly reduce the crude protein level of the diet by 2-4% based on the NRC (1998) recommended crude protein level for each stage of fattening pigs. However, the crude protein level of the NRC is based on the average results obtained from the pig herds studied, and may not be completely suitable for the growth performance of different pig herds in my country. Moreover, the crude protein level does not take into account the standard ileal digestibility (SID) characteristics of Chinese pig herds. SID is a key indicator for accurately matching the amino acid absorption efficiency of pigs, which directly affects the nutrient utilization rate of the diet and the breeding benefits.

[0005] 2. The pig weight stage division in the NRC nutritional requirements is fixed rather than dynamic, and is not entirely suitable for the stage-based feeding pattern of pig herds in my country.

[0006] 3. Existing technologies (including factorial models applied to other species such as poultry) fail to provide a factorial model specifically applicable to pigs that can dynamically analyze the relationship between growth performance and protein deposition. Factorial models for species such as poultry are based on their unique physiological indicators (such as pectoral muscle rate and leg muscle rate), and their formulas and parameters are completely unsuitable for the lean meat growth pattern of pigs. Furthermore, they only focus on predicting requirements and do not address the issues of low soybean meal substitution and formula cost optimization, which are fundamentally different from the technical objectives and application scenarios of this invention.

[0007] 4. Currently, most domestic and foreign scholars' research focuses on low-protein diets. Even though low-protein diets reduce protein levels, soybean meal is still traditionally needed to meet the remaining protein levels. However, my country is severely lacking in protein raw material resources, and there is still relatively little research on diversified low-protein, low-soybean meal diets for pigs based on my country's national conditions.

[0008] 5. More importantly, existing technologies lack a systematic solution that connects accurate dynamic nutrient requirement prediction with the design of diversified diet formulations with low soybean meal, making it impossible to achieve full-process automation and precision from "requirement prediction" to "cost-optimal formulation generation".

[0009] Therefore, the technical problem to be solved by this invention is: how to develop a highly accurate and reliable method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement breakdown method, and to use SID protein as a nutritional indicator in the diet design, taking into account the standard ileal digestibility of pigs. Summary of the Invention

[0010] The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies and provide a method for predicting low-protein, low-soybean meal diversified diets specifically for growing-finishing pigs, based on a pig-specific requirement segmentation model. This method establishes a dynamic function relating pig-specific body weight, cumulative net energy intake, lean meat weight, and age to accurately segment its maintenance and deposition requirements. Finally, through linear programming, it integrates the cost and nutritional indicators of soybean meal substitutes to generate a cost-optimal low-soybean meal diversified diet formula. This method differs fundamentally from nutritional prediction methods for other species such as poultry in terms of model, parameters, and final application.

[0011] A method for predicting low-protein, low-soybean meal diversified diets for pigs based on requirement decomposition, specifically designed for growing-finishing pigs, includes the following steps:

[0012] Step 1: Obtain production data and feed formulation data for growing-finishing pigs;

[0013] Step 2: Organize and clean the collected growing-finishing pig production data and feed formulation data to form standardized production sample datasets and formulation sample datasets;

[0014] Step 3: Analyze the formula sample dataset and the production sample dataset to obtain the body weight, cumulative net energy intake, and lean meat weight at each stage; conduct regression modeling analysis on the production sample dataset to establish a dynamic functional relationship between body weight (BW), cumulative net energy intake (NEi), lean meat weight (FFL), and age (d) for growing-finishing pigs.

[0015] Step 4: Based on the functional relationship between the lean meat weight of growing-finishing pigs and age, calculate the body protein content of growing-finishing pigs at each growth stage, and adopt a requirement breakdown method suitable for pigs. The requirement breakdown method formula is: Total protein requirement = Maintenance requirement ((intestinal loss + fur loss) ÷ Maintenance efficiency coefficient) + Deposition requirement (body protein deposition ÷ Deposition efficiency coefficient). Establish a dynamic model of protein and lysine requirements of pigs at different growth stages.

[0016] Step 5: Based on the actual production process of dividing growing and finishing pigs into weight stages and setting the net energy level of the diet, calculate the corresponding dietary protein and lysine levels for each weight stage.

[0017] Step 6: Based on the theory of ideal protein and amino acid balance, construct a low-protein diet model for different weight stages of growing-finishing pigs;

[0018] Step 7: Integrating low-protein diet models, shadow prices of soybean meal substitutes, raw material usage limits, growth performance constraints, and multi-dimensional indices for minimizing diet costs, a linear programming method is used to automatically construct diversified low-protein, low-soybean meal diet formulas for growing-finishing pigs.

[0019] Furthermore, the production data for the growing-finishing pigs includes sex, age, weight, feed intake, backfat thickness, and eye muscle area; the feed formulation data is the net energy level of the diet.

[0020] Furthermore, step 2 includes detecting whether the production data of growing-finishing pigs contains null values ​​and outliers. If null values ​​and outliers are found, the null values ​​and outliers are deleted.

[0021] Furthermore, step 3 includes the following sub-steps:

[0022] Step A1: Calculate the cumulative net energy intake of growing-finishing pigs based on the net energy level and feed intake of the diet at each stage.

[0023] Step A2: Calculate the lean meat weight of the growing-finishing pigs based on their sex, weight at each stage, backfat thickness, and eye muscle area.

[0024] Step A3: Use the prediction model y=a+bx+cx 2 +dx 3 Modeling is performed where y represents body weight, cumulative net energy intake, or body protein deposition, x represents age in days, a is a constant term, b is the coefficient of the first term, c is the coefficient of the second term, and d is the coefficient of the third term. Abnormal samples appearing during the modeling process are identified and removed using studentized residuals (greater than 2.0 or less than -2.0). A function of body weight, cumulative net energy intake, lean meat weight, and age in days is established.

[0025] Furthermore, step 4 includes the following sub-steps:

[0026] Step B1: Calculate the body protein content of growing-finishing pigs at each stage based on the function of lean meat weight and age. Calculate the amount of lysine deposited in the body protein deposition of growing-finishing pigs at each stage based on the lysine content per gram of body protein.

[0027] Step B2: The requirement breakdown method divides the protein and amino acid requirements into maintenance requirements and deposition requirements. The maintenance requirements of protein and lysine are calculated based on the body weight, dry matter intake and maintenance efficiency of feed at each stage of growing and finishing pigs. The deposition requirements of protein and lysine are calculated based on the amount and efficiency of lysine deposition in body protein at each stage of growing and finishing pigs.

[0028] Step B3: Add the maintenance requirements and deposition requirements of protein and lysine to obtain the total requirements of protein and lysine for each stage of growing-finishing pigs.

[0029] Furthermore, step 5 includes the following sub-steps:

[0030] Step C1: Based on the actual production needs of the pig farm, set the weight division and net energy level of the diet for the growing and finishing pig stages;

[0031] Step C2: Based on the cumulative net energy intake and daily net energy level of growing-finishing pigs at each stage, calculate the feed intake of growing-finishing pigs at each weight stage.

[0032] Step C3: Based on the total protein and lysine requirements of growing-finishing pigs at each stage, calculate the dietary protein and lysine levels for each weight stage of growing-finishing pigs.

[0033] Furthermore, in step 6, based on the obtained dietary protein and lysine levels of growing-finishing pigs at each weight stage, and according to the amino acid balance theory, the levels are further reduced by 1-2 percentage points to formulate a low-protein diet model for growing-finishing pigs at different weight stages.

[0034] Furthermore, in step 7, based on the low-protein diet model for pigs at different stages, the shadow price of soybean meal substitute raw materials, raw material usage limits, growth performance constraints, and cost indicators, a multi-dimensional index is constructed using linear programming to build a diversified diet for growing and finishing pigs that is low in protein and low in soybean meal.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] This proposed low-protein, low-soybean meal diversified diet for swine utilizes a mathematical model. By constructing functions relating body weight, cumulative net energy intake, body protein deposition, and age for swine pigs with different growth performance, the growth characteristics of this breed of swine pig are obtained. Based on this, a requirement decomposition model is used to obtain the protein and amino acid requirements of swine pigs with different growth performance. Furthermore, based on a comprehensive index including low-protein diet models for different swine stages, shadow prices of soybean meal substitutes, and growth performance requirements, a linear programming method is applied to construct a low-protein, low-soybean meal diversified diet for swine pigs. Addressing the growth differences among different swine pig breeds, selected core production indicators are used as input variables to establish a growth model. The requirement decomposition method is then used to obtain the corresponding protein and amino acid requirements under this growth model, improving the accuracy and reliability of the low-protein, low-soybean meal diversified diet for swine pigs. This method overcomes the limitations of the linear assumptions in traditional nutritional models, dynamically analyzing the protein and amino acid requirements of swine pigs with different growth performance, providing data-driven decision support for precise diet formulation.

[0037] This invention provides for the first time a complete technical system from predicting the dynamic nutritional requirements of pigs to automatically generating diversified diet formulas with low soybean meal. It differs fundamentally from the factorial model used for other species such as poultry in terms of its foundation, technical complexity, and final application results. It provides an irreplaceable and dedicated solution for cost reduction and efficiency improvement in pig farming and food security in my country. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the embodiments of this application. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] Example 1 (Comparison of the same variety at different stages):

[0040] Step 1: Obtain production data and feed formulation data for growing-finishing pigs;

[0041] In this embodiment, the production data for growing-finishing pigs of the same breed includes sex, age, weight, feed intake, backfat thickness, and eye muscle area; the feed formulation data is the net energy level of the diet.

[0042] We collected 360 production data sets of growing-finishing pigs through production practice. Table 1 shows some examples of the data.

[0043] surface Summary of growth performance data

[0044]

[0045] Step 2: Process the production data of growing-finishing pigs, and check whether the production data of growing-finishing pigs contains null values ​​and outliers. If it contains null values ​​and outliers, delete the null values ​​and outliers to obtain the production sample dataset.

[0046] Step 3 includes the following sub-steps:

[0047] Step A1: Based on the net energy level and feed intake of the diet at each stage of growing-finishing pigs, calculate the cumulative net energy intake of half male and half female growing-finishing pigs;

[0048] In this embodiment, the calculation results are shown in Table 2:

[0049] Table 2 Cumulative Net Energy Intake of Growing-Finishing Pigs

[0050]

[0051] Step A2: Based on the sex, weight at each stage, backfat thickness, and eye muscle area of ​​the growing-finishing pigs, calculate the lean meat weight of the half-male and half-female growing-finishing pigs at each stage;

[0052] In this embodiment, the calculation results are shown in Table 3:

[0053] Table 3. Lean meat weight without fat in growing-finishing pigs

[0054]

[0055] Step A3: Use the prediction model y=a+bx+cx 2 +dx 3 A model was constructed, where y represents body weight, cumulative net energy intake, or body protein deposition, x represents age in days, a is a constant term, b is the coefficient of the linear term, c is the coefficient of the quadratic term, and d is the coefficient of the cubic term. Outliers were identified and removed during the modeling process using studentized residuals (greater than 2.0 or less than -2.0). A function was established relating body weight, cumulative net energy intake, body protein deposition, and age in days.

[0056] (1) Relationship between body weight and age:

[0057] y=-0.000033x 3 +0.012128x 2 -0.494802x+12.829298(R²=0.973110)

[0058] (2) Relationship between cumulative net energy intake and age:

[0059] y=-0.000116x 3 +0.061186x 2 -3.327895x+56.641862 (R²=0.991142)

[0060] (3) Relationship between lean meat weight and age:

[0061] y=-0.000017x 3 +0.005984x 2 -0.264462x+6.023537 (R²=0.967094)

[0062] Step 4 includes the following sub-steps:

[0063] Step B1: Calculate the body protein content of growing-finishing pigs at each stage based on the function of lean meat weight and age. Calculate the lysine content in the body protein deposition of growing-finishing pigs at each stage based on the lysine content per gram of body protein.

[0064] In this embodiment, the calculation results are shown in Table 4:

[0065] Table 4. Protein and Lysine Content in Growing-Finishing Pigs

[0066]

[0067] Step B2: The requirement breakdown method divides the protein and amino acid requirements into maintenance requirements and deposition requirements. The maintenance requirements of protein and lysine are calculated based on the body weight, dry matter intake and maintenance efficiency of feed at each stage of growing and finishing pigs. The deposition requirements of protein and lysine are calculated based on the amount and efficiency of lysine deposition in body protein at each stage of growing and finishing pigs.

[0068] Step B3: Add the maintenance requirements and deposition requirements of protein and lysine to obtain the total requirements of protein and lysine for each stage of growing-finishing pigs.

[0069] In this embodiment, the calculation results are shown in Table 5:

[0070] Table 5 Protein and Lysine Requirements for Growing-Finishing Pigs

[0071]

[0072] Step 5 includes the following sub-steps:

[0073] Step C1: Based on the actual production needs of the pig farm, set the weight division and net energy level of the diet for the growing and finishing pig stages;

[0074] In this embodiment, based on the actual production needs of the pig farm, the 60kg fattening pig stage is divided into three stages: 60-75kg, 75-90kg, and 90-120kg, with corresponding net energy levels of 2450, 2450, and 2550 kcal / kg, respectively.

[0075] Step C2: Based on the cumulative net energy intake and daily net energy level of growing-finishing pigs at each stage, calculate the feed intake of growing-finishing pigs at each weight stage.

[0076] In this embodiment, the calculation results are shown in Table 6:

[0077] Table 6 Net energy intake and feed intake of growing-finishing pigs

[0078]

[0079] Step C3: Based on the total protein and amino acid requirements of growing-finishing pigs at each stage, calculate the dietary protein and lysine levels for each weight stage of growing-finishing pigs.

[0080] In this embodiment, the calculation results are shown in Table 7:

[0081] Table 7. SID protein and lysine levels in diets for growing-finishing pigs

[0082]

[0083] Step 6: Based on the obtained dietary protein and amino acid levels of growing and finishing pigs at different weight stages, and according to the amino acid balance theory, further reduce the protein level by 1-2 percentage points to formulate a low-protein diet model for pigs at different weight stages.

[0084] Step 7: Based on the low-protein diet model for pigs at different stages, the shadow price of soybean meal substitute raw materials, growth performance requirements, and other comprehensive indices, use linear programming to construct a diversified low-protein, low-soybean meal diet for growing-finishing pigs.

[0085] Nutritional constraints: Net energy levels of diets for 60-75kg, 75-90kg, and 90-120kg should be greater than or equal to 2480, 2490, and 2540 kcal / kg, respectively; SID protein of diets for 60-75kg, 75-90kg, and 90-120kg should be greater than or equal to 10.77%, 10.14%, and 9.12%, respectively; SID lysine of diets for 60-75kg, 75-90kg, and 90-120kg should be greater than or equal to 0.82%, 0.73%, and 0.64%, respectively; and calcium, phosphorus, etc. should not be lower than the set values.

[0086] Raw material usage limits: such as rapeseed meal ≤13%, sunflower seed meal ≤10%, cottonseed meal ≤8%, etc.

[0087] Growth performance constraints: Daily weight gain (ADG) and feed conversion ratio (F / G) are within the expected range;

[0088] Non-negative constraint: All raw material usage ≥ 0%, total feed formulation = 100%.

[0089] Objective function:

[0090]

[0091] in:

[0092] Z: Total cost of daily ration (RMB / ton);

[0093] P i : Shadow price of raw material i (yuan / ton);

[0094] X i : Amount of raw material i used (kg / ton).

[0095] By solving this linear programming model, the optimal low-protein, low-soybean meal diet formulation is automatically output.

[0096] In this embodiment, the low-protein, low-soybean meal diversified diet formulation for pigs established based on the requirement segmentation method is shown in Table 8:

[0097] Table 8. Raw material composition and nutrient levels of low-protein, low-soybean meal diversified diets for pigs established based on requirement segmentation method.

[0098]

[0099] In this embodiment, compared with the control diet, the proportion of soybean meal in the low-protein, low-soybean meal diversified diet groups (60-75 kg, 75-90 kg, and 90-120 kg) was reduced by 82.22%, 100.00%, and 100.00%, respectively; the crude protein level was reduced by 2.00%, 2.00%, and 2.00%, respectively; and the digestible protein level was reduced by 1.90%, 1.93%, and 1.85%, respectively.

[0100] In this embodiment, the feeding effect of a low-protein, low-soybean meal diversified diet for pigs established based on the requirement segmentation method is shown in Table 9:

[0101] Table 9. Feeding Effects of Low-Protein, Low-Soybean Meal Diversified Diets on Pigs Based on Demand Profile Method

[0102]

[0103] In this embodiment, there was no significant difference in growth performance between the low-protein, low-soybean meal diversified diet group and the control group. However, the feed cost per unit weight gain in the low-protein, low-soybean meal diversified diet group was significantly lower than that in the control group at 108-120 days, 121-139 days, and 108-179 days of age (P<0.05).

[0104] Example 2 (Comparison of different varieties at the same stage):

[0105] A method for predicting low-protein, low-soybean meal diversified diets for pigs based on requirement partitioning includes the following steps:

[0106] Step 1: Obtain production data and feed formulation data for growing-finishing pigs;

[0107] In this embodiment, the production data for high-lean-type and low-lean-type growing-finishing pigs includes sex, age, weight, feed intake, backfat thickness, and eye muscle area; the feed formulation data is the net energy level of the diet.

[0108] We collected 360 production data sets of growing-finishing pigs through production practice. Table 10 shows some of the data examples.

[0109] Table 10 Summary of Growth Performance Data

[0110]

[0111] Step 2: Process the production data of growing-finishing pigs, and check whether the production data of growing-finishing pigs contains null values ​​and outliers. If it contains null values ​​and outliers, delete the null values ​​and outliers to obtain the production sample dataset.

[0112] Step 3 includes the following sub-steps:

[0113] Step A1: Based on the net energy level and feed intake of the diet at each stage of growing-finishing pigs, calculate the cumulative net energy intake of high-lean-type and low-lean-type growing-finishing pigs;

[0114] In this embodiment, the calculation results are shown in Table 11:

[0115] Table 11 Cumulative Net Energy Intake of Growing-Finishing Pigs

[0116]

[0117] Step A2: Based on the sex, weight at each stage, backfat thickness, and eye muscle area of ​​the growing-finishing pigs, calculate the lean meat weight at each stage for high-lean-meat and low-lean-meat growing-finishing pigs;

[0118] In this embodiment, the calculation results are shown in Table 12:

[0119] Table 12 Lean Meat Weight of Growing-Finishing Pigs (Fat-Free)

[0120]

[0121] Step A3: Use the prediction model y=a+bx+cx 2 +dx 3 A model was constructed, where y represents body weight, cumulative net energy intake, or body protein deposition, x represents age in days, a is a constant term, b is the coefficient of the linear term, c is the coefficient of the quadratic term, and d is the coefficient of the cubic term. Outlier samples were identified and removed during the modeling process using studentized residuals (greater than 2.0 or less than -2.0). Functions relating body weight, cumulative net energy intake, and body protein deposition to age in both lean and lean growth-finishing pigs were established.

[0122] (1) Relationship between body weight and age:

[0123] Tall and lean meat type: y = -0.000022x 3 +0.007963x 2 -0.065042x+4.597004 (R²=0.962240)

[0124] Low lean meat type: y = -0.000025x 3 +0.009303x 2 -0.231838x+7.812879 (R²=0.970445)

[0125] (2) Relationship between cumulative net energy intake and age:

[0126] Tall and lean meat type: y=-0.000061x 3 +0.041042x 2 -1.347796x+17.725908 (R²=0.985849)

[0127] Low lean meat type: y=-0.000089x 3 +0.051059x 2 -2.491602x+42.900671 (R²=0.992944)

[0128] (3) Relationship between lean meat weight and age:

[0129] Tall and lean meat type: y=-0.000008x 3 +0.002853x 2 +0.025118x+0.084092 (R²=0.962500)

[0130] Low lean meat type: y=-0.000009x 3 +0.003100x 2 -0.017885x+0.682973 (R²=0.964347)

[0131] Step 4 includes the following sub-steps:

[0132] Step B1: Calculate the body protein content of high-lean-meat and low-lean-meat-meat-type growing-finishing pigs at each stage based on the function of lean meat weight and age. Calculate the lysine content in the body protein deposition of growing-finishing pigs at each stage based on the lysine content per gram of body protein.

[0133] In this embodiment, the calculation results are shown in Table 13:

[0134] Table 13. Protein and Lysine Content in Lean Meat and Low Lean Meat Growing-Finishing Pigs

[0135]

[0136] Step B2: The requirement breakdown method divides the protein and amino acid requirements into maintenance requirements and deposition requirements. The maintenance requirements of protein and lysine are calculated based on the body weight, dry matter intake and maintenance efficiency of feed at each stage of growing and finishing pigs. The deposition requirements of protein and lysine are calculated based on the amount and efficiency of lysine deposition in body protein at each stage of growing and finishing pigs.

[0137] Step B3: Add the maintenance requirements and deposition requirements of protein and amino acids to obtain the total requirements of protein and lysine for each stage of growing-finishing pigs.

[0138] In this embodiment, the calculation results are shown in Table 14:

[0139] Table 14 Protein and Lysine Requirements for Lean-Tooth and Lean-Tooth Growing-Finishing Pigs

[0140]

[0141] Step 5 includes the following sub-steps:

[0142] Step C1: Based on the actual production needs of the pig farm, set the weight division and net energy level of the diet for the growing and finishing pig stages;

[0143] In this embodiment, based on the actual production needs of the pig farm, the fattening pig stage with a weight of 100-130kg is regarded as one stage, and the net energy water of the diet is set at 2450kcal / kg.

[0144] Step C2: Based on the cumulative net energy intake and daily net energy level of lean-type and lean-type growing-finishing pigs, calculate the feed intake of lean-type and lean-type growing-finishing pigs at this stage.

[0145] In this embodiment, the calculation results are shown in Table 15:

[0146] Table 15 Net Energy Intake and Feed Intake of Lean-Type and Low-Lean-Type Growing-Finishing Pigs

[0147]

[0148] Step C3: Based on the total protein and lysine requirements of high-lean-type and low-lean-type growing-finishing pigs at each stage, calculate the dietary protein and lysine levels for each type of growing-finishing pig.

[0149] In this embodiment, the calculation results are shown in Table 16:

[0150] Table 16 SID Protein and Lysine Levels in Diets of Lean-Rich and Lean-Potential Growing-Finishing Pigs

[0151]

[0152] Step 6: Based on the obtained dietary protein and amino acid levels of growing and finishing pigs at different weight stages, and according to the amino acid balance theory, further reduce the protein level by 1-2 percentage points to formulate a low-protein diet model for pigs at different weight stages.

[0153] Step 7: Based on the low-protein diet model for pigs at different stages, the shadow price of soybean meal substitute raw materials, growth performance requirements, and other comprehensive indices, use linear programming to construct a diversified low-protein, low-soybean meal diet for growing-finishing pigs.

[0154] Nutritional constraints: The net energy level of the 100-130kg diet for both lean and low lean meat types should be greater than or equal to 2450kcal / kg; the SID protein of the 100-130kg diet for both lean and low lean meat types should be greater than or equal to 8.34% and 7.50%, respectively; the SID lysine of the 100-130kg diet for both lean and low lean meat types should be greater than or equal to 0.61% and 0.50%, respectively; and calcium, phosphorus, etc. should not be lower than the set values.

[0155] Raw material usage limits: such as rapeseed meal ≤13%, sunflower seed meal ≤10%, cottonseed meal ≤8%, etc.

[0156] Growth performance constraints: Daily weight gain (ADG) and feed conversion ratio (F / G) are within the expected range;

[0157] Non-negative constraint: All raw material usage ≥ 0%, total feed formulation = 100%.

[0158] Objective function:

[0159]

[0160] in:

[0161] Z: Total cost of daily ration (RMB / ton);

[0162] P i : Shadow price of raw material i (yuan / ton);

[0163] X i : Amount of raw material i used (kg / ton).

[0164] By solving this linear programming model, the optimal low-protein, low-soybean meal diet formulation is automatically output.

[0165] In this embodiment, the diversified diet formulations for high-lean-type and low-lean-type pigs, based on the requirement segmentation method, are shown in Table 17:

[0166] Table 17. Ingredient composition and nutrient levels of low-protein, low-soybean meal diversified diets for lean and lean hogs, established based on requirement segmentation method.

[0167]

[0168] In this embodiment, compared with the control diet, the proportion of soybean meal in the high-lean-meat and low-lean-meat, low-protein, low-soybean-meat diversified diet groups was reduced by 100.00% and 100.00%, respectively; the crude protein level was reduced by 1.52% and 2.53%, respectively; and the digestible protein level was reduced by 1.66% and 2.50%, respectively. Compared with the high-lean-meat diet, the crude protein level in the low-lean-meat diet for fattening pigs was reduced by 1.01%, and the digestible protein level was reduced by 0.84%, respectively.

[0169] In this embodiment, the feeding effect of a low-protein, low-soybean meal diversified diet for pigs established based on the requirement segmentation method is shown in Table 18:

[0170] Table 18 shows the effects of low-protein, low-soybean meal diversified diets on pigs, established based on the requirement segmentation method.

[0171]

[0172] In this embodiment, compared with lean-type finishing pigs, lean-type finishing pigs showed a significantly lower average daily weight gain, a significantly higher feed conversion ratio, and a significantly higher feed cost per unit weight gain (P<0.05). Compared with the control diet, there were no significant differences in average daily weight gain, average daily feed intake, and feed conversion ratio between lean-type and lean-type pigs fed a low-protein, low-soybean meal diversified diet (P>0.05). For lean-type pigs, the feed cost per unit weight gain of finishing pigs fed a low-protein, low-soybean meal diversified diet was significantly lower than that of finishing pigs fed a control diet, decreasing by 6.31% (P<0.05). For lean-type pigs, although the feed cost per unit weight gain of finishing pigs fed a low-protein, low-soybean meal diversified diet was not significantly different from that of finishing pigs fed a control diet, it was still reduced by 3.57% (P>0.05).

[0173] This invention was verified through two parallel experiments (different stages of the same breed and the same stage of different breeds). Compared with the control group, the low-protein, low-soybean meal diet group reduced the amount of soybean meal by 82%-100%, reduced the feed cost per unit weight gain by 3%-10%, and had no significant difference in pig growth performance (P>0.05).

[0174] Obviously, the above embodiments of this application are merely examples for clear illustration and are not intended to limit the implementation of the application. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of the claims of this application.

Claims

1. A method for predicting diversified diets for pigs with low protein and low soybean meal based on a requirement decomposition method, characterized in that, The method is specifically designed for growing-finishing pigs and includes the following steps: Step 1: Obtain production data and feed formulation data for growing-finishing pigs; Step 2: Organize and clean the collected growing-finishing pig production data and feed formulation data to form standardized production sample datasets and formulation sample datasets; Step 3: Analyze the formula sample dataset and the production sample dataset to obtain the body weight, cumulative net energy intake, and lean meat weight at each stage; conduct regression modeling analysis on the production sample dataset to establish a dynamic functional relationship between body weight (BW), cumulative net energy intake (NEi), lean meat weight (FFL), and age (d) for growing-finishing pigs. Step 4: Based on the functional relationship between the lean meat weight of growing-finishing pigs and age, calculate the body protein content of growing-finishing pigs at each growth stage, and adopt a requirement breakdown method suitable for pigs. The requirement breakdown method formula is: Total protein requirement = Maintenance requirement ((intestinal loss + fur loss) ÷ Maintenance efficiency coefficient) + Deposition requirement (body protein deposition ÷ Deposition efficiency coefficient). Establish a dynamic model of protein and lysine requirements of pigs at different growth stages. Step 5: Based on the actual production process of dividing growing and finishing pigs into weight stages and setting the net energy level of the diet, calculate the corresponding dietary protein and lysine levels for each weight stage. Step 6: Based on the theory of ideal protein and amino acid balance, construct a low-protein diet model for different weight stages of growing-finishing pigs; Step 7: Integrating low-protein diet models, shadow prices of soybean meal substitutes, raw material usage limits, growth performance constraints, and multi-dimensional indices for minimizing diet costs, a linear programming method is used to automatically construct diversified low-protein, low-soybean meal diet formulas for growing-finishing pigs.

2. The method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement segmentation method according to claim 1, characterized in that, The production data for growing-finishing pigs includes sex, age, weight, feed intake, backfat thickness, and eye muscle area; the feed formulation data is the net energy level of the diet.

3. The method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement segmentation method according to claim 1, characterized in that, Step 2 includes detecting whether the production data of growing-finishing pigs contains null values ​​and outliers. If null values ​​and outliers are found, they are deleted.

4. The method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement segmentation method according to claim 1, characterized in that, Step 3 includes the following sub-steps: Step A1: Calculate the cumulative net energy intake of growing-finishing pigs based on the net energy level and feed intake of the diet at each stage. Step A2: Calculate the lean meat weight of the growing-finishing pigs based on their sex, weight at each stage, backfat thickness, and eye muscle area. Step A3: Use the prediction model y=a+bx+cx 2 +dx 3 Modeling is performed where y represents body weight, cumulative net energy intake, or body protein deposition, x represents age in days, a is a constant term, b is the coefficient of the first term, c is the coefficient of the second term, and d is the coefficient of the third term. Abnormal samples appearing during the modeling process are identified and removed using studentized residuals (greater than 2.0 or less than -2.0). A function of body weight, cumulative net energy intake, lean meat weight, and age in days is established.

5. The method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement segmentation method according to claim 1, characterized in that, Step 4 includes the following sub-steps: Step B1: Calculate the body protein content of growing-finishing pigs at each stage based on the function of lean meat weight and age. Calculate the amount of lysine deposited in the body protein deposition of growing-finishing pigs at each stage based on the lysine content per gram of body protein. Step B2: The requirement breakdown method divides the protein and amino acid requirements into maintenance requirements and deposition requirements. The maintenance requirements of protein and lysine are calculated based on the body weight, dry matter intake and maintenance efficiency of feed at each stage of growing and finishing pigs. The deposition requirements of protein and lysine are calculated based on the amount and efficiency of lysine deposition in body protein at each stage of growing and finishing pigs. Step B3: Add the maintenance requirements and deposition requirements of protein and lysine to obtain the total requirements of protein and lysine for each stage of growing-finishing pigs.

6. The method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement segmentation method according to claim 1, characterized in that, Step 5 includes the following sub-steps: Step C1: Based on the actual production needs of the pig farm, set the weight division and net energy level of the diet for the growing and finishing pig stages; Step C2: Based on the cumulative net energy intake and daily net energy level of growing-finishing pigs at each stage, calculate the feed intake of growing-finishing pigs at each weight stage. Step C3: Based on the total protein and lysine requirements of growing-finishing pigs at each stage, calculate the dietary protein and lysine levels for each weight stage of growing-finishing pigs.

7. The method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement segmentation method according to claim 1, characterized in that, In step 6, based on the obtained dietary protein and lysine levels of growing-finishing pigs at each weight stage, and according to the amino acid balance theory, the levels are further reduced by 1-2 percentage points to formulate a low-protein diet model for growing-finishing pigs at different weight stages.

8. The method for predicting low-protein, low-soybean meal diversified diets for pigs based on the requirement segmentation method according to claim 1, characterized in that, In step 7, based on the low-protein diet model for pigs at different stages, the shadow price of soybean meal substitute raw materials, raw material usage limits, growth performance constraints, and cost indicators, a multi-dimensional index is constructed using linear programming to build a diversified diet for growing and finishing pigs that is low in protein and low in soybean meal.

Images