A method and system for evaluating and grading quality of unconventional feed

Abstract

The present invention discloses a method and system for grading the quality of unconventional feed, which specifically relates to the technical field of feed quality assessment. Conventional nutrition, functional nutrition and harmful substance components are comprehensively analyzed through high-precision detection technology. A multi-factor synergistic toxicity assessment model is constructed, and the synergistic effects of multiple harmful substances are quantified based on toxic equivalent factors and matrix interaction analysis, and a synergistic toxicity index is output. The overall quality of the feed sample is evaluated in combination with the detection results and the toxicity index, and the sample is divided into high-quality, qualified and unqualified grades. Targeted application suggestions and risk control measures are provided, a cloud database is constructed to record assessment data, and Internet of Things sensors are used to monitor supply chain quality fluctuations in real time. Anomalies are automatically identified and early warnings are triggered to achieve dynamic management of feed quality, effectively improve the scientific nature and accuracy of unconventional feed quality assessment, reduce potential safety hazards, and facilitate its safe promotion and efficient application in the breeding industry.

Description

Technical Field

[0001] The present invention relates to the technical field of feed quality assessment, and in particular to a non-conventional feed quality assessment and grading method and system. Background Art

[0002] With the rapid development of the animal husbandry industry, feed resource shortages are becoming increasingly prominent. Unconventional feeds (such as agricultural and sideline products, industrial by-products, and wild plant resources) are becoming an important supplement to the feed industry due to their abundant resources and low cost. However, compared with traditional feeds, unconventional feeds have complex ingredients, fluctuate greatly in nutritional value, and may contain harmful substances such as heavy metals, toxins, and anti-nutritional factors. These characteristics make the quality evaluation and grading of unconventional feeds a major topic in feed science. There is an urgent need to establish systematic and scientific evaluation methods to ensure the health and stable production performance of farmed animals.

[0003] Currently, the quality assessment of unconventional feeds is primarily based on conventional nutritional indicators (such as protein, energy, and minerals) and functional nutritional indicators (such as immune enhancement and antioxidant properties), while also focusing on the potential impact of heavy metals, toxins, and anti-nutritional factors (such as phytates and tannins). However, due to the complex interactions between these indicators and the significant compositional differences among unconventional feeds from different sources, existing evaluation systems struggle to fully reflect their overall quality and potential risks, hindering the development and application of unconventional feeds.

[0004] The existing technology has the following shortcomings:

[0005] Existing unconventional feed quality assessment methods are unable to accurately evaluate the combined effects of complex ingredients on animal health and production performance, especially the lack of effective evaluation mechanisms for the interactions between different feed ingredients and the cumulative effects of toxic factors. For example, when certain unconventional feeds contain low doses of heavy metals (such as lead and cadmium) and toxins (such as aflatoxins), these substances may produce synergistic toxic effects. In addition, if existing technologies cannot accurately predict or quantify their long-term harm to animals. This technical limitation may not only lead to potential safety hazards in feed use, but also limit the widespread promotion of unconventional feed in the industry. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for evaluating and grading the quality of unconventional feeds, so as to overcome the shortcomings of the background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for evaluating and grading the quality of unconventional feed, comprising the following steps:

[0008] S1: Unconventional feed samples were collected from industrial by-products, agricultural by-products and wild plant materials. The samples were pre-treated using low-temperature drying technology, crushed to a particle diameter of less than 1 mm using high-precision crushing equipment, and evenly mixed using a three-dimensional mixer;

[0009] S2: Component testing is performed on the pre-treated samples. Component testing includes three categories of indicators: general nutrition, functional nutrition, and harmful substances. Protein, fat, immune-enhancing factors, and antioxidant components in the samples are tested respectively. For heavy metals, toxins, and anti-nutritional factors, their content is accurately determined using ICP-MS, GC-MS, and HPLC techniques.

[0010] S3: Construct a multi-factor synergistic toxicity assessment model, calculate the toxicity contribution of a single toxin based on the toxicity equivalent factor, use the matrix interaction analysis method to simulate the synergistic effect of multiple harmful substances, and output the synergistic toxicity index;

[0011] S4: Evaluate the overall quality of feed samples based on conventional nutrition, functional nutrition, and synergistic toxicity indices. Based on the evaluation results, classify unconventional feeds into three levels: high quality, qualified, and unqualified. Provide corresponding application recommendations for each level, including scope of application and risk control measures.

[0012] S5: Build a cloud-based database to record the evaluation and grading data of unconventional feeds in real time, and use IoT sensors to monitor feed quality fluctuations in the raw material supply chain in real time, automatically identify abnormal samples and trigger early warnings.

[0013] Preferably, in S3, a multi-factor synergistic toxicity assessment model is constructed, the toxicity contribution of a single toxin is calculated based on the toxicity equivalent factor, and the synergistic effect of multiple harmful substances is simulated using the matrix interaction analysis method to output the synergistic toxicity index, specifically:

[0014] Determine the toxicity equivalent factor TEF, which represents the relative toxicity intensity of a substance to a reference substance. The toxicity contribution of a single toxin is calculated as: T i =C i ×TEF i Where, T i is the toxicity contribution value of toxin i, C i is the concentration of toxin i; TEF i is the toxicity equivalent factor of toxin i, and T is calculated for each toxin i. i And summarize to form a list of single toxin toxicity values, define the synergy matrix M, where the element M ij represents the synergistic effect between substances i and j; M ij >0: synergistic effect, indicating that the toxicity of the two substances is enhanced; M ij=0: no synergistic effect, indicating that the toxicity of the two substances is independent; M ij <0: Antagonistic effect, indicating that the toxicity of the two substances is weakened. The synergistic effect calculation formula is: T total is the comprehensive toxicity value, T i is the toxicity contribution value of a single toxin i, n is the total number of toxin types, and the normalized comprehensive toxicity value is: STI is the synergistic toxicity index, ranging from 0 to 100; T max The maximum toxicity value is used as a reference for normalization.

[0015] Preferably, in S4, the overall quality of the feed sample is evaluated based on conventional nutrition, functional nutrition and synergistic toxicity index, specifically:

[0016] Collect ingredient test results, including conventional nutrition, functional nutrition and synergistic toxicity index, and normalize each index so that its value is between 0 and 1: X norm is the normalized index value; X is the original index value; X min and X max The minimum and maximum values ​​of the indicators are set according to their importance: the weight of the conventional nutrition indicator is 50%; the weight of the functional nutrition indicator is 30%; the weight of the synergistic toxicity index is 20%; the comprehensive score calculation formula is: Q = 0.5 Q 常规营养 +0.3·Q 功能营养 -0.2 STI norm .

[0017] Preferably, in S4, Q is the comprehensive quality score of the feed sample; Q 常规营养 It is a comprehensive score of conventional nutritional indicators. W i is the weight of the i-th nutritional indicator; N i is the normalized value of the ith nutritional index; b is the number of conventional nutritional indicators; Q 功能营养 It is a comprehensive score of functional nutrition index; STI norm is the normalized synergistic toxicity index; Q 功能营养 is the comprehensive score of functional nutrition indicators, expressed as: W j is the weight of the jth functional nutrition index, F j is the normalized value of the jth functional nutritional index; m is the number of functional nutritional indicators.

[0018] Preferably, in S4, the unconventional feed is divided into three levels: high quality, qualified, and unqualified according to the evaluation results, specifically:

[0019] According to the range of the comprehensive score Q, the feed is divided into three levels: high quality Q≥0.8: comprehensive nutritional ingredients, strong functionality, and low synergistic toxicity; qualified 0.5≤Q<0.8: nutritional indicators meet the requirements, but functionality or toxicity is insufficient; unqualified Q<0.5: low nutritional value and high synergistic toxicity, not suitable for use.

[0020] Preferably, in S5, the feed quality fluctuations in the raw material supply chain are monitored in real time through IoT sensors, abnormal samples are automatically identified and early warnings are triggered, specifically:

[0021] IoT sensors collect real-time raw material characteristic data R at supply chain nodes, including: temperature R T , humidity R H , Volatile organic compounds R VOC , weight R w ; Assume that the historical range of raw material characteristic data is [R i,min ,R i,max ], real-time data R i The abnormal judgment conditions are: R i is a characteristic data collected in real time; R i,min 、R i,max The historical upper and lower limits of the characteristic data. The output value 1 indicates abnormality and 0 indicates normality.

[0022] When multiple data indicators are collected simultaneously, a comprehensive scoring method is used to determine the degree of sample abnormality: Where a is the total number of monitoring indicators, R i,mean is the historical mean of the characteristic data, R i,std is the standard deviation of the feature data; the anomaly score is higher than the set threshold.

[0023] Preferably, the real-time monitoring results and abnormal conditions are recorded in the cloud database, and the historical data range is updated. i,min ,R i,max ] and mean and standard deviation;

[0024] Update history includes: Where, The updated historical minimum value indicates that the current real-time data R i and the historical minimum R i,min After comparison, take the smaller value. The updated historical maximum value indicates that the current real-time data R i The historical maximum R i,max After comparison, take the larger value;

[0025] Update statistical parameters: Where R i,kis the value of the i-th characteristic data in the k-th sample, G is the number of new samples, is the updated historical mean, which represents the average value of the characteristic data in G historical samples; is the updated historical standard deviation, which indicates the degree of fluctuation of the characteristic data relative to the mean in G samples.

[0026] The present invention also provides a non-conventional feed quality assessment and grading system, including a sample collection and pretreatment module, a component detection and analysis module, a collaborative toxicity assessment module, a quality assessment and grading module, and a management module:

[0027] Sample collection and pretreatment module: Unconventional feed samples are collected from industrial by-products, agricultural by-products and wild plant materials. The samples are pretreated using low-temperature drying technology, crushed to particles less than 1 mm in diameter using high-precision crushing equipment, and evenly mixed using a three-dimensional mixer;

[0028] Component detection and analysis module: Component detection is performed on pre-treated samples. Component detection includes three categories of indicators: general nutrition, functional nutrition, and harmful substances. Protein, fat, immune-enhancing factors, and antioxidant components in the samples are detected respectively. For heavy metals, toxins, and anti-nutritional factors, ICP-MS, GC-MS, and HPLC technologies are used to accurately determine their content.

[0029] Synergistic toxicity assessment module: Constructs a multi-factor synergistic toxicity assessment model, calculates the toxicity contribution of a single toxin based on the toxicity equivalent factor, uses matrix interaction analysis to simulate the synergistic effect of multiple hazardous substances, and outputs a synergistic toxicity index;

[0030] Quality Assessment and Grading Module: Evaluates the overall quality of feed samples based on conventional nutrition, functional nutrition, and synergistic toxicity index. Based on the assessment results, unconventional feeds are classified into three levels: high quality, qualified, and unqualified. Corresponding application recommendations are provided for each level, including scope of application and risk control measures.

[0031] Management module: Build a cloud database to record the evaluation and grading data of unconventional feeds in real time, and use IoT sensors to monitor feed quality fluctuations in the raw material supply chain in real time, automatically identify abnormal samples and trigger early warnings.

[0032] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0033] 1. The present invention adopts standardized sample pretreatment technology (low-temperature drying, fine grinding and uniform mixing), combined with high-precision detection technologies such as ICP-MS, GC-MS and HPLC, to achieve a comprehensive analysis of conventional nutrition, functional nutrition and harmful substances. By constructing a multi-factor synergistic toxicity assessment model, the toxicity equivalent factor and matrix interaction analysis method are used to quantify the synergistic effects of multiple harmful substances, and a synergistic toxicity index (STI) is generated to scientifically evaluate the potential risks of the feed. Combined with the nutritional score and toxicity index, the quality grading of the sample is achieved through comprehensive weight calculation, and application recommendations and risk control measures are provided to ensure the safety and applicability of the feed.

[0034] 2. The present invention achieves dynamic management of feed quality by constructing a cloud database and Internet of Things monitoring system. By using sensors to monitor quality fluctuations in the raw material supply chain in real time, and combining it with anomaly detection algorithms to identify risk samples and trigger early warnings, it can effectively prevent potential risks in feed production and use. The system also supports dynamic updating of historical data ranges and statistical parameters to ensure strong model adaptability and reliable monitoring results. In summary, the present invention significantly improves the scientific nature, accuracy, and management efficiency of unconventional feed quality assessment, providing technical support for its safe promotion and efficient application. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0036] Figure 1 Flow chart of the method of the present invention.

[0037] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0038] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0039] Example 1, please refer to Figure 1 As shown, the method for evaluating and grading the quality of unconventional feeds described in this embodiment includes the following steps:

[0040] S1: Unconventional feed samples were collected from industrial by-products, agricultural by-products and wild plant materials. The samples were pre-treated using low-temperature drying technology, crushed to a particle diameter of less than 1 mm using high-precision crushing equipment, and evenly mixed using a three-dimensional mixer;

[0041] S2: Component testing is performed on the pre-treated samples. Component testing includes three categories of indicators: general nutrition, functional nutrition, and harmful substances. Protein, fat, immune-enhancing factors, and antioxidant components in the samples are tested respectively. For heavy metals, toxins, and anti-nutritional factors, their content is accurately determined using ICP-MS, GC-MS, and HPLC techniques.

[0042] S3: Construct a multi-factor synergistic toxicity assessment model, calculate the toxicity contribution of a single toxin based on the toxicity equivalent factor, use the matrix interaction analysis method to simulate the synergistic effect of multiple harmful substances, and output the synergistic toxicity index;

[0043] S4: Evaluate the overall quality of feed samples based on conventional nutrition, functional nutrition, and synergistic toxicity indices. Based on the evaluation results, classify unconventional feeds into three levels: high quality, qualified, and unqualified. Provide corresponding application recommendations for each level, including scope of application and risk control measures.

[0044] S5: Build a cloud-based database to record the evaluation and grading data of unconventional feeds in real time, and use IoT sensors to monitor feed quality fluctuations in the raw material supply chain in real time, automatically identify abnormal samples and trigger early warnings.

[0045] In S1, representative sample sources are selected based on the potential application scenarios of unconventional feeds, including but not limited to: industrial by-products: such as distiller's grains, soybean meal, cottonseed cake, and sugarcane bagasse; agricultural by-products: such as corn straw, rice husks, and fruit and vegetable waste; and wild plant raw materials: such as wild forage grass, seaweed, and weed seeds.

[0046] Use a multi-point random sampling method to collect several samples from different batches, areas or raw material piles to ensure representativeness; samples from each sampling point are mixed to form an initial comprehensive sample.

[0047] Low-temperature drying technology removes moisture, reducing the risk of sample spoilage while preventing high temperatures from damaging active ingredients (such as vitamins and antioxidants). Use a vacuum freeze dryer to dry samples at temperatures between -50°C and -30°C until the moisture content is below 10%. If vacuum equipment is unavailable, use a forced-air drying oven at temperatures below 40°C to ensure uniform drying and stable ingredient activity.

[0048] Use high-precision pulverization equipment (such as a high-speed pulverizer or vibration mill) to ensure sample particle size refinement, facilitating uniformity and representativeness in subsequent testing. Place dried samples in batches into the pulverizer, adjusting the speed to at least 10,000 rpm. Continue pulverizing for 1-3 minutes, until the sample particles are less than 1 mm in diameter (verify using a standard sieve). Intermittent pulverization may be used to prevent damage to sample components due to equipment overheating.

[0049] Through mixing, the components of each part of the sample are evenly distributed, eliminating detection errors caused by differences in particle size or distribution. Use a three-dimensional mixer to place the crushed sample into the rotating mixing chamber; set the mixing time to 15-20 minutes to ensure that the sample is fully flipped and mixed in the three-dimensional direction; after mixing is completed, take some samples for uniformity testing to ensure the consistency of the sample in the distribution of nutrients and harmful substances. The processed samples are placed in moisture-proof and oxidation-proof sealed bags, marked with batch, source and date; the samples are stored in a low-temperature environment (below 4°C) to prevent changes in the sample composition before subsequent testing.

[0050] S2: The pre-treated samples are subjected to component testing. The component testing includes three categories of indicators: conventional nutrition, functional nutrition and harmful substances. The protein, fat, immune enhancement factors and antioxidant components in the samples are tested respectively. For heavy metals, toxins and anti-nutritional factors, ICP-MS, GC-MS and HPLC technologies are used to accurately determine their content.

[0051] Protein detection: Use a fully automatic near-infrared spectrometer (NIR) for rapid detection to directly determine the protein content of the sample; if further verification is required, the Kjeldahl method can be used to calculate the protein content (total nitrogen content multiplied by 6.25). Fat detection: Use the Soxhlet extraction method to extract the crude fat in the sample, and calculate the fat content using the residual amount after solvent evaporation; or directly use NIR for rapid detection to achieve accurate analysis. Fiber content: Use the acid-base digestion method (Weende method) to detect crude fiber content; Energy value: Based on the chemical composition results, use a near-infrared prediction model or calculation formula (such as the total digestible nutrient method) to estimate metabolizable energy. Minerals: Use flame photometry to determine the potassium and sodium content, and atomic absorption spectrometry to determine minerals such as calcium and phosphorus; Vitamins: Use spectrophotometry to determine the content of vitamin A, vitamin E, and vitamin C in the sample.

[0052] Enzyme-linked immunosorbent assay (ELISA) was used to quantitatively detect immune-enhancing factors such as β-glucans and nucleotides in the samples. Standard curves were used to calibrate the results to ensure data accuracy and reliability. Carotenoid and vitamin E content in the samples was determined by high-performance liquid chromatography (HPLC). The peak areas of antioxidant compounds were compared with those of standards to calculate concentration levels. Heavy metals such as lead, cadmium, arsenic, and mercury were detected using inductively coupled plasma mass spectrometry (ICP-MS). Samples were subjected to wet digestion (e.g., nitric acid-hydrogen peroxide system) to ensure complete release of heavy metals. Aflatoxins, ochratoxins, and vomitoxin were analyzed by gas chromatography-mass spectrometry (GC-MS). Internal standard calibration was used to improve detection sensitivity and accuracy. Phytate and tannin content were determined by high-performance liquid chromatography (HPLC). After appropriate extraction, the samples were separated by HPLC and quantitatively analyzed using standards.

[0053] All test results are collated and entered into a data management system, categorized into three categories: general nutrition, functional nutrition, and harmful substances. Statistical methods (such as mean and variance analysis) are used to verify the data to ensure the repeatability and reliability of the test results.

[0054] S3: Construct a multi-factor synergistic toxicity assessment model, calculate the toxicity contribution of a single toxin based on the toxic equivalent factor, use the matrix interaction analysis method to simulate the synergistic effects of multiple harmful substances, and output the synergistic toxicity index.

[0055] Determine the Toxicity Equivalency Factor (TEF). The toxicity equivalence factor represents the relative toxicity intensity of a substance relative to a reference substance (such as aflatoxin B1). It is derived from authoritative databases or experimental results. TEF example: If a toxin has a TEF of 2, it is twice as toxic as the reference substance. The formula for calculating the toxicity contribution of a single toxin is: T i =C i ×TEF i Where, T i is the toxicity contribution value of toxin i, C i is the concentration of toxin i (e.g. mg / kg); TEF i is the toxicity equivalent factor of toxin i, and T is calculated for each toxin i. i And summarize to form a list of single toxin toxicity values, define the synergy matrix M, where the element M ij represents the synergistic effect between substances i and j; M ij >0: synergistic effect, indicating that the toxicity of the two substances is enhanced; M ij =0: no synergistic effect, indicating that the toxicity of the two substances is independent; M ij <0: Antagonistic effect, indicating that the toxicity of the two substances is weakened. The synergistic effect calculation formula is: T total is the comprehensive toxicity value, T i is the toxicity contribution value of a single toxin i, n is the total number of toxin types, and the normalized comprehensive toxicity value is: STI is the synergistic toxicity index, ranging from 0 to 100; T max The maximum toxicity value is used as a reference for normalization.

[0056] S4: Evaluate the overall quality of feed samples based on conventional nutrition, functional nutrition and synergistic toxicity index. Based on the evaluation results, classify unconventional feed into three levels: high quality, qualified and unqualified. Provide corresponding application recommendations for each level, including scope of application and risk control measures.

[0057] Collect ingredient test results, including conventional nutrition, functional nutrition and synergistic toxicity index, and normalize each index so that its value is between 0 and 1: X norm is the normalized index value; X is the original index value; X min and X max are the minimum and maximum values ​​of the indicator.

[0058] The weights are set according to the importance of the indicators: general nutrition indicator weight: 50%; functional nutrition indicator weight: 30%; synergistic toxicity index weight: 20%. The comprehensive score calculation formula is: Q = 0.5·Q 常规营养 +0.3·Q 功能营养 -0.2 STI norm ; Where, Q is the comprehensive quality score of feed sample; Q 常规营养 It is a comprehensive score of conventional nutritional indicators. W i is the weight of the i-th nutritional indicator; N i is the normalized value of the i-th nutritional indicator; b is the amount of conventional nutritional indicators (such as protein, fat, etc.). The weights can be set according to actual needs. For example: protein: W1 = 0.4, fat: W2 = 0.2, fiber: W3 = 0.2, minerals: W4 = 0.1, vitamins: W5 = 0.1. Q 功能营养 It is a comprehensive score of functional nutrition index; STI norm Q is the normalized synergistic toxicity index. 功能营养 is the comprehensive score of functional nutrition indicators, expressed as: W j is the weight of the jth functional nutrition index, F jis the normalized value of the jth functional nutritional indicator; m is the number of functional nutritional indicators (such as immune-enhancing factors, antioxidants, etc.). Weights can be assigned based on the importance of the functional nutritional indicators. For example: β-glucan: W1 = 0.6; nucleotides: W2 = 0.2; antioxidants: W3 = 0.2.

[0059] Feeds are divided into three grades based on the range of the comprehensive score (Q): High Quality (Q ≥ 0.8): Comprehensive nutritional profile, strong functionality, and extremely low synergistic toxicity; Acceptable (Q ≤ 0.5 < 0.8): Meets basic nutritional requirements but slightly deficient in functionality or toxicity; Unacceptable (Q < 0.5): Low nutritional value and high synergistic toxicity, making it unsuitable for use. Each feed sample is labeled with the grading results and the basis for the grading, including general nutritional, functional nutritional, and synergistic toxicity data.

[0060] Application recommendations and risk control measures include: High-quality feed (Q≥0.8) applicable scope: high-end farms, efficient production models; risk control measures: only routine quality monitoring is required, no special risk intervention is required.

[0061] Scope of application of qualified feed (0.5≤Q<0.8): ordinary breeding environment, can be used as regular feed; risk control measures: limit the usage ratio, avoid long-term single feeding, and strengthen the dynamic monitoring of synergistic toxicity indicators.

[0062] Scope of application of unqualified feed (Q<0.5): Direct use is not recommended; Risk control measures: It is recommended to abandon it or optimize the raw materials, and combine it with toxicity reduction treatment (such as adsorbents or detoxification processes) when necessary.

[0063] The output assessment report includes: Quality Grade: Indicates the grade of the sample (high quality, qualified, or unqualified). Data Support: Provides detailed test data and the calculation process for the comprehensive score. Recommendations and Measures: Specific usage recommendations and risk control plans are provided for different sample grades.

[0064] S5: Build a cloud-based database to record the evaluation and grading data of unconventional feeds in real time, and use IoT sensors to monitor feed quality fluctuations in the raw material supply chain in real time, automatically identify abnormal samples and trigger early warnings.

[0065] The system will evaluate the data of non-conventional feeds, including Q 常规营养 , Q 功能营养 And STI are uploaded to the cloud database through the network interface. Each data includes: sample number ID, timestamp T, conventional nutrition score Q 常规营养 , functional nutritional score Q 功能营养 , Synergistic Toxicity Index STI, grading result L (high quality, qualified, unqualified).

[0066] Data is stored in relational database or NoSQL database: Table structure example: SampleData(ID, T, Q 常规营养 ,Q 功能营养 ,STI,L).

[0067] IoT sensors collect real-time raw material characteristic data R at key nodes in the supply chain: temperature R T , humidity R H 、Gas concentration (such as volatile organic compounds R VOC ), weight R w The data is transmitted to the cloud via wireless network for real-time analysis.

[0068] Assume that the historical range of raw material characteristic data is [R i,min ,R i,max ], real-time data R i The abnormal judgment conditions are: R i is a characteristic data collected in real time; R i,min 、R i,max The historical upper and lower limits of the characteristic data (set based on historical data or industry standards). An output value of 1 indicates abnormality, and 0 indicates normality.

[0069] When multiple data indicators are collected simultaneously, a comprehensive scoring method is used to determine the degree of sample abnormality: Where a is the total number of monitoring indicators, R i,mean is the historical mean of the characteristic data, R i,std is the standard deviation of the characteristic data; an early warning is triggered when the anomaly score is higher than the set threshold (such as 3).

[0070] When the abnormal conditions are met, the system automatically generates an early warning message and sends it to the management end: early warning message = {timestamp, ID, abnormality type, abnormality score}; after receiving it, the management end can take measures, such as suspending the use of raw materials or further analyzing the cause of the abnormality.

[0071] Record real-time monitoring results and abnormal conditions to the cloud database and update historical data range [R i,min ,R i,max ] and statistical parameters (such as mean and standard deviation).

[0072] Update history includes: max(R i ,R i,max ); where The updated historical minimum value indicates that the current real-time data R i and the historical minimum R i,min After comparison, the smaller value is taken to dynamically adjust the historical lower limit range to reflect new trends in the data. The updated historical maximum value indicates that the current real-time data R i The historical maximum R i,max After comparison, the larger value is taken to dynamically adjust the historical upper limit range and capture new peak changes in the data; update statistical parameters: Where R i,k is the value of the i-th characteristic data in the k-th sample, G is the number of new samples, is the updated historical mean, which represents the average value of the characteristic data in G historical samples; is the updated historical standard deviation, which indicates the degree of fluctuation of the characteristic data relative to the mean in G samples.

[0073] In this embodiment, first, samples are collected from industrial by-products, agricultural by-products and wild plant raw materials, and pretreated by low-temperature drying, fine grinding and three-dimensional mixing technology; then, the samples are tested for composition, covering conventional nutrition (such as protein, fat), functional nutrition (such as immune enhancing factors, antioxidants) and harmful substances (such as heavy metals, toxins, anti-nutritional factors), and accurate analysis is performed using ICP-MS, GC-MS and HPLC technologies; then, a multi-factor synergistic toxicity assessment model is constructed to calculate the toxic contribution of a single toxin, and the synergistic effect is simulated using the matrix interaction analysis method to output a synergistic toxicity index; then, the overall quality of the sample is evaluated by combining conventional nutrition, functional nutrition and synergistic toxicity index, and the feed is divided into three grades: high quality, qualified and unqualified, and application suggestions and risk control measures are provided; finally, a cloud database is constructed to record evaluation and grading data in real time, and IoT sensors are combined to monitor feed quality fluctuations in the supply chain, automatically identify anomalies and trigger early warnings to ensure dynamic management and risk control of feed quality.

[0074] Example 2, please refer to Figure 2 As shown, the unconventional feed quality assessment and grading system described in this embodiment includes a sample collection and pretreatment module, a component detection and analysis module, a collaborative toxicity assessment module, a quality assessment and grading module, and a management module:

[0075] Sample collection and pretreatment module: Unconventional feed samples are collected from industrial by-products, agricultural by-products and wild plant materials. The samples are pretreated using low-temperature drying technology, crushed to particles less than 1 mm in diameter using high-precision crushing equipment, and evenly mixed using a three-dimensional mixer;

[0076] Component detection and analysis module: Component detection is performed on pre-treated samples. Component detection includes three categories of indicators: general nutrition, functional nutrition, and harmful substances. Protein, fat, immune-enhancing factors, and antioxidant components in the samples are detected respectively. For heavy metals, toxins, and anti-nutritional factors, ICP-MS, GC-MS, and HPLC technologies are used to accurately determine their content.

[0077] Synergistic toxicity assessment module: Constructs a multi-factor synergistic toxicity assessment model, calculates the toxicity contribution of a single toxin based on the toxicity equivalent factor, uses matrix interaction analysis to simulate the synergistic effect of multiple hazardous substances, and outputs a synergistic toxicity index;

[0078] Quality Assessment and Grading Module: Evaluates the overall quality of feed samples based on conventional nutrition, functional nutrition, and synergistic toxicity index. Based on the assessment results, unconventional feeds are classified into three levels: high quality, qualified, and unqualified. Corresponding application recommendations are provided for each level, including scope of application and risk control measures.

[0079] Management module: Build a cloud database to record the evaluation and grading data of unconventional feeds in real time, and use IoT sensors to monitor feed quality fluctuations in the raw material supply chain in real time, automatically identify abnormal samples and trigger early warnings.

[0080] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0081] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0082] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0083] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A method for evaluating and grading the quality of unconventional feeds, characterized by: The following steps are involved: S1: Unconventional feed samples were collected from industrial by-products, agricultural by-products and wild plant materials. The samples were pre-treated using low-temperature drying technology, crushed to a particle diameter of less than 1 mm using high-precision crushing equipment, and evenly mixed using a three-dimensional mixer; S2: Component testing is performed on the pre-treated samples. Component testing includes three categories of indicators: general nutrition, functional nutrition, and harmful substances. Protein, fat, immune-enhancing factors, and antioxidant components in the samples are tested respectively. For heavy metals, toxins, and anti-nutritional factors, their content is accurately determined using ICP-MS, GC-MS, and HPLC techniques. S3: Construct a multi-factor synergistic toxicity assessment model, calculate the toxicity contribution of a single toxin based on the toxicity equivalent factor, use the matrix interaction analysis method to simulate the synergistic effect of multiple harmful substances, and output the synergistic toxicity index; Specifically: determine the toxicity equivalent factor TEF, which represents the relative toxicity intensity of a substance to a reference substance. The calculation formula for the toxicity contribution of a single toxin is: T i =C i ×TEF i Where, T i is the toxicity contribution value of toxin i, C i is the concentration of toxin i; TEF i is the toxicity equivalent factor of toxin i, and T is calculated for each toxin i. i And summarize to form a list of single toxin toxicity values, define the synergy matrix M, where the element M ij represents the synergistic effect between substances i and j; M ij >0: synergistic effect, indicating that the toxicity of the two substances is enhanced; M ij =0: no synergistic effect, indicating that the toxicity of the two substances is independent; M ij <0: Antagonistic effect, indicating that the toxicity of the two substances is weakened. The synergistic effect calculation formula is: T total is the comprehensive toxicity value, T i is the toxicity contribution value of a single toxin i, n is the total number of toxin types, and the normalized comprehensive toxicity value is: STI is the synergistic toxicity index, ranging from 0 to 100; T max It is the reference maximum toxicity value used for normalization; S4: Evaluate the overall quality of feed samples based on conventional nutrition, functional nutrition, and synergistic toxicity indices. Based on the evaluation results, classify unconventional feeds into three levels: high quality, qualified, and unqualified. Provide corresponding application recommendations for each level, including scope of application and risk control measures. S5: Build a cloud-based database to record the evaluation and grading data of unconventional feeds in real time. Use IoT sensors to monitor feed quality fluctuations in the raw material supply chain in real time, automatically identify abnormal samples, and trigger alerts. Specifically: IoT sensors collect real-time raw material characteristic data R at supply chain nodes, including: temperature R T , humidity R H , Volatile organic compounds R VOC , weight R w ; Assume that the historical range of raw material characteristic data is [R i,min ,R i,max ], real-time data R i The abnormal judgment conditions are: R i is a characteristic data collected in real time; R i,min 、R i,max The historical upper and lower limits of the characteristic data. The output value 1 indicates abnormality and 0 indicates normality. When multiple data indicators are collected simultaneously, a comprehensive scoring method is used to determine the degree of sample abnormality: Where a is the total number of monitoring indicators, R i,mean is the historical mean of the characteristic data, R i,std is the standard deviation of the feature data; the anomaly score is higher than the set threshold; Record real-time monitoring results and abnormal conditions to the cloud database and update the historical data range R i,min ,R i,max ] and mean and standard deviation; Update history includes: Where, The updated historical minimum value indicates that the current real-time data R i and the historical minimum R i,min After comparison, take the smaller value. The updated historical maximum value indicates that the current real-time data R i The historical maximum R i,max After comparison, take the larger value; Update statistical parameters: Where R i,k is the value of the i-th characteristic data in the k-th sample, G is the number of new samples, is the updated historical mean, which represents the average value of the characteristic data in G historical samples; R i,newstd is the updated historical standard deviation, which indicates the degree of fluctuation of the characteristic data relative to the mean in G samples.

2. The method for evaluating and grading the quality of unconventional feed according to claim 1, wherein: In S4, the overall quality of the feed sample is evaluated based on general nutrition, functional nutrition and synergistic toxicity index, specifically: Collect ingredient test results, including conventional nutrition, functional nutrition and synergistic toxicity index, and normalize each index so that its value is between 0 and 1: X norm is the normalized index value; X is the original index value; X min and X max The minimum and maximum values ​​of the indicators are set; weights are set according to the importance of the indicators: weight of conventional nutrition indicators: 50%; weight of functional nutrition indicators: 30%; Synergistic toxicity index weight: 20%; The comprehensive score calculation formula is: Q = 0.5 Q 常规营养 +0.3·Q 功能营养 -0.2 STI norm .

3. The method for evaluating and grading the quality of unconventional feed according to claim 1, wherein: In S4, Q is the comprehensive quality score of the feed sample; Q 常规营养 It is a comprehensive score of conventional nutritional indicators. W i is the weight of the i-th nutritional indicator; N i is the normalized value of the i-th nutritional index; b is the number of conventional nutritional indicators; Q 功能营养 It is a comprehensive score of functional nutrition index; STI norm is the normalized synergistic toxicity index; Q 功能营养 is the comprehensive score of functional nutrition indicators, expressed as: W j is the weight of the jth functional nutrition index, F j is the normalized value of the jth functional nutritional index; m is the number of functional nutritional indicators.

4. The method for evaluating and grading the quality of unconventional feed according to claim 3, wherein: In S4, based on the evaluation results, unconventional feeds are divided into three levels: high quality, qualified and unqualified. Specifically: According to the range of the comprehensive score Q, the feed is divided into three levels: high quality Q≥0.8: comprehensive nutritional ingredients, strong functionality, and low synergistic toxicity; qualified 0.5≤Q<0.8: nutritional indicators meet the requirements, but functionality or toxicity is insufficient; unqualified Q<0.5: low nutritional value and high synergistic toxicity, not suitable for use.

5. A non-conventional feed quality assessment and grading system, for implementing the non-conventional feed quality assessment and grading method according to any one of claims 1 to 4, characterized in that: It includes sample collection and pretreatment module, component detection and analysis module, collaborative toxicity assessment module, quality assessment and grading module and management module: Sample collection and pretreatment module: Unconventional feed samples are collected from industrial by-products, agricultural by-products and wild plant materials. The samples are pretreated using low-temperature drying technology, crushed to particles less than 1 mm in diameter using high-precision crushing equipment, and evenly mixed using a three-dimensional mixer; Component detection and analysis module: Component detection is performed on pre-treated samples. Component detection includes three categories of indicators: general nutrition, functional nutrition, and harmful substances. Protein, fat, immune-enhancing factors, and antioxidant components in the samples are detected respectively. For heavy metals, toxins, and anti-nutritional factors, ICP-MS, GC-MS, and HPLC technologies are used to accurately determine their content. Synergistic toxicity assessment module: Constructs a multi-factor synergistic toxicity assessment model, calculates the toxicity contribution of a single toxin based on the toxicity equivalent factor, uses matrix interaction analysis to simulate the synergistic effect of multiple hazardous substances, and outputs a synergistic toxicity index; Quality Assessment and Grading Module: Evaluates the overall quality of feed samples based on conventional nutrition, functional nutrition, and synergistic toxicity index. Based on the assessment results, unconventional feeds are classified into three levels: high quality, qualified, and unqualified. Corresponding application recommendations are provided for each level, including scope of application and risk control measures. Management module: Build a cloud database to record the evaluation and grading data of unconventional feeds in real time, and use IoT sensors to monitor feed quality fluctuations in the raw material supply chain in real time, automatically identify abnormal samples and trigger early warnings.

Images