Intelligent management method for quality control of selenium-rich egg chicken breeding in whole cycle

Abstract

This invention relates to the field of intelligent management and control of agricultural production and quality regulation of livestock and poultry, specifically an intelligent management method for the whole-cycle quality control of selenium-enriched laying hen breeding. The method divides laying hens into five stages, from brooding to late laying, and configures benchmark parameters. It collects multi-source data on environment, feeding, microbial community, and materials through multiple terminals and performs edge preprocessing. Based on a two-compartment kinetic model, it dynamically estimates selenium accumulation in the body, using accumulation safety as a rigid constraint, and employs a rolling optimization algorithm to solve for the optimal selenium feeding ratio. An LSTM model is constructed to predict egg selenium content and implement graded management. Combined with fecal microbial community data, it supplements probiotics in stages to regulate intestinal selenium conversion efficiency. This method establishes full-chain traceability for selenium raw materials, feed, and feeding, controls selenium emissions in manure through material balance, and builds digital twin survival data. After batch culling, iterative optimization of management parameters is performed. This invention balances egg selenium stability, selenium accumulation safety in the body, cost reduction in breeding, and manure reduction.

Description

Technical Field

[0001] This invention relates to the field of intelligent management and control of agricultural production and quality regulation of livestock and poultry, specifically to an intelligent management method for the whole-cycle quality control of selenium-enriched laying hen breeding. Background Technology

[0002] Selenium is an essential trace element for maintaining normal human physiological functions. It participates in the synthesis of various antioxidant enzymes and plays an important role in immune regulation and delaying aging. Selenium-enriched eggs convert inorganic selenium into organic selenium that is easily absorbed by the human body through the biotransformation process of laying hens. It is a safe and efficient carrier of selenium nutrition, and market demand has been increasing year by year, making it an important high-value-added category in the laying hen farming industry.

[0003] Currently, there are still many technical shortcomings in the quality control of selenium-enriched egg-laying chicken farming, which makes it difficult to support the high-quality development of the industry.

[0004] First, the logic of selenium feeding regulation is simplistic, generally focusing on "sufficient addition" as the core idea, ignoring the differences in the absorption and conversion capacity of selenium in the intestines of laying hens. Fluctuations in intestinal flora and changes in feed composition during the breeding process can significantly affect the bioavailability of selenium. Simply relying on increasing the amount added is not only costly, but also easily leads to selenium waste and excessive manure discharge. Existing solutions have not yet established a linkage regulation mechanism between intestinal microecological status and selenium feeding strategies.

[0005] Secondly, the safety risks of selenium accumulation in the body have been overlooked. Selenium is accumulative in animals, and long-term continuous supplementation will gradually accumulate in tissues such as the liver and muscles. In the later stages of egg production, the metabolic capacity of laying hens decreases, making it very easy for the selenium content in eggs to exceed the standard or even the selenium content in tissues to exceed the standard. Current regulations only focus on the daily content of egg products and do not dynamically estimate or impose safety constraints on the total amount accumulated in the body, which poses a long-term food safety hazard.

[0006] Third, the quality traceability system is incomplete. Existing traceability systems for selenium-enriched products mostly only record production data from the breeding end, failing to cover the entire material chain from selenium raw materials entering storage, feed formulation to on-site feeding. This makes it impossible to verify the authenticity and accuracy of selenium addition, and problems such as incorrect or omitted addition, or the use of inferior materials, are difficult to detect in a timely manner, hindering consumer trust in the quality of selenium-enriched products. Fourth, environmental control is lacking. Selenium not utilized by laying hens is excreted in feces. Excessive addition can lead to excessive selenium content in manure, posing environmental risks to farmland application. Current solutions only aim to meet egg product standards, failing to include selenium reduction in manure in the control objectives, which is inconsistent with the development direction of green farming.

[0007] Furthermore, existing intelligent aquaculture management systems mostly rely on conventional proportional control or experience-based adjustments, which are insufficient in terms of precision and robustness for complex systems like egg-laying hen farming, which are characterized by significant time lags, numerous disturbances, and multiple constraints. Some solutions introduce general predictive control algorithms, but these are not deeply integrated with the specific physiological and metabolic models of egg-laying hens, easily falling into the technical limitations of simply applying general algorithms and failing to fundamentally address the core pain points of selenium-enriched aquaculture. Currently, the industry lacks a safety management solution based on in vivo metabolic kinetics, and a comprehensive intelligent management system that balances quality stability, in vivo safety, material reliability, and environmental emission reduction throughout the entire lifecycle has not been established, hindering the standardization and large-scale upgrading of the selenium-enriched egg industry. Summary of the Invention

[0008] This invention addresses the shortcomings of existing technologies, such as low selenium absorption and conversion efficiency, lack of safety control over selenium accumulation, incomplete material traceability, and lack of control over manure emissions. It provides an intelligent management method for the entire lifecycle quality control of selenium-enriched egg-laying hen farming. The method constructs a closed-loop control system from multiple dimensions, including absorption and conversion, accumulation safety, material traceability, and environmental protection and emission reduction. While ensuring that the selenium content in eggs meets the standards, it improves selenium bioavailability, reduces farming costs, avoids accumulation safety risks, and reduces selenium emissions from manure, thereby achieving synergistic optimization of the quality, safety, and environmental protection of selenium-enriched farming.

[0009] The technical solution adopted by this invention to solve its technical problem is: an intelligent management method for quality control throughout the entire life cycle of selenium-enriched layer chicken farming, comprising the following steps: S1. Full life cycle stage division and benchmark parameter configuration: The entire life cycle of laying hens is divided into five control stages: brooding period, growing period, pre-laying period, peak laying period and post-laying period. Selenium nutrition benchmark parameters, production performance benchmarks, environmental thresholds, health benchmarks and in vivo selenium accumulation safety thresholds are configured for each stage. A basic parameter library for control is constructed, and bioavailability conversion coefficients and intestinal conversion benchmark values ​​for different selenium sources are configured. S2. Real-time acquisition and preprocessing of multi-source data: Through the chicken house environment sensor group, intelligent feeding terminal, production acquisition equipment, visual and acoustic device, fecal near-infrared detection module and selenium material traceability terminal, environmental data, feeding data, production data, chicken health data, intestinal flora characteristic data and selenium material flow data are collected in real time. After being cleaned and normalized by the edge node, the data is uploaded to the management and control platform. S3. Dynamic optimization of selenium feeding amount with accumulation safety constraints: Based on the two-compartment kinetic model, the selenium content and total accumulation in the central compartment and peripheral compartment of the laying hen are calculated through differential equation iteration; with the target egg selenium content as the control target and the in vivo accumulation safety threshold as the rigid constraint, combined with the daily production performance parameters, intestinal selenium conversion efficiency coefficient and stress deposition efficiency decay coefficient, the optimal total amount of selenium added and the ratio of organic selenium to inorganic selenium are calculated by the rolling optimization algorithm and sent to the intelligent feeding terminal for execution; S4. Egg Selenium Content Time-Series Prediction and Graded Closed-Loop Management: A long short-term memory network time-series prediction model is constructed, integrating multi-dimensional features such as age, selenium feeding amount, feed intake, environmental parameters, stress index, intestinal conversion efficiency, and body accumulation, to output the predicted selenium content of eggs for the current day and the next 3 days; three levels of control thresholds—target value, warning value, and exceedance value—are set to match corresponding control strategies, and measured values ​​are obtained regularly through near-infrared spectroscopy of egg products to correct the prediction and control model parameters; S5. Targeted intervention and synergistic regulation of gut microbiota: The gut selenium conversion efficiency coefficient is calculated based on the abundance characteristics of the gut microbiota obtained by near-infrared detection of feces. When the conversion efficiency is lower than the gut conversion benchmark value of the corresponding stage, the corresponding gradient of probiotic targeted supplementation program is automatically triggered according to the decline level. At the same time, the health of the flock is assessed based on visual voiceprint data. When the health is abnormal, synergistic intervention of nutrition and environment is initiated to maintain the stability of selenium deposition efficiency. S6. Full-chain traceability of selenium materials and coordinated management of manure emission reduction: Collect material flow data for the entire process of selenium raw material warehousing, feed formulation, and on-site feeding; realize automatic verification of addition amount through three-party material balance verification; estimate the selenium emission of manure based on selenium element material balance calculation; optimize feeding strategy and manure treatment parameters in conjunction with selenium material balance calculation; construct batch digital twins to support simulation and simulation of control parameters and reliable data storage throughout the entire chain. S7. Batch Cycle Review and Strategy Iteration: After culling a single batch of laying hens, review and analyze the quality, storage safety, material loss and emission reduction data throughout the entire cycle, optimize the benchmark parameters and model coefficients at each stage, and update the basic control parameter library.

[0010] Specifically, in step S1, the in vivo selenium accumulation safety threshold includes the upper limit of total selenium content in laying hens at each stage and the maximum allowable daily accumulation increment, which is set according to the safety limit standards for trace elements in livestock and poultry; the intestinal conversion benchmark value is the selenium absorption and conversion rate under normal flora conditions in laying hens of the corresponding age, which serves as the judgment benchmark for intervention triggering.

[0011] Specifically, in step S3, the two-compartment dynamic model divides the laying hen's body into a central compartment and a peripheral compartment, and the corresponding differential equation is: ; in , These represent the selenium concentrations in the central chamber and the peripheral chamber, respectively. The apparent distribution volume of the central chamber. This represents the daily selenium intake. , The inter-compartmental transport rate constant, Let f_t be the fecal excretion rate constant. The egg deposition rate constant is used; the initial parameters of the model are calibrated through variety metabolism experiments and iteratively corrected during operation based on measured data.

[0012] Specifically, in step S3, the intestinal selenium conversion efficiency coefficient is obtained by extracting the abundance of three types of bacterial groups—Firmwallis, Bacteroidetes, and Bifidobacteria—from fecal near-infrared spectroscopy and inputting them into a pre-trained partial least squares regression model. The coefficient ranges from 0.3 to 0.8, with a lower coefficient indicating a poorer intestinal selenium absorption and conversion capacity.

[0013] Specifically, in step S3, the rolling optimization algorithm takes the minimum deviation of egg selenium content and the minimum feeding cost as the objective function, and takes the egg selenium content meeting the standard, the total body accumulation not exceeding the safety threshold, and the daily addition not exceeding the safety upper limit as the constraints. It uses a quadratic programming algorithm to solve for the optimal solution, and the control cycle is 1 day.

[0014] Specifically, in step S4, the three-level control threshold is divided into target value range, warning range and exceedance range; when the predicted value is in the warning range, the feeding parameters are automatically fine-tuned and the prediction update cycle is shortened to once every 12 hours; when the predicted value exceeds the standard, the corresponding egg batch is automatically locked and offline near-infrared re-inspection is triggered.

[0015] Specifically, in step S5, the probiotic targeted supplementation program is set up with three levels of gradient according to the degree of decline in conversion efficiency: when the decline is mild, supplement with a compound preparation of Bacillus subtilis and Lactobacillus plantarum; when the decline is moderate, increase Saccharomyces cerevisiae and increase the total dose; when the decline is severe, further increase the organic selenium ratio and add probiotics to drinking water; after intervention, continuously monitor the conversion efficiency, and gradually reduce and stop the medication after it recovers to the baseline range.

[0016] Specifically, in step S6, the tripartite material balance verification involves comparing the theoretical consumption of selenium raw materials, the actual amount added in feed formulation, and the actual amount fed in the chicken house on the same day. An alarm is automatically triggered when the deviation exceeds a set threshold, and material loss and formulation errors are checked.

[0017] Specifically, in step S6, the amount of selenium discharged from feces is calculated by subtracting the total amount of egg deposits and the increase in body accumulation from the total daily selenium intake; when the estimated discharge exceeds the environmental protection threshold, the system increases the proportion of organic selenium and reduces the total amount added, thereby reducing selenium discharge from the source, while ensuring that the selenium in eggs meets the standards.

[0018] The beneficial effects of this invention are: Dual improvement in conversion efficiency and safety control: The innovative introduction of targeted regulation of gut microbiota and two-compartment dynamic accumulation estimation improves selenium bioavailability and reduces ineffective addition by improving gut microbiota. On the other hand, it avoids the risk of long-term overdose by dynamically constraining the amount of selenium in the body, achieving a synergy of "efficient utilization and safe accumulation". Compared with traditional feeding methods, selenium utilization can be improved, while completely eliminating the risk of excessive selenium in the later stage of egg production.

[0019] End-to-end material traceability ensures reliable quality: By connecting the material flow data of selenium raw materials, feed formulation, and on-site feeding, and realizing automatic verification of the amount added through three-party material balance verification, the pain point of "difficulty in proving the authenticity of selenium-enriched products" is solved. Combined with blockchain evidence storage, it greatly improves product credibility and market premium space.

[0020] Quality and environmental protection are optimized in synergy: Selenium emission reduction in manure is incorporated into the management and control objectives. Based on material balance, the emission volume is dynamically estimated. Selenium emission is reduced from the source through feed optimization, which takes into account both egg quality standards and environmental protection requirements for breeding. This is in line with the development direction of green agriculture and expands the application value of the management and control system.

[0021] Multi-constraint dynamic optimization enhances management robustness: By incorporating multiple dimensions such as accumulation safety, conversion efficiency, stress disturbances, and costs into a unified optimization framework, this approach is more adaptable to the complex and ever-changing scenarios of egg-laying hen farming compared to single-objective management solutions, and can maintain stable management effects under different growth stages and environmental conditions. Attached Figure Description

[0022] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0023] Figure 1 A flowchart of the intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens provided by the present invention; Figure 2 The overall architecture diagram of the intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens provided by the present invention is shown below. Figure 3 The following is a logic block diagram of the dynamic optimization of selenium feeding with accumulation safety constraints in the intelligent management method for quality control of the entire process of selenium-enriched layer chicken breeding provided by the present invention. Figure 4 This is a schematic diagram of the intestinal microecological targeted intervention and health synergistic regulation process in the intelligent management method for full-cycle quality control of selenium-enriched layer chicken breeding provided by the present invention. Detailed Implementation

[0024] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0025] like Figures 1-4As shown, the intelligent management method for quality control throughout the entire lifecycle of selenium-enriched egg-laying hen farming described in this invention includes seven core steps: full-cycle stage division and benchmark configuration, multi-source data acquisition and preprocessing, dynamic feeding optimization with accumulation safety constraints, egg selenium time-series prediction and graded control, synergistic intervention of gut microbiota and health, full-chain material traceability and emission reduction synergy, and batch review and iteration. The specific implementation method is as follows: (1) Full life cycle stage division and benchmark parameter configuration: Based on the physiological laws of egg production and development, the entire life cycle of laying hens is divided into five control stages: brooding period (0-6 weeks old), growing period (7-18 weeks old), pre-laying period (19-25 weeks old), peak laying period (26-45 weeks old), and post-laying period (46 weeks old to culling). Selenium nutrition benchmark parameters, production performance benchmarks, environmental thresholds, health benchmarks, and in vivo selenium accumulation safety thresholds are configured for each stage to construct a complete database of basic control parameters.

[0026] The selenium nutrition baseline parameters include daily selenium intake baselines for each stage, target selenium deposition in eggs, and safe upper limits for selenium addition in feed. These parameters comply with national feed hygiene and food safety standards. Bioavailability conversion factors for different selenium sources were configured, with a selenite bioavailability of 1.0 as the baseline, setting the relative bioavailability of yeast selenium at 1.8 and methionine selenium at 2.0, for dosage conversion when different selenium sources are mixed. Intestinal conversion baselines for each stage were also configured, representing the absorption and conversion efficiency of feed selenium in laying hens under normal physiological conditions: approximately 0.45 for the brooding period and approximately 0.6 for peak laying period, serving as the criteria for determining intestinal intervention.

[0027] The safe threshold for selenium accumulation in the body includes the upper limit of total selenium content in laying hens at each stage and the maximum allowable daily accumulation increase, set according to the safety limits for trace elements in livestock and poultry. The upper limit of total selenium content during peak egg production is set at 0.5 mg per kilogram of body weight, serving as a rigid constraint for feed optimization. Production performance benchmarks include standard feed intake, standard egg production rate, and standard egg weight at each stage; environmental thresholds include suitable temperature and humidity ranges, upper limit of ammonia concentration, and light parameters; health benchmarks include thresholds for normal flock behavior characteristics and the health qualification line.

[0028] (2) Real-time acquisition and preprocessing of multi-source data: Construct a sensing and acquisition system covering six dimensions: environment, feeding, production, health, gut health, and materials, to achieve real-time acquisition of all elements of data. The environmental sensor group is evenly deployed at a density of no less than 3 monitoring points per chicken house, and collects data on temperature, humidity, ammonia concentration, and light intensity every 5 minutes; the intelligent feeding terminal is deployed in each feeding unit to record the daily feed amount, selenium additive amount, and actual feed intake, with a data accuracy of no less than 10g; the production acquisition equipment is connected to the egg collection system to acquire daily data on egg production rate and average egg weight; the visual and acoustic acquisition devices are deployed in the chicken house aisles and in front of the cages to collect data on the appearance, behavior, and calls of the flock for health status assessment.

[0029] A near-infrared detection module for feces is deployed at the front end of the manure removal system. This module automatically collects the near-infrared spectrum of fresh fecal samples daily. Spectral analysis reveals the abundance of three core bacterial groups: Firmicutes, Bacteroidetes, and Bifidobacteria, which are used to calculate intestinal selenium conversion efficiency. Selenium material traceability terminals are deployed at the raw material warehouse, feed production workshop, and chicken house feeding area. These terminals collect data on selenium raw material batches, feed formulation records, and on-site feeding consumption, enabling end-to-end material traceability.

[0030] All collected data undergoes preprocessing at edge computing nodes, including using the 3σ criterion to remove outliers, using linear interpolation to fill in missing values, and using maximum-minimum normalization to eliminate dimensional differences between different data dimensions. The preprocessed data is then uploaded to the cloud management platform via an IoT gateway, providing a unified data foundation for intelligent decision-making.

[0031] (3) Dynamic optimization of selenium feeding amount with accumulation safety constraint: This step constructs a multi-constraint dynamic feeding optimization model. The core is to introduce a two-compartment kinetic model to realize the dynamic estimation of selenium accumulation in the body. Accumulation safety is used as a rigid constraint. At the same time, the intestinal conversion efficiency and stress attenuation coefficient are coupled to achieve precise feeding.

[0032] A two-compartment kinetic model was used to simulate the metabolism of selenium in laying hens: metabolically active tissues such as blood circulation and liver were designated as the central compartment, while slower-metabolizing tissues such as muscle and feathers were designated as the peripheral compartment. The corresponding differential equations are as follows: ; In the formula, For the central chamber selenium concentration, This represents the selenium concentration in the peripheral chamber, in mg / L. The apparent volume of the central chamber is expressed in liters (L). This represents the daily intake of selenium absorbed into the central compartment via the intestines, expressed in mg / d. The rate constant for transport from the central chamber to the peripheral chamber. Let be the rate constant for transport from the peripheral chamber to the central chamber. Let be the rate constant for selenium excretion via feces. The rate constant for selenium deposition into eggs is given in d. -1 .

[0033] The initial parameters of the model were calibrated through selenium metabolism experiments on the corresponding breed of laying hens. An example value is taken as follows: peak egg production period... During operation, based on the previous day's selenium intake, egg production, and excretion, the selenium content and total accumulation in each compartment of the body are calculated iteratively using numerical methods. At the same time, the rate constant is periodically corrected by combining measured data of selenium content in blood and eggs to improve the estimation accuracy.

[0034] Using the target selenium content in eggs as the control objective and the safe threshold for selenium accumulation in the body as a rigid constraint, and combining basic production parameters such as age on the day of birth, actual feed intake, egg production rate, and average egg weight, a dual correction was performed using the intestinal selenium conversion efficiency coefficient and the stress deposition efficiency decay coefficient. The intestinal selenium conversion efficiency coefficient was calculated by inputting the characteristics of three types of microbiota obtained from fecal near-infrared spectroscopy into a pre-trained partial least squares regression (PLSR) model. The coefficient ranged from 0.3 to 0.8, with lower coefficients indicating poorer intestinal absorption and conversion capacity, corresponding to an increase in the proportion of organic selenium. The stress deposition efficiency decay coefficient was obtained by querying the stress index calculated from temperature and humidity indices, ammonia concentration, and behavioral characteristics. The stronger the stress, the greater the decay, corresponding to the amount of selenium supplemented.

[0035] The rolling optimization algorithm uses the minimum deviation in egg selenium content and the lowest feeding cost as its objective functions, and the constraints of egg selenium content within the target range, total selenium accumulation not exceeding the safety threshold, and daily addition not exceeding the safety upper limit. It employs a quadratic programming algorithm to find the optimal solution, with a control cycle of one day. Every morning, the optimal total selenium addition and the ratio of organic to inorganic selenium are calculated and sent to the intelligent feeding terminal for execution. When the selenium accumulation approaches 90% of the safety threshold, the system prioritizes safety, automatically reducing the total addition and increasing the proportion of organic selenium to maintain stable egg selenium levels without exceeding the limits.

[0036] (4) Temporal prediction and graded closed-loop management of egg selenium content: A temporal prediction model for egg selenium content based on Long Short-Term Memory (LSTM) network was constructed. The input layer has 7 dimensions, including age, daily selenium feeding amount, average daily feed intake, average daily temperature and humidity, stress index, intestinal conversion efficiency, and body accumulation. Using historical data from the past 14 days as the input window, the model outputs predicted values ​​of egg selenium content for the current day and the next 3 days, enabling early prediction of quality status. The model is trained using the Adam optimizer with mean squared error as the loss function. It was pre-trained using historical aquaculture data before deployment and operation.

[0037] The system sets three levels of control thresholds: target value, warning value, and exceedance value, and implements a tiered control strategy: when the predicted value is within the target value range, the current feeding strategy remains unchanged; when the predicted value enters the warning range, the system automatically fine-tunes the amount of selenium added and the ratio of organic selenium, and shortens the prediction update cycle from once a day to once every 12 hours to closely track the trend of quality changes; when the predicted value reaches the exceedance value, the system automatically locks the egg batch in the corresponding time period, prohibits it from leaving the warehouse, and triggers offline near-infrared rapid detection for re-inspection and confirmation.

[0038] The egg collection line is equipped with a near-infrared spectroscopy rapid detection module. Twenty eggs are randomly sampled daily to obtain the actual selenium content. The measured data is regularly fed back to the LSTM prediction model and the feeding optimization model to iteratively correct the model parameters, continuously improve the accuracy of prediction and control, and form a complete quality closed loop of "prediction-control-detection-iteration".

[0039] (5) Targeted intervention and synergistic regulation of gut microbiota: Establish a linkage mechanism between gut microbiota status and selenium nutrition regulation, stabilize selenium deposition efficiency from the perspective of improving absorption and conversion capacity, rather than simply relying on increasing feeding amount. Based on the microbiota characteristic data obtained by fecal near-infrared detection, the gut selenium conversion efficiency coefficient is calculated daily. When the coefficient is lower than the gut conversion benchmark value of the corresponding stage, the probiotic targeted supplementation program is automatically triggered.

[0040] The probiotic supplementation program is structured in three tiers: For mild decline (efficiency at 80%-90% of baseline), add 100g of a compound preparation of Bacillus subtilis and Lactobacillus plantarum per ton of feed, in a 2:1 ratio; for moderate decline (efficiency at 60%-80% of baseline), add 50g / t of Saccharomyces cerevisiae, increasing the total probiotic dose to 150g / t; for severe decline (efficiency below 60% of baseline), increase the total probiotic dose to 200g / t, simultaneously supplementing with drinking water and appropriately increasing the proportion of organic selenium. After intervention, intestinal conversion efficiency is continuously monitored. Once the coefficient recovers to the baseline range and stabilizes for 3 days, the dosage is gradually reduced and eventually discontinued to avoid long-term excessive supplementation.

[0041] Simultaneously, coordinated regulation of flock health was implemented. Based on visual and vocal data, four characteristics were extracted: comb color saturation, fecal morphology classification, feed intake frequency per unit time, and the proportion of abnormal respiratory sounds. A weighted health score of 0-100 was calculated. When the health score was below 70, the proportion of organic selenium added was appropriately increased by 5%-10% within the safe limit. Vitamin E and probiotics were supplemented simultaneously, and environmental control equipment was used to optimize the temperature, humidity, and ventilation conditions in the coop. Through dual intervention of nutrition and environment, the health status of the flock was restored, avoiding a decrease in selenium deposition efficiency caused by health fluctuations.

[0042] (6) Full-chain traceability of selenium materials and coordinated management of manure and wastewater reduction: Construct a full-chain material traceability system from selenium raw materials to egg production to ensure the authenticity and verifiability of selenium-enriched quality. In the selenium raw material warehousing stage, record the raw material batch, origin, selenium content test report and warehousing weight to generate a unique raw material batch file; in the feed formulation stage, record the selenium addition amount, formulation time and production batch number of each batch of feed, and automatically calculate the selenium concentration of the feed; in the chicken house feeding stage, record the daily actual feeding amount and selenium consumption of each chicken house. The data of the three stages are automatically linked and matched, and automatic verification is achieved through three-party material balance verification: compare the theoretical consumption of selenium raw materials on the day, the actual addition amount of feed formulation, and the actual feeding amount of chicken houses. When the deviation exceeds 2%, an alarm is automatically triggered to notify the management personnel to check the material loss and formulation error.

[0043] Simultaneously, a coordinated management system for selenium reduction in manure is implemented. Based on the principle of selenium element material balance, the daily selenium emission in manure is estimated by subtracting the total amount of selenium deposited in eggs and the increase in selenium accumulation in the body from the total daily selenium intake. When the estimated emission exceeds the set environmental threshold, the system, while ensuring that the selenium content in eggs meets the standards, improves selenium utilization by increasing the proportion of organic selenium and optimizing the feeding time window, thereby reducing the total amount of selenium added and reducing selenium emissions in manure from the source. At the same time, the system coordinates with the manure treatment system to adjust composting parameters to ensure that the selenium content in the treated manure meets the standards.

[0044] A full-cycle digital twin is constructed using a single batch of laying hens as the unit, comprehensively mapping all data related to feeding, environment, health, materials, testing, and emission reduction, and supporting simulation and extrapolation of control parameters. Operators can simulate and adjust parameters such as selenium addition and probiotic supplementation. The system outputs trends in egg selenium content, selenium accumulation in the body, manure emissions, and breeding costs to assist in production decision-making. Core control data, material data, and test results are uploaded to a blockchain node for storage and evidence, generating a unique traceability QR code to ensure the immutability of data throughout the entire chain.

[0045] (7) Batch cycle review and strategy iteration: After a batch of laying hens is culled, the system automatically performs a multi-dimensional review and analysis of the entire cycle data of that batch, and statistically analyzes core indicators such as the fluctuation range of selenium content in eggs, quality qualification rate, trend of changes in selenium accumulation in the body, selenium bioavailability, material loss rate, and reduction of manure emissions, and quantitatively evaluates the implementation effect of each link of intestinal intervention, stress correction, and accumulation control. Based on the review results, the baseline parameters, model rate constants and intervention thresholds of each stage are iteratively optimized to continuously improve the control accuracy and economy, and realize the self-evolution of the control system.

[0046] Example 1: Application of Gut Microecology-Driven Quality Control During Peak Egg Production: This example was applied in a large-scale egg-laying farm during peak egg production to verify the effect of gut microecology-driven regulation on improving selenium utilization and quality stability. Three buildings of 30-week-old Hy-Line Brown chickens were selected, each housing 12,000 birds. Different control schemes were used in a comparative 8-week trial. During the trial, the overall health of the flocks was good, and no large-scale disease outbreaks occurred.

[0047] The specific implementation process of the test building using this method is as follows: First, baseline parameters for peak egg production were configured, setting the target selenium deposition value at 30 μg per egg, with a target range of 27-33 μg / egg. The baseline intestinal conversion rate for peak egg production was set at 0.6, and the safe upper limit for total selenium accumulation was set at 0.5 mg per kilogram of body weight. Based on a sodium selenite bioavailability of 1.0, the relative bioavailability of yeast selenium was set at 1.8. A comprehensive data acquisition system was then implemented: four environmental sensors were evenly deployed within the chicken house, collecting data on temperature, humidity, ammonia concentration, and light intensity every 5 minutes; the intelligent feeding terminal recorded daily feed and selenium additive consumption with a data accuracy of 5g; the egg collection system uploaded daily data on egg production rate and average egg weight; the near-infrared fecal detection module at the front end of the manure removal system automatically collected fecal samples daily, analyzing the abundance of Firmicutes, Bacteroides, and Bifidobacteria, and calculating the daily intestinal selenium conversion efficiency coefficient using a pre-trained partial least squares regression model; visual and voiceprint devices monitored the flock's health status daily and calculated a health score. All data is preprocessed by the edge gateway before being uploaded to the management platform.

[0048] Every day at dawn, dynamic optimization calculations of feeding amount are performed: First, the two-compartment kinetic model is called. Based on the selenium intake, egg deposition and excretion of the previous day, the selenium concentration and total accumulation in the central compartment and peripheral compartment of the body are calculated iteratively by numerical solution of differential equations to confirm that they are within the safe threshold range. Then, the production parameters of the day, the intestinal conversion efficiency coefficient and the decay coefficient corresponding to the stress index are input. With the goal of minimizing the deviation of egg selenium content and the lowest feeding cost, and with the constraints of egg selenium reaching the standard, accumulation safety and the addition amount not exceeding the limit, the quadratic programming algorithm is used to solve the optimal total amount of selenium added and the ratio of yeast selenium and sodium selenite for the day, and then the solution is sent to the feeding terminal for execution.

[0049] Daily time-series prediction of egg selenium content was performed synchronously. Based on 7-dimensional feature data from the past 14 days, an LSTM model was used to predict the selenium content of eggs for the current day and the next 3 days, and a three-level control threshold was set to implement tiered control. Twenty eggs were sampled daily for selenium content detection using a near-infrared spectrometer, and the measured data was fed back to correct the parameters of the prediction model and the compartmentalized model. During the second week of the experiment, continuous rainy weather occurred, causing fluctuations in the gut microbiota of the chickens. The intestinal conversion efficiency coefficient dropped to 0.48, below 80% of the baseline value. The system automatically triggered a moderate intervention program, adding 150g of a compound preparation of Bacillus subtilis, Lactobacillus plantarum, and Saccharomyces cerevisiae per ton of feed. After 5 days of continuous intervention, the intestinal conversion efficiency recovered to 0.58, and the system gradually reduced the dosage until the intervention was discontinued.

[0050] Control Example 1: The traditional fixed selenium addition method was used, with yeast selenium added at a fixed ratio throughout the process without dynamic adjustment or intestinal intervention. Daily feeding management was consistent with that of the experimental building.

[0051] Example 2: A conventional dynamic feeding program was adopted, and the amount of selenium added was adjusted only according to the feed intake and egg production rate. Intestinal conversion efficiency and in vivo accumulation control were not considered. All other conditions were the same as those in the experimental building.

[0052] The experimental results showed that during the 8-week experimental period, the average selenium content of eggs in the experimental unit was 30.1 μg / egg, with a fluctuation range of ±1.9 μg / egg, and the quality qualification rate was 99.1%. The total selenium consumption in the feed was 18.7% lower than that in control example 1, and the selenium bioavailability was significantly improved. The average selenium content of eggs in control example 1 was 29.7 μg / egg, with a fluctuation range of ±5.8 μg / egg, and the quality qualification rate was 81.5%. The selenium content of eggs in control example 2 declined significantly during rainy weather. The average selenium content of eggs in control example 2 was 30.0 μg / egg, with a fluctuation range of ±4.0 μg / egg, and the quality qualification rate was 88.2%. The selenium consumption was 16.3% higher than that in the experimental unit.

[0053] The comparison shows that this method significantly improves selenium bioavailability through the coordinated regulation of the gut microbiota, achieving higher quality stability with a lower total addition amount, greatly reducing breeding costs, and avoiding waste and emission pressure caused by blindly increasing the dosage. During the experiment, the overall gut health of the chicken flock improved, and the incidence of digestive tract diseases decreased compared to the control flock, verifying the positive synergistic effect of gut regulation and quality control. The system operates stably, with daily optimization calculations completed within seconds, and data acquisition and command issuance delays within acceptable ranges, meeting the daily production rhythm of large-scale farms.

[0054] Example 2: Application of Full-Cycle In vivo Selenium Accumulation Safety Management: This example focuses on the full-cycle application of this method for a single batch of laying hens, specifically verifying the effectiveness of in vivo selenium accumulation management in ensuring the quality and safety of eggs during the later stages of egg production. The experimental batch consisted of 15,000 one-day-old Hy-Line Brown chicks. This method was used throughout the entire brooding period until the end of the later laying period (60 weeks of age), fully covering the entire life cycle of the laying hen. The experimental site was a standardized chicken house in a large-scale laying hen farm.

[0055] The full-cycle management and control implementation process is as follows: The brooding and rearing periods focus on growth and development, configuring corresponding selenium nutritional benchmarks and accumulation safety thresholds for each stage. The two-compartment kinetic model is calibrated to initial values ​​based on chick metabolic parameters. (Brooding period) , The selenium intake is gradually adjusted as the chicks grow older. The system collects daily data on weight, feed intake, and gut microbiota to dynamically optimize selenium feeding amounts, ensuring normal growth and selenium reserves in chicks while strictly controlling selenium accumulation within a safe range. During this stage, blood selenium levels are tested every two weeks, and the measured values ​​are fed back to correct the rate constant of the compartmentalized model, improving the accuracy of accumulation estimation.

[0056] After entering the pre-laying stage, the system gradually transitions to a laying period management mode, increasing selenium supplementation targets and initiating egg selenium content prediction and closed-loop management. During peak laying, precise dynamic regulation is maintained, combined with intestinal intervention and stress correction to ensure stable quality. Simultaneously, changes in selenium accumulation are continuously monitored to ensure the total accumulation remains within 70%-80% of the safe threshold. In the late laying stage, the hen's metabolic capacity declines, egg production gradually decreases, and selenium excretion slows. The system, through a compartmentalized model, detects a gradual increase in selenium accumulation. When this accumulation approaches 90% of the safe threshold, a constraint mechanism is automatically triggered: while ensuring egg selenium levels meet standards, the total selenium supplementation is gradually reduced, while the proportion of organic selenium is increased from 60% to 80%. Higher bioavailability maintains egg selenium deposition efficiency, ensuring stable egg selenium content while preventing continuous increases in selenium accumulation from exceeding the safe upper limit.

[0057] Throughout the entire cycle, the system outputs daily estimates and safety margins for selenium accumulation in the body, allowing managers to monitor real-time trends through the management platform. When the accumulation enters the warning range, the system automatically issues alerts and adjusts feeding strategies to consistently maintain total selenium levels within safe limits. During the later stages of egg production, five chickens are sampled monthly to test liver and muscle selenium content, verifying the accuracy of the accumulation estimate and the effectiveness of the management. The deviation between the measured values ​​and the model estimates is within 8%.

[0058] A control group was set up: laying hens from adjacent batches in the same farm were fed a standard, fixed selenium supplementation program with no control over selenium accumulation in their bodies. The amount of selenium added remained constant throughout the process, and all other feeding and management conditions were exactly the same as those of the experimental batch.

[0059] Statistical data after the batch was completed showed that the average selenium content of eggs in the experimental batch throughout the entire laying period was 29.9 μg / egg, with a fluctuation range of ±2.3 μg / egg in the later stages of laying, and no exceedances were observed throughout the entire process. The selenium content in the sampled body tissues all met the safety limits for livestock and poultry, and the control effect on the accumulation in the body was good. The total selenium consumption throughout the entire cycle was reduced by 15.2% compared with the control group. The selenium content of eggs in the control batch was basically within the standard in the early stages of laying, but the selenium content in eggs gradually increased in the later stages of laying, with an exceedance rate of 11.3% in the later stages. The selenium content in the body tissues was close to the safety limit, posing a significant food safety hazard.

[0060] This embodiment verifies the effectiveness of compartmentalized kinetic accumulation control in two-compartment farming, addressing the long-standing industry pain point of excessive selenium levels in the later stages of egg production, and achieving dual assurance of quality and safety throughout the entire lifecycle. Compared to methods that only focus on the selenium content of eggs on a given day, this method starts from the underlying logic of metabolism in the body, making the control more forward-looking and safer. It can effectively avoid the accumulation risks caused by long-term feeding, ensuring the food safety compliance of the product, and is especially suitable for long-term large-scale farming applications.

[0061] Example 3: Synergistic Application of End-to-End Material Traceability and Manure Reduction: This example was applied in a large-scale egg-laying hen farm to verify the actual effect of synergistic management of end-to-end selenium material traceability and manure reduction. Two peak-laying hen houses, each with a capacity of 15,000 birds, were selected for the experiment. This method and a conventional management plan were used respectively. The experimental period was 10 weeks. During the experiment, feed was uniformly prepared throughout the farm, and the basic daily ration formula was consistent.

[0062] The specific implementation process of the test building using this method is as follows: A full-chain traceability system for selenium materials was established: Traceability terminals were deployed in the raw material warehouse, recording batch number, supplier, selenium content test report, and weight upon receipt of selenium raw materials. The system automatically generates a unique batch file for each raw material. Data acquisition terminals were deployed in the feed production workshop, recording the preparation time, selenium additive dosage, and feed output for each batch of feed. The system automatically calculates the theoretical selenium concentration in the feed. Feeding terminals in each chicken house record the actual daily feed intake and selenium consumption. Data from these three stages are uploaded in real time and automatically correlated. The system performs daily material balance checks, comparing the deviations between the theoretical raw material consumption, the actual feed addition, and the actual feed intake in the chicken houses. An alarm is automatically triggered when the deviation exceeds 2%, notifying management for verification. During the trial, a feed preparation and weighing deviation occurred once; the system triggered an alarm on the same day, and management corrected it promptly, preventing quality fluctuations.

[0063] Simultaneous coordinated management and control of manure emission reduction: The system estimates daily manure selenium emissions based on total daily selenium intake, minus total egg deposition and increased selenium accumulation in the body, using material balance. When the estimated emissions approach the set environmental threshold, the system automatically optimizes feeding strategies, increasing the proportion of organic selenium to improve deposition efficiency and reducing total selenium addition, while ensuring selenium content in eggs meets standards, thereby reducing manure selenium emissions at the source. Simultaneously, it coordinates with the manure treatment system to adjust composting fermentation parameters to ensure that the selenium content of treated manure meets standards.

[0064] A batch digital twin is constructed, fully integrating data across all dimensions of materials, feeding, environment, quality, and emission reduction, and supporting simulation and extrapolation functions. Managers can simulate and adjust the selenium source ratio within the twin interface. The system simultaneously outputs quantitative results of changes in egg selenium content, selenium utilization rate, and manure emissions, providing data support for production decisions. All core material and testing data are uploaded to the blockchain for evidence storage. Each batch of product is accompanied by a unique traceability QR code, which consumers can scan to view selenium raw material batch information, feed formulation records, daily feeding amounts, and egg selenium testing reports.

[0065] Example 1 (Control): Conventional farming management is adopted, with no full-chain traceability of materials. Only the total feed consumption is recorded, and material balance verification is not performed. Quality is controlled by weekly random inspections.

[0066] Example 2: A conventional quality control scheme was adopted, which only aimed to achieve the selenium standard for eggs, without considering the control of selenium discharge from manure and sewage, and the amount of selenium added was set according to the conventional standard.

[0067] The test results showed that during the 10-week test period, the material balance deviation in the experimental building was consistently controlled within 1.5%, and no quality incidents occurred due to missed or incorrect addition of selenium, indicating a significant improvement in material management precision. The average selenium emission from manure was reduced by 22.6% compared to Control Example 2, and the selenium content after manure treatment remained stable and met the environmental protection standards for farmland application. The egg selenium quality qualification rate reached 98.8%, maintaining excellent quality stability. Market research showed that consumers' trust in selenium-enriched egg products with full-chain traceability significantly increased, and the market price was about 20% higher than that of ordinary selenium-enriched eggs. Control Example 1 experienced a deviation in the amount of selenium added in the feed formulation, which was not discovered until a sampling inspection three days later, resulting in fluctuations in the selenium content of the eggs over those three days, and the traceability and verification took more than half a day. Control Example 2 had a higher selenium content in manure, posing an environmental compliance risk.

[0068] This embodiment verifies the practical value of the coordinated management of end-to-end material traceability and manure emission reduction. It not only ensures the authenticity and controllability of selenium-enriched quality through closed-loop material management, but also aligns with the policy requirements of green farming through source emission reduction. This achieves multiple benefits, including quality improvement, brand enhancement, and environmental compliance, and is suitable for promotion and application in large-scale farming parks.

[0069] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of protection claimed by the present invention. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent management method for quality control of the whole cycle of selenium-rich laying hen breeding, characterized in that, Includes the following steps: S1. Full life cycle stage division and benchmark parameter configuration: The entire life cycle of laying hens is divided into five control stages: brooding period, growing period, pre-laying period, peak laying period and post-laying period. Selenium nutrition benchmark parameters, production performance benchmarks, environmental thresholds, health benchmarks and in vivo selenium accumulation safety thresholds are configured for each stage. A basic parameter library for control is constructed, and bioavailability conversion coefficients and intestinal conversion benchmark values ​​for different selenium sources are configured. S2. Real-time acquisition and preprocessing of multi-source data: Through the chicken house environment sensor group, intelligent feeding terminal, production acquisition equipment, visual and acoustic device, fecal near-infrared detection module and selenium material traceability terminal, environmental data, feeding data, production data, chicken health data, intestinal flora characteristic data and selenium material flow data are collected in real time. After being cleaned and normalized by the edge node, the data is uploaded to the management and control platform. S3. Dynamic optimization of selenium feeding amount with accumulation safety constraints: Based on the two-compartment kinetic model, the selenium content and total accumulation in the central compartment and peripheral compartment of the laying hen are calculated through differential equation iteration; with the target egg selenium content as the control target and the in vivo accumulation safety threshold as the rigid constraint, combined with the daily production performance parameters, intestinal selenium conversion efficiency coefficient and stress deposition efficiency decay coefficient, the optimal total amount of selenium added and the ratio of organic selenium to inorganic selenium are calculated by the rolling optimization algorithm and sent to the intelligent feeding terminal for execution; S4. Egg Selenium Content Time-Series Prediction and Graded Closed-Loop Management: A long short-term memory network time-series prediction model is constructed, integrating multi-dimensional features such as age, selenium feeding amount, feed intake, environmental parameters, stress index, intestinal conversion efficiency, and body accumulation, to output the predicted selenium content of eggs for the current day and the next 3 days; three levels of control thresholds—target value, warning value, and exceedance value—are set to match corresponding control strategies, and measured values ​​are obtained regularly through near-infrared spectroscopy of egg products to correct the prediction and control model parameters; S5. Targeted intervention and synergistic regulation of gut microbiota: The gut selenium conversion efficiency coefficient is calculated based on the abundance characteristics of the gut microbiota obtained by near-infrared detection of feces. When the conversion efficiency is lower than the gut conversion benchmark value of the corresponding stage, the corresponding gradient of probiotic targeted supplementation program is automatically triggered according to the decline level. At the same time, the health of the flock is assessed based on visual voiceprint data. When the health is abnormal, synergistic intervention of nutrition and environment is initiated to maintain the stability of selenium deposition efficiency. S6. Full-chain traceability of selenium materials and coordinated management of manure emission reduction: Collect material flow data for the entire process of selenium raw material warehousing, feed formulation, and on-site feeding; realize automatic verification of addition amount through three-party material balance verification; estimate the selenium emission of manure based on selenium element material balance calculation; optimize feeding strategy and manure treatment parameters in conjunction with selenium material balance calculation; construct batch digital twins to support simulation and simulation of control parameters and reliable data storage throughout the entire chain. S7. Batch Cycle Review and Strategy Iteration: After culling a single batch of laying hens, review and analyze the quality, storage safety, material loss and emission reduction data throughout the entire cycle, optimize the benchmark parameters and model coefficients at each stage, and update the basic control parameter library. 2.The intelligent management method for quality control of the whole cycle of selenium-enriched laying hen breeding according to claim 1, characterized in that: In step S1, the in vivo selenium accumulation safety threshold includes the upper limit of total selenium content in laying hens at each stage and the maximum allowable daily accumulation increment, which is set according to the safety limit standards for trace elements in livestock and poultry. The intestinal conversion benchmark value is the selenium absorption and conversion rate under normal flora conditions in laying hens of the corresponding age, which serves as the criterion for determining intervention trigger. 3.The intelligent management method for quality control of the whole cycle of selenium-enriched laying hen breeding according to claim 1, characterized in that: In step S3, the two-compartment kinetic model divides the laying hen's body into a central compartment and a peripheral compartment, and the corresponding differential equation is: ； in , These represent the selenium concentrations in the central chamber and the peripheral chamber, respectively. The apparent distribution volume of the central chamber. This represents the daily selenium intake. , The inter-compartmental transport rate constant, Let f_t be the fecal excretion rate constant. The egg deposition rate constant is used; the initial parameters of the model are calibrated through variety metabolism experiments and iteratively corrected during operation based on measured data.

4. The intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens according to claim 1, characterized in that: In step S3, the intestinal selenium conversion efficiency coefficient is obtained by extracting the abundance of three types of bacterial groups—Firmwallis, Bacteroidetes, and Bifidobacteria—from fecal near-infrared spectroscopy and inputting them into a pre-trained partial least squares regression model. The coefficient ranges from 0.3 to 0.8, with a lower coefficient indicating a poorer intestinal selenium absorption and conversion capacity.

5. The intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens according to claim 1, characterized in that: In step S3, the rolling optimization algorithm takes the minimum deviation of egg selenium content and the minimum feeding cost as the objective function, and takes the egg selenium content meeting the standard, the total body accumulation not exceeding the safety threshold, and the daily addition not exceeding the safety upper limit as the constraints. It uses a quadratic programming algorithm to solve for the optimal solution, and the control cycle is 1 day.

6. The intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens according to claim 1, characterized in that: In step S4, the three-level control threshold is divided into target value range, warning range, and exceedance range; When the predicted value is in the warning range, the feeding parameters are automatically fine-tuned and the prediction update cycle is shortened to once every 12 hours. When the predicted value exceeds the standard, the corresponding egg batch will be automatically locked and an offline near-infrared re-inspection will be triggered.

7. The intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens according to claim 1, characterized in that: In step S5, the probiotic targeted supplementation program is set with a three-level gradient according to the degree of decline in conversion efficiency: For mild decline, supplement with a compound preparation of Bacillus subtilis and Lactobacillus plantarum; When the level decreases moderately, increase the amount of brewer's yeast and increase the total dosage; When the level drops significantly, further increase the organic selenium ratio and add probiotics to the drinking water; After intervention, the conversion efficiency was continuously monitored, and the dosage was gradually reduced and discontinued after the conversion efficiency returned to the baseline range.

8. The intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens according to claim 1, characterized in that: In step S6, the tripartite material balance verification involves comparing the theoretical consumption of selenium raw materials, the actual amount added in feed formulation, and the actual feeding amount in the chicken house on the same day. An alarm is automatically triggered when the deviation exceeds a set threshold, and material loss and formulation errors are checked.

9. The intelligent management method for quality control throughout the entire breeding cycle of selenium-enriched laying hens according to claim 1, characterized in that: In step S6, the amount of selenium discharged from feces is calculated by subtracting the total amount of egg deposits and the increase in selenium accumulation in the body from the total daily selenium intake. When the estimated emissions exceed the environmental threshold, the system increases the proportion of organic selenium and reduces the total amount added, while ensuring that the selenium content of the eggs meets the standards, thereby reducing selenium emissions from the source.

Images