A duck feed formula optimization method and system

Abstract

The application provides a duck feed formula optimization method and system, relates to the technical field of feed formula optimization, and has the technical scheme as follows: acquiring nutrition component detection data of feed, the nutrition component detection data including first batch history detection data and second batch history detection data; obtaining a first detection confidence according to the first batch history detection data, adjusting the first detection confidence according to the second batch history detection data, and obtaining a second detection confidence; in the case of feed formula preparation, determining a first adjustment coefficient based on the second detection confidence, and adjusting the nutrition component content and cost of the feed by the first adjustment coefficient to realize feed formula optimization. The application aims to accurately evaluate the authenticity of the nutrition components of the feed, so that more scientific and economic feed formula optimization is realized, and hidden nutritional deficiency and cost waste are avoided.

Description

Technical Field

[0001] This application relates to the field of feed formulation optimization technology, and more specifically, to a method and system for optimizing duck feed formulation. Background Technology

[0002] In modern duck farming, feed formulation optimization typically achieves a balance between nutrition and cost based on pre-set supplier nutrient composition tables and raw material prices. However, the key nutrient content of feed ingredients may fluctuate slightly within the contractually permissible range, for example, consistently trending towards the minimum value, leading to persistent nutritional deficiencies. Farms often have limited precision and insufficient sampling frequency in their raw material testing equipment, making it difficult to detect small, persistently low trends within the acceptable range. This results in outdated databases and overestimation of nutrient values. In pursuing the mathematical goal of lowest cost, formulation programs tend to select ingredient combinations that appear low in cost but are actually overestimated in nutrition, further exacerbating nutrient deficiencies, leading to decreased feed utilization efficiency and hidden increases in farming costs. Summary of the Invention

[0003] The purpose of this application is to provide a method and system for optimizing duck feed formulation, which aims to accurately assess the authenticity of feed nutrient components, thereby achieving more scientific and economical feed formulation optimization and avoiding hidden nutritional deficiencies and cost waste.

[0004] Firstly, this application provides a method for optimizing duck feed formulation, including: Obtain nutritional component testing data of feed, wherein the nutritional component testing data includes the first batch of historical testing data and the second batch of historical testing data, wherein the first batch of historical testing data has a larger batch number than the second batch of historical testing data and the first batch of historical testing data includes the second batch of historical testing data. A first detection confidence level is obtained based on the first batch of historical detection data, and a second detection confidence level is obtained by adjusting the first detection confidence level based on the second batch of historical detection data. When formulating feed formulations, a first adjustment coefficient is determined based on the second detection confidence level. The nutrient content and cost of the feed are adjusted using the first adjustment coefficient to optimize the feed formulation.

[0005] Furthermore, the method for optimizing duck feed formulation also includes: The growth performance parameters of the duck flock are monitored. If the changes in the growth performance parameters match the preset changes, and it is determined based on the second detection confidence level that the changes in the growth performance parameters are related to the nutrient content of the feed, the deficiency of feed quantity or the deficiency of nutrient content in the feed is determined based on the second detection confidence level, and a corresponding formula adjustment plan is generated based on the deficiency of feed quantity or the deficiency of nutrient content in the feed.

[0006] Further, the step of obtaining a first detection confidence level based on the first batch of historical detection data, and adjusting the first detection confidence level based on the second batch of historical detection data to obtain a second detection confidence level, includes: Obtain the nutritional component standard data corresponding to the first batch of historical test data, identify the difference between the first batch of historical test data and the nutritional component standard data, and obtain the first test confidence level; The trend of the second batch of data changes is determined based on the historical test data of the second batch. The first detection confidence level is adjusted based on the trend of the second batch of data changes and the preset trend change threshold to obtain the second detection confidence level.

[0007] Furthermore, the method for optimizing duck feed formulation also includes: Obtain current nutrient composition test data and feed moisture test data, and obtain the first dry matter nutrient content based on the feed moisture test data and current nutrient composition test data; Obtain the historical dry matter nutrient content fluctuation range. If the first dry matter nutrient content exceeds the historical dry matter nutrient content fluctuation range, adjust the first dry matter nutrient content to obtain the second dry matter nutrient content. The second dry matter nutrient content is converted into nutrient detection data and stored in a preset feed database so that, in the case of feed formulation, a second detection confidence level can be determined based on the nutrient detection data and the feed formulation can be optimized.

[0008] Further, the step of acquiring current nutrient composition detection data and feed moisture detection data, and obtaining the first dry matter nutrient content based on the feed moisture detection data and current nutrient composition detection data, includes: Acquire current nutrient composition detection data, environmental humidity data, and feed moisture detection data; determine the moisture detection confidence level of the feed moisture detection data based on the environmental humidity data, the feed moisture detection data, and a preset moisture range. The feed moisture detection data is adjusted based on the moisture detection confidence level to obtain corrected feed moisture detection data. The current nutrient component detection data is then adjusted based on the corrected feed moisture detection data and the environmental humidity data to obtain corrected nutrient component detection data. The first dry matter nutrient content is obtained based on the corrected feed moisture detection data and the corrected nutrient component detection data.

[0009] Furthermore, in the process of formulating feed formulation, determining a first adjustment coefficient based on the second detection confidence level, and adjusting the nutrient content and cost of the feed through the first adjustment coefficient to achieve feed formulation optimization includes: When the current feed is supplied by a newly introduced feed supplier, a new supplier confidence level is determined based on the current supply batches of the newly introduced feed supplier to assess the reliability of the second detection confidence level. The newly introduced feed supplier is a feed supplier whose supply batches are within a preset batch quantity. The second detection confidence level corresponds to the nutrient composition detection data of the current feed supplied by the newly introduced feed supplier. When formulating feed formulations, a first adjustment coefficient and a second adjustment coefficient are determined based on the second detection confidence level and the new supplier confidence level, respectively. The nutrient content and cost of the feed are adjusted using the first adjustment coefficient and the second adjustment coefficient to optimize the feed formulation.

[0010] Further, determining the new supplier confidence level for assessing the reliability of the second test confidence level based on the current supply batches of the newly introduced feed supplier includes: If the current supply batch is within the first batch range, the confidence level of the new supplier is determined to be the preset initial confidence level of the new supplier; When the current supply batch is within the second batch range, the current confidence level is obtained. Based on the current supply batch, the second detection confidence level of the second range of historical data is extracted as the new supplier confidence level assessment data. When the average deviation determined based on the new supplier confidence level assessment data meets the preset average range condition, the current confidence level is adjusted based on the preset second range adjustment amount to obtain the new supplier confidence level. When the current supply batch is within the range of the third batch, the current confidence level is obtained. Based on the current supply batch, the second detection confidence level of the number of historical third-range quantities is extracted as the new supplier confidence level assessment data. If the standard deviation determined based on the new supplier confidence level assessment data meets the preset standard deviation range condition, the current confidence level is adjusted based on the preset third-range adjustment amount to obtain the new supplier confidence level. Wherein, the range of the third batch is greater than the range of the second batch, the range of the second batch is greater than the range of the first batch, and the adjustment amount of the third range is less than the adjustment amount of the second range.

[0011] Furthermore, the method for optimizing duck feed formulation also includes: The growth performance parameters of the duck flock are monitored. If the changes in the growth performance parameters match the preset changes and the confidence level of the new supplier is lower than the preset confidence level threshold, the suppliers of different feed components and the corresponding supply ratio of each supplier are determined based on the second detection confidence level and the corresponding new supplier confidence level. A corresponding formula adjustment plan is generated according to the suppliers of different feed components and the corresponding supply ratio of each supplier.

[0012] Furthermore, the first batch of historical test data includes multiple sets of first batch historical test data provided by different time periods, different test methods, or different test institutions; The step of obtaining a first detection confidence level based on the first batch of historical detection data, and adjusting the first detection confidence level based on the second batch of historical detection data to obtain a second detection confidence level includes: Identify systematic biases in the multiple sets of first-batch historical test data, including persistent differences in the test results of the same nutrient components between different data sources; Based on the systematic bias, the multiple sets of first-batch historical detection data are normalized so that detection data from different data sources can be compared on a unified benchmark. The first detection confidence level is obtained based on the normalized batch of historical detection data from the first batch. The second detection confidence level is then adjusted based on the second batch of historical detection data to obtain the second detection confidence level.

[0013] Secondly, embodiments of this application provide a duck feed formulation optimization system, comprising: The data acquisition module is used to acquire the nutritional component testing data of the feed. The nutritional component testing data includes the first batch of historical testing data and the second batch of historical testing data. The first batch of historical testing data has a larger batch number than the second batch of historical testing data, and the first batch of historical testing data includes the second batch of historical testing data. The confidence module is used to obtain a first detection confidence based on the first batch of historical detection data, and to adjust the first detection confidence based on the second batch of historical detection data to obtain a second detection confidence. The optimization module is used to determine a first adjustment coefficient based on the second detection confidence level when formulating feed formulation, and to adjust the nutrient content and cost of the feed through the first adjustment coefficient to achieve feed formulation optimization.

[0014] As can be seen from the above, the duck feed formulation optimization method of this application obtains multiple batches of historical test data and uses this data to obtain a first test confidence level. Then, it adjusts the first test confidence level by combining it with a second batch of historical test data to obtain a more reliable second test confidence level. When formulating the feed formulation, a first adjustment coefficient is determined based on the second test confidence level, and the nutrient content and cost of the feed are adjusted through the adjustment coefficient, thereby achieving feed formulation optimization. This method can effectively solve the problems of inaccurate formulation optimization, decreased feed utilization efficiency, and hidden increases in breeding costs caused by slight fluctuations in the nutrient content of feed raw materials. Through the confidence level assessment mechanism, the true situation of feed nutrient content can be reflected more accurately, avoiding overestimation of nutrient content. This makes the formulation optimization more scientific and reasonable, ensuring that the duck flock receives sufficient nutrition while effectively controlling feed costs and achieving a balance between nutritional needs and cost control. Attached Figure Description

[0015] Figure 1 This is a flowchart of a duck feed formulation optimization method provided in one embodiment of this application; Figure 2 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application; Figure 3 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application; Figure 4 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application; Figure 5 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application; Figure 6 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application; Figure 7 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application; Figure 8 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application; Figure 9 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Detailed Implementation

[0016] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0017] like Figure 1 As shown, Figure 1 This is a flowchart of a duck feed formulation optimization method provided in one embodiment of this application. The method may include, but is not limited to, steps S110 to S130.

[0018] Step S110: Obtain nutritional component testing data of the feed. The nutritional component testing data includes the first batch of historical testing data and the second batch of historical testing data. The batch number of the first batch of historical testing data is greater than that of the second batch of historical testing data, and the first batch of historical testing data includes the second batch of historical testing data. Step S120: Obtain the first detection confidence level based on the first batch of historical detection data, and adjust the first detection confidence level based on the second batch of historical detection data to obtain the second detection confidence level; Step S130: When formulating a feed formulation, determine a first adjustment coefficient based on a second detection confidence level, and adjust the nutrient content and cost of the feed through the first adjustment coefficient to achieve feed formulation optimization.

[0019] It is understandable that duck feed formulation optimization systems in related technologies have some limitations in practical applications. The actual content of key nutrients in feed ingredients provided by feed suppliers may fluctuate slightly. While these fluctuations may be within the limits allowed by the purchase contract, a persistent bias towards the lowest values ​​can lead to hidden nutritional deficiencies. Simultaneously, the raw material testing equipment and rapid testing procedures available at farms often lack the precision and sampling frequency to capture these persistent, small-scale trends of low nutrient levels within acceptable limits. This results in the database recording raw material nutrient information not being updated promptly and accurately. Consequently, when optimizing feed formulations, the calculation program may rely on overestimated nutrient values, leading to insufficient actual nutrient intake in the duck flock and potentially increasing farming costs. The embodiments of this application, by acquiring multiple batches of historical testing data and adjusting the confidence level, can more accurately assess the actual content of feed nutrients. This allows for optimization of feed formulations based on more reliable nutrient information, effectively avoiding nutritional deficiencies and hidden cost increases caused by overestimation of nutrient content.

[0020] For example, nutritional component testing data of feed can be obtained through various testing methods, which can be information on the content of various nutrients (protein, fat, vitamins, minerals, etc.) in the feed. Nutritional component testing data is the basis for feed formulation optimization. Nutritional component testing data includes the first batch of historical testing data and the second batch of historical testing data. The first batch of historical testing data and the second batch of historical testing data are sets of historical testing data with different batch sizes. The first batch of historical testing data has a larger batch size and includes the second batch of historical testing data, indicating that the first batch of data provides more macroscopic and long-term historical information, while the second batch of data represents the test results under recent or specific conditions, which is used to refine the confidence level established by the first batch of data.

[0021] The first batch of historical testing data can be testing data accumulated over a long period of time and in a large number of batches, such as nutritional component testing reports for all batches of feed within the past year; the second batch of historical testing data can be testing data obtained recently in a relatively small number of batches, such as nutritional component testing reports within the most recent month.

[0022] For example, a first confidence level can be obtained based on the first batch of historical test data. Statistical analysis can be performed on this data to calculate the average value, standard deviation, and other statistical parameters for each nutrient content. Combined with preset nutrient standard ranges, the reliability of the test results can be preliminarily assessed, thus obtaining the first confidence level. Subsequently, the first confidence level is adjusted based on the second batch of historical test data to obtain a second confidence level. For instance, if the second batch of historical test data shows a continuous deviation or fluctuation trend in certain nutrient components compared to the first batch, the first confidence level can be corrected based on this trend to better reflect the current situation, thus obtaining the second confidence level.

[0023] When formulating feed formulations, a first adjustment coefficient is determined based on a second detection confidence level. This first adjustment coefficient is then used to adjust the nutrient content and cost of the feed to optimize the feed formulation. In other words, when a new feed formulation needs to be developed, the current nutrient content testing data and corresponding second detection confidence levels can be referenced. If the second detection confidence level is low, it indicates that the current testing data may be biased. A first adjustment coefficient can be determined based on the second detection confidence level using a preset algorithm. This first adjustment coefficient can be a multiplier or a modifier, used to correct the content of specific nutrients in the feed or adjust the cost of feed ingredients. For example, if the detection confidence level of a certain nutrient is low, the corresponding content can be adjusted downwards to ensure that the actual nutrient intake of the ducks is not lower than expected. Simultaneously, the cost of related raw materials can also be adjusted to reflect their true value.

[0024] It is understood that the embodiments of this application obtain the first and second batches of historical test data. The first batch of historical test data, as macro data accumulated over a long period of time, provides a basis for the preliminary assessment of the reliability of feed nutrient composition. For example, test data of protein, energy and other nutrients of major raw materials such as corn and soybean meal provided by all suppliers in the past year can be collected, and statistical characteristics such as the average value and fluctuation range of each nutrient can be calculated. Combined with industry standards or the nutritional needs of duck flocks, a preliminary first test confidence level can be obtained to reflect the overall degree of trust in the test results of feed nutrient composition in the context of long-term and large-scale data.

[0025] However, relying solely on long-term data may not be able to capture subtle changes in the recent period or specific suppliers. Therefore, by using a second batch of historical testing data, for example, collecting protein content testing data of a specific batch of soybean meal provided by a particular supplier within the past month, and comparing and analyzing the second batch of historical testing data with the first testing confidence level, the first testing confidence level can be finely adjusted to obtain the second testing confidence level. This allows for timely reflection of the latest trends in feed nutrient composition, making the confidence level assessment more dynamic and accurate.

[0026] When formulating feed formulations, a first adjustment coefficient can be determined based on the second detection confidence level. This first adjustment coefficient comprehensively considers the reliability of the detection data. For example, if the detection confidence level of a certain ingredient's protein content is low, its actual contribution to the formulation can be reduced based on the first adjustment coefficient, or a risk premium can be added during cost calculation. Thus, even with slight deviations in the detection data, the formulation optimization process can effectively avoid nutritional deficiencies in ducks due to overestimation of nutrient content. Furthermore, adjusting the nutrient content and cost of the feed through the first adjustment coefficient enables more precise feed formulation optimization, ensuring that ducks receive sufficient nutrition and effectively controlling feed costs.

[0027] like Figure 2 As shown, Figure 2 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding the above method, the method may also include, but is not limited to, step S210.

[0028] Step S210: Monitor the growth performance parameters of the duck flock. If the changes in the growth performance parameters match the preset changes, and the changes in the growth performance parameters are determined to be related to the nutrient content of the feed based on the second detection confidence level, determine the amount of feed deficiency or the amount of feed nutrient deficiency based on the second detection confidence level, and generate a corresponding formula adjustment plan based on the amount of feed deficiency or the amount of feed nutrient deficiency.

[0029] Understandably, duck feed formulation optimization methods in related technologies usually focus on adjusting the formulation based on the nutritional component test data of the feed to achieve theoretical nutritional balance and cost optimization. However, they do not fully consider the actual physiological response and growth performance of the duck flock. Therefore, even if the feed formulation seems reasonable at the data level, if the actual growth condition of the duck flock is poor, it may mean that the current formulation has failed to truly meet its growth needs, thereby affecting the breeding efficiency.

[0030] Therefore, growth performance parameters of duck flocks can be monitored. These parameters can be parameters that continuously track and record various physiological indicators and production performance of duck flocks during a specific breeding cycle, including but not limited to the average weight, daily weight gain, feed conversion rate, survival rate, and health status of the duck flock.

[0031] Matching the changes in growth performance parameters with preset changes can be achieved by comparing the actual monitored growth performance parameters of the duck flock with preset target values, standard curves, or ideal trends. When the actual trend of the parameters deviates significantly from the preset situation, it is considered that the changes in growth performance parameters match the preset situation.

[0032] Subsequently, the correlation between changes in growth performance parameters and feed nutrient content is determined based on the second test confidence level. This can be achieved by using the obtained second test confidence level to assess the association between the abnormality and the feed nutrient content when abnormal growth performance is detected in duck flocks. For example, if the second test confidence level is high, it indicates that the reliability of the feed nutrient test data is high, and the abnormal growth performance is more likely to be directly attributed to the actual lack or imbalance of feed nutrients. If the second test confidence level is low, it may be necessary to further verify the accuracy of the feed test data or consider the possibility of other non-feed factors causing abnormal growth, in order to improve the accuracy of problem diagnosis and avoid blindly adjusting the formula.

[0033] Therefore, when the changes in growth performance parameters match the preset changes, and the changes in growth performance parameters are determined to be related to the nutrient content of the feed based on the second detection confidence level, the amount of feed deficiency or the amount of nutrient deficiency in the feed can be determined based on the second detection confidence level. By combining the actual growth needs of the duck flock, the current nutrient content of the feed, and the second detection confidence level, the total amount of feed required by the duck flock or the degree of deficiency of specific nutrients can be quantified. For example, based on the weight gain curve and feed conversion rate of the duck flock, the amount of additional feed required to achieve the target growth performance can be calculated, or the amount of supplementation of one or more nutrients such as protein, energy, and amino acids can be specified.

[0034] Based on the deficiencies in feed quantity or nutrient content, corresponding formula adjustment plans can be generated. Specific feed formula adjustment suggestions can be formulated based on the quantified deficiencies, including increasing the overall feed supply, adjusting the proportion of a certain ingredient in the existing feed, adding specific nutritional supplements, and replacing some feed ingredients.

[0035] Therefore, this application embodiment introduces the monitoring of duck flock growth performance parameters to construct a closed-loop feedback mechanism from feed input to duck flock output. This allows formula adjustment to no longer be based solely on theoretical data, but on actual biological feedback and data reliability assessment. This enables a more precise determination of whether the problem stems from insufficient total feed volume or a deficiency in specific nutrients, and allows for the generation of targeted adjustment plans, improving the accuracy and effectiveness of duck feed formula optimization. Furthermore, compared to methods that rely solely on feed component detection data for optimization, this application embodiment utilizes feedback from the actual growth performance of the duck flock, allowing formula adjustments to more directly respond to the flock's physiological needs. This avoids losses in farming efficiency due to a disconnect between theoretical optimization and actual results. Moreover, by combining a second detection confidence level, the cause of abnormal growth can be identified, thereby avoiding unnecessary or erroneous formula adjustments. This ensures that the duck flock receives the most suitable nutritional supply, promotes healthy growth, improves feed conversion rate, and ultimately achieves higher economic benefits and a more sustainable production model.

[0036] In some embodiments, a farm obtains a preliminary optimized feed formula based on historical and current testing data, and calculates the corresponding second detection confidence level, indicating that the reliability of the current feed nutrient composition testing data is high. During the duck farming process, by continuously monitoring the ducks' growth performance parameters, it was found that the average daily weight gain of the ducks was consistently lower than the preset ideal growth curve, meaning that the changes in growth performance parameters matched the preset changes. At this point, due to the high second detection confidence level, it can be determined that the current feed nutrient composition testing data is reliable; therefore, the abnormal growth performance of the ducks is highly correlated with the nutrient content of the feed. Based on this, combined with the actual feed intake and growth status of the ducks, it was calculated that the ducks were deficient in protein. Therefore, based on the protein deficiency, a formula adjustment plan was automatically generated, suggesting adding a certain proportion of soybean meal to the existing feed to increase the protein content of the feed to the target level. After the farm adjusted according to the plan, and after subsequent monitoring, the daily weight gain of the ducks gradually recovered to the ideal level, thus achieving dynamic optimization of the feed formula and healthy and efficient growth of the ducks.

[0037] like Figure 3 As shown, Figure 3 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding step S120 above, the method may also include, but is not limited to, steps S310 and S320.

[0038] Step S310: Obtain the nutritional component standard data corresponding to the first batch of historical test data, identify the difference between the first batch of historical test data and the nutritional component standard data, and obtain the first test confidence level. Step S320: Determine the data change trend of the second batch based on the historical detection data of the second batch, and adjust the first detection confidence level according to the data change trend of the second batch and the preset trend change threshold to obtain the second detection confidence level.

[0039] For example, obtaining the nutrient composition standard data corresponding to the first batch of historical test data can be achieved by retrieving the ideal or benchmark content of various nutrients corresponding to the first batch of historical test data from a pre-set database or industry standards, serving as a reference for assessing the accuracy of the actual test data. Then, by comparing the first batch of historical test data with the nutrient composition standard data, the degree of deviation between the two is quantified, resulting in the difference between the first batch of historical test data and the nutrient composition standard data. This difference can be calculated using absolute difference, relative difference, or deviation. The resulting first test confidence level can be a preliminary assessment of the reliability of the feed nutrient composition test results based on a comparison of a large amount of historical data with standard data, reflecting the degree of consistency between the test results and standard values ​​in the context of long-term, large-scale data.

[0040] Determining the trend of the second batch of data based on historical testing data can be achieved by analyzing recent small batches of historical testing data to identify the patterns or directions of changes in nutrient content over time. For example, statistical methods such as moving average, exponential smoothing, or regression analysis can be used to obtain the trend of the second batch of data.

[0041] Because the second batch of historical test data is relatively small and included in the first batch of historical test data, it can more sensitively reflect the fluctuations in recent test results. Therefore, the first test confidence level is adjusted based on the trend of the second batch of data and a preset trend change threshold. That is, when the identified trend of the second batch of data exceeds the preset trend change threshold, it indicates that there may be a systematic change or anomaly in the recent test results, and the initial assessment of the first test confidence level needs to be corrected. Here, the preset trend change threshold can be determined based on experience, the range of historical data fluctuations, or statistical methods, and is used to judge the significance of the trend change.

[0042] It is understood that the embodiments of this application use a first batch of historical test data with a larger batch size and a second batch of historical test data with a smaller batch size but more timely data. This allows the obtained second test confidence level to not only have the stability of long-term data, but also take into account the dynamic changes of recent data. This improves the accuracy and real-time performance of the reliability assessment of the current feed nutrient composition test results, thereby avoiding assessment bias caused by data lag, more accurately reflecting the true reliability of the current feed nutrient composition test, and improving the accuracy and effectiveness of feed formulation optimization.

[0043] like Figure 4 As shown, Figure 4 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding the above method, the method may also include, but is not limited to, steps S410 to S430.

[0044] Step S410: Obtain the current nutrient composition test data and feed moisture test data. Based on the feed moisture test data and the current nutrient composition test data, obtain the first dry matter nutrient content. Step S420: Obtain the historical dry matter nutrient content fluctuation range. If the first dry matter nutrient content exceeds the historical dry matter nutrient content fluctuation range, adjust the first dry matter nutrient content to obtain the second dry matter nutrient content. Step S430: Convert the second dry matter nutrient content into nutrient detection data and store it in a preset feed database so that, when formulating feed formulation, the second detection confidence level can be determined based on the nutrient detection data and the feed formulation can be optimized.

[0045] It is understandable that feed nutrient content testing data, especially for the current batch, is often affected by environmental factors such as moisture content, and may contain measurement errors or abnormal fluctuations. Directly using this data for formula optimization could lead to inaccurate nutrient content assessments, thus affecting the effectiveness of formula optimization and the growth performance of the duck flock. Therefore, it is necessary to obtain current nutrient content testing data and feed moisture testing data. The current nutrient content testing data can be the content data of each nutrient obtained from testing the current batch of feed, and the feed moisture testing data can be the data obtained from testing the moisture content of the current batch of feed. Since the moisture content in the feed affects the actual concentration of its nutrients, to obtain more accurate nutrient content, the nutrient content is converted to a dry matter basis based on the feed moisture testing data and the current nutrient content testing data to obtain the first dry matter nutrient content. The first dry matter nutrient content is the nutrient content calculated based on dry matter after removing the influence of moisture.

[0046] Obtaining the historical dry matter nutrient content fluctuation range is crucial. This range represents the normal fluctuation range of various nutrients on a dry matter basis, derived from long-term historical data analysis. It reflects the inherent variability of feed under normal production and testing conditions. When the first dry matter nutrient content exceeds the preset historical fluctuation range, it indicates potential anomalies in the current testing data, such as testing errors, batch quality abnormalities, or excessive environmental influences. In this case, the first dry matter nutrient content needs adjustment to eliminate the adverse effects of abnormal data on subsequent formulation optimization, thereby obtaining a more reliable second dry matter nutrient content. Adjustment methods may include, but are not limited to, correcting it to the boundary value of the fluctuation range, using historical averages or trend predictions as substitutes, and correction based on statistical models.

[0047] Subsequently, the adjusted second dry matter nutrient content is converted into nutrient composition testing data and stored in a pre-set feed database. This conversion can involve transforming the second dry matter nutrient content back into a conventional nutrient composition testing data format, based on actual application needs and in conjunction with standard or current moisture content, to ensure consistency with other data in the database. The processed and calibrated nutrient composition testing data will be used in subsequent feed formulation processes as a basis for determining the confidence level of the second test, thereby achieving more precise feed formulation optimization.

[0048] It is understood that the embodiments of this application, by preprocessing the current feed nutrient content detection data and moisture detection data, make the data from different batches and under different environments comparable, improve the accuracy and reliability of the feed nutrient content detection data, ensure that the nutrient content data used for formula optimization more accurately reflects the actual quality of the feed, and make the determined second detection confidence level more valuable. This allows for more precise guidance on feed formula adjustments, ultimately achieving more efficient, economical, and duck flock growth-compliant feed formula optimization, effectively reducing the risks caused by data quality issues.

[0049] In some embodiments, current nutrient composition test data for a batch of duck feed shows a protein content of 20% and a moisture content of 15%. The dry matter content of the protein can be calculated as 20% / (1%-15%)≈23.53%. Then, a preset feed database is queried to obtain the historical dry matter nutrient content fluctuation range of this type of feed, ranging from 22% to 24%. In this case, the calculated 23.53% is within the fluctuation range, indicating that the data is normal and does not require adjustment; it is directly used as the second dry matter nutrient content. If the current batch of feed shows a protein content of 18% and a moisture content of 10%, the calculated first dry matter nutrient content is 18% / (1-10%)=20%, which is lower than the lower limit of the historical fluctuation range of 22%. In this case, the data anomaly is identified and adjusted, for example, to the lower limit of the fluctuation range of 22%, resulting in a second dry matter nutrient content of 22%. Ultimately, whether it is the unadjusted 23.53% or the adjusted 22%, the second dry matter nutrient content will be converted into nutrient analysis data and stored in the feed database. This data will be used to determine the confidence level of the second analysis and optimize the feed formulation when formulating feed.

[0050] like Figure 5 As shown, Figure 5 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding step S410 above, the method may also include, but is not limited to, steps S510 to S530.

[0051] Step S510: Obtain current nutrient composition detection data, environmental humidity data, and feed moisture detection data. Based on the environmental humidity data, feed moisture detection data, and preset moisture range, determine the moisture detection confidence level of the feed moisture detection data. Step S520: Adjust the feed moisture detection data according to the moisture detection confidence level to obtain corrected feed moisture detection data; adjust the current nutrient component detection data according to the corrected feed moisture detection data and the environmental humidity data to obtain corrected nutrient component detection data. Step S530: Based on the corrected feed moisture detection data and corrected nutrient component detection data, obtain the first dry matter nutrient content.

[0052] For example, the current nutrient content detection data can be the result of nutrient content determination of the feed sample at the current time point; the environmental humidity data can be the relative humidity of the environment in which the feed sample is located, which can affect the actual moisture content of the feed; the feed moisture detection data can be the moisture content of the feed sample obtained through a moisture detection device; the preset moisture range can be the reasonable range of feed moisture content under specific environmental conditions, and when the environmental humidity is high, the upper limit of the reasonable moisture range of the feed will be increased accordingly.

[0053] Therefore, the confidence level of feed moisture detection data can be determined based on environmental humidity data, feed moisture detection data, and a preset moisture range. The confidence level of moisture detection can be an assessment of the reliability of feed moisture detection data. It can be judged comprehensively based on the degree of matching between feed moisture detection data and preset moisture range, as well as environmental humidity data. For example, when feed moisture detection data is within the preset moisture range and its trend is consistent with that of environmental humidity data, the confidence level of moisture detection is high.

[0054] For example, corrected feed moisture detection data can be obtained by adjusting the original feed moisture detection data after considering the confidence level of the moisture detection, in order to eliminate or reduce possible errors in the original detection data. For instance, when the confidence level of the moisture detection is low, the original feed moisture detection data can be appropriately corrected to be closer to the actual value. Corrected nutrient component detection data can be obtained by adjusting the original current nutrient component detection data after considering the corrected feed moisture detection data and environmental humidity data, in order to unify the nutrient component content to a dry matter basis and eliminate the influence of moisture changes on the nutrient component content.

[0055] It is understandable that the embodiments of this application can obtain the first dry matter nutrient content more accurately because they fully consider the key factors affecting the accuracy of feed moisture and nutrient content detection. First, by obtaining environmental humidity data, an external reference can be provided to assess the reliability of feed moisture detection data. Second, by comparing the feed moisture detection data with a preset moisture range and combining it with environmental humidity data, the confidence level of the feed moisture detection data can be determined. After obtaining the moisture detection confidence level, the original feed moisture detection data can be adjusted in a targeted manner to obtain corrected feed moisture detection data. Then, based on the corrected feed moisture detection data and environmental humidity data, the current nutrient content detection data is adjusted to obtain corrected nutrient content detection data, thereby obtaining the first dry matter nutrient content. This makes the obtained first dry matter nutrient content more accurate and reliable, ensuring that the feed formulation is more in line with the actual nutritional needs of the duck flock for subsequent feed formulation optimization, thereby improving feed utilization and reducing breeding costs.

[0056] In some embodiments, during a feed formulation process, it is necessary to obtain the dry matter nutrient content of a batch of feed. First, acquire the current nutrient content detection data (protein content 20%) and feed moisture detection data (moisture content 15%). Simultaneously, the sensor detects the current ambient humidity as 80%, with a preset moisture range of 10%–14%. At this point, the confidence level of the feed moisture detection data can be determined based on the ambient humidity data, the feed moisture detection data, and the preset moisture range. Since the current ambient humidity is high, and the feed moisture detection data is slightly higher than the upper limit of the preset normal moisture range, the confidence level of the moisture detection data can be judged to be moderately low. Based on this moderately low confidence level, the original feed moisture detection data can be adjusted, for example, by correcting the 15% moisture detection data to 14.5%. Subsequently, the current nutrient content test data is adjusted based on the corrected feed moisture test data and environmental humidity data. For example, if the original protein content is calculated based on wet weight, it needs to be converted to a dry matter basis after moisture content correction. That is, if the corrected moisture content is 14.5%, the corrected nutrient content test data will be calculated as 20% / (1%—14.5%)≈23.39%. Finally, the first dry matter nutrient content is obtained based on the corrected feed moisture test data and the corrected nutrient content test data.

[0057] like Figure 6 As shown, Figure 6 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding step S130 above, the method may also include, but is not limited to, steps S610 to S620.

[0058] Step S610: When the current feed is supplied by a newly introduced feed supplier, determine the new supplier confidence level for assessing the reliability of the second test confidence level based on the current supply batch of the newly introduced feed supplier. The newly introduced feed supplier is the feed supplier whose supply batch is within the preset batch quantity. The second test confidence level corresponds to the nutrient composition test data of the current feed supplied by the newly introduced feed supplier. Step S620: When formulating a feed formulation, determine a first adjustment coefficient and a second adjustment coefficient based on the second test confidence level and the new supplier confidence level, respectively. Adjust the nutrient content and cost of the feed using the first adjustment coefficient and the second adjustment coefficient to optimize the feed formulation.

[0059] Understandably, in practical applications, when feed is supplied by a newly introduced feed supplier, the lack of historical data accumulation from the new supplier may mean that adjustments based solely on the confidence level of a second test may not adequately reflect the actual quality fluctuations and reliability of the supplied feed. This could affect the accuracy of formulation optimization, leading to risks of nutritional deviations or inadequate cost control when dealing with new suppliers. To address this, the embodiments of this application introduce a new supplier confidence level to more comprehensively assess the reliability of feed quality and make more refined adjustments accordingly.

[0060] Specifically, when feed is currently supplied through a newly introduced feed supplier, the confidence level of the new supplier is determined based on the current batches supplied by the new supplier to assess the reliability of the second test confidence level. The new supplier can be one that supplies feed batches within a preset batch quantity. The nutritional component testing data of the current feed supplied by the new supplier corresponds to the second test confidence level. In other words, when a feed supplier's first batch supply, or a batch supply within a short period, has not yet reached the preset stable batch quantity, that supplier is considered a newly introduced feed supplier. The determination of the new supplier confidence level aims to quantify the reliability or credibility of the testing data (i.e., the corresponding second test confidence level) of the feed provided by the new supplier, reflecting the inadequacy of historical data accumulation or the uncertainty of quality stability.

[0061] When formulating feed formulations, a first adjustment coefficient and a second adjustment coefficient are determined based on the confidence level of the second test and the confidence level of the new supplier, respectively. The nutrient content and cost of the feed are adjusted using the first and second adjustment coefficients to optimize the feed formulation. The first adjustment coefficient can be mainly based on the confidence level of the second test, reflecting the reliability of overall historical data; while the second adjustment coefficient is based on the confidence level of the new supplier to compensate for the potential risks caused by the insufficient reliability of the new supplier's data. By introducing two independent adjustment coefficients, the nutrient content and cost of the feed can be adjusted more flexibly to adapt to the uncertainties brought by the new supplier.

[0062] It is understood that the embodiments of this application can quantify the reliability of the second test confidence level of the new supplier by using the new supplier confidence level. This allows the formulation of feed formulas to consider not only the second test confidence level obtained based on overall historical data to determine the first adjustment coefficient, but also the new supplier confidence level to determine the second adjustment coefficient. Through the synergistic effect of the two adjustment coefficients, the adjustment of the nutrient content and cost of the feed can simultaneously take into account the overall quality reliability of the feed and the uncertainty unique to the new supplier. This results in more robust and accurate feed formula optimization and improves the accuracy of duck feed formula optimization.

[0063] In some embodiments, a farm introduces a new corn supplier that has never supplied the farm before, resulting in very little historical testing data, far below a preset batch quantity threshold. In this case, the supplier is identified as a newly introduced feed supplier. When the new supplier supplies its first batch of corn, although nutritional component testing is performed and a corresponding second confidence level is obtained, the supply batch is still within the first batch range. Therefore, a lower confidence level for the new supplier can be determined based on the current supply batch.

[0064] When formulating duck feed, the confidence levels of the second test for this batch of corn and the confidence levels of the new supplier are considered simultaneously. The first adjustment coefficient is determined based on the confidence level of the second test, and the second adjustment coefficient is determined based on the confidence level of the new supplier. The second adjustment coefficient is used to avoid the potential quality instability of the new supplier. For example, the energy value of the corn can be adjusted more conservatively, or a certain amount of safety margin can be added to the formula to ensure the overall nutritional level of the feed.

[0065] like Figure 7 As shown, Figure 7 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding step S610 above, the method may also include, but is not limited to, steps S710 to S730.

[0066] Step S710: If the current supply batch is within the first batch range, determine the new supplier confidence level as the preset initial new supplier confidence level; Step S720: When the current supply batch is within the second batch range, obtain the current confidence level, extract the second detection confidence level of the historical second range quantity based on the current supply batch as the new supplier confidence level assessment data, and when the average deviation determined based on the new supplier confidence level assessment data meets the preset average range condition, adjust the current confidence level based on the preset second range adjustment amount to obtain the new supplier confidence level. Step S730: When the current supply batch is within the range of the third batch, obtain the current confidence level, extract the second detection confidence level of the historical third range quantity based on the current supply batch as the new supplier confidence level assessment data, and adjust the current confidence level based on the preset standard deviation range condition based on the preset third range adjustment amount to obtain the new supplier confidence level when the standard deviation determined based on the new supplier confidence level assessment data meets the preset standard deviation range condition. Among them, the range of the third batch is larger than that of the second batch, the range of the second batch is larger than that of the first batch, and the adjustment amount of the third range is smaller than that of the second range.

[0067] For example, the first, second, and third batch ranges represent different stages in the supply volume of a newly introduced feed supplier, reflecting a gradual increase in the amount of historical data obtained from that supplier. The first batch range typically corresponds to the initial supply stage of a new supplier, when the amount of historical data is limited; the second batch range corresponds to a stage with moderate data volume; and the third batch range corresponds to a stage with relatively abundant and stable data volume. The initial confidence level for a new supplier can be a preset, typically conservative confidence level value set for the new supplier when the amount of data is extremely limited (i.e., within the first batch range), providing a basic assessment when sufficient historical data is lacking. The current confidence level can be the confidence level value currently held by the system for that supplier when conducting a new supplier confidence assessment; this value will be adjusted based on new data. The mean deviation and standard deviation are statistical indicators used to evaluate the confidence assessment data for new suppliers (i.e., historical second-test confidence levels). The mean deviation measures the degree of deviation of the data from a benchmark mean, while the standard deviation measures the dispersion or volatility of the data itself. The preset mean range and preset standard deviation range are used to determine whether these statistical indicators are within acceptable ranges, thus deciding whether to adjust the confidence level. The second and third range adjustment amounts are the values ​​used to adjust the current confidence level when the corresponding conditions are met. The third range adjustment amount is smaller than the second range adjustment amount, reflecting that as the amount of data increases and the stability of the data improves, the adjustment of the confidence level should be more refined and conservative.

[0068] It is understandable that this application embodiment dynamically employs different methods to determine the confidence level of a new feed supplier based on the current batch volume of the newly introduced supplier. When the batch volume is small, due to a lack of sufficient data for statistical analysis, a preset initial confidence level for the new supplier is directly used to ensure a certain degree of conservatism in the assessment when data is insufficient. As the batch volume increases, a certain amount of historical second-detection confidence level data begins to accumulate. At this point, the stability of the new supplier's supply data can be initially assessed. If the conditions are met, the current confidence level is adjusted based on a second-range adjustment, so that the confidence level can reflect the initial data performance. When the batch volume further increases and reaches a large scale, the amount of historical data becomes richer. At this point, by calculating the standard deviation of the historical second-detection confidence level data, the volatility and consistency of the data can be measured more accurately. When the standard deviation meets the preset standard deviation range conditions, the current confidence level is adjusted based on a smaller third-range adjustment, indicating that as the amount of data increases and the data quality stabilizes, the adjustment of the confidence level should be more cautious and refined.

[0069] Therefore, by adopting a phased adaptive evaluation mechanism, the process of determining the confidence level of new suppliers can make full use of historical data information at different stages, thereby improving the accuracy of the evaluation. This ensures that the reliability evaluation of new suppliers is more accurate at different data accumulation stages, avoiding the risk of over-reliance on limited data for inaccurate evaluation when data is insufficient, and also avoiding the failure to make full use of data for refined evaluation when data is sufficient.

[0070] In some embodiments, a duck feed producer introduces a new corn supplier. When the new supplier's batches are in the first batch range, due to a lack of sufficient historical data, the new supplier's confidence level is directly set to a preset initial new supplier confidence level, for example, 0.6. When the new supplier's batches enter the second batch range (batch 4 to batch 10), a certain amount of historical second-test confidence level data has been accumulated. At this time, the current confidence level is obtained, and the second-test confidence levels of batches 4 to 10 are extracted as the new supplier confidence level assessment data. If the determined average deviation meets the preset average range condition, for example, the average deviation is less than 0.05, the current confidence level is adjusted based on the preset second range adjustment amount. When the new supplier's batches enter the third batch range (batch 11 and beyond), with richer historical data, the current confidence level is obtained again, and the second-test confidence levels of batches 11 and beyond in the third range are extracted as assessment data. If the determined standard deviation meets the preset standard deviation range condition, the current confidence level is adjusted based on the preset third range adjustment amount.

[0071] like Figure 8 As shown, Figure 8This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding the above method, the method may also include, but is not limited to, step S810.

[0072] Step S810: Monitor the growth performance parameters of the duck flock. If the changes in the growth performance parameters match the preset changes and the confidence level of the new supplier is lower than the preset confidence level threshold, determine the suppliers of different feed components and the corresponding supply ratio of each supplier based on the second detection confidence level and the corresponding new supplier confidence level. Generate a corresponding formula adjustment plan based on the suppliers of different feed components and the corresponding supply ratio of each supplier.

[0073] For example, monitoring the growth performance parameters of duck flocks can involve continuously tracking and recording various physiological indicators and production performance of duck flocks during the feeding process, including but not limited to the average daily weight gain, feed conversion rate, mortality rate, health status score, and digestibility and absorption rate of specific nutrients, thereby obtaining the actual physiological response data of duck flocks to the current feed formula, which can serve as a basis for evaluating feed effectiveness.

[0074] Changes in growth performance parameters must match preset changes. This means that when monitored growth performance parameters of the duck flock show negative changes that deviate from expected targets, such as significantly lower daily weight gain than the target value, deterioration in feed conversion rate, abnormally high mortality rate, or a decline in health status score, and the trend of these changes matches the preset abnormal change pattern. A new supplier's confidence level must be lower than a preset confidence threshold. In other words, after introducing a new feed supplier, the reliability assessment result of the testing data for the feed supplied by that new supplier (i.e., the new supplier's confidence level) is lower than the preset minimum acceptable level. The confidence threshold is set based on experience, historical data, or risk preference and is used to define the risk level of feed supplied by the new supplier.

[0075] Subsequently, based on the second detection confidence level and the corresponding new supplier confidence level, suppliers for different feed components and their respective supply ratios are determined. This can be done when duck flock growth performance is abnormal and new suppliers pose a high risk. A comprehensive analysis is conducted, considering the detection confidence levels of each nutrient in the feed (second detection confidence level) and the overall reliability of the new supplier (new supplier confidence level), to identify potentially problematic feed components and find alternative or supplementary suppliers. Thus, based on the suppliers of different feed components and their corresponding supply ratios, a corresponding formula adjustment plan is generated. After determining the new supplier combination and supply ratios, a new feed formula can be recalculated and generated.

[0076] This application embodiment constructs a closed-loop feedback mechanism from feed input to animal output by real-time monitoring of duck flock growth performance parameters. Specifically, when the duck flock's growth performance parameters show changes matching preset abnormal conditions, it indicates a potential problem with the current feed formulation or feed quality. If, simultaneously, the confidence level of a new supplier is found to be below a preset confidence threshold, it further clarifies that the problem may be related to the newly introduced feed supplier, meaning that the feed supplied by the new supplier carries high uncertainty or risk, leading to abnormal duck growth. Subsequently, by comprehensively utilizing the second detection confidence level and the new supplier's confidence level, it is possible to analyze and determine which feed components may be problematic, and adjust the suppliers and their supply ratios for different feed components accordingly. This optimizes the source structure of feed components from the source, effectively reducing the potential risks brought about by the introduction of new suppliers, ensuring the healthy growth and production efficiency of the duck flock, and improving the intelligent level of feed formulation optimization, achieving an upgrade from data-driven to a dual-driven approach of data and biological feedback.

[0077] In some embodiments, a farm introduced a new corn supplier and tested the supplier's corn, obtaining a corresponding second confidence level. During initial formulation, the formula was adjusted based on this second confidence level and the new supplier's confidence level. However, during subsequent feeding, monitoring of the duck flock's growth performance parameters revealed that the average daily weight gain was below the preset target value for two consecutive weeks, and the feed conversion ratio also showed a significant decline, matching the preset abnormal changes. Simultaneously, a reassessment of the new corn supplier's confidence level showed it had fallen below the preset confidence threshold, indicating a high risk of instability in the quality of the new supplier's corn.

[0078] In this scenario, based on the current confidence level of the second detection and the low confidence level of the new supplier, it can be analyzed that corn quality issues may be one of the main reasons for the abnormal growth of the duck flock. Therefore, it is recommended to reduce the proportion of corn supplied by this new supplier and increase the proportion of corn purchased from another supplier with a good historical track record and high confidence level. It is also suggested to introduce a small amount of corn from a third alternative supplier to further diversify the risk.

[0079] In another embodiment of the duck feed formulation optimization method provided in this application, the first batch of historical test data includes multiple sets of first batch historical test data provided by different time periods, different test methods, or different test institutions; such as Figure 9 As shown, Figure 9 This is a flowchart of a duck feed formulation optimization method provided in another embodiment of this application. Regarding step S120 above, the method may also include, but is not limited to, steps S910 to S930.

[0080] Step S910: Identify systematic biases in multiple sets of first-batch historical test data. Systematic biases include persistent differences in the test results of the same nutrient components between different data sources. Step S920: Based on systematic bias, normalize multiple sets of first-batch historical detection data to make the detection data from different data sources comparable on a unified benchmark. Step S930: Obtain the first detection confidence level based on the normalized sets of first-batch historical detection data, and adjust the first detection confidence level based on the second-batch historical detection data to obtain the second detection confidence level.

[0081] For example, the first batch of historical testing data can be feed nutrient composition test results collected through multiple means over a relatively long time span. This includes test results conducted by different laboratories on different dates, or results obtained using different analytical instruments and standard methods. Identifying systematic biases in multiple sets of first batch historical testing data is crucial. Systematic bias can be a persistent and predictable difference between results from different data sources when testing the same nutrient composition. For example, a testing institution might consistently detect protein content that is a fixed percentage or value higher than another institution due to its specific equipment calibration or operating habits.

[0082] After identifying systematic biases, multiple sets of first-batch historical test data need to be normalized to eliminate or reduce their impact, ensuring that test data from different data sources can be compared and analyzed on a unified benchmark. For example, an authoritative testing institution or a standard testing method can be selected as the benchmark, and then test results from other data sources can be adjusted to align with the identified systematic bias. Thus, after normalizing multiple sets of first-batch historical test data, a more accurate and reliable first-batch confidence level can be obtained. Subsequently, this first-batch confidence level is adjusted based on second-batch historical test data to obtain the final second-batch confidence level.

[0083] It is understood that by introducing the identification and normalization of systematic deviations in multiple sets of first-batch historical test data, the embodiments of this application can unify the test results from different data sources onto a common benchmark, thereby eliminating the influence of heterogeneity between data. This allows the subsequently calculated first test confidence level to more accurately reflect the reliability of feed nutrient component testing, and enables the adjustment coefficient determined based on a more accurate test confidence level to more accurately reflect the actual situation and cost of feed nutrients when formulating feed formulations.

[0084] In some embodiments, a duck feed production company has been obtaining nutritional component testing data for its feed from three different testing institutions, A, B, and C, as the first batch of historical testing data. After a period of accumulation, it was found that the crude protein content tested by institution A was generally 0.5% higher than that of institution B, while the crude fat content tested by institution C was generally 0.3% lower than that of institution A, which is a systematic bias.

[0085] Subsequently, a benchmark was selected, and the data from other institutions were normalized. For example, the test results from institution B were used as the benchmark for crude protein, and the test results from institution A were used as the benchmark for crude fat. All crude protein test data provided by institution A were uniformly subtracted by 0.5%, and all crude fat test data provided by institution C were uniformly added by 0.3%. After normalization, all crude protein and crude fat test data from institutions A, B, and C were adjusted to a unified benchmark, eliminating systematic differences between institutions.

[0086] Secondly, embodiments of this application provide a duck feed formulation optimization system, comprising: The data acquisition module is used to acquire the nutritional component testing data of the feed. The nutritional component testing data includes the first batch of historical testing data and the second batch of historical testing data. The first batch of historical testing data has a larger batch number than the second batch of historical testing data, and the first batch of historical testing data includes the second batch of historical testing data. The confidence module is used to obtain a first detection confidence based on the first batch of historical detection data, and to adjust the first detection confidence based on the second batch of historical detection data to obtain a second detection confidence. The optimization module is used to determine a first adjustment coefficient based on the second detection confidence level when formulating feed formulation, and to adjust the nutrient content and cost of the feed through the first adjustment coefficient to achieve feed formulation optimization.

[0087] For example, the data acquisition module can be configured to connect to external testing equipment, databases, or manual input terminals through various interfaces to ensure that the required historical and real-time testing data can be collected comprehensively and in a timely manner. For instance, the data acquisition module can integrate a data acquisition interface to receive output data from testing equipment such as spectrometers and chemical analyzers; it can also access cloud databases or supplier data platforms through network interfaces; or it can provide a user interface to allow operators to manually enter test reports.

[0088] The confidence module can incorporate statistical analysis algorithms and machine learning models to process large amounts of historical data, identify patterns, trends, and anomalies in the data, and thus quantify the reliability of the detection results.

[0089] The optimization module can integrate optimization algorithms such as linear programming and nonlinear programming, and combine them with the duck flock's nutritional requirements model, raw material cost model and market supply and demand information to dynamically adjust the feed formula based on the second detection confidence level.

[0090] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for optimizing duck feed formulation, characterized in that, include: Obtain nutritional component testing data of feed, wherein the nutritional component testing data includes the first batch of historical testing data and the second batch of historical testing data, wherein the first batch of historical testing data has a larger batch number than the second batch of historical testing data and the first batch of historical testing data includes the second batch of historical testing data. A first detection confidence level is obtained based on the first batch of historical detection data, and a second detection confidence level is obtained by adjusting the first detection confidence level based on the second batch of historical detection data. When formulating feed formulations, a first adjustment coefficient is determined based on the second detection confidence level. The nutrient content and cost of the feed are adjusted using the first adjustment coefficient to optimize the feed formulation.

2. The method for optimizing duck feed formulation according to claim 1, characterized in that, The duck feed formulation optimization method also includes: The growth performance parameters of the duck flock are monitored. If the changes in the growth performance parameters match the preset changes, and it is determined based on the second detection confidence level that the changes in the growth performance parameters are related to the nutrient content of the feed, the deficiency of feed quantity or the deficiency of nutrient content in the feed is determined based on the second detection confidence level, and a corresponding formula adjustment plan is generated based on the deficiency of feed quantity or the deficiency of nutrient content in the feed.

3. The method for optimizing duck feed formulation according to claim 1, characterized in that, The step of obtaining a first detection confidence level based on the first batch of historical detection data, and adjusting the first detection confidence level based on the second batch of historical detection data to obtain a second detection confidence level includes: Obtain the nutritional component standard data corresponding to the first batch of historical test data, identify the difference between the first batch of historical test data and the nutritional component standard data, and obtain the first test confidence level; The trend of the second batch of data changes is determined based on the historical test data of the second batch. The first detection confidence level is adjusted based on the trend of the second batch of data changes and the preset trend change threshold to obtain the second detection confidence level.

4. The method for optimizing duck feed formulation according to claim 1, characterized in that, The duck feed formulation optimization method also includes: Obtain current nutrient composition test data and feed moisture test data, and obtain the first dry matter nutrient content based on the feed moisture test data and current nutrient composition test data; Obtain the historical dry matter nutrient content fluctuation range. If the first dry matter nutrient content exceeds the historical dry matter nutrient content fluctuation range, adjust the first dry matter nutrient content to obtain the second dry matter nutrient content. The second dry matter nutrient content is converted into nutrient detection data and stored in a preset feed database so that, in the case of feed formulation, a second detection confidence level can be determined based on the nutrient detection data and the feed formulation can be optimized.

5. The method for optimizing duck feed formulation according to claim 4, characterized in that, The process of acquiring current nutrient composition detection data and feed moisture detection data, and obtaining the first dry matter nutrient content based on the feed moisture detection data and current nutrient composition detection data, includes: Acquire current nutrient composition detection data, environmental humidity data, and feed moisture detection data; determine the moisture detection confidence level of the feed moisture detection data based on the environmental humidity data, the feed moisture detection data, and a preset moisture range. The feed moisture detection data is adjusted based on the moisture detection confidence level to obtain corrected feed moisture detection data. The current nutrient component detection data is then adjusted based on the corrected feed moisture detection data and the environmental humidity data to obtain corrected nutrient component detection data. The first dry matter nutrient content is obtained based on the corrected feed moisture detection data and the corrected nutrient component detection data.

6. The method for optimizing duck feed formulation according to claim 1, characterized in that, In the process of formulating a feed formulation, a first adjustment coefficient is determined based on the second detection confidence level. The nutrient content and cost of the feed are adjusted using the first adjustment coefficient to optimize the feed formulation, including: When the current feed is supplied by a newly introduced feed supplier, a new supplier confidence level is determined based on the current supply batches of the newly introduced feed supplier to assess the reliability of the second detection confidence level. The newly introduced feed supplier is a feed supplier whose supply batches are within a preset batch quantity. The second detection confidence level corresponds to the nutrient composition detection data of the current feed supplied by the newly introduced feed supplier. When formulating feed formulations, a first adjustment coefficient and a second adjustment coefficient are determined based on the second detection confidence level and the new supplier confidence level, respectively. The nutrient content and cost of the feed are adjusted using the first adjustment coefficient and the second adjustment coefficient to optimize the feed formulation.

7. The method for optimizing duck feed formulation according to claim 6, characterized in that, The step of determining the new supplier confidence level for assessing the reliability of the second test confidence level based on the current supply batches of the newly introduced feed supplier includes: If the current supply batch is within the first batch range, the confidence level of the new supplier is determined to be the preset initial confidence level of the new supplier; When the current supply batch is within the second batch range, the current confidence level is obtained. Based on the current supply batch, the second detection confidence level of the second range of historical data is extracted as the new supplier confidence level assessment data. When the average deviation determined based on the new supplier confidence level assessment data meets the preset average range condition, the current confidence level is adjusted based on the preset second range adjustment amount to obtain the new supplier confidence level. When the current supply batch is within the range of the third batch, the current confidence level is obtained. Based on the current supply batch, the second detection confidence level of the number of historical third-range quantities is extracted as the new supplier confidence level assessment data. If the standard deviation determined based on the new supplier confidence level assessment data meets the preset standard deviation range condition, the current confidence level is adjusted based on the preset third-range adjustment amount to obtain the new supplier confidence level. Wherein, the range of the third batch is greater than the range of the second batch, the range of the second batch is greater than the range of the first batch, and the adjustment amount of the third range is less than the adjustment amount of the second range.

8. The method for optimizing duck feed formulation according to claim 6, characterized in that, The duck feed formulation optimization method also includes: The growth performance parameters of the duck flock are monitored. If the changes in the growth performance parameters match the preset changes and the confidence level of the new supplier is lower than the preset confidence level threshold, the suppliers of different feed components and the corresponding supply ratio of each supplier are determined based on the second detection confidence level and the corresponding new supplier confidence level. A corresponding formula adjustment plan is generated according to the suppliers of different feed components and the corresponding supply ratio of each supplier.

9. The method for optimizing duck feed formulation according to claim 1, characterized in that, The first batch of historical test data includes multiple sets of first batch historical test data provided by different time periods, different test methods, or different test institutions; The step of obtaining a first detection confidence level based on the first batch of historical detection data, and adjusting the first detection confidence level based on the second batch of historical detection data to obtain a second detection confidence level includes: Identify systematic biases in the multiple sets of first-batch historical test data, including persistent differences in the test results of the same nutrient components between different data sources; Based on the systematic bias, the multiple sets of first-batch historical detection data are normalized so that detection data from different data sources can be compared on a unified benchmark. The first detection confidence level is obtained based on the normalized batch of historical detection data from the first batch. The second detection confidence level is then adjusted based on the second batch of historical detection data to obtain the second detection confidence level.

10. A duck feed formulation optimization system, characterized in that, include: The data acquisition module is used to acquire the nutritional component testing data of the feed. The nutritional component testing data includes the first batch of historical testing data and the second batch of historical testing data. The first batch of historical testing data has a larger batch number than the second batch of historical testing data, and the first batch of historical testing data includes the second batch of historical testing data. The confidence module is used to obtain a first detection confidence based on the first batch of historical detection data, and to adjust the first detection confidence based on the second batch of historical detection data to obtain a second detection confidence. The optimization module is used to determine a first adjustment coefficient based on the second detection confidence level when formulating feed formulation, and to adjust the nutrient content and cost of the feed through the first adjustment coefficient to achieve feed formulation optimization.

Images