Pig feed formula data optimization processing method and system based on computer

Abstract

The invention discloses a computer-based pig feed formula data optimization processing method and a computer-based pig feed formula data optimization processing system, and relates to the technical field of breeding management. According to the method, by constructing a growth curve expansion algorithm model and combining individual data of pork pigs and microbial metabolic activity data, the weight gain of the pork pigs can be predicted; and further calculating the nutritional requirement Dn of each growth stage. Compared with a traditional method for formulating a feed formula depending on experience or a static model, the method has the advantages that the daily nutritional requirements are calculated in real time through the growth curve extension algorithm model, so that the feed formula is adjusted more accurately. And when the nutritional demand Dn exceeds a preset nutritional demand threshold value, the system can automatically trigger a feed formula optimization mechanism and output the adjusted feed demand Fi, so that the growth demand of the pork pigs is met. Through the dynamic adjustment, the pork pigs can obtain the nutrition proportion most suitable for the current growth stage, and the problem that health and growth speed are affected due to overnutrition or insufficient nutrition is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of breeding management, and in particular to a computer-based pig feed formula data optimization processing method and system. Background Art

[0002] With the widespread adoption and application of intelligent technologies, precision feeding has become a key development direction in modern animal husbandry. The widespread use of intelligent devices enables feed formula optimization through real-time data analysis and calculation to achieve precise feeding. In feed demand analysis based on growth curve predictions, the system can automatically collect and analyze pig weight, health data, metabolic levels, and other data to dynamically adjust feed formulas in real time, achieving more efficient and precise breeding management.

[0003] In current traditional animal husbandry management, feed formulation is often based on a fixed formula, lacking precise individualized feeding for pigs at different growth stages. This results in low feed utilization and is prone to nutritional over- or undernutrition. Furthermore, the metabolic state and microbial activity levels of pigs are influenced by a variety of factors, including the environment, health status, and stress. These changes are often overlooked, resulting in feed formulas that fail to accurately meet the pigs' physiological needs. This fixed formula model is unable to adapt to the dynamic nutritional needs of pigs during growth, leading to wasted feed resources and reduced growth efficiency. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides a computer-based pig feed formula data optimization processing method and system, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: comprising the following steps: S1. Install a detection device set and a feces collector on the farmed pigs to collect individual pig data and microbial metabolic activity data in real time. At the same time, set up a wireless communication gateway to wirelessly transmit the individual pig data collected by the detection device set and feces collector to a cloud server, where the data is pre-processed to obtain a pig feeding dataset. S2. Build a growth curve extension algorithm model, extract a pig feeding data set, input it into the growth curve extension algorithm model, calculate and output the predicted pig weight W; S3. Based on the growth curve expansion algorithm model, the predicted pig weight W is output, and the nutritional requirement Dn of the pig at each growth stage is further calculated. A nutritional requirement threshold D is preset, and the nutritional requirement threshold D is initially compared and evaluated with the nutritional requirement Dn. If the initial comparison and evaluation indicates that a feed formula optimization mechanism is implemented, the feed formula component ratio of the pig is further analyzed to output the feed requirement F; S4. Based on the predicted weight W and the actual weight Ws of the pigs, a feedback error calculation is performed to obtain a predicted weight error value E of the pigs. Based on the output of the predicted weight error value E, the nutritional intake of the pigs due to the pig feed formula is evaluated, and the growth curve expansion algorithm model is adaptively adjusted and optimized. S5. After adaptive adjustment and optimization, a feed efficiency algorithm formula is constructed to calculate and output the feed efficiency ratio (FER) of the pigs, and an efficiency threshold X is preset. The efficiency threshold X is then compared with the feed efficiency ratio (FER) for a second evaluation to analyze the feed formula optimization. If the utilization efficiency is assessed to be abnormal, further feed adjustments are made.

[0006] Preferably, S1 includes S11, S12 and S13; S11. Collect individual pig data in real time by installing a detection device set and a feces collector on the pigs. The detection device set includes a body temperature sensor, an ambient temperature sensor, an acceleration sensor, a gyroscope, a positioning sensor, a heart rate detector, and a respiratory rate sensor. The feces collector is used to regularly collect feces samples from pigs. The feces are then analyzed by a portable microbial analysis device built into the feces collector to analyze the microbial conditions in the collected feces and output microbial metabolic activity data. Individual pig data includes pig body temperature Tbo, pig house ambient temperature Ten, horizontal step count data (X1, X2), forward and backward step count data (Y1, Y2), vertical movement data (Z1, Z2), pig respiratory rate BR(t) and pig heart rate HR(t) on day t. Microbial metabolic activity data include fecal bacterial concentration Cba, bacterial metabolite concentration Ame, and fecal total organic matter content Cto, which represents the undigested portion of feed ingested by pigs; S12. Install a WIFI wireless gateway in the pig house area to connect to the detection equipment and feces collector, and send the real-time collected pig individual data and microbial metabolic activity data to the cloud server.

[0007] Preferably, S13, receiving the collected individual pig data and microbial metabolic activity data in real time in the cloud server, and preprocessing the individual pig data and microbial metabolic activity data to obtain a pig feeding data set; The pig feeding data set includes metabolic heat change rate △M, exercise change △A, stress level Sstr and in vivo microbial metabolic rate Rmic; The metabolic heat change rate △M is obtained by calculating the pig body temperature Tbo and the pig house ambient temperature Ten in the individual pig data. The specific algorithm formula is: , where Indicates the body surface metabolic constant, which is related to the pig's body shape and fur thickness and is set by the user; The change in movement volume △A is obtained by calculating and processing the horizontal movement step data (X1, X2), the forward and backward movement step data (Y1, Y2), and the vertical movement data (Z1, Z2) of the pig in the individual data. The specific algorithm formula is: ; The stress level Sstr is obtained by calculating the respiratory rate BR(t) and heart rate HR(t) of the pig on day t in the individual data of the pig. The specific algorithm formula is: ; In the formula, HR0 represents the normal heart rate baseline, BR0 represents the normal respiratory rate baseline, represents a constant regulating physiological markers of stress; The in vivo microbial metabolic rate Rmic is calculated and obtained by processing the fecal bacterial concentration Cba, the concentration of bacterial metabolites Ame and the total organic matter content Cto in the microbial metabolic activity data. The specific algorithm formula is: Where SW i represents the weight of the i-th feed, Q i It represents the nutrient content of the i-th feed, and n represents the total number of feed types.

[0008] Preferably, S2 includes S21; S21. Construct a growth curve extension algorithm model, extract a pig feeding data set, input it into the growth curve extension algorithm model, predict the growth requirements of the pigs, and output the predicted pig weight W; The predicted weight W of the pig is calculated and output by the following growth curve expansion algorithm model; ; Where W(t) represents the predicted weight of the pig on day t, W0 represents the initial weight of the pig, A represents the expected value of weight growth, k represents the growth rate coefficient, t0 represents the inflection point of the growth curve, and e represents the exponential function. represents the metabolic coefficient, △M(t) represents the metabolic heat change rate on the tth day, △A(t) represents the change in exercise volume on the tth day, Sstr(t) represents the stress level on the tth day, and Rmic(t) represents the in vivo microbial metabolic rate on the tth day.

[0009] Preferably, S3 includes S31, S32 and S33; S31. Based on the predicted weight W(t) of the pig on day t predicted by the growth curve expansion algorithm model, further calculate and output the daily nutritional requirement Dn of the pig at each growth stage; The nutritional requirement Dn is calculated and output by the following algorithm formula; ; Where Dn(t) represents the nutritional requirement on day t, Indicates the first constant parameter used to adjust the protein requirement ratio, Represents the second constant parameter, adjustment coefficients representing metabolic and exercise effects; S32. The user presets a nutritional requirement threshold D based on the physiological requirement standards of the pig. The physiological requirement standards include the pig's growth stage, weight, reproductive status, and feeding goals. The nutritional requirement threshold D is then compared with the obtained nutritional requirement Dn(t) on day t to analyze the current feed requirements of the pig. The specific evaluation content is as follows: When the nutritional requirement Dn(t) on day t is greater than the nutritional requirement threshold D, the feed formula optimization mechanism is executed; When twice the nutritional requirement threshold D < the nutritional requirement on day t Dn(t) ≤ the nutritional requirement threshold D, there is no need to adjust the feed formula; When the nutritional requirement Dn(t) on day t is less than or equal to twice the nutritional requirement threshold D, an early warning is generated to prompt the breeder to conduct disease screening on the pigs.

[0010] Preferably, S33, when it is identified based on the initial comparative evaluation results that the current feed formula nutrition does not meet the nutritional requirements of the current pig growth stage, the formula optimization mechanism is automatically triggered, and the formula optimization mechanism calculates and outputs the feed requirements F of different raw materials by constructing a feed formula optimization algorithm formula; The feed requirement F is calculated and output by the following feed formula optimization algorithm formula; ; Where, F i (t) represents the amount of feed required for the i-th type of feeding on the t-th day, R i (t) represents the digestion efficiency of the i-th feed ingredient on the t-th day.

[0011] Preferably, S4 includes S41 and S42; S41. After the feed formula is optimized and adjusted, the pig feeding data set of the pigs is collected in real time every day to update the growth curve expansion algorithm model, and the weight of the pigs on the tth day after the output adjustment is predicted. , perform feedback error calculation to obtain the pig weight prediction error value E; The pig weight prediction error value E is calculated and output by the following algorithm formula;

[0012] Where E(t) represents the prediction error of the pig weight on day t, which is the difference between the weight predicted by the system and the actual weight; Ws(t) represents the actual weight of the pig on day t; S42. Based on the output of the predicted error value E(t) of the pig weight on day t, evaluate and analyze the current feed formula adjustment. The specific evaluation contents are as follows: When the predicted error value E(t) of the pig weight on day t is ≥0, it means that the predicted weight exceeds the actual weight, the feed formula adjustment is in line with expectations, and no secondary adjustment is required; When the predicted weight error value E(t) of the pig on day t is less than 0, it means that the predicted weight does not match the actual weight and the predicted weight is inaccurate. At this time, the gradient descent method is used to adjust the growth rate coefficient k and metabolic coefficient in the growth curve expansion algorithm model. Perform adaptive adjustment and optimization.

[0013] Preferably, S5 includes S51 and S52; S51. After adaptive adjustment and optimization, a feed efficiency algorithm formula is constructed to calculate and output the feed efficiency ratio (FER) of the pigs, and to monitor the optimization of the current feed formula; The feed efficiency ratio FER is calculated and output by the following feed efficiency algorithm formula; ; Where Ftotal(t) represents the total feed intake on day t.

[0014] Preferably, in S52, a user presets an efficacy threshold X, and a secondary comparative evaluation is performed between the efficacy threshold X and the feed efficiency ratio FER, the feed formula optimization is analyzed, and the feed formula is further optimized based on the evaluation results. The specific evaluation contents are as follows: When the feed efficiency ratio FER ≥ the efficiency threshold X, it means that the utilization efficiency of the feed formula is normal and there is no need to adjust the feed formula; When the feed efficiency ratio FER is less than the efficiency threshold X, it indicates that the utilization efficiency of the feed formula is abnormal. At this time, the feed requirement F of the i-th type of feed in the t-th day obtained by the secondary calculation after adaptive adjustment and optimization is further sent to the automatic feed formula machine. i (t), the automatic feed formula machine automatically adjusts the amount of various feed raw materials.

[0015] A computer-based pig feed formula data optimization processing system includes a data acquisition and processing module, a growth curve analysis module, a nutritional demand prediction module, a feedback adjustment optimization module, and an optimization evaluation and analysis module; The data acquisition and processing module collects individual pig data and microbial metabolic activity data in real time by installing a detection device group and a feces collector on the farmed pigs. At the same time, a wireless communication gateway is set up to wirelessly transmit the individual pig data collected by the detection device group and the feces collector to the cloud server, where it is pre-processed to obtain the pig feeding data set. The growth curve analysis module constructs a growth curve extension algorithm model, extracts the pig feeding data set, inputs it into the growth curve extension algorithm model, and calculates and outputs the predicted pig weight W; The nutritional requirement prediction module outputs the predicted weight W of the pig by expanding the algorithm model according to the growth curve, further calculates the nutritional requirement Dn of the pig at each growth stage, and presets the nutritional requirement threshold D. Then, the nutritional requirement threshold D is compared with the nutritional requirement Dn for the first time. If the initial comparison and evaluation shows that the feed formula optimization mechanism is implemented, the feed formula component ratio of the pig is further analyzed to output the feed requirement F; The feedback adjustment optimization module calculates the feedback error based on the predicted weight W and the actual weight Ws of the pigs to obtain the predicted weight error value E of the pigs. Based on the output of the predicted weight error value E, the module evaluates the nutritional intake of the pigs from the pig feed formula and adaptively adjusts and optimizes the growth curve expansion algorithm model. The optimization evaluation and analysis module constructs a feed efficiency algorithm formula after adaptive adjustment optimization, calculates and outputs the feed efficiency ratio (FER) of meat pigs, and presets the efficiency threshold X. Then, the efficiency threshold X is compared with the feed efficiency ratio (FER) for a second evaluation to analyze the feed formula optimization. If the utilization efficiency is evaluated to be abnormal, further feed adjustments will be made.

[0016] The present invention provides a computer-based pig feed formula data optimization processing method and system. It has the following beneficial effects: (1) This method collects individual pig data and microbial metabolic activity data in real time by installing a detection device set and a feces collector on the pigs, and transmits this data to the cloud server for preprocessing via a wireless communication gateway. This process ensures that the data source in feeding management is real-time and accurate, including not only individual physiological data but also detailed data such as microbial metabolic activity, so that the health status and metabolic level of each pig can be understood more accurately. In this way, the adjustment of feed formula is no longer based on experience or rough judgment, but on scientific decision-making based on the actual needs of the pigs, thereby improving feed utilization efficiency and reducing unnecessary feed waste.

[0017] (2) This method can predict the weight gain of pigs and further calculate the nutritional requirement Dn at each growth stage by constructing a growth curve extension algorithm model and combining it with individual pig data and microbial metabolic activity data. Compared with the traditional method of formulating feed formulas based on experience or static models, the real-time calculation of daily nutritional requirements through the growth curve extension algorithm model makes the adjustment of feed formulas more accurate. When the nutritional requirement Dn exceeds the preset nutritional requirement threshold, the system will automatically trigger the feed formula optimization mechanism and output the adjusted feed requirement F i Through this dynamic adjustment, pigs can obtain the most suitable nutritional ratio for their current growth stage, avoiding the problem of excessive or insufficient nutrition affecting their health and growth rate.

[0018] (3) After the actual weight Ws of the pig is collected by the detection equipment, it is compared with the predicted weight W calculated by the growth curve expansion algorithm model and evaluated by the feedback error E. If there is a difference between the predicted and actual weights, the system will automatically adjust the parameters of the growth curve model such as the growth rate coefficient k, the metabolic coefficient The system automatically adjusts the feed formula to ensure that the adjusted formula better meets the pigs' actual needs. Furthermore, the feed efficiency ratio (FER) is calculated and compared with a preset efficiency threshold (X) to further evaluate the effectiveness of the feed formula optimization. If the FER falls below the threshold, the system triggers an automatic optimization mechanism to readjust the feed ingredients. This not only ensures more precise nutrient intake for the pigs and avoids feed waste, but also improves overall feed utilization efficiency, enabling the pigs to achieve their desired growth goals more efficiently and profitably. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a schematic diagram of the steps of a computer-based pig feed formula data optimization processing method of the present invention; Figure 2 The present invention is a flowchart of a computer-based pig feed formula data optimization processing system. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] Example 1 See also Figure 1The present invention provides a computer-based method for optimizing and processing pig feed formula data. To achieve the above objectives, the present invention is implemented through the following technical solutions: comprising the following steps: S1. Install a detection device set and a feces collector on the farmed pigs to collect individual pig data and microbial metabolic activity data in real time. At the same time, set up a wireless communication gateway to wirelessly transmit the individual pig data collected by the detection device set and feces collector to a cloud server, where the data is pre-processed to obtain a pig feeding dataset. S2. Build a growth curve extension algorithm model, extract a pig feeding data set, input it into the growth curve extension algorithm model, calculate and output the predicted pig weight W; S3. Based on the growth curve expansion algorithm model, the predicted pig weight W is output, and the nutritional requirement Dn of the pig at each growth stage is further calculated. A nutritional requirement threshold D is preset, and the nutritional requirement threshold D is initially compared and evaluated with the nutritional requirement Dn. If the initial comparison and evaluation indicates that a feed formula optimization mechanism is implemented, the feed formula component ratio of the pig is further analyzed to output the feed requirement F; S4. Based on the predicted weight W and the actual weight Ws of the pigs, a feedback error calculation is performed to obtain a predicted weight error value E of the pigs. Based on the output of the predicted weight error value E, the nutritional intake of the pigs due to the pig feed formula is evaluated, and the growth curve expansion algorithm model is adaptively adjusted and optimized. S5. After adaptive adjustment and optimization, a feed efficiency algorithm formula is constructed to calculate and output the feed efficiency ratio (FER) of the pigs, and an efficiency threshold X is preset. The efficiency threshold X is then compared with the feed efficiency ratio (FER) for a second evaluation to analyze the feed formula optimization. If the utilization efficiency is assessed to be abnormal, further feed adjustments are made.

[0022] In this embodiment, the method, by installing a detection equipment set and a feces collector, enables real-time collection of individual pig data and microbial metabolic activity data. The data is then wirelessly transmitted to a cloud server for processing, thereby forming an accurate pig feeding dataset. Secondly, by utilizing a growth curve-based extended algorithm model and analyzing this data, the weight changes of pigs can be dynamically predicted, generating accurate weight prediction data W and further calculating the nutritional requirements Dn at different growth stages. By comparing and analyzing the data with a preset nutritional requirement threshold, the method can automatically adjust the feed formula and output an accurate feed requirement F, ensuring the scientific nature of the pig's nutritional intake. Furthermore, through feedback error calculation and an adaptive optimization mechanism, the method can dynamically adjust the feed formula based on the error value E between the actual weight and the predicted pig weight, optimizing nutritional supply and ensuring that the pig's actual needs are fully met. Finally, the feed efficiency ratio (FER) is calculated using a feed efficiency algorithm formula and compared and evaluated with a set efficiency threshold X, further improving feed utilization efficiency. Compared with traditional technical means that rely on experience or fixed formulas, this optimization method based on real-time data and feedback mechanism significantly improves the accuracy of feed formula, reduces feed waste, and improves the growth efficiency and health level of pigs, ultimately achieving a dual improvement in economic benefits and production efficiency.

[0023] Example 2 This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: S1 includes S11, S12 and S13; S11. Collect individual pig data in real time by installing a detection device set and a feces collector on the pigs. The detection device set includes a body temperature sensor, an ambient temperature sensor, an acceleration sensor, a gyroscope, a positioning sensor, a heart rate detector, and a respiratory rate sensor. The feces collector is used to regularly collect feces samples from pigs. The feces are then analyzed by a portable microbial analysis device built into the feces collector to analyze the microbial conditions in the collected feces and output microbial metabolic activity data. Individual pig data includes the pig's body temperature Tbo, the pig's ambient temperature Ten, the pig's horizontal step count data (X1, X2), the pig's forward and backward step count data (Y1, Y2), and the pig's vertical movement data (Z1, Z2), representing the pig's movement status, the pig's respiratory rate BR(t) on day t, and the pig's heart rate HR(t) on day t. Microbial metabolic activity data include fecal bacterial concentration Cba, bacterial metabolite concentration Ame, and fecal total organic matter content Cto, which represents the undigested portion of feed ingested by pigs; S12. Install a WIFI wireless gateway in the pig house area to connect to the detection equipment and feces collector, and send the real-time collected pig individual data and microbial metabolic activity data to the cloud server.

[0024] S13. Receiving the collected individual pig data and microbial metabolic activity data in real time in the cloud server, and preprocessing the individual pig data and microbial metabolic activity data to obtain a pig feeding data set; The pig feeding data set includes metabolic heat change rate △M, exercise change △A, stress level Sstr and in vivo microbial metabolic rate Rmic; The metabolic heat change rate △M is obtained by calculating the pig body temperature Tbo and the pig house ambient temperature Ten in the individual pig data. The specific algorithm formula is: , where Indicates the body surface metabolic constant, which is related to the pig's body shape and fur thickness and is set by the user; The change in movement volume △A is obtained by calculating and processing the horizontal movement step data (X1, X2), the forward and backward movement step data (Y1, Y2), and the vertical movement data (Z1, Z2) of the pig in the individual data. The specific algorithm formula is: ; The stress level Sstr is obtained by calculating the respiratory rate BR(t) and heart rate HR(t) of the pig on day t in the individual data of the pig. The specific algorithm formula is: ; In the formula, HR0 represents the normal heart rate baseline, BR0 represents the normal respiratory rate baseline, represents a constant regulating physiological markers of stress; The in vivo microbial metabolic rate Rmic is calculated and obtained by processing the fecal bacterial concentration Cba, the concentration of bacterial metabolites Ame and the total organic matter content Cto in the microbial metabolic activity data. The specific algorithm formula is: Where SW i represents the weight of the i-th feed, Q i It represents the nutrient content of the i-th feed, and n represents the total number of feed types.

[0025] In this embodiment, the method realizes the real-time collection and analysis of individual data of meat pigs and microbial metabolic activity data through a series of advanced sensor equipment and feces collectors. A variety of sensors including body temperature, movement, heart rate, etc. are installed to provide comprehensive data on the health status and exercise behavior of the pigs, and the microbial analysis equipment built into the feces collector conducts an in-depth analysis of the pigs' digestive system and generates microbial metabolic data reflecting feed utilization. These data are then uploaded to the cloud via a wireless network, and data preprocessing is performed in the cloud server to generate core parameters such as the metabolic heat change rate, exercise volume change, stress level and microbial metabolic rate of the meat pigs. These parameters provide accurate data support for subsequent feed formula optimization. This step significantly improves the accuracy and dynamic adjustment ability of the feed formula. Through real-time monitoring of pigs and analysis of microbial metabolic rate, the system can respond to the physiological changes and nutritional needs of pigs in a timely manner, thereby optimizing feed feeding.

[0026] Example 3 This embodiment is explained in Example 1, please refer to Figure 1 , specifically: S2 includes S21; S21. Construct a growth curve extension algorithm model, extract a pig feeding data set, input it into the growth curve extension algorithm model, predict the growth requirements of the pigs, and output the predicted pig weight W; The predicted weight W of the pig is calculated and output by the following growth curve expansion algorithm model; ; Where W(t) represents the predicted weight of the pig on day t, W0 represents the initial weight of the pig, A represents the expected weight growth value, the specific value is set by the user according to the pig breed, k represents the growth rate coefficient, t0 represents the inflection point of the growth curve, and e represents the exponential function. It represents the metabolic coefficient, which is used to control the effects of indirect factors such as metabolism, exercise, and microorganisms on body weight. △M(t) represents the metabolic heat change rate on the tth day. △A(t) represents the change in exercise volume on the tth day. Sstr(t) represents the stress level on the tth day, which indicates the inhibitory effect of stress level on body weight gain. Rmic(t) represents the metabolic rate of microorganisms in the body on the tth day, which indicates the efficiency of feed digestion by animals.

[0027] In this embodiment, the method accurately predicts the weight change W of pigs at each growth stage based on the feeding data set of pigs by constructing a growth curve expansion algorithm model, and calculates their nutritional needs accordingly. The system comprehensively considers a variety of factors that have a direct or indirect impact on the growth of pigs. Through this growth curve expansion algorithm, the system can accurately output the daily predicted weight W of pigs and dynamically adjust the feed supply through a feedback mechanism. The algorithm model achieves accurate prediction of pig growth needs by introducing more real-time data and personalized parameters. This improvement enables feed formulas to more flexibly adapt to the changes of pigs at different growth stages, prevent overfeeding or underfeeding, avoid waste of resources, and identify growth abnormalities or health problems in advance. This accurate prediction and dynamic adjustment mechanism significantly improves the growth efficiency of pigs, optimizes feed utilization, and ultimately achieves the purpose of improving production efficiency and economic benefits.

[0028] Example 4 This embodiment is explained in Example 1, please refer to Figure 1 , specifically: S3 includes S31, S32 and S33; S31. Based on the predicted weight W(t) of the pig on day t predicted by the growth curve expansion algorithm model, further calculate and output the daily nutritional requirement Dn of the pig at each growth stage; The nutritional requirement Dn is calculated and output by the following algorithm formula; ; Where Dn(t) represents the nutritional requirement on day t, Indicates the first constant parameter used to adjust the protein requirement ratio, Represents the second constant parameter, which is used to adjust the ratio of energy demand, The adjustment coefficient for the effects of metabolism and exercise represents the additional effects of indirect factors on nutritional requirements; S32. The user presets a nutritional requirement threshold D based on the physiological requirement standards of the pig. The physiological requirement standards include the pig's growth stage, weight, reproductive status, and feeding goals. The nutritional requirement threshold D is then compared with the obtained nutritional requirement Dn(t) on day t to analyze the current feed requirements of the pig. The specific evaluation content is as follows: When the nutritional requirement Dn(t) on day t is greater than the nutritional requirement threshold D, it means that the current feed formula nutrition does not meet the nutritional requirements of the current pig growth stage. At this time, the feed formula optimization mechanism is executed; When twice the nutritional requirement threshold D < the nutritional requirement on day t Dn(t) ≤ the nutritional requirement threshold D, it means that the current feed formula nutrition meets the nutritional requirements of the current pig growth stage, and there is no need to adjust the feed formula at this time; When the nutritional requirement Dn(t) on day t is less than or equal to twice the nutritional requirement threshold D, it indicates that the nutritional requirement of the pig is abnormal. At this time, an early warning is generated to prompt the breeder to check the pig for diseases.

[0029] S33. When it is determined based on the initial comparative evaluation results that the current feed formula nutrition does not meet the nutritional requirements of the current pig growth stage, a formula optimization mechanism is automatically triggered. The formula optimization mechanism calculates and outputs the feed requirements F of different raw materials by constructing a feed formula optimization algorithm formula; The feed requirement F is calculated and output by the following feed formula optimization algorithm formula; ; Where, F i (t) represents the amount of feed required for the i-th type of feeding on the t-th day, R i (t) represents the digestion efficiency of the i-th feed ingredient on the t-th day.

[0030] In this embodiment, the method calculates the daily nutritional requirements Dn of the pigs and compares it with the preset nutritional requirement threshold D to ensure that the feed formula matches the nutritional requirements of the pigs in their actual growth stage. Based on the growth curve expansion algorithm model, the system can accurately calculate the daily protein, energy and other nutritional requirements of the pigs at different growth stages, and take into account the impact of factors such as metabolism and exercise on the requirements. By initially comparing and evaluating the matching degree between the current feed formula and the nutritional requirements, the deficiencies in the formula can be discovered in a timely manner. If the nutritional requirements exceed the preset threshold, the system will automatically start the optimization mechanism to ensure that the nutrition consumed by the pigs is consistent with the requirements of their growth stage. The required amount F of each feed ingredient is then dynamically calculated through the optimization algorithm. i , combined with the digestion efficiency R i Ensure the accuracy and scientific nature of the formula.

[0031] Example 5 This embodiment is explained in Example 1, please refer to Figure 1 , specifically: S4 includes S41 and S42; S41. After the feed formula is optimized and adjusted, the pig feeding data set of the pigs is collected in real time every day to update the growth curve expansion algorithm model, and the weight of the pigs on the tth day after the output adjustment is predicted. , perform feedback error calculation to obtain the pig weight prediction error value E; The pig weight prediction error value E is calculated and output by the following algorithm formula;

[0032] Where E(t) represents the prediction error of the pig weight on day t, which is the difference between the weight predicted by the system and the actual weight; Ws(t) represents the actual weight of the pig on day t; S42. Based on the output of the predicted error value E(t) of the pig weight on day t, evaluate and analyze the current feed formula adjustment. The specific evaluation contents are as follows: When the predicted error value E(t) of the pig weight on day t is ≥0, it means that the predicted weight exceeds the actual weight, the feed formula adjustment is in line with expectations, and no secondary adjustment is required; When the predicted weight error value E(t) of the pig on day t is less than 0, it means that the predicted weight does not match the actual weight and the predicted weight is inaccurate. At this time, the gradient descent method is used to adjust the growth rate coefficient k and metabolic coefficient in the growth curve expansion algorithm model. Perform adaptive adjustment and optimization.

[0033] In this embodiment, the method further improves the accuracy and adaptability of feed management through a feedback mechanism. The system collects the latest data of the pigs in real time every day, updates the growth curve extension algorithm model, and performs feedback adjustment based on the predicted error value E(t) of the pig weight on the tth day between the actual weight Ws and the predicted weight W. Through this error calculation, the system can determine whether the current feed formula meets expectations and ensure that the nutrition consumed by the pigs matches their actual needs. By analyzing the prediction error, if the predicted error value E(t) of the pig weight on the tth day is positive, it means that the feed formula effect meets expectations and no further adjustment is required; if the predicted error value E(t) of the pig weight on the tth day is negative, the system uses the gradient descent method to automatically adjust the growth rate coefficient k and metabolic coefficient in the growth curve extension algorithm. , ensuring more accurate models and feed formulations. This method effectively reduces weight prediction bias caused by inaccurate nutrient supply through a dynamic feedback mechanism, enabling real-time optimization. The advantage of this adaptive adjustment is that it not only promptly identifies deficiencies in the feed formulation but also automatically corrects the formulation through a data-driven approach, ensuring optimal growth and nutrient intake for pigs.

[0034] Example 6 This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: S5 includes S51 and S52; S51. After adaptive adjustment and optimization, a feed efficiency algorithm formula is constructed to calculate and output the feed efficiency ratio (FER) of the pigs, and to monitor the optimization of the current feed formula; The feed efficiency ratio FER is calculated and output by the following feed efficiency algorithm formula; ; Where Ftotal(t) represents the total feed intake on day t.

[0035] S52: Based on the user's preset efficiency threshold X, a secondary comparative evaluation is performed between the efficiency threshold X and the feed efficiency ratio FER, the feed formula optimization is analyzed, and the feed formula is further optimized based on the evaluation results. The specific evaluation contents are as follows: When the feed efficiency ratio FER ≥ the efficiency threshold X, it means that the utilization efficiency of the feed formula is normal and there is no need to adjust the feed formula; When the feed efficiency ratio FER is less than the efficiency threshold X, it indicates that the utilization efficiency of the feed formula is abnormal. At this time, the feed requirement F of the i-th type of feed in the t-th day obtained by the secondary calculation after adaptive adjustment and optimization is further sent to the automatic feed formula machine. i (t), the automatic feed formula machine automatically adjusts the amount of various feed raw materials.

[0036] In this embodiment, the method further improves the accuracy and utilization efficiency of feed management by constructing a feed efficiency algorithm. Based on the data collected in real time, the system calculates the feed efficiency ratio FER of the meat pigs and monitors the optimization effect of the current feed formula. The feed efficiency ratio FER provides a direct evaluation indicator of feed utilization by combining the total feed intake Ftotal with key indicators such as the weight change of the pigs. A secondary comparative evaluation of FER is performed through the preset efficiency threshold X to determine the utilization efficiency of the current feed formula. When the feed efficiency ratio FER is higher than the efficiency threshold X, it indicates that the feed utilization efficiency is normal and no further adjustment is required; when the feed efficiency ratio FER is lower than the efficiency threshold X, the system will automatically adjust the formula through the feed formula machine to ensure that the required amount of each feed F i This method significantly improves feed utilization efficiency through real-time monitoring of feed efficiency and a data-driven secondary evaluation mechanism. It can not only determine the efficiency of the current feed in real time, but also automatically optimize feed delivery when efficiency is insufficient, thus avoiding feed waste and increased breeding costs. This dynamic adjustment method based on efficiency feedback makes the nutritional intake of pigs more accurate, ensures the stability of their growth rate and health status, and ultimately brings higher economic benefits, breeding efficiency and resource utilization. Example 7 See also Figure 1 and Figure 2 , a computer-based pig feed formula data optimization processing system, including a data acquisition and processing module, a growth curve analysis module, a nutritional demand prediction module, a feedback adjustment optimization module and an optimization evaluation analysis module; The data acquisition and processing module collects individual pig data and microbial metabolic activity data in real time by installing a detection device group and a feces collector on the farmed pigs. At the same time, a wireless communication gateway is set up to wirelessly transmit the individual pig data collected by the detection device group and the feces collector to the cloud server, where it is pre-processed to obtain the pig feeding data set. The growth curve analysis module constructs a growth curve extension algorithm model, extracts the pig feeding data set, inputs it into the growth curve extension algorithm model, and calculates and outputs the predicted pig weight W; The nutritional requirement prediction module outputs the predicted weight W of the pig by expanding the algorithm model according to the growth curve, further calculates the nutritional requirement Dn of the pig at each growth stage, and presets the nutritional requirement threshold D. Then, the nutritional requirement threshold D is compared with the nutritional requirement Dn for the first time. If the initial comparison and evaluation shows that the feed formula optimization mechanism is implemented, the feed formula component ratio of the pig is further analyzed to output the feed requirement F; The feedback adjustment optimization module calculates the feedback error based on the predicted weight W and the actual weight Ws of the pigs to obtain the predicted weight error value E of the pigs. Based on the output of the predicted weight error value E, the module evaluates the nutritional intake of the pigs from the pig feed formula and adaptively adjusts and optimizes the growth curve expansion algorithm model. The optimization evaluation and analysis module constructs a feed efficiency algorithm formula after adaptive adjustment optimization, calculates and outputs the feed efficiency ratio (FER) of meat pigs, and presets the efficiency threshold X. Then, the efficiency threshold X is compared with the feed efficiency ratio (FER) for a second evaluation to analyze the feed formula optimization. If the utilization efficiency is evaluated to be abnormal, further feed adjustments will be made.

[0037] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions, and alterations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A computer-based method for optimizing pig feed formula data, characterized in that: The following steps are involved: S1. Install a detection device set and a feces collector on the farmed pigs to collect individual pig data and microbial metabolic activity data in real time. Set up a wireless communication gateway to wirelessly transmit the individual pig data and microbial metabolic activity data collected by the detection device set and feces collector to a cloud server. Preprocess the individual pig data and microbial metabolic activity data in the cloud server to obtain a pig feeding dataset. S2. Build a growth curve extension algorithm model, extract a pig feeding data set, input it into the growth curve extension algorithm model, calculate and output the predicted pig weight W; S3. Based on the growth curve expansion algorithm model, the predicted pig weight W is output, and the nutritional requirement Dn of the pig at each growth stage is further calculated. A nutritional requirement threshold D is preset, and the nutritional requirement threshold D is initially compared and evaluated with the nutritional requirement Dn. If the initial comparison and evaluation indicates that a feed formula optimization mechanism is implemented, the feed formula component ratio of the pig is further analyzed to output the feed requirement F; S4. Based on the predicted weight W and the actual weight Ws of the pigs, a feedback error calculation is performed to obtain a predicted weight error value E of the pigs. Based on the output of the predicted weight error value E, the nutritional intake of the pigs due to the pig feed formula is evaluated, and the growth curve expansion algorithm model is adaptively adjusted and optimized. S5. After adaptive adjustment and optimization, a feed efficiency algorithm formula is constructed to calculate and output the feed efficiency ratio (FER) of the pigs, and an efficiency threshold X is preset. The efficiency threshold X is then compared with the feed efficiency ratio (FER) for a second evaluation to analyze the feed formula optimization. If the utilization efficiency is assessed to be abnormal, further feed adjustments are made.

2. A computer-based pig feed formula data optimization processing method according to claim 1, characterized in that: Said S1 includes S11, S12 and S13; S11. Collect individual pig data in real time by installing a detection device group and a feces collector on the pigs. The detection device group includes a body temperature sensor, an ambient temperature sensor, an acceleration sensor, a gyroscope, a positioning sensor, a heart rate detector, and a respiratory rate sensor. The feces collector is used to regularly collect feces samples from pigs, and then the feces collector is equipped with a portable microbial analysis device to analyze the microbial conditions in the collected feces and output microbial metabolic activity data; The individual pig data includes the pig's body temperature Tbo, the pig's ambient temperature Ten, the pig's horizontal step count data (X1, X2), the pig's forward and backward step count data (Y1, Y2), the pig's vertical movement data (Z1, Z2), the pig's respiratory rate BR(t) on day t, and the pig's heart rate HR(t) on day t; The microbial metabolic activity data include fecal bacterial concentration Cba, bacterial metabolite concentration Ame and fecal total organic matter content Cto; S12. Install a WIFI wireless gateway in the pig house area to connect to the detection equipment and feces collector, and send the real-time collected pig individual data and microbial metabolic activity data to the cloud server.

3. A computer-based pig feed formula data optimization processing method according to claim 2, characterized in that: S13. Receiving the collected individual pig data and microbial metabolic activity data in real time in the cloud server, and preprocessing the individual pig data and microbial metabolic activity data to obtain a pig feeding data set; The pig feeding data set includes metabolic heat change rate ΔM, exercise change ΔA, stress level Sstr and in vivo microbial metabolic rate Rmic; The metabolic heat change rate ΔM is obtained by calculating and processing the pig body temperature Tbo and the pig house ambient temperature Ten in the individual pig data. The specific algorithm formula is: , where Indicates the body surface metabolic constant, which is set by the user; The movement amount change ΔA is obtained by calculating and processing the horizontal movement step data (X1, X2), the front-back movement step data (Y1, Y2), and the vertical movement data (Z1, Z2) of the pig in the individual data of the pig. The specific algorithm formula is: ; The stress level Sstr is obtained by calculating and processing the respiratory rate BR(t) and heart rate HR(t) of the pig on day t in the individual data of the pig. The specific algorithm formula is: ; In the formula, HR0 represents the normal heart rate baseline, BR0 represents the normal respiratory rate baseline, represents a constant regulating physiological markers of stress; The in vivo microbial metabolic rate Rmic is obtained by calculating the fecal bacterial concentration Cba, the concentration of bacterial metabolites Ame and the fecal total organic matter content Cto in the microbial metabolic activity data. The specific algorithm formula is: Where SW i represents the weight of the i-th feed, Q i It represents the nutrient content of the i-th feed, and n represents the total number of feed types.

4. The computer-based pig feed formula data optimization processing method according to claim 1, characterized in that: Said S2 includes S21; S21. Construct a growth curve extension algorithm model, extract a pig feeding data set, input it into the growth curve extension algorithm model, predict the growth requirements of the pigs, and output the predicted pig weight W; The predicted pig weight W is calculated and output by the following growth curve expansion algorithm model; ； Where W(t) represents the predicted weight of the pig on day t, W0 represents the initial weight of the pig, A represents the expected value of weight growth, k represents the growth rate coefficient, t0 represents the inflection point of the growth curve, and e represents the exponential function. represents the metabolic coefficient, △M(t) represents the metabolic heat change rate on the tth day, △A(t) represents the change in exercise volume on the tth day, Sstr(t) represents the stress level on the tth day, and Rmic(t) represents the in vivo microbial metabolic rate on the tth day.

5. The computer-based pig feed formula data optimization processing method according to claim 1, characterized in that: Said S3 includes S31, S32 and S33; S31. Based on the predicted weight W(t) of the pig on day t predicted by the growth curve expansion algorithm model, further calculate and output the daily nutritional requirement Dn of the pig at each growth stage; The nutritional requirement Dn is calculated and outputted by the following algorithm formula; ； Where Dn(t) represents the nutritional requirement on day t, Indicates the first constant parameter used to adjust the protein requirement ratio, Represents the second constant parameter, adjustment coefficients representing metabolic and exercise effects; S32. The user presets a nutritional requirement threshold D based on the physiological requirement standards of the pig. The physiological requirement standards include the pig's growth stage, weight, reproductive status, and feeding goals. The nutritional requirement threshold D is then compared with the obtained nutritional requirement Dn(t) on day t to analyze the current feed requirements of the pig. The specific evaluation content is as follows: When the nutritional requirement Dn(t) on day t is greater than the nutritional requirement threshold D, the feed formula optimization mechanism is executed; When twice the nutritional requirement threshold D < the nutritional requirement on day t Dn(t) ≤ the nutritional requirement threshold D, there is no need to adjust the feed formula; When the nutritional requirement Dn(t) on day t is less than or equal to twice the nutritional requirement threshold D, an early warning is generated to prompt the breeder to conduct disease screening on the pigs.

6. A computer-based pig feed formula data optimization processing method according to claim 5, characterized in that: S33. When it is determined based on the initial comparative evaluation results that the current feed formula nutrition does not meet the nutritional requirements of the current pig growth stage, a formula optimization mechanism is automatically triggered. The formula optimization mechanism calculates and outputs the feed requirements F of different raw materials by constructing a feed formula optimization algorithm formula; The feed requirement F is calculated and outputted by the following feed formula optimization algorithm formula; ； Where, F i (t) represents the amount of feed required for the i-th type of feeding on the t-th day, R i (t) represents the digestion efficiency of the i-th feed ingredient on the t-th day.

7. The computer-based pig feed formula data optimization processing method according to claim 1, characterized in that: Said S4 includes S41 and S42; S41. After the feed formula is optimized and adjusted, the pig feeding data set of the pigs is collected in real time every day to update the growth curve expansion algorithm model, and the weight of the pigs on the tth day after the output adjustment is predicted. , perform feedback error calculation to obtain the pig weight prediction error value E; The pig weight prediction error value E is calculated and output by the following algorithm formula: Where E(t) represents the prediction error of the pig weight on day t, which is the difference between the weight predicted by the system and the actual weight; Ws(t) represents the actual weight of the pig on day t; S42. Based on the output of the predicted error value E(t) of the pig weight on day t, evaluate and analyze the current feed formula adjustment. The specific evaluation contents are as follows: When the predicted error value E(t) of the pig weight on day t is ≥0, it means that the predicted weight exceeds the actual weight, the feed formula adjustment is in line with expectations, and no secondary adjustment is required; When the predicted weight error value E(t) of the pig on day t is less than 0, it means that the predicted weight does not match the actual weight and the predicted weight is inaccurate. At this time, the gradient descent method is used to adjust the growth rate coefficient k and metabolic coefficient in the growth curve expansion algorithm model. Perform adaptive adjustment and optimization.

8. The computer-based pig feed formula data optimization processing method according to claim 6, characterized in that: Said S5 includes S51 and S52; S51. After adaptive adjustment and optimization, a feed efficiency algorithm formula is constructed to calculate and output the feed efficiency ratio (FER) of the pigs, and to monitor the optimization of the current feed formula; The feed efficiency ratio FER is calculated and outputted by the following feed efficiency algorithm formula; ； Where Ftotal(t) represents the total feed intake on day t.

9. A computer-based pig feed formula data optimization processing method according to claim 8, characterized in that: S52: Based on the user's preset efficiency threshold X, a secondary comparative evaluation is performed between the efficiency threshold X and the feed efficiency ratio FER, the feed formula optimization is analyzed, and the feed formula is further optimized based on the evaluation results. The specific evaluation contents are as follows: When the feed efficiency ratio FER ≥ the efficiency threshold X, it means that the utilization efficiency of the feed formula is normal and there is no need to adjust the feed formula; When the feed efficiency ratio FER is less than the efficiency threshold X, it indicates that the utilization efficiency of the feed formula is abnormal. At this time, the feed requirement F of the i-th type of feed in the t-th day obtained by the secondary calculation after adaptive adjustment and optimization is further sent to the automatic feed formula machine. i (t), the automatic feed formula machine automatically adjusts the amount of various feed raw materials.

10. A computer-based pig feed formula data optimization processing system, applied to a computer-based pig feed formula data optimization processing method according to any one of claims 1 to 9, characterized in that: It includes data acquisition and processing module, growth curve analysis module, nutritional demand prediction module, feedback adjustment and optimization module and optimization evaluation and analysis module; The data acquisition and processing module collects individual pig data and microbial metabolic activity data in real time by installing a detection device group and a feces collector on the farmed pigs. At the same time, a wireless communication gateway is set up to wirelessly transmit the individual pig data collected by the detection device group and the feces collector to the cloud server, where the data is pre-processed to obtain the pig feeding data set. The growth curve analysis module constructs a growth curve extension algorithm model, extracts a pig feeding data set, inputs the data into the growth curve extension algorithm model, calculates and outputs the predicted pig weight W; The nutritional requirement prediction module further calculates the nutritional requirement Dn of the pig at each growth stage by expanding the algorithm model according to the growth curve, outputting the predicted weight W of the pig, and presetting the nutritional requirement threshold D. The nutritional requirement threshold D is then compared and evaluated with the nutritional requirement Dn for the first time. If the initial comparison and evaluation indicates that the feed formula optimization mechanism is implemented, the feed formula component ratio of the pig is further analyzed to output the feed requirement F. The feedback adjustment and optimization module performs feedback error calculation based on the predicted weight W and the actual weight Ws of the pig to obtain the predicted weight error value E of the pig, and evaluates the nutritional intake of the pigs by the pig feed formula based on the output result of the predicted weight error value E, and adaptively adjusts and optimizes the growth curve expansion algorithm model; The optimization evaluation and analysis module constructs a feed efficiency algorithm formula after adaptive adjustment optimization, calculates and outputs the feed efficiency ratio FER of the pig, and presets the efficiency threshold X. Then, the efficiency threshold X and the feed efficiency ratio FER are compared and evaluated twice to analyze the feed formula optimization. If the utilization efficiency is evaluated to be abnormal, further feed adjustment is performed.

Images