An automatic grading and weighing conveyor system on a production line

Abstract

This invention relates to the field of feed weighing information technology, specifically disclosing an automatic grading, weighing, and conveying system for production lines, comprising six modules: data acquisition, core data processing, early warning output, uniformity prediction, economic decision-making, and central interaction. The system optimizes livestock farming efficiency through a three-layer progressive data processing chain: First, based on routine feeding and individual weighing data, a dynamic correlation analysis algorithm is used to identify hidden, inefficient feed-consuming individuals; second, this data drives a population uniformity decay model to predict the distribution of slaughter quality grades; finally, combining cost and market price, a multi-objective optimization algorithm recommends the optimal slaughter time and differentiated pricing strategies. This invention achieves end-to-end automation from production data to management insights and economic decision-making, effectively improving the refined management capabilities and economic benefits of livestock farms.

Description

Technical Field

[0001] This invention relates to the field of feed weighing information technology, specifically an automatic grading, weighing, and conveying system for a production line. Background Technology

[0002] Automated grading, weighing, and conveying systems are essential equipment in modern large-scale duck farms for improving feeding management and growth monitoring efficiency. Their core function is to automatically and precisely deliver feed to each pen and weigh individual ducks in a streamlined process, recording the corresponding weight data. Based on this data, the system can calculate the overall feed conversion ratio, reflecting feed efficiency and providing a fundamental efficiency indicator for farm production management.

[0003] For example, application number "CN202211518333.4" discloses a feed delivery system that requires only one operator in the control room to complete the entire breeding process, eliminating the need for additional personnel and significantly reducing labor input, thereby lowering breeding costs. Simultaneously, with the control room located outside the chicken coop, operators do not need to enter or leave the coop, reducing the time spent in contact with germ-free chickens and thus reducing the probability of infection, allowing the chickens to grow better and further lowering breeding costs. However, existing systems of this type have significant limitations in their data processing logic and application depth. Firstly, they typically only calculate the average feed conversion ratio (FCR) of all ducks in the coop over a breeding cycle, failing to penetrate the group average to identify individual ducks with low actual feeding efficiency and potential health risks that go undetected. This leads to management blind spots and hidden feed waste. Secondly, due to the lack of in-depth analysis of the correlation between individual abnormalities and group performance, existing technologies cannot make reliable quantitative predictions on the final body uniformity of the entire batch of ducks and the proportion of high-value commercial ducks before slaughter. This results in an information gap between production management and market sales preparation. Finally, existing systems are completely lacking in the ability to conduct dynamic economic analysis based on real-time growth data and market conditions. They cannot provide farms with decision support regarding the optimal slaughter time and differentiated pricing strategies, making slaughter decisions still highly dependent on experience and making it difficult to maximize the economic benefits of farming when facing cost fluctuations and market price changes. Summary of the Invention

[0004] This invention provides an automatic grading, weighing, and conveying system for production lines, which solves the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an automatic grading, weighing, and conveying system for a production line, comprising: The data acquisition module is used to automatically collect feeding and individual weight data. The core data processing module is used to perform progressive calculations on the data collected by the data acquisition module to achieve inefficient individual identification, quality uniformity prediction, and economic benefit decision-making, respectively. The early warning output module is used to generate and output early warning information for inefficient foraging individuals based on the first-level calculation results of the core data processing module. The uniformity prediction module is used to quantitatively evaluate the distribution and uniformity of the final slaughter quality grade of the duck flock based on the second-layer calculation results of the core data processing module. The economic decision-making module is used to analyze the optimal timing for slaughtering and formulate differentiated pricing strategies based on the third-level calculation results of the core data processing module. The central interaction module is used to integrate and display the output results of the early warning output module, the uniformity prediction module and the economic decision-making module, and provide a human-computer interaction interface.

[0006] Furthermore, the data acquisition module includes: The feeding metering unit is used to accurately measure and record the total weight of feed for each feeding operation according to the breeding pen. The individual weighing unit is used to automatically weigh individual ducks and small batches of ducks at preset growth nodes or triggering conditions, and record their individual identification and weight data.

[0007] Furthermore, the core data processing module includes: The correlation analysis unit is used to perform the first-level calculation. The correlation analysis unit is connected to the data acquisition module and has a preset dynamic feed intake and weight gain correlation analysis algorithm. This algorithm establishes a short-term expected weight gain model for each duck based on its historical weight gain trend, current growth curve stage, and real-time total feed amount and flock density. By continuously comparing the actual weight gain data obtained by the individual weighing unit with the expected value of the short-term expected weight gain model, it identifies individuals whose actual weight gain is consistently significantly lower than the expected value for multiple consecutive periods, thereby generating a list of inefficient feed-consuming individuals.

[0008] Furthermore, the core data processing module also includes: The uniformity prediction unit is used to perform the second-level calculation. The uniformity prediction unit is connected to the correlation analysis unit. It has a preset population uniformity decay prediction model. The model uses individuals in the list of inefficient feeders output by the correlation analysis unit as perturbation factors. Based on their weight deviation value, number proportion and remaining growth cycle of the duck flock, it simulates and calculates the dynamic impact trajectory of their impact on the standard deviation of the weight of all ducks in the pen. Combined with the mapping relationship between uniformity and quality grade in historical slaughter data, it predicts the estimated quality grade distribution map and overall uniformity coefficient of the ducks in the pen on the planned slaughter date.

[0009] Furthermore, the core data processing module also includes: The decision optimization unit is used to perform the third-level calculation. The decision optimization unit is connected to the uniform prediction unit and has a preset multi-objective economic decision optimization algorithm. This algorithm is based on the estimated quality grade distribution map and overall uniformity coefficient output by the uniform prediction unit. It combines the remaining feeding days, the current feed conversion ratio, real-time feed costs and market graded purchase prices. With the core objective of maximizing expected net profit, it performs cost-benefit simulation calculations and comparisons at multiple potential slaughter time points, thereby outputting the recommended optimal economic slaughter date, the expected net profit on that date, and a differentiated pricing baseline based on the predicted quality grade.

[0010] Furthermore, the early warning output module is connected to the correlation analysis unit in the core data processing module, and is used to receive the list of inefficient feeders, and format the individuals in the list according to their breeding pen, individual identifier and the degree of weight gain deviation, to generate a visual alarm or structured data report containing clear early warning level and specific location information, and transmit it to the central interaction module for display.

[0011] Furthermore, the uniformity prediction module is connected to the uniformity prediction unit in the core data processing module, and is used to receive the estimated quality grade distribution map and the overall uniformity coefficient, and convert the data into an intuitive chart form. The chart includes a pie chart of the proportion of premium grade rate, first grade rate and second grade rate, and a uniformity trend curve reflecting the dispersion of weight distribution. At the same time, it supports users to dynamically query and refresh the prediction results based on different slaughter date assumptions.

[0012] Furthermore, the economic decision-making module is connected to the decision optimization unit in the core data processing module. It is used to receive the recommended optimal economic slaughter date, expected net profit, and differentiated pricing baseline, and generate a strategy proposal containing detailed economic analysis reports. The strategy proposal specifically shows the feed cost, estimated total revenue, net profit comparison, and suggested selling price range for ducks of different quality grades corresponding to different slaughter time points. If the system determines that continuing to raise ducks will lead to a decrease in expected net profit, the module will simultaneously generate and highlight a strong warning message suggesting early slaughter to stop losses.

[0013] Furthermore, the central interaction module integrates a data cockpit interface, which dynamically displays real-time early warning panels from the early warning output module, quality prediction charts from the uniformity prediction module, and economic decision reports from the economic decision module. It also provides a parameter configuration entry that allows users to adjust key thresholds and model parameters of various algorithms in the core data processing module, while supporting query, export, and retrospective analysis of all historical data and decision records.

[0014] Furthermore, the data acquisition module, core data processing module, early warning output module, uniformity prediction module, economic decision-making module, and central interaction module realize the interaction of instructions and data through the internal data bus of the system. The calculation process of the core data processing module strictly follows the progressive relationship of the data flow, that is, the output of the correlation analysis unit serves as the only input of the uniformity prediction unit, and the output of the uniformity prediction unit serves as the only input of the decision optimization unit, ensuring the chain transformation from basic data to management decisions.

[0015] This invention provides an automatic grading, weighing, and conveying system for production lines. It offers the following advantages: (I) The automatic grading, weighing, and conveying system on this production line, through the first-level data processing, utilizes the system's conventionally collected feeding amount and individual time-series weight data to run a dynamic feed intake and weight gain correlation analysis algorithm. This establishes a short-term expected weight gain model for each duck based on multiple growth environmental factors. By continuously comparing the deviation trend between the actual weight gain and the model's expected value, it achieves early, accurate, and automated identification and warning of hidden inefficient feed-consuming individuals or potentially sub-healthy groups in mixed-species duck farming. This effect allows farm managers to identify individuals that appear to be feeding normally but have low feed conversion efficiency or hidden health problems several weeks before the ducks are ready for market, without relying on additional dedicated health monitoring sensors or frequent manual observation. This refines the management granularity from the traditional pen group to specific individuals, providing precise targets for implementing targeted isolation, diagnosis, or nutritional intervention measures. It fundamentally changes the passive model of health management based solely on the overall feed conversion ratio and post-hoc observation, effectively curbing the continuous drag of inefficient individuals on overall feed efficiency and the risk of potential disease spread, and achieving predictability and proactivity in the management of the farming process.

[0016] (II) The automatic grading, weighing, and conveying system on this production line, based on the successful identification of inefficient individuals, uses the list data representing individual anomalies as the core input through a second layer of data processing. It runs a population evenness decay prediction model to quantitatively assess the dynamic impact trajectory of these disturbance factors on the overall weight distribution dispersion of the duck flock during the remaining growth cycle. With the help of historical data mapping relationships, it finally outputs a quantitative prediction of the distribution and evenness coefficient of duck quality grades at future slaughter. This effect enables the system to scientifically extend and transform physiological abnormality signals at the individual level into a forward-looking prediction of the final commercial quality and consistency of the entire batch of products. As a result, managers can predict the expected premium grade rate, first grade rate, and other key market value indicators of the batch of ducks several weeks in advance during the critical fattening stage before slaughter. This allows the management perspective to shift from focusing on the growth process itself to intervening in the market value assessment of the output results in advance. It provides crucial, data-driven intermediate decision-making basis for whether to adjust the feeding strategy, whether to implement group management, and how to plan sales channels. This breaks the lag and uncontrollability of the traditional method of only knowing the product quality results at slaughter and grading.

[0017] (III) The automatic grading, weighing, and conveying system on this production line further utilizes a third layer of data processing. Based on the accurate prediction of quality uniformity, it runs a multi-objective economic decision optimization algorithm. By dynamically integrating real-time feed costs and market purchase prices, it simulates and calculates costs, revenues, and net profits at different slaughter times. This intelligently recommends the optimal economic slaughter time and generates differentiated pricing strategy suggestions. The final effect completes the full-chain automation from production data perception to business management decision-making. The system is no longer just a recording and monitoring tool, but has been transformed into a decision support system that can directly provide optimal economic benefit solutions. It enables farms to scientifically quantify and compare the economic trade-offs between continuing to raise and increase weight gain versus early slaughter to save costs and lock in the value of existing quality grades when facing complex situations such as unsatisfactory population uniformity, market price fluctuations, or changes in feed costs. This allows them to make slaughter decisions that maximize net profit. At the same time, the grading and pricing baseline provides accurate data support for sales negotiations, fundamentally improving the refined management level and market risk response capabilities of breeding enterprises, and realizing a leap from experience-driven breeding to data-driven intelligent operation. Attached Figure Description

[0018] Figure 1 This is the overall system flowchart of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] First embodiment: as follows Figure 1 As shown, the present invention provides a technical solution: an automatic grading, weighing, and conveying system for a production line, comprising: The data acquisition module is used to automatically collect feeding and individual weight data. The core data processing module is used to perform progressive calculations on the data collected by the data acquisition module to achieve inefficient individual identification, quality uniformity prediction, and economic benefit decision-making, respectively. The early warning output module is used to generate and output early warning information for inefficient foraging individuals based on the first-level calculation results of the core data processing module. The uniformity prediction module is used to quantitatively evaluate the distribution and uniformity of the final slaughter quality grade of the duck flock based on the second-level calculation results of the core data processing module. The economic decision-making module is used to analyze the optimal timing for slaughtering and formulate differentiated pricing strategies based on the third-level calculation results of the core data processing module. The central interaction module is used to integrate and display the output results of the early warning output module, the uniformity prediction module, and the economic decision-making module, and to provide a human-computer interaction interface.

[0021] The data acquisition module includes: The feeding metering unit is used to accurately measure and record the total weight of feed for each feeding operation according to the breeding pen. The individual weighing unit is used to automatically weigh individual ducks and small batches of ducks at preset growth nodes or triggering conditions, and record their individual identification and weight data.

[0022] The core data processing module includes: The correlation analysis unit is used to perform the first-level calculation. The correlation analysis unit is connected to the data acquisition module. It has a preset dynamic feed intake and weight gain correlation analysis algorithm. This algorithm establishes a short-term expected weight gain model for each duck based on its historical weight gain trend, current growth curve stage, and real-time total feed amount and flock density. By continuously comparing the actual weight gain data obtained from the individual weighing unit with the expected value of the short-term expected weight gain model, it identifies individuals whose actual weight gain is consistently significantly lower than the expected value for multiple consecutive periods, thereby generating a list of inefficient feed-consuming individuals.

[0023] The core data processing module also includes: The uniformity prediction unit is used to perform the second-level calculation. The uniformity prediction unit is connected to the correlation analysis unit. It has a preset population uniformity decay prediction model. This model uses individuals in the list of inefficient feeders output by the correlation analysis unit as perturbation factors. Based on their weight deviation value, number proportion and remaining growth cycle of the duck flock, it simulates and calculates the dynamic impact trajectory of their impact on the standard deviation of the weight of all ducks in the pen. Combined with the mapping relationship between uniformity and quality grade in historical slaughter data, it predicts the estimated quality grade distribution map and overall uniformity coefficient of the ducks in the pen on the planned slaughter date.

[0024] The core data processing module also includes: The decision optimization unit is used to perform the third-level calculation. It is connected to the uniform prediction unit and has a pre-set multi-objective economic decision optimization algorithm. Based on the estimated quality grade distribution map and overall uniformity coefficient output by the uniform prediction unit, the algorithm combines the remaining feeding days, current feed conversion ratio, real-time feed cost and market graded purchase price. With the core objective of maximizing expected net profit, it performs cost-benefit simulation calculations and comparisons at multiple potential slaughter time points, thereby outputting the recommended optimal economic slaughter date, the expected net profit on that date, and a differentiated pricing baseline based on the predicted quality grade.

[0025] The early warning output module is connected to the correlation analysis unit in the core data processing module. It is used to receive a list of individuals with inefficient feeding, and to format the individuals in the list according to their breeding pen, individual identifier and the degree of deviation in weight gain. It generates a visual alarm or structured data report containing clear early warning level and specific location information, and transmits it to the central interactive module for display.

[0026] During operation, the data acquisition module first starts working. Its feeding and metering unit automatically and accurately measures and records the total weight of the feed each time it is fed to a specific breeding pen. At the same time, the individual weighing unit weighs each duck individually and binds its unique identification to a pre-set key node in the duck's growth (e.g., weekly) or when triggered by the system, through the automatic weighing and identification station integrated on the production line, thereby continuously acquiring the time-series weight data of each duck. Next, these raw feeding amounts and individual weight data are transmitted to the correlation analysis unit of the core data processing module. This unit runs a dynamic feed intake-weight gain correlation analysis algorithm, which dynamically establishes a short-term expected weight gain model for each duck. This model comprehensively considers the individual's weight gain history of the previous week, its current weight stage in the standard growth curve, the total feed amount of its pen on that day, and real-time duck density information. Then, the algorithm compares and analyzes the latest actual weight gain data collected by the individual weighing unit with the predicted value of the individual's short-term expected weight gain model in real time. When the system identifies that a duck's actual weight gain is more than 15% lower than its model's expected value for three or more consecutive short-term cycles (each cycle is 5 days), and the feeding record of its pen shows that the feed is sufficient, the individual is marked as an invisible inefficient feeder and a list containing its identifier, pen, and degree of deviation is generated. Finally, the list is immediately sent to the early warning output module, which formats the list information into high-priority visual alarms and structured reports, and displays them in real time through a pop-up window on the data cockpit interface of the central interactive module, notifying the breeding managers of the specific column and individual number so that isolation inspection or medical intervention can be carried out in a timely manner. By performing in-depth correlation analysis on the routinely collected feeding and weighing data, it is possible to accurately identify individuals that appear to be feeding but have low conversion efficiency or potential health risks before slaughter without additional sensors. This achieves a breakthrough from population average data to individual anomaly identification, providing a direct basis for refined health management.

[0027] Second embodiment: as follows Figure 1 As shown, the uniformity prediction module is connected to the uniformity prediction unit in the core data processing module. It is used to receive the estimated quality grade distribution map and the overall uniformity coefficient, and convert the data into an intuitive chart. The chart includes a pie chart of the proportion of premium grade, first grade and second grade, as well as a uniformity trend curve reflecting the dispersion of weight distribution. It also supports users to dynamically query and refresh the prediction results based on different slaughter date assumptions.

[0028] During operation, calculations are further performed based on the list of inefficient feeders generated in Example 1. First, the uniformity prediction unit of the core data processing module receives the list from the correlation analysis unit and defines all inefficient individuals in the list as perturbation factors affecting the evenness of the flock. Then, the uniformity prediction unit starts the flock evenness decay prediction model. This model first calculates the initial influence weight of each inefficient individual in the list on the dispersion of the overall weight distribution in the pen based on the weight deviation value and the proportion of each individual in the pen. Then, combined with the remaining expected growth days of the ducks in the pen, it dynamically simulates the evolution trajectory of these inefficient individuals on the standard deviation of the weight of the entire pen of ducks due to their continuous abnormal growth during the remaining growth cycle. Then, the model calls the historical database, which stores the correlation between the weight evenness of specific days before slaughter and the distribution of the official grade quality (special grade, first grade, second grade) at the final slaughter in multiple past breeding batches. The mapping relationship is used to transform the predicted weight uniformity coefficient on the planned slaughter date into a specific quality grade prediction distribution map. For example, it outputs a quantitative prediction that premium grade ducks account for 35%, first grade ducks account for 50%, and second grade ducks account for 15%. Finally, this prediction result is transmitted to the uniformity prediction module. This module generates the predicted quality grade distribution map in the form of a pie chart and a bar chart, along with the uniformity change trend curve, and pushes it to a special panel of the central interactive module for dynamic display. Managers can intuitively see the expected output quality of the ducks in the current pen, effectively extending the identification of individual anomalies to the quantitative prediction of the commercial quality of the group. This allows managers to predict the grade composition and market value of the batch of products long before the ducks are slaughtered, thus shifting production management from focusing on the growth process to predicting the outcome, providing a crucial intermediate data link for subsequent economic decisions.

[0029] Third embodiment: as follows Figure 1 As shown, the economic decision-making module is connected to the decision optimization unit in the core data processing module. It is used to receive the recommended optimal economic slaughter date, expected net profit, and differentiated pricing baseline, and generate a strategy proposal containing detailed economic analysis reports. The strategy proposal specifically shows the feed cost, estimated total revenue, net profit comparison, and suggested selling price range for ducks of different quality grades corresponding to different slaughter time points. If the system determines that continuing to raise ducks will lead to a decrease in expected net profit, this module will simultaneously generate and highlight a strong warning message suggesting early slaughter to stop losses.

[0030] The central interaction module integrates a data cockpit interface, which dynamically displays real-time early warning panels from the early warning output module, quality prediction charts from the uniformity prediction module, and economic decision reports from the economic decision module. It also provides a parameter configuration entry that allows users to adjust key thresholds and model parameters of various algorithms in the core data processing module. At the same time, it supports querying, exporting, and retrospective analysis of all historical data and decision records.

[0031] The data acquisition module, core data processing module, early warning output module, uniformity prediction module, economic decision-making module, and central interaction module realize the interaction of instructions and data through the internal data bus of the system. The calculation process of the core data processing module strictly follows the progressive relationship of data flow, that is, the output of the correlation analysis unit serves as the sole input of the uniformity prediction unit, and the output of the uniformity prediction unit serves as the sole input of the decision optimization unit, ensuring the chain transformation from basic data to management decisions.

[0032] During operation, the system takes the estimated quality grade distribution map and overall uniformity coefficient output by the uniformity prediction module as direct input. First, the decision optimization unit of the core data processing module calls a multi-objective economic decision optimization algorithm. This algorithm integrates a real-time external data interface to obtain the current feed market price, the market purchase price table for different grades of duck meat products, and the fixed cost parameters of the farm. Next, the algorithm sets multiple assessment time points forward (e.g., 5 days, 10 days, 15 days earlier) based on the planned slaughter date, and performs simulation calculations for each time point: on the one hand, it calculates the total increase in feeding costs up to that time point based on the remaining feeding days, the current average feed conversion ratio, and the feed price; on the other hand, it calculates the expected sales revenue based on the quality grade distribution, which has been fine-tuned for the specific early slaughter time point, provided by the uniformity prediction unit, combined with the market purchase price. Then, the core algorithm compares the expected net profit at all simulated time points, and may use the return on investment indicator for cross-validation, automatically selecting the time point that maximizes net profit as the optimal economic slaughter date recommended by the system. Finally, the economic decision-making module receives the decision results and generates a detailed economic decision-making report. The report not only clearly lists the recommended slaughter date and its corresponding expected net profit, cost and revenue details, but also calculates a comprehensive benchmark price for the entire flock of ducks based on the predicted quality distribution on that date, and provides a competitive suggested price fluctuation range for different grades such as premium and first-grade ducks. If the simulation calculation shows that the net profit from slaughtering according to the original plan will be lower than that from slaughtering earlier, the system will generate and highlight a strong warning in this report: "It is recommended to slaughter earlier on [specific date] to stop losses and increase efficiency." All information is summarized in the central interactive module for decision-makers to review, completing the final closed loop from production data to business decisions. Through dynamic economic model simulation, the two key business decisions of when to sell and how to price are quantified and made scientific, directly serving the goal of maximizing breeding efficiency. Especially when facing market price fluctuations and poor uniformity within the flock, it can provide precise guidance for loss-stopping or efficiency-increasing operations, realizing intelligent breeding management.

[0033] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0034] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An automatic grading, weighing, and conveying system for a production line, characterized in that, include: The data acquisition module is used to automatically collect feeding and individual weight data. The core data processing module is used to perform progressive calculations on the data collected by the data acquisition module to achieve inefficient individual identification, quality uniformity prediction, and economic benefit decision-making, respectively. The early warning output module is used to generate and output early warning information for inefficient foraging individuals based on the first-level calculation results of the core data processing module. The uniformity prediction module is used to quantitatively evaluate the distribution and uniformity of the final slaughter quality grade of the duck flock based on the second-layer calculation results of the core data processing module. The economic decision-making module is used to analyze the optimal timing for slaughtering and formulate differentiated pricing strategies based on the third-level calculation results of the core data processing module. The central interaction module is used to integrate and display the output results of the early warning output module, the uniformity prediction module and the economic decision-making module, and provide a human-computer interaction interface.

2. The automatic grading, weighing, and conveying system for a production line according to claim 1, characterized in that, The data acquisition module includes: The feeding metering unit is used to accurately measure and record the total weight of feed for each feeding operation according to the breeding pen. The individual weighing unit is used to automatically weigh individual ducks and small batches of ducks at preset growth nodes or triggering conditions, and record their individual identification and weight data.

3. The automatic grading, weighing, and conveying system for a production line according to claim 1, characterized in that, The core data processing module includes: The correlation analysis unit is used to perform the first-level calculation. The correlation analysis unit is connected to the data acquisition module and has a preset dynamic feed intake and weight gain correlation analysis algorithm. This algorithm establishes a short-term expected weight gain model for each duck based on its historical weight gain trend, current growth curve stage, and real-time total feed amount and flock density. By continuously comparing the actual weight gain data obtained by the individual weighing unit with the expected value of the short-term expected weight gain model, it identifies individuals whose actual weight gain is consistently significantly lower than the expected value for multiple consecutive periods, thereby generating a list of inefficient feed-consuming individuals.

4. The automatic grading, weighing, and conveying system for a production line according to claim 3, characterized in that, The core data processing module also includes: The uniformity prediction unit is used to perform the second-level calculation. The uniformity prediction unit is connected to the correlation analysis unit. It has a preset population uniformity decay prediction model. The model uses individuals in the list of inefficient feeders output by the correlation analysis unit as perturbation factors. Based on their weight deviation value, number proportion and remaining growth cycle of the duck flock, it simulates and calculates the dynamic impact trajectory of their impact on the standard deviation of the weight of all ducks in the pen. Combined with the mapping relationship between uniformity and quality grade in historical slaughter data, it predicts the estimated quality grade distribution map and overall uniformity coefficient of the ducks in the pen on the planned slaughter date.

5. An automatic grading, weighing, and conveying system for a production line according to claim 4, characterized in that, The core data processing module also includes: The decision optimization unit is used to perform the third-level calculation. The decision optimization unit is connected to the uniform prediction unit and has a preset multi-objective economic decision optimization algorithm. This algorithm is based on the estimated quality grade distribution map and overall uniformity coefficient output by the uniform prediction unit. It combines the remaining feeding days, the current feed conversion ratio, real-time feed costs and market graded purchase prices. With the core objective of maximizing expected net profit, it performs cost-benefit simulation calculations and comparisons at multiple potential slaughter time points, thereby outputting the recommended optimal economic slaughter date, the expected net profit on that date, and a differentiated pricing baseline based on the predicted quality grade.

6. The automatic grading, weighing, and conveying system for a production line according to claim 1, characterized in that, The early warning output module is connected to the correlation analysis unit in the core data processing module. It is used to receive the list of inefficient feeders, format the individuals in the list according to their breeding pen, individual identifier and the degree of weight gain deviation, generate a visual alarm or structured data report containing clear early warning level and specific location information, and transmit it to the central interaction module for display.

7. The automatic grading, weighing, and conveying system for a production line according to claim 1, characterized in that, The uniformity prediction module is connected to the uniformity prediction unit in the core data processing module. It is used to receive the estimated quality grade distribution map and the overall uniformity coefficient, and convert the data into an intuitive chart. The chart includes a pie chart showing the proportion of premium grade, first grade, and second grade, as well as a uniformity trend curve reflecting the dispersion of weight distribution. It also supports users to dynamically query and refresh the prediction results based on different slaughter date assumptions.

8. An automatic grading, weighing, and conveying system for a production line according to claim 1, characterized in that, The economic decision-making module is connected to the decision optimization unit in the core data processing module. It is used to receive the recommended optimal economic slaughter date, expected net profit, and differentiated pricing baseline, and generate a strategy proposal containing detailed economic analysis reports. The strategy proposal specifically shows the feed cost, estimated total revenue, net profit comparison, and suggested selling price range for ducks of different quality grades corresponding to different slaughter time points. If the system determines that continuing to raise ducks will lead to a decrease in expected net profit, the module will simultaneously generate and highlight a strong warning message suggesting early slaughter to stop losses.

9. An automatic grading, weighing, and conveying system for a production line according to claim 1, characterized in that, The central interaction module integrates a data cockpit interface, which dynamically displays real-time early warning panels from the early warning output module, quality prediction charts from the uniformity prediction module, and economic decision reports from the economic decision module. It also provides a parameter configuration entry that allows users to adjust key thresholds and model parameters of various algorithms in the core data processing module. At the same time, it supports querying, exporting, and retrospective analysis of all historical data and decision records.

10. An automatic grading, weighing, and conveying system for a production line according to any one of claims 1 to 9, characterized in that, The data acquisition module, core data processing module, early warning output module, uniformity prediction module, economic decision-making module, and central interaction module realize the interaction of instructions and data through the internal data bus of the system. The calculation process of the core data processing module strictly follows the progressive relationship of the data flow, that is, the output of the correlation analysis unit serves as the only input of the uniformity prediction unit, and the output of the uniformity prediction unit serves as the only input of the decision optimization unit.

Images