Feed online production intelligent monitoring method and system based on big data analysis
Through big data analysis technology, an intelligent monitoring system for online feed production is built, which solves the problem of lack of real-time monitoring in traditional feed production, and realizes efficient, accurate detection and optimization of feed quality, ensuring the stability and efficiency of the production process.
Patent Information
- Application Number
- CN202410883373.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-03
- Publication Date
- 2024-11-26
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The lack of real-time monitoring methods in the production process of traditional feeds leads to unstable feed quality and difficulty in time to discover and correct problems in production, affecting animal health and breeding benefits.
Adopt the online feed production intelligent monitoring system based on big data analysis, including feed sampling and detection module, data processing and storage module, big data analysis module and comprehensive evaluation module. Through automatic sampling, data preprocessing, algorithm model construction and real-time analysis, it calculates The particle size coefficient, moisture content coefficient and nutritional content coefficient generate a comprehensive quality index to achieve real-time monitoring and optimization.
Improve detection efficiency and accuracy, ensure the consistency and reliability of product quality, timely discover production abnormalities, improve the flexibility and response speed of the production process, realize real-time monitoring and optimization of feed quality, and ensure every batch The feed meets preset quality standards.
Smart Images

Figure CN119027036A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of feed production monitoring, and in particular to an online feed production intelligent monitoring method and system based on big data analysis. Background Art
[0002] In the modern feed production process, the quality of feed directly affects the breeding benefits and animal health. High-quality feed can not only increase the growth rate and enhance the immunity of animals, but also improve the cost-effectiveness of feeding. Therefore, ensuring that the feed meets the nutritional standards and safety requirements is one of the core tasks of feed production companies. With the development of information technology and big data analysis technology, online production intelligent monitoring systems based on big data analysis have gradually become the mainstream in feed production.
[0003] In the current traditional feed production process, due to the lack of effective real-time monitoring methods, it is often difficult to detect and correct problems in production in a timely manner, resulting in unstable feed quality. For example, problems such as uneven feed mixing, excessive or low moisture content, and substandard nutrients will not only affect the taste and palatability of the feed, but may also have a negative impact on the health of the animals. Traditional detection methods mainly rely on offline detection and manual judgment, which are prone to problems such as detection lag and inaccurate data. It is not easy to meet the high efficiency and high standards of modern feed production, and it is not easy to comprehensively analyze the problems caused by the feed re-mixing process. Summary of the invention
[0004] In view of the deficiencies in the prior art, the present invention provides an intelligent monitoring method and system for online feed production based on big data analysis, which solves the problems mentioned in the background technology.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: including a feed sampling and detection module, a data processing and storage module, a big data analysis module, a data calculation module and a comprehensive evaluation module;
[0006] The feed sampling and detection module is used to install an automatic sampler on the feed mixer. After the feed is mixed, the feed is automatically sampled and tested, and the feed moisture data, feed nutrition data and particle size distribution data are detected, and then summarized to generate a comprehensive feed mixing data set;
[0007] The data processing and storage unit is used to transmit the collected feed comprehensive mixing data set to the data processing and storage unit through a network communication protocol for preprocessing, and to construct a database to integrate and store the preprocessed feed comprehensive mixing data set;
[0008] The big data analysis module is used to construct a big data analysis model, and to set a particle size algorithm model, a moisture content algorithm model, and a nutrient composition algorithm model through the big data analysis model;
[0009] The data calculation module is used to import the feed comprehensive mixing data set in the database into the relevant algorithm model, calculate and obtain the particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, and then associate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY to obtain the comprehensive quality index QI;
[0010] The comprehensive evaluation module is used to preset the stirring threshold J and conduct preliminary comparative evaluation with the obtained particle size coefficient KL, analyze the preliminary stirring condition of the feed, and then preset the quality threshold Pz based on the actual detection data and historical data, and then conduct a secondary evaluation with the obtained comprehensive quality index QI to analyze the comprehensive quality of the feed in the stirring production process.
[0011] Preferably, the feed sampling and detection module includes a sampling and detection unit and a data acquisition unit;
[0012] The sampling and detection unit automatically samples the stirred feed through an automatic sampler installed near the discharge port of the feed mixer, and after sampling, the sampled feed is scanned by a laser particle size analyzer and an infrared spectrum to obtain particle size distribution data and feed nutritional data, and then the scanned sampled feed is weighed, and dried after weighing, and the data before and after drying of the sampled feed are recorded respectively and summarized into feed moisture data;
[0013] The data collection unit is used to collect the detected feed moisture data, feed nutrition data and particle size distribution data in real time, and summarize the collected feed moisture data, feed nutrition data and particle size distribution data to generate a feed comprehensive mixing data set, wherein the feed comprehensive mixing data set includes a feed moisture data set, a feed nutrition data set and a particle size distribution data set;
[0014] The feed moisture data set includes the sampled feed wet weight Sz and the sampled feed dry weight Gz;
[0015] The feed nutrition data set includes a sample weight W, fat weight Zf, nitrogen content N and fiber weight big data analysis module w;
[0016] The particle size distribution data set includes a particle size distribution rate Fb and a particle diameter Fz.
[0017] Preferably, the data processing and storage module includes a data preprocessing unit and a data storage unit;
[0018] The data preprocessing unit is used to transmit the collected feed comprehensive mixing data set to the data preprocessing unit through the network communication IP protocol for preprocessing, and the preprocessing method includes processing missing values, processing abnormal values, data integration and normalization processing;
[0019] The data storage unit is used to construct a relational database MySQL, store the preprocessed feed comprehensive mixing data set in the database, divide the relational database MySQL into three different data models, and integrate data from different data sources into a unified data model.
[0020] Preferably, the big data analysis module is used to collect historical data on particle size, moisture content and nutritional components of various feed samples, and then construct a particle size algorithm model, a moisture content algorithm model and a nutritional component algorithm model according to demand, and then parameterize the constructed algorithm model, and use historical data to train the constructed particle size algorithm model, moisture content algorithm model and nutritional component algorithm model.
[0021] Preferably, the data calculation module includes a particle size analysis algorithm unit, a moisture content algorithm unit and a nutrient component determination algorithm unit;
[0022] The particle size analysis algorithm unit is used to measure the feed sample after mixing using a laser particle size analyzer to obtain a particle size distribution data set, and then based on the particle size algorithm model established by big data analysis, the particle size distribution data set collected in real time is transmitted to the particle size algorithm model for analysis and calculation to obtain the particle size coefficient KL;
[0023] The particle size coefficient KL is calculated and obtained by the following particle size algorithm model;
[0024]
[0025] In the formula, i represents the i-th particle, Fz i represents the diameter of the corresponding i-th particle, Fb i represents the diameter of the i-th particle Fz i The percentage.
[0026] Preferably, the moisture content algorithm unit constructs a moisture content algorithm model based on big data, and then inputs the preprocessed feed moisture data set into the moisture content algorithm model to analyze and calculate the moisture content of the sampled feed to obtain the moisture content coefficient SF;
[0027] The moisture content coefficient SF is calculated and obtained by the following moisture content algorithm model;
[0028]
[0029] In the formula, a1 and a2 represent the preset weight values of the sampled feed wet weight Sz and the sampled feed dry weight Gz, respectively, and a1+a2=1, and the specific values are set by the user according to actual conditions.
[0030] Preferably, the nutrient component determination algorithm unit constructs a nutrient component algorithm model based on big data, and inputs the preprocessed feed nutrient data set into the nutrient component algorithm model, analyzes and calculates the nutrient components of the sampled feed, and obtains the nutrient component content coefficient YY;
[0031] The nutrient content coefficient YY is obtained through the following nutrient algorithm model;
[0032]
[0033] Preferably, the comprehensive evaluation module includes a comprehensive quality calculation unit, a first evaluation unit and a second evaluation unit;
[0034] The comprehensive quality calculation unit is used to calculate and associate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY after dimensionless processing to obtain a comprehensive quality index QI;
[0035] The comprehensive quality index QI is obtained by the following algorithm formula:
[0036]
[0037] Wherein, b1, b2 and b3 represent the preset weight values of the particle size coefficient KL, the nutrient content coefficient YY and the moisture content coefficient SF respectively, and b1+b2+b3=1, the specific value is set by the user, and A represents the correction parameter.
[0038] Preferably, the first evaluation unit is used to preset the stirring threshold J according to the particle size of the historical feed after stirring, and then make a preliminary comparative evaluation with the obtained particle size coefficient KL to analyze the stirring condition of the current sampled feed. The specific evaluation scheme is as follows;
[0039] When the particle size coefficient KL> stirring threshold J, it means that the current sampled feed is not stirred evenly and there are large particles. At this time, a comprehensive quality assessment is further performed;
[0040] When the particle coefficient KL ≤ stirring threshold J, it means that the current sampled feed is stirred evenly;
[0041] The second evaluation unit is used to perform a secondary comparative evaluation of the preset quality threshold value Pz and the obtained comprehensive quality index QI when detecting that the current sampled feed is unevenly stirred, and further evaluate and analyze the comprehensive quality of the feed. The specific evaluation scheme is as follows;
[0042] When the comprehensive quality index QI>quality threshold Pz, it means that although the current sampled feed is not mixed evenly, the moisture content is sufficient, resulting in normal fusion of the feed raw materials and normal nutritional components. At this time, the current feed is marked as "first-level feed";
[0043] When the comprehensive quality index QI = quality threshold Pz, it means that the current sampled feed is not mixed evenly because of insufficient water content. At this time, it is marked as a "secondary feed" label, and the first warning information is generated to remind the relevant staff to add water content in the mixer in proportion;
[0044] When the comprehensive quality index QI is less than the quality threshold Pz, it means that the current sampled feed is not mixed evenly and the nutritional content is abnormal. At this time, a "level 3 feed" label is generated and a second warning message is generated to prompt the staff to improve the feed mixing process.
[0045] The feed online production intelligent monitoring method based on big data analysis includes the following steps:
[0046] S1. First, an automatic sampler is installed on the feed mixer. After the feed is mixed, the feed is automatically sampled and tested, and the feed moisture data, feed nutrition data and particle size distribution data are detected, and the data are summarized to generate a comprehensive feed mixing data set;
[0047] S2, transmitting the collected feed comprehensive mixing data set to the data processing and storage unit through a network communication protocol for preprocessing, and constructing a database to integrate and store the preprocessed feed comprehensive mixing data set;
[0048] S3. Build a big data analysis model, and use the big data analysis model to set a particle size algorithm model, a moisture content algorithm model, and a nutrient composition algorithm model;
[0049] S4, bringing the feed comprehensive mixing data set in the database into the relevant algorithm model, respectively, to calculate and obtain the particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, and then to correlate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY to obtain the comprehensive quality index QI;
[0050] S5. Finally, the preset stirring threshold J is compared with the obtained particle size coefficient KL for preliminary evaluation, and the preliminary stirring condition of the feed is analyzed. Then, based on the actual detection data and historical data, the quality threshold Pz is preset, and then a secondary evaluation is performed with the obtained comprehensive quality index QI to analyze the comprehensive quality of the feed during the mixing production process.
[0051] The present invention provides an intelligent monitoring method and system for online feed production based on big data analysis, which has the following beneficial effects:
[0052] (1) The system uses the feed sampling and detection module to install an automatic sampler on the feed mixer, and automatically performs sampling and detection after mixing. This process greatly reduces the need for manual intervention, making the sampling and detection process more efficient. In addition, through advanced technical means such as laser particle size analyzer and infrared spectrum scanning, the system can quickly and accurately obtain the moisture content, nutritional components and particle size distribution data of the feed, thereby ensuring the high accuracy of the test results. This intelligent detection process effectively improves the detection efficiency and accuracy of the feed production process and ensures the consistency and reliability of product quality.
[0053] (2) Through the data processing and storage module, the system can pre-process, store and manage the collected feed comprehensive mixing data set, and build a variety of algorithm models based on big data analysis, including particle size algorithm model, moisture content algorithm model and nutrient content algorithm model. These algorithm models can analyze production data in real time, calculate the particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, and further calculate the comprehensive quality index QI. Through these real-time analyses and calculations, the system can promptly detect abnormal conditions in the production process and make corresponding adjustments and optimizations to ensure that each batch of feed meets the preset quality standards. This real-time monitoring and optimization capability greatly improves the flexibility and response speed of the production process.
[0054] (3) The system uses a comprehensive evaluation module to compare and evaluate the preset stirring threshold J and quality threshold Pz with the actual detected particle size coefficient KL and comprehensive quality index QI, thereby comprehensively analyzing the stirring condition and comprehensive quality of the feed. Through preliminary comparative evaluation, the system can quickly determine whether the stirring of the current sampled feed is normal, and if uneven stirring is found, further comprehensive quality evaluation is performed to determine the quality grade of the feed. In addition, the system also has an early warning mechanism. When feed quality problems are detected, corresponding early warning information will be generated to prompt staff to make necessary adjustments and processing. This comprehensive evaluation and early warning mechanism not only ensures the production quality of the feed, but also provides a scientific basis and guidance for the continuous improvement of the production process. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 This is a schematic diagram of the process of the online feed production intelligent monitoring system based on big data analysis of the present invention;
[0056] Figure 2 This is a schematic diagram of the steps of the online feed production intelligent monitoring method based on big data analysis of the present invention. DETAILED DESCRIPTION
[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0058] Example 1
[0059] See also Figure 1 The present invention provides an online feed production intelligent monitoring system based on big data analysis. To achieve the above purpose, the present invention is implemented through the following technical solutions: including a feed sampling and detection module, a data processing and storage module, a big data analysis module, a data calculation module and a comprehensive evaluation module;
[0060] The feed sampling and detection module is used to install an automatic sampler on the feed mixer. After the feed is mixed, it automatically samples and detects the feed moisture data, feed nutrition data and particle size distribution data, and summarizes them to generate a comprehensive feed mixing data set;
[0061] The data processing and storage unit is used to transmit the collected feed comprehensive mixing data set to the data processing and storage unit through a network communication protocol for preprocessing, and to build a database to integrate and store the preprocessed feed comprehensive mixing data set;
[0062] The big data analysis module is used to build a big data analysis model, and to set the particle size algorithm model, the moisture content algorithm model and the nutrient composition algorithm model through the big data analysis model;
[0063] The data calculation module is used to import the feed comprehensive mixing data set in the database into the relevant algorithm model, calculate and obtain the particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, and then associate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY to obtain the comprehensive quality index QI;
[0064] The comprehensive evaluation module is used to pre-set the stirring threshold J and make a preliminary comparative evaluation with the obtained particle size coefficient KL, analyze the preliminary stirring condition of the feed, and then pre-set the quality threshold Pz based on the actual detection data and historical data, and then make a secondary evaluation with the obtained comprehensive quality index QI to analyze the comprehensive quality of the feed during the stirring production process.
[0065] In this embodiment, the system samples the mixed feed by relying on the automatic sampler installed on the feed mixer through the feed sampling and detection module, obtains feed moisture data, feed nutrition data and particle size distribution data, and generates a comprehensive feed mixing data set. This module greatly reduces the need for manual operation, improves detection efficiency and accuracy, and ensures the real-time and reliability of data. The data processing and storage module preprocesses the collected comprehensive data set, including processing missing values, outliers, data integration and normalization, and stores it in a relational database. This module provides a solid foundation for subsequent big data analysis by building and managing large-scale data sets. The big data analysis module uses these data sets to build and train a variety of algorithm models, such as particle size algorithm models, moisture content algorithm models and nutrient content algorithm models, thereby providing accurate analysis tools for the data calculation module. The data calculation module inputs the preprocessed data into these algorithm models, calculates key quality indicators, including particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, and finally generates a comprehensive quality index QI by correlation. The comprehensive evaluation module sets the stirring threshold J and quality threshold Pz, compares them with the actual calculated quality indicators, and conducts preliminary and secondary evaluations, thereby comprehensively analyzing the stirring uniformity and comprehensive quality of the feed. This module can not only quickly detect and feedback quality problems in the production process, but also provide clear quality grading and early warning information to guide production adjustments and optimization. Compared with traditional manual testing and single data analysis methods, this system significantly improves the detection speed, accuracy and comprehensive evaluation capabilities, thereby improving the overall quality and efficiency of feed production and ensuring the maximization of animal health and breeding benefits.
[0066] Example 2
[0067] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the feed sampling and detection module includes a sampling and detection unit and a data acquisition unit;
[0068] The sampling and detection unit automatically samples the stirred feed through an automatic sampler installed near the discharge port of the feed mixer. After sampling, the sampled feed is scanned by a laser particle size analyzer and an infrared spectrum to obtain particle size distribution data and feed nutritional data. The scanned sampled feed is then weighed and dried after weighing. The data before and after drying of the sampled feed are recorded and summarized into feed moisture data.
[0069] The data collection unit is used to collect the detected feed moisture data, feed nutrition data and particle size distribution data in real time, and summarize the collected feed moisture data, feed nutrition data and particle size distribution data to generate a feed comprehensive mixing data set, which includes a feed moisture data set, a feed nutrition data set and a particle size distribution data set;
[0070] The feed moisture data set includes the sampled feed wet weight Sz and the sampled feed dry weight Gz;
[0071] The feed nutrition data set includes sample weight W, fat weight Zf, nitrogen content N and fiber weight big data analysis module w;
[0072] The particle size distribution data set includes the particle size distribution rate Fb and the particle diameter Fz.
[0073] The data processing and storage module includes a data preprocessing unit and a data storage unit;
[0074] The data preprocessing unit is used to transmit the collected feed comprehensive mixing data set to the data preprocessing unit through the network communication IP protocol for preprocessing. The preprocessing methods include processing missing values, processing abnormal values, data integration and normalization processing;
[0075] The data storage unit is used to construct a relational database MySQL, store the preprocessed feed comprehensive mixing data set in the database, divide the relational database MySQL into three different data models, and integrate data from different data sources into a unified data model.
[0076] In this embodiment, the system automatically samples the stirred feed through an automatic sampler installed near the discharge port of the feed mixer through a sampling and detection unit. Then, through the application of laser particle size analyzer and infrared spectrum scanning technology, the particle size distribution and nutritional component data of the feed can be accurately obtained, and the moisture data of the feed can be accurately recorded through weighing and drying. This high-precision automated sampling and detection technology not only improves the real-time and accuracy of the data, but also reduces the error of manual operation and ensures the reliability of the feed quality data. The data acquisition unit collects and summarizes the feed moisture data, nutritional data and particle size distribution data in real time to generate a comprehensive feed mixing data set. Through the detailed record of the feed moisture data set, nutritional data set and particle size distribution data set, the system can comprehensively and meticulously grasp the physical and chemical properties of the feed. These detailed and accurate data provide a solid foundation for subsequent analysis and processing. The data processing and storage module further improves the efficiency and data processing capability of the system through the data preprocessing unit and the data storage unit. The data preprocessing unit transmits the collected feed comprehensive mixing data set to the preprocessing unit through the network communication IP protocol to process missing values, outliers, data integration and normalization to ensure data consistency and integrity. The preprocessed data is stored in the constructed relational database MySQL, and by dividing different data models, data from different data sources are integrated into a unified data model. This systematic data management method not only improves the efficiency of data storage and retrieval, but also provides high-quality data support for subsequent big data analysis. Compared with traditional manual records and single data processing methods, the system has achieved comprehensive automation and intelligence in data collection, processing and storage, greatly improving the quality control capabilities in the feed production process.
[0077] Example 3
[0078] This embodiment is explained in Example 2. Please refer to Figure 1 Specifically: The big data analysis module is used to collect historical data on particle size, moisture content and nutritional components of various feed samples, and then construct particle size algorithm model, moisture content algorithm model and nutritional component algorithm model according to demand, and then parameterize the constructed algorithm model, and use historical data to train the constructed particle size algorithm model, moisture content algorithm model and nutritional component algorithm model.
[0079] The data calculation module includes a particle size analysis algorithm unit, a moisture content algorithm unit and a nutrient component determination algorithm unit;
[0080] The particle size analysis algorithm unit is used to measure the feed sample after mixing using a laser particle size analyzer to obtain a particle size distribution data set, and then based on the particle size algorithm model established by big data analysis, the particle size distribution data set collected in real time is transmitted to the particle size algorithm model for analysis and calculation to obtain the particle size coefficient KL;
[0081] The particle size coefficient KL is calculated and obtained through the following particle size algorithm model;
[0082]
[0083] Where i represents the diameter of the i-th particle Fz i represents the diameter of the corresponding i-th particle, Fb i represents the diameter of the i-th particle Fz i The percentage.
[0084] The moisture content algorithm unit constructs a moisture content algorithm model based on big data, and then inputs the pre-processed feed moisture data set into the moisture content algorithm model to analyze and calculate the moisture content of the sampled feed to obtain the moisture content coefficient SF;
[0085] The moisture content coefficient SF is calculated by the following moisture content algorithm model;
[0086]
[0087] In the formula, a1 and a2 represent the preset weight values of the sampled feed wet weight Sz and the sampled feed dry weight Gz, respectively, and a1+a2=1, and the specific values are set by the user according to the actual situation.
[0088] The nutrient component determination algorithm unit constructs a nutrient component algorithm model based on big data, and inputs the preprocessed feed nutrient data set into the nutrient component algorithm model to analyze and calculate the nutrient components of the sampled feed to obtain the nutrient component content coefficient YY;
[0089] The nutrient content coefficient YY is obtained through the following nutrient algorithm model;
[0090]
[0091] In this embodiment, the scientificity and accuracy of feed quality detection are significantly improved through the precise algorithm model of the big data analysis module and the data calculation module. The big data analysis module collects historical data of various feed samples to construct and parameterize the particle size algorithm model, the moisture content algorithm model and the nutrient composition algorithm model. The model is trained using historical data so that the model can accurately reflect the actual situation of the feed sample. This process not only improves the efficiency of data analysis, but also enhances the decision-making support capability in the feed production process. In the data calculation module, various algorithm units perform detailed calculations on particle size, moisture content and nutrient composition respectively. The particle size analysis algorithm unit uses a laser particle size analyzer to measure the stirred feed sample, obtains the particle size distribution data, and calculates it through the particle size algorithm model to obtain the particle size coefficient KL. This coefficient is calculated by comprehensively considering different particle diameters and their proportions, accurately reflecting the distribution of feed particles. The moisture content algorithm unit is based on the wet weight and dry weight data of the sampled feed, and calculates by the set weight value to obtain the moisture content coefficient SF. This calculation method ensures the accuracy of the moisture content analysis of the feed. The nutrient content coefficient YY is obtained by analyzing the nutrient data, which fully reflects the nutrient status of the feed.
[0092] Example 4
[0093] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the comprehensive evaluation module includes a comprehensive quality calculation unit, a first evaluation unit, and a second evaluation unit;
[0094] The comprehensive quality calculation unit is used to calculate and associate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY after dimensionless processing to obtain a comprehensive quality index QI;
[0095] The comprehensive quality index QI is obtained through the following algorithm formula;
[0096]
[0097] Wherein, b1, b2 and b3 represent the preset weight values of the particle size coefficient KL, the nutrient content coefficient YY and the moisture content coefficient SF respectively, and b1+b2+b3=1, the specific value is set by the user, and A represents the correction parameter.
[0098] In this embodiment, the system can generate a comprehensive quality index QI through correlation analysis and calculation of the particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, providing a scientific basis for the evaluation of feed quality. Compared with traditional detection methods, this system significantly improves the accuracy and reliability of feed quality detection, ensuring that every link in the feed production process is carried out under strict quality control. Overall, the online feed production intelligent monitoring system based on big data analysis not only improves production efficiency and quality control levels, but also provides feed production companies with strong data support and decision-making basis, greatly improving overall production efficiency.
[0099] Example 5
[0100] This embodiment is explained in Example 4. Please refer to Figure 1 Specifically: the first evaluation unit is used to preset the stirring threshold J according to the particle size of the historical feed after stirring, and then make a preliminary comparative evaluation with the obtained particle size coefficient KL to analyze the stirring condition of the current sampled feed. The specific evaluation scheme is as follows;
[0101] When the particle size coefficient KL> stirring threshold J, it means that the current sampled feed is not stirred evenly and there are large particles. At this time, a comprehensive quality assessment is further performed;
[0102] When the particle coefficient KL ≤ stirring threshold J, it means that the current sampled feed is stirred evenly;
[0103] The second evaluation unit is used to perform a secondary comparative evaluation of the preset quality threshold value Pz and the obtained comprehensive quality index QI when detecting that the current sampled feed is unevenly stirred, and further evaluate and analyze the comprehensive quality of the feed. The specific evaluation scheme is as follows;
[0104] When the comprehensive quality index QI>quality threshold Pz, it means that although the current sampled feed is not mixed evenly, the moisture content is sufficient, resulting in normal fusion of the feed raw materials and normal nutritional components. At this time, the current feed is marked as "first-level feed";
[0105] When the comprehensive quality index QI = quality threshold Pz, it means that the current sampled feed is not mixed evenly because of insufficient water content. At this time, it is marked as a "secondary feed" label, and the first warning information is generated to remind the relevant staff to add water content in the mixer in proportion;
[0106] When the comprehensive quality index QI is less than the quality threshold Pz, it means that the current sampled feed is not mixed evenly and the nutritional content is abnormal. At this time, a "level 3 feed" label is generated and a second warning message is generated to prompt the staff to improve the feed mixing process.
[0107] In this embodiment, the system greatly improves the scientificity and accuracy of feed quality detection and evaluation through the hierarchical evaluation mechanism of the first evaluation unit and the second evaluation unit. First, the first evaluation unit can quickly determine the mixing uniformity of the feed by comparing the particle size coefficient KL with the historical mixing threshold J. If it is detected that the particle size coefficient KL is greater than the threshold J, the system will promptly identify the uneven mixing of the feed and start further comprehensive quality evaluation. This preliminary evaluation mechanism helps to timely discover and deal with problems in the mixing process to ensure the basic quality of the feed. In the further comprehensive quality evaluation, the second evaluation unit deeply analyzes the overall quality of the feed by comparing the comprehensive quality index QI with the preset quality threshold Pz. When the comprehensive quality index QI is higher than the quality threshold Pz, even if the feed is not mixed evenly, the raw materials are normally fused due to sufficient moisture content, resulting in normal nutritional components of the feed, and the system will mark the feed as "first-level feed". This classification mechanism ensures that the quality of the feed can be accurately evaluated and marked in various situations. When the comprehensive quality index QI is equal to the quality threshold Pz, the system will identify that the root cause of the problem is insufficient moisture content, and prompt the staff to adjust the moisture content to ensure the stability of the quality of the feed. When the comprehensive quality index QI is lower than the quality threshold Pz, the system will generate a "third-level feed" label and issue an early warning, prompting the staff to improve the mixing process and fundamentally solve the problem. Through such a hierarchical evaluation mechanism, the system can not only accurately detect quality problems in feed, but also provide clear directions and measures for improvement. Compared with traditional testing methods, this system significantly improves the efficiency and accuracy of feed quality testing, ensuring that each batch of feed meets high quality standards, thereby improving the overall benefits of feed production and the economic benefits of breeding. This improvement provides feed production companies with a powerful quality control tool, effectively improving their competitiveness in the market.
[0108] Example 6
[0109] See also Figure 1 and Figure 2 , the feed online production intelligent monitoring method based on big data analysis includes the following steps:
[0110] S1. First, an integrated sensor group is installed on the photovoltaic panel to detect and collect the operation data of the photovoltaic panel in real time, and the collected operation data is pre-processed;
[0111] S2. After preprocessing the operation data, extract the features of the operation data, extract the instantaneous features and power features of the operation data, and summarize them into an operation feature set;
[0112] S3. Calculate the instantaneous efficiency S(t) and fill factor FF of the photovoltaic panel according to the operation feature set, then perform dimensionless processing on the instantaneous efficiency S(t) and fill factor FF and perform correlation calculation to obtain the health index HI, and preset the first health threshold J for comparative evaluation;
[0113] S4, when an abnormality is detected, the influence of temperature on the performance of the photovoltaic panel is analyzed, and the temperature influence factor Tf is calculated. The health index HI and the temperature influence factor Tf are combined to calculate the multivariate health index HI2;
[0114] S5. Finally, based on the mean of the historical health data of the photovoltaic panels, a secondary comparative evaluation is conducted between the historical health threshold L and the obtained multivariate health index HI2, a multivariate result analysis of the health of the photovoltaic panels is performed, and a corresponding early warning strategy is generated based on the evaluation results.
[0115] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention.
Claims
1. The feed online production intelligent monitoring system based on big data analysis is characterized by: It includes feed sampling and detection module, data processing and storage module, big data analysis module, data calculation module and comprehensive evaluation module; The feed sampling and detection module is used to install an automatic sampler on the feed mixer. After the feed is mixed, the feed is automatically sampled and tested, and the feed moisture data, feed nutrition data and particle size distribution data are detected, and then summarized to generate a comprehensive feed mixing data set; The data processing and storage unit is used to transmit the collected feed comprehensive mixing data set to the data processing and storage unit through a network communication protocol for preprocessing, and to construct a database to integrate and store the preprocessed feed comprehensive mixing data set; The big data analysis module is used to construct a big data analysis model, and to set a particle size algorithm model, a moisture content algorithm model, and a nutrient composition algorithm model through the big data analysis model; The data calculation module is used to import the feed comprehensive mixing data set in the database into the relevant algorithm model, calculate and obtain the particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, and then associate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY to obtain the comprehensive quality index QI; The comprehensive evaluation module is used to preset the stirring threshold J and conduct preliminary comparative evaluation with the obtained particle size coefficient KL, analyze the preliminary stirring condition of the feed, and then preset the quality threshold Pz based on the actual detection data and historical data, and then conduct a secondary evaluation with the obtained comprehensive quality index QI to analyze the comprehensive quality of the feed in the stirring production process.
2. The feed online production intelligent monitoring system based on big data analysis according to claim 1 is characterized by: The feed sampling and detection module includes a sampling and detection unit and a data acquisition unit; The sampling and detection unit automatically samples the stirred feed through an automatic sampling machine installed near the discharge port of the feed mixer, and after sampling, the sampled feed is scanned by a laser particle size analyzer and an infrared spectrum to obtain particle size distribution data and feed nutritional data, and then the scanned sampled feed is weighed, and dried after weighing, and the data before and after drying of the sampled feed are recorded respectively and summarized into feed moisture data; The data collection unit is used to collect the detected feed moisture data, feed nutrition data and particle size distribution data in real time, and summarize the collected feed moisture data, feed nutrition data and particle size distribution data to generate a feed comprehensive mixing data set, wherein the feed comprehensive mixing data set includes a feed moisture data set, a feed nutrition data set and a particle size distribution data set; The feed moisture data set includes the sampled feed wet weight Sz and the sampled feed dry weight Gz; The feed nutrition data set includes a sample weight W, fat weight Zf, nitrogen content N and fiber weight big data analysis module w; The particle size distribution data set includes a particle size distribution rate Fb and a particle diameter Fz.
3. The feed online production intelligent monitoring system based on big data analysis according to claim 2 is characterized in that: The data processing and storage module includes a data preprocessing unit and a data storage unit; The data preprocessing unit is used to transmit the collected feed comprehensive mixing data set to the data preprocessing unit through the network communication IP protocol for preprocessing, and the preprocessing method includes processing missing values, processing abnormal values, data integration and normalization processing; The data storage unit is used to construct a relational database MySQL, store the preprocessed feed comprehensive mixing data set in the database, divide the relational database MySQL into three different data models, and integrate data from different data sources into a unified data model.
4. The feed online production intelligent monitoring system based on big data analysis according to claim 3 is characterized by: The big data analysis module is used to collect historical data on the particle size, moisture content and nutritional components of various feed samples, and then construct a particle size algorithm model, a moisture content algorithm model and a nutritional component algorithm model according to demand, and then parameterize the constructed algorithm model, and use historical data to train the constructed particle size algorithm model, moisture content algorithm model and nutritional component algorithm model.
5. The feed online production intelligent monitoring system based on big data analysis according to claim 4 is characterized in that: The data calculation module includes a particle size analysis algorithm unit, a moisture content algorithm unit and a nutrient component determination algorithm unit; The particle size analysis algorithm unit is used to measure the feed sample after mixing using a laser particle size analyzer to obtain a particle size distribution data set, and then based on the particle size algorithm model established by big data analysis, the particle size distribution data set collected in real time is transmitted to the particle size algorithm model for analysis and calculation to obtain the particle size coefficient KL; The particle size coefficient KL is calculated and obtained by the following particle size algorithm model; In the formula, i represents the i-th particle, Fz i represents the diameter of the corresponding i-th particle, Fb i represents the diameter of the i-th particle Fz i The percentage.
6. The feed online production intelligent monitoring system based on big data analysis according to claim 5 is characterized by: The moisture content algorithm unit constructs a moisture content algorithm model based on big data, and then inputs the preprocessed feed moisture data set into the moisture content algorithm model to analyze and calculate the moisture content of the sampled feed to obtain the moisture content coefficient SF; The moisture content coefficient SF is calculated and obtained by the following moisture content algorithm model; In the formula, a1 and a2 represent the preset weight values of the sampled feed wet weight Sz and the sampled feed dry weight Gz, respectively, and a1+a2=1, and the specific values are set by the user according to actual conditions.
7. The feed online production intelligent monitoring system based on big data analysis according to claim 5 is characterized by: The nutrient component determination algorithm unit constructs a nutrient component algorithm model based on big data, and inputs the preprocessed feed nutrient data set into the nutrient component algorithm model, analyzes and calculates the nutrient components of the sampled feed, and obtains the nutrient component content coefficient YY; The nutrient content coefficient YY is obtained through the following nutrient algorithm model; 8. The feed online production intelligent monitoring system based on big data analysis according to claim 6 is characterized by: The comprehensive evaluation module includes a comprehensive quality calculation unit, a first evaluation unit and a second evaluation unit; The comprehensive quality calculation unit is used to calculate and associate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY after dimensionless processing to obtain a comprehensive quality index QI; The comprehensive quality index QI is obtained by the following algorithm formula: Wherein, b1, b2 and b3 represent the preset weight values of the particle size coefficient KL, the nutrient content coefficient YY and the moisture content coefficient SF respectively, and b1+b2+b3=1, the specific value is set by the user, and A represents the correction parameter.
9. The feed online production intelligent monitoring system based on big data analysis according to claim 8 is characterized in that: The first evaluation unit is used to preset the stirring threshold J according to the particle size of the historical feed after stirring, and then make a preliminary comparative evaluation with the obtained particle size coefficient KL to analyze the stirring condition of the current sampled feed. The specific evaluation scheme is as follows; When the particle size coefficient KL> stirring threshold J, it means that the current sampled feed is not stirred evenly and there are large particles. At this time, a comprehensive quality assessment is further performed; When the particle coefficient KL ≤ stirring threshold J, it means that the current sampled feed is stirred evenly; The second evaluation unit is used to perform a secondary comparative evaluation of the preset quality threshold value Pz and the obtained comprehensive quality index QI when detecting that the current sampled feed is unevenly stirred, and further evaluate and analyze the comprehensive quality of the feed. The specific evaluation scheme is as follows; When the comprehensive quality index QI>quality threshold Pz, it means that although the current sampled feed is not stirred evenly, the moisture content is sufficient, resulting in normal fusion of the feed raw materials, and thus the nutritional components are normal. At this time, the current feed is marked as "first-level feed"; When the comprehensive quality index QI = quality threshold Pz, it means that the current sample feed is not mixed evenly because of insufficient water content. At this time, it is marked as "secondary feed" and the first warning information is generated to remind the relevant staff to add water content in the mixer in proportion; When the comprehensive quality index QI is less than the quality threshold Pz, it means that the current sampled feed is not mixed evenly and the nutritional content is abnormal. At this time, a "level 3 feed" label is generated and a second warning message is generated to prompt the staff to improve the feed mixing process.
10. A method for intelligent monitoring of online feed production based on big data analysis, comprising the intelligent monitoring system for online feed production based on big data analysis according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. First, an automatic sampler is installed on the feed mixer. After the feed is mixed, the feed is automatically sampled and tested, and the feed moisture data, feed nutrition data and particle size distribution data are detected, and the data are summarized to generate a comprehensive feed mixing data set; S2, transmitting the collected feed comprehensive mixing data set to the data processing and storage unit through a network communication protocol for preprocessing, and constructing a database to integrate and store the preprocessed feed comprehensive mixing data set; S3. Build a big data analysis model, and use the big data analysis model to set a particle size algorithm model, a moisture content algorithm model, and a nutrient composition algorithm model; S4, bringing the feed comprehensive mixing data set in the database into the relevant algorithm model, respectively, to calculate and obtain the particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY, and then to correlate the obtained particle size coefficient KL, moisture content coefficient SF and nutrient content coefficient YY to obtain the comprehensive quality index QI; S5. Finally, the preset stirring threshold J is compared with the obtained particle size coefficient KL for preliminary evaluation, and the preliminary stirring condition of the feed is analyzed. Then, based on the actual detection data and historical data, the quality threshold Pz is preset, and then a secondary evaluation is performed with the obtained comprehensive quality index QI to analyze the comprehensive quality of the feed during the mixing production process.
Citation Information
Patent Citations
Near-infrared online feed production system
CN107280049A
Live pig feed production data optimization acquisition monitoring method
CN116993527A
Feed online production monitoring system based on big data analysis
CN117482827A
Automatic feed production control system and method
CN117970884A
Feed production control method and system and storage medium
CN118244727A
