Improved agricultural production process traceability method based on artificial intelligence

Abstract

The invention discloses an improved agricultural production process traceability method based on artificial intelligence. The method comprises the following steps: S1, collecting multi-dimensional agricultural production data; s2, performing data cleaning, noise removal, abnormal value detection and format conversion on the multi-dimensional agricultural production data; s3, dividing the multi-dimensional agricultural production data into short-term data, medium-term data and long-term data according to time attributes of the multi-dimensional agricultural production data; s4, dividing the multi-dimensional agricultural production data according to land parcels, areas and crop type space positions; s5, classifying the multi-dimensional agricultural production data according to a plurality of production stages of crops from sowing, fertilizing, watering, weeding and harvesting; s6, information of the abnormal agricultural production nodes is fed back to the management system; s7, generating a traceability report; and S8, dynamically adjusting the irrigation frequency, the fertilization amount or the equipment maintenance plan of the crops according to the feedback information. According to the invention, flexible management of the agricultural production process is realized, and the overall efficiency and quality of agricultural production are improved.

Description

Technical Field

[0001] The present invention relates to the field of agricultural technology, and in particular to an improved agricultural production process tracing method based on artificial intelligence. Background Art

[0002] In the existing technology, the data management and traceability methods of smart agricultural production processes mainly rely on traditional data collection and analysis tools, such as tracking crop growth, soil conditions, meteorological environment and equipment operating status through fixed rules or manual monitoring. Traditional methods are usually based on distributed sensor networks for data collection, and then processed manually or through simple algorithms. However, with the continuous improvement of the intelligence and automation of agricultural production, the data generated in the agricultural production process is becoming more and more complex, dynamic and diverse. The agricultural data generated by different equipment is highly heterogeneous and uncertain. Traditional processing methods are often inefficient and unable to cope with dynamic changes in data when faced with such complex data scenarios, which can easily lead to inaccurate data classification and untimely identification of abnormal nodes.

[0003] Existing agricultural data classification methods usually rely on fixed algorithm models, such as rule-based classification or simple clustering algorithms. Simple clustering algorithms are difficult to automatically adapt to the changing characteristics of data when processing heterogeneous data. When faced with emergencies in the agricultural production process (climate change, pest and disease outbreaks, equipment failures), manual intervention is often required, and it is difficult to adjust model parameters in time to reflect real-time changes. Traditional algorithms have poor robustness and are easily limited by the limitations of a single data source. They cannot effectively integrate and process multidimensional data, resulting in insufficient accuracy in classification results and abnormal node identification, which may result in missing key data and causing potential production risks.

[0004] Another shortcoming of the existing technology is that in the multi-stage process management of agricultural production, traditional methods are difficult to flexibly adapt to the needs of different production stages. Whether it is from sowing, fertilizing to harvesting, the data of each link has significant differences. Traditional processing tools cannot be dynamically adjusted according to the specific characteristics of each stage and cannot accurately identify the key characteristics of the production process, resulting in inefficient monitoring and management of the production process.

[0005] To sum up, the existing technology has the following main shortcomings: First, the traditional agricultural data classification algorithm is inefficient in processing multi-source and heterogeneous data and cannot adapt to the dynamic changes of data; second, the existing technology has limited capabilities in identifying abnormal nodes and real-time feedback, and cannot respond to emergencies in the agricultural production process in a timely manner; third, the existing production management tools cannot adapt to the multi-stage process of agricultural production, lack a flexible dynamic adjustment mechanism, and affect the management and monitoring efficiency of the production process. Summary of the invention

[0006] One purpose of the present invention is to propose an improved agricultural production process traceability method based on artificial intelligence. The present invention realizes flexible management of the agricultural production process and improves the overall efficiency and quality of agricultural production.

[0007] An improved agricultural production process traceability method based on artificial intelligence according to an embodiment of the present invention comprises the following steps:

[0008] S1. Real-time collection of multi-dimensional agricultural production data on crop growth, soil conditions, meteorological environment and equipment operation status through equipment deployed in agricultural production environments;

[0009] S2, perform data cleaning, noise removal, outlier detection and format conversion on multi-dimensional agricultural production data;

[0010] S3. According to the time attribute of multidimensional agricultural production data, the multidimensional agricultural production data is divided into short-term data, medium-term data and long-term data, and the data characteristics of different time periods are analyzed by using an adaptive clustering model;

[0011] S4. According to the spatial distribution characteristics of multidimensional agricultural production data, the multidimensional agricultural production data is divided according to the spatial locations of plots, regions and crop types, and adaptive clustering analysis is performed on the data of each spatial location, and the parameters of the adaptive clustering model are dynamically adjusted so that the classification results of the multidimensional agricultural production data reflect the crop growth, environmental conditions and production equipment status of different regions;

[0012] S5. According to the specific links of agricultural production, the multi-dimensional agricultural production data is classified according to the multiple production stages of crops from sowing, fertilization, watering, weeding and harvesting, and the data of each production stage is dynamically clustered to identify the key features in each stage, and the adaptive clustering model parameters are adjusted in real time to cope with the data changes in the agricultural production links;

[0013] S6. Automatically identify abnormal nodes in the agricultural production process based on the multi-level clustering results of steps S3 to S5, including abnormal crop growth, abnormal soil conditions, sudden changes in the meteorological environment, and equipment failures, and issue abnormal alarms in real time, and feed back information on abnormal agricultural production nodes to the management system;

[0014] S7. When an abnormality occurs in the agricultural production process, based on the multi-level clustering results of multi-dimensional agricultural production data, automatic traceability analysis is performed to determine the time, location and cause of the abnormality by tracing the data source and historical changes of the abnormal node, and generate a traceability report;

[0015] S8. Feedback abnormal nodes and their causes in traceability management to the agricultural production management system in real time, and dynamically adjust the irrigation frequency, fertilizer application amount or equipment maintenance plan of crops based on the feedback information.

[0016] Optionally, the S1 includes:

[0017] S11. Real-time collection of multi-dimensional data on crop growth through sensor networks deployed in agricultural production environments c (t), including leaf area, stem length, photosynthesis rate and nutrient content, D c is the multidimensional vector of crop growth parameters, and t is the collection time point;

[0018] S12. Use sensor equipment deployed in the soil to collect multi-dimensional data of soil conditions in real time s (t), including soil temperature, moisture, pH value and nutrient content, D s is the multidimensional vector of soil conditions;

[0019] S13. Collect meteorological environment data in real time through weather stations w (t), including air temperature, humidity, wind speed, light intensity and rainfall, D w is the multidimensional vector of meteorological environment;

[0020] S14. Real-time collection of equipment operation status data related to agricultural production equipment through the equipment operation status monitoring system e (t), including the equipment's operating current, operating voltage, temperature and vibration, D e is a multidimensional vector of the equipment operation status;

[0021] S15, integrating the multidimensional agricultural production data collected in steps S11 to S14 into multidimensional agricultural production data D(t):

[0022] D(t)={D c (t),D s (t),D w (t),D e (t)}.

[0023] Optionally, S2 includes:

[0024] S21. Clean the collected multidimensional agricultural production data D(t) to remove invalid data and duplicate data, and obtain the cleaned multidimensional agricultural production data D clean (t);

[0025] S22. Cleaned multidimensional agricultural production data D clean (t) Perform noise removal to eliminate the noise generated by sensor equipment and environmental interference, and obtain the multi-dimensional agricultural production data D after noise removal denoise (t);

[0026] S23, multidimensional agricultural production data after noise removal D denoise (t) performs outlier detection, using a distribution-based detection method to identify abnormal data points in multidimensional agricultural production data. Abnormal data points include crop growth parameters, soil conditions, and meteorological environment data that exceed a reasonable range. The outlier detection result is expressed by the function E(t):

[0027] E(t)={e1(t),e2(t),…,e n (t)};

[0028] Among them, e n (t) is an abnormal data point;

[0029] S24, analyze the abnormal data in E(t), automatically correct or delete the abnormal data, and obtain the multi-dimensional agricultural production data D after abnormal processing corrected (t);

[0030] S25, the multi-dimensional agricultural production data D after abnormal processing corrected (t) Perform format conversion, convert the multidimensional agricultural production data into a unified format according to the requirements of data analysis and clustering models, and obtain the converted multidimensional agricultural production data D formatted (t).

[0031] Optionally, the S3 includes:

[0032] S31. According to the time attribute of the multidimensional agricultural production data, the multidimensional agricultural production data D formatted (t) is divided into short-term data D short (t), mid-term data D mid (t) and long-term data D long (t):

[0033] D short (t) represents agricultural production data collected within a short time frame for real-time monitoring and feedback;

[0034] D mid (t) represents agricultural production data collected over a medium-term period for trend identification;

[0035] D long (t) represents agricultural production data collected over a long period of time for periodic pattern analysis;

[0036] S32. For short-term data D short (t) Adopting adaptive clustering model for classification analysis to identify the changing patterns of crop growth, soil conditions, meteorological environment and equipment operation status in agricultural production in the short term:

[0037]

[0038] Among them, C short (t) is the short-term clustering result, represents the i-th short-term data point, μ j (t) is the cluster center, the mean vector of the jth class, and w ij (t) is the sample The membership function belonging to the jth class represents the distance weight between the short-term data point and the cluster center;

[0039] S33. Interim data D mid (t) Trend identification: Use the adaptive clustering model to analyze the data change trend. By integrating the trend in the time interval [t0, t1], the data change trend over time is captured and used for the analysis of crop growth trend, soil nutrient fluctuation trend and meteorological change trend:

[0040]

[0041] Among them, T mid (t) is the mid-term trend identification result, represents the i-th mid-term data point, μ j (t) is the changing trend of cluster center, w ij (t) is the membership function;

[0042] S34. For long-term data D long (t) Analyze the periodicity law and use the adaptive clustering model to detect and analyze the periodic fluctuations of long-term data in agricultural production. j (t-t0)) captures cyclical fluctuations in long-term data:

[0043]

[0044] Among them, P long (t) is the result of long-term periodic law analysis, is the i-th long-term data point, μ j (t) is the cluster center, f j represents the periodic frequency, w ij (t) is the membership function.

[0045] Optionally, the S4 includes:

[0046] S41. Based on multidimensional agricultural production data formatted The spatial distribution characteristics of (t) are used to divide the multidimensional agricultural production data into spatial distributions according to plots, regions and crop types, and obtain agricultural production data D at different spatial locations. loc (t):

[0047] D loc1 (t) represents the dataset of plot 1,

[0048] D loc2 (t) represents the dataset of plot 2,

[0049] D locn (t) represents the dataset of plot n;

[0050] S42, agricultural production data D for each spatial location locn (t) Adopting adaptive clustering model for analysis, the clustering process dynamically adjusts the cluster center μ according to the different spatial positions. j (t) and membership function w ij (t), so that the adaptive clustering model reflects the crop growth, environmental conditions and production equipment status of the corresponding spatial location;

[0051] S43. Based on the differences in spatial locations, agricultural production data D loc (t) Dynamic parameter adjustment is performed to make the clustering results reflect the agricultural production characteristics of each region. The dynamic adjustment of the adaptive clustering model is achieved by minimizing the function L loc (t) Implement:

[0052]

[0053] in, represents the multidimensional agricultural production data point at the i-th spatial location, w ij (t) is a multi-dimensional agricultural production data point The membership function of cluster center j, λ is the regularization parameter, Represents the dynamic adjustment of the cluster center over time, which adapts to the changes of agricultural production data in different time periods, and captures the acceleration changes of production data at different spatial locations through the second-order derivative term;

[0054] S44. Based on the results of cluster analysis, the crop growth conditions, soil conditions, meteorological environment and production equipment status in different regions are reflected in real time, and the spatial location classification results C are generated. spatial (t):

[0055]

[0056] in, Producing data points for multi-dimensional agriculture and cluster center μ j (t) reflects the spatial difference of agricultural production data at time t. Used to describe dynamic adjustments in the time dimension, involving multi-dimensional agricultural production data points The time derivative of Reflects the growth rate of crops, β is the adjustment coefficient, and the angle parameter θ jij (t) is used to capture the relative angle changes between different cluster centers. It reflects the influence of the second-order time derivative on the cluster center and captures the acceleration change of crop growth or equipment status. δ is the spatial attenuation coefficient. Represents that with the spatial distance |x i -x j |Increasing and decaying acceleration effects.

[0057] Optionally, the S5 includes:

[0058] S51. According to the specific links of agricultural production, multi-dimensional agricultural production data D formatted (t) Classify crops according to their production stages, including sowing, fertilizing, watering, weeding and harvesting, and obtain the production stage dataset D stage (t), where D seed (t) represents the data of the sowing stage, D fertil (t) represents the data of fertilization stage, D water (t) represents the data of the watering stage, D weed (t) represents the data of weeding stage, D harvest (t) represents the data of the harvesting stage;

[0059] S52, for each production stage data set D stage (t) Adopting adaptive clustering model to conduct dynamic clustering analysis, identify key features in each production stage, and dynamically adjust the parameters of adaptive clustering model according to the changes of data in agricultural production process;

[0060] S53. Identify key features of the agricultural production process based on the clustering results of each production stage, including soil conditions in the sowing stage, nutrient requirements in the fertilization stage, humidity changes in the watering stage, weed coverage in the weeding stage, and crop maturity in the harvesting stage. j (t) and w ij (t) respond to changes in data during the production phase;

[0061] S54, real-time adjustment of the adaptive clustering model parameters at each production stage, by minimizing the objective function L stage (t) Optimize the clustering process:

[0062]

[0063] Among them, λ is the adjustment parameter, Represents the dynamic changes of cluster centers over time.

[0064] Optionally, the S6 includes:

[0065] S61, based on the multi-level clustering results of steps S3 to S5, automatically identify abnormal nodes in the agricultural production process, including abnormal crop growth, abnormal soil conditions, sudden meteorological environment changes and equipment failures, and short (t), mid-term trend identification data T mid (t) and long-term periodic data P long (t), different spatial position data C spatial (t) and data C of each production stage stage The clustering results of (t) are compared to detect deviations from the normal production mode. The result of abnormal node detection is expressed by function A(t):

[0066] A(t)={a1(t),a2(t),…,a m (t)};

[0067] Among them, aa m (t) represents the mth abnormal node. The detected abnormal nodes include abnormalities in crop growth, soil conditions, meteorological environment, and equipment status;

[0068] S62: for each abnormal node a m (t) Classify and combine various clustering results to determine the type of anomaly, including crop growth anomaly, soil condition anomaly, meteorological environment anomaly and equipment failure. According to different types of anomalies, automatically generate classification labels L m (t):

[0069]

[0070] Among them, L m (t) is the classification label of the abnormal node, They represent the variance of the cluster data at each stage, respectively. The classification process depends on the difference from the normal cluster center at each stage;

[0071] S63. Based on the classification results of abnormal nodes, combined with short-term, medium-term and long-term clustering data and clustering results of different spatial locations and production stages, the abnormality degree E is automatically calculated. m (t) and assess the degree of deviation from normal production conditions;

[0072] S64: detect abnormal node a m (t), exception type L m (t) and abnormality degree E m(t) Real-time feedback is provided to the agricultural production management system. The system generates an alarm based on the feedback information and, combined with the data clustering characteristics of different time periods, spatial locations and production stages, prompts managers to deal with abnormal situations and provides corresponding treatment suggestions based on the severity of the abnormality.

[0073] The beneficial effects of the present invention are:

[0074] (1) The present invention adopts an improved adaptive clustering algorithm that can dynamically adjust the clustering model parameters according to the changes in the time, space and production stage of agricultural production data. By dividing the multidimensional agricultural data into short-term, medium-term and long-term data, and adaptively classifying and trending the data at each stage, it can more accurately handle data changes at different time scales, overcome the limitation that traditional clustering algorithms are difficult to adapt to data variability, and in the agricultural production process, the clustering model can be updated in real time in scenarios where the meteorological environment, soil conditions and crop growth data change rapidly, ensuring the accuracy and sensitivity of data classification, and effectively avoiding the production risks caused by data lag or failure to identify abnormal points in time in traditional algorithms.

[0075] (2) The present invention automatically identifies abnormal nodes in agricultural production through a multi-level clustering model, including abnormalities in crop growth, soil conditions, meteorological environment and equipment operating status. Unlike the monitoring methods in the prior art that rely on manual or fixed rules, the improved adaptive clustering algorithm can identify abnormalities by combining the clustering results of short-term, medium-term and long-term data in real time with the data dynamics of spatial location and production stage. It can not only improve the recognition accuracy of abnormal nodes, but also classify, quantify and provide real-time feedback on abnormal situations by calculating the degree of deviation between abnormalities and normal states, generate automated alarm information, reduce manual intervention, and improve the efficiency of responding to emergencies in the agricultural production process.

[0076] (3) The present invention proposes to classify and dynamically cluster data according to different links of agricultural production (seeding, fertilizing, watering, weeding and harvesting). The data characteristics of each production stage are different, and traditional methods are difficult to cope with the differences between stages. By adjusting the clustering model parameters in real time at different production stages to make it adapt to the data characteristics of each stage, flexible management of the agricultural production process is achieved. The dynamic adjustment mechanism can ensure that the data classification of each production stage is accurate, and the production management is more targeted and timely, thereby improving the overall efficiency and quality of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0078] Figure 1A flowchart of an improved agricultural production process traceability method based on artificial intelligence proposed by the present invention;

[0079] Figure 2 This is a flow chart of dynamic clustering analysis of data at various stages of agricultural production (seeding, fertilizing, watering, weeding, and harvesting) in an improved agricultural production process traceability method based on artificial intelligence proposed by the present invention. DETAILED DESCRIPTION

[0080] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0081] refer to Figure 1-2 , an improved agricultural production process traceability method based on artificial intelligence, comprising the following steps:

[0082] S1. Real-time collection of multi-dimensional agricultural production data on crop growth, soil conditions, meteorological environment and equipment operation status through equipment deployed in agricultural production environments;

[0083] S2, perform data cleaning, noise removal, outlier detection and format conversion on multi-dimensional agricultural production data;

[0084] S3. According to the time attribute of multidimensional agricultural production data, the multidimensional agricultural production data is divided into short-term data, medium-term data and long-term data, and the data characteristics of different time periods are analyzed by using an adaptive clustering model;

[0085] S4. According to the spatial distribution characteristics of multidimensional agricultural production data, the multidimensional agricultural production data is divided according to the spatial locations of plots, regions and crop types, and adaptive clustering analysis is performed on the data of each spatial location, and the parameters of the adaptive clustering model are dynamically adjusted so that the classification results of the multidimensional agricultural production data reflect the crop growth, environmental conditions and production equipment status of different regions;

[0086] S5. According to the specific links of agricultural production, the multi-dimensional agricultural production data is classified according to the multiple production stages of crops from sowing, fertilization, watering, weeding and harvesting, and the data of each production stage is dynamically clustered to identify the key features in each stage, and the adaptive clustering model parameters are adjusted in real time to cope with the data changes in the agricultural production links;

[0087] S6. Automatically identify abnormal nodes in the agricultural production process based on the multi-level clustering results of steps S3 to S5, including abnormal crop growth, abnormal soil conditions, sudden changes in the meteorological environment, and equipment failures, and issue abnormal alarms in real time, and feed back information on abnormal agricultural production nodes to the management system;

[0088] S7. When an abnormality occurs in the agricultural production process, based on the multi-level clustering results of multi-dimensional agricultural production data, automatic traceability analysis is performed to determine the time, location and cause of the abnormality by tracing the data source and historical changes of the abnormal node, and generate a traceability report;

[0089] S8. Feedback abnormal nodes and their causes in traceability management to the agricultural production management system in real time, and dynamically adjust the irrigation frequency, fertilizer application amount or equipment maintenance plan of crops based on the feedback information.

[0090] In this implementation, S1 includes:

[0091] S11. Real-time collection of multi-dimensional data on crop growth through sensor networks deployed in agricultural production environments c (t), including leaf area, stem length, photosynthesis rate and nutrient content, D c is the multidimensional vector of crop growth parameters, and t is the collection time point;

[0092] S12. Use sensor equipment deployed in the soil to collect multi-dimensional data of soil conditions in real time s (t), including soil temperature, moisture, pH value and nutrient content, D s is the multidimensional vector of soil conditions;

[0093] S13. Collect meteorological environment data in real time through weather stations w (t), including air temperature, humidity, wind speed, light intensity and rainfall, D w is the multidimensional vector of meteorological environment;

[0094] S14. Real-time collection of equipment operation status data related to agricultural production equipment through the equipment operation status monitoring system e (t), including the equipment's operating current, operating voltage, temperature and vibration, D e is a multidimensional vector of the equipment operation status;

[0095] S15, integrating the multidimensional agricultural production data collected in steps S11 to S14 into multidimensional agricultural production data D(t):

[0096] D(t)={D c (t),D s (t),D w (t),D e (t)}.

[0097] In this implementation, S2 includes:

[0098] S21. Clean the collected multidimensional agricultural production data D(t) to remove invalid data and duplicate data, and obtain the cleaned multidimensional agricultural production data D clean (t);

[0099] S22. Cleaned multidimensional agricultural production data D clean (t) Perform noise removal to eliminate the noise generated by sensor equipment and environmental interference, and obtain the multi-dimensional agricultural production data D after noise removal denoise (t);

[0100] S23, multidimensional agricultural production data after noise removal D denoise (t) performs outlier detection, using a distribution-based detection method to identify abnormal data points in multidimensional agricultural production data. Abnormal data points include crop growth parameters, soil conditions, and meteorological environment data that exceed a reasonable range. The outlier detection result is expressed by the function E(t):

[0101] E(t)={e1(t),e2(t),…,e n (t)};

[0102] Among them, e n (t) is an abnormal data point;

[0103] S24, analyze the abnormal data in E(t), automatically correct or delete the abnormal data, and obtain the multi-dimensional agricultural production data D after abnormal processing corrected (t);

[0104] S25, the multi-dimensional agricultural production data D after abnormal processing corrected (t) Perform format conversion, convert the multidimensional agricultural production data into a unified format according to the requirements of data analysis and clustering models, and obtain the converted multidimensional agricultural production data D formatted (t).

[0105] In this implementation, S3 includes:

[0106] S31. According to the time attribute of the multidimensional agricultural production data, the multidimensional agricultural production data D formatted (t) is divided into short-term data D short (t), mid-term data D mid (t) and long-term data D long (t):

[0107] D short (t) represents agricultural production data collected within a short time frame for real-time monitoring and feedback;

[0108] D mid (t) represents agricultural production data collected over a medium-term period for trend identification;

[0109] D long (t) represents agricultural production data collected over a long period of time for periodic pattern analysis;

[0110] S32. For short-term data D short (t) Adopting adaptive clustering model for classification analysis to identify the changing patterns of crop growth, soil conditions, meteorological environment and equipment operation status in agricultural production in the short term:

[0111]

[0112] Among them, C short (t) is the short-term clustering result, represents the i-th short-term data point, μ j (t) is the cluster center, the mean vector of the jth class, and w ij (t) is the sample The membership function belonging to the jth class represents the distance weight between the short-term data point and the cluster center;

[0113] S33. Interim data D mid (t) Trend identification: Use the adaptive clustering model to analyze the data change trend. By integrating the trend in the time interval [t0, t1], the data change trend over time is captured and used for the analysis of crop growth trend, soil nutrient fluctuation trend and meteorological change trend:

[0114]

[0115] Among them, T mid (t) is the mid-term trend identification result, represents the i-th mid-term data point, μ j (t) is the changing trend of cluster center, w ij (t) is the membership function;

[0116] S34. For long-term data D long (t) Analyze the periodicity law and use the adaptive clustering model to detect and analyze the periodic fluctuations of long-term data in agricultural production. j (t-t0)) captures cyclical fluctuations in long-term data:

[0117]

[0118] Among them, P long (t) is the result of long-term periodic law analysis, is the i-th long-term data point, μ j (t) is the cluster center, fj represents the periodic frequency, w ij (t) is the membership function.

[0119] In this implementation, S4 includes:

[0120] S41. Based on multidimensional agricultural production data formatted The spatial distribution characteristics of (t) are used to divide the multidimensional agricultural production data into spatial distributions according to plots, regions and crop types, and obtain agricultural production data D at different spatial locations. loc (t):

[0121] D loc1 (t) represents the dataset of plot 1,

[0122] D loc2 (t) represents the dataset of plot 2,

[0123] D locn (t) represents the dataset of plot n;

[0124] S42, agricultural production data D for each spatial location locn (t) Adopting adaptive clustering model for analysis, the clustering process dynamically adjusts the cluster center μ according to the different spatial positions. j (t) and membership function w ij (t), so that the adaptive clustering model reflects the crop growth, environmental conditions and production equipment status of the corresponding spatial location;

[0125] S43. Based on the differences in spatial locations, agricultural production data D loc (t) Dynamic parameter adjustment is performed to make the clustering results reflect the agricultural production characteristics of each region. The dynamic adjustment of the adaptive clustering model is achieved by minimizing the function L loc (t) Implement:

[0126]

[0127] in, represents the multidimensional agricultural production data point at the i-th spatial location, w ij (t) is a multi-dimensional agricultural production data point The membership function of cluster center j, λ is the regularization parameter, Represents the dynamic adjustment of the cluster center over time, which adapts to the changes of agricultural production data in different time periods, and captures the acceleration changes of production data at different spatial locations through the second-order derivative term;

[0128] S44. Based on the results of cluster analysis, the crop growth conditions, soil conditions, meteorological environment and production equipment status in different regions are reflected in real time, and the classification results of spatial locations are generated C spatial (t):

[0129]

[0130] in, Producing data points for multi-dimensional agriculture and cluster center μ j (t) reflects the spatial difference of agricultural production data at time t. Used to describe dynamic adjustments in the time dimension, involving multi-dimensional agricultural production data points The time derivative of Reflects the growth rate of crops, β is the adjustment coefficient, and the angle parameter θ ij (t) is used to capture the relative angle changes between different cluster centers. It reflects the influence of the second-order time derivative on the cluster center and captures the acceleration change of crop growth or equipment status. δ is the spatial attenuation coefficient. Represents that with the spatial distance |x i -x j |Increasing and decaying acceleration effects.

[0131] In this implementation, S5 includes:

[0132] S51. According to the specific links of agricultural production, multi-dimensional agricultural production data D formatted (t) Classify crops according to their production stages, including sowing, fertilizing, watering, weeding and harvesting, and obtain the production stage dataset D stage (t), where D seed (t) represents the data of the sowing stage, D fertil (t) represents the data of fertilization stage, D water (t) represents the data of the watering stage, D weed (t) represents the data of weeding stage, D harvest (t) represents the data of the harvesting stage;

[0133] S52, for each production stage data set D stage (t) Adopting adaptive clustering model to conduct dynamic clustering analysis, identify key features in each production stage, and dynamically adjust the parameters of adaptive clustering model according to the changes of data in agricultural production process;

[0134] S53. Identify key features of the agricultural production process based on the clustering results of each production stage, including soil conditions in the sowing stage, nutrient requirements in the fertilization stage, humidity changes in the watering stage, weed coverage in the weeding stage, and crop maturity in the harvesting stage. j (t) and w ij (t) respond to changes in data during the production phase;

[0135] S54, real-time adjustment of the adaptive clustering model parameters at each production stage, by minimizing the objective function L stage (t) Optimize the clustering process:

[0136]

[0137] Among them, λ is the adjustment parameter, Represents the dynamic changes of cluster centers over time.

[0138] In this implementation, S6 includes:

[0139] S61, automatically identifying the existence of agricultural production processes based on the multi-level clustering results of steps S3 to S5.

[0140] In the abnormal nodes, including abnormal crop growth, abnormal soil conditions, sudden meteorological environment changes and equipment failures, by analyzing the short-term data C short (t), mid-term trend identification data T mid (t) and long-term periodic data P long (t), different spatial position data C spatial (t) and data C of each production stage stage The clustering results of (t) are compared to detect deviations from the normal production mode. The result of abnormal node detection is expressed by function A(t):

[0141] A(t)={a1(t),a2(t),…,a m (t)};

[0142] Among them, aa m (t) represents the mth abnormal node. The detected abnormal nodes include abnormalities in crop growth, soil conditions, meteorological environment, and equipment status;

[0143] S62: for each abnormal node a m (t) Classify and combine various clustering results to determine the type of anomaly, including crop growth anomaly, soil condition anomaly, meteorological environment anomaly and equipment failure. According to different types of anomalies, automatically generate classification labels L m (t):

[0144]

[0145] Among them, L m (t) is the classification label of the abnormal node, They represent the variance of the cluster data at each stage, respectively. The classification process depends on the difference from the normal cluster center at each stage;

[0146] S63. Based on the classification results of abnormal nodes, combined with short-term, medium-term and long-term clustering data and clustering results of different spatial locations and production stages, the abnormality degree E is automatically calculated. m (t) and assess the degree of deviation from normal production conditions;

[0147] S64: detect abnormal node a m (t), exception type L m (t) and abnormality degree E m (t) Real-time feedback is provided to the agricultural production management system. The system generates an alarm based on the feedback information and, combined with the data clustering characteristics of different time periods, spatial locations and production stages, prompts managers to deal with abnormal situations and provides corresponding treatment suggestions based on the severity of the abnormality.

[0148] Embodiment 1:

[0149] In this embodiment, data collection, abnormal node identification and traceability management are performed on the smart agricultural production process of an agricultural base based on an improved adaptive clustering algorithm. The base is located in City A, covers an area of ​​about 1,200 acres, and grows soybeans and corn. By deploying agricultural Internet of Things equipment, including soil sensors, weather stations and drone systems, the base can monitor crop growth, soil conditions and meteorological environment in real time, and track the operating status of agricultural equipment. The entire production process is compared and tested by the method of the present invention and the traditional method to verify the advantages of the present invention in the smart agricultural production process.

[0150] At a certain stage, the farm management system received a piece of data from the soil sensor, indicating that the soil moisture in a certain area continued to decrease. In the next 10 hours, the soil moisture data failed to return to the normal range. The system found through the clustering model that the soil moisture in the area deviated significantly from other areas, and the anomaly score reached 0.92, exceeding the threshold of 0.85 set by the system. Further analysis showed that the anomaly of the humidity data did not spread to the adjacent areas, and it was speculated that it might be due to the failure of the irrigation equipment. The system generated an anomaly report, recording the location information of the abnormal humidity area (plot ID: #045), the abnormal period from 15:30 on May 15, 2024 to 01:30 on May 16, 2024, the anomaly score was 0.92, and the predicted cause of the anomaly was: the irrigation equipment might fail. The report was sent to the agricultural management center in real time through the Internet of Things platform.

[0151] The technicians of the Agricultural Management Center went to the site to check within 15 minutes after receiving the report and found that there was indeed an irrigation device that had failed. The equipment had not been working properly in the past 12 hours, resulting in the root system of the crops not being properly irrigated for a long time. After the equipment was repaired, the soil moisture data gradually returned to normal. In contrast, the adjacent areas using traditional monitoring systems failed to detect the equipment failure in time, resulting in damage to crops, and the regional soil moisture failed to recover, affecting the growth of crops. It is expected that the yield will be reduced by about 8%.

[0152] Equipment fault detection comparison:

[0153] Implementation of the method of the present invention: At 15:30 on May 15, the system detected abnormal soil moisture of plot ID#045, with an abnormality score of 0.92. The system automatically generated a report and sent it to the management center. The technicians repaired the fault at 16:00, and the soil moisture data returned to the normal range.

[0154] Using the traditional method: The equipment failure in the same area was not discovered until the next day during manual inspection. The repair time was 11:00 on May 16, which was a 15-hour delay. Crops were severely affected by water shortage, and the regional output was expected to decrease by 8%.

[0155] Comparison of crop growth abnormality identification:

[0156] Implementation of the method of the present invention: On June 3, 2024, crop height monitoring data showed that the corn in plot ID#078 grew slowly, with the plant height being only 80% of the average height. The system found through cluster analysis that the soil nutrient content in the area was lower than that in other areas, with an abnormality score of 0.87, and it was recommended to increase the amount of fertilizer. After receiving the notification, the management personnel immediately adjusted the fertilization strategy in the area. Four days later, the growth rate of crops returned to normal.

[0157] Using traditional methods: Because the traditional system failed to detect growth abnormalities in time, it was not until the routine inspection on June 12 that it was discovered that the growth of crops in the area was significantly lagging behind. By the time remedial measures were taken, irreversible growth losses had already occurred, and the yield was expected to drop by 5%.

[0158] Comparison of abnormal responses to meteorological environment:

[0159] Implementation of the method of the present invention: On July 10, 2024, the weather station recorded heavy rainfall, with rainfall exceeding 100 mm in a short period of time. The system predicted through adaptive clustering model analysis that some low-lying areas would face the risk of waterlogging, generated an early warning and suggested that managers start the drainage system. The report included the specific location of the low-lying areas (plot ID: #011 and #017), rainfall estimates, and possible waterlogging ranges. Managers immediately took measures to prevent crops from being soaked in water.

[0160] Using traditional methods: Due to the failure to detect the risk of water accumulation in time, some areas accumulated too much water after the rain, causing the roots of crops to be flooded. The management staff only discovered this during the inspection the next day. It took nearly 10 hours to clear the water. Crops in the area were damaged and the output was expected to drop by 6%.

[0161] Table 1 Comparative data of smart agricultural production

[0162]

[0163] It can be clearly seen from this Example 1 that the present invention significantly improves the efficiency and accuracy of data processing in the process of intelligent agricultural production, especially in abnormal node identification, crop growth monitoring and abnormal event response, which can handle emergencies more quickly than traditional methods and reduce crop losses caused by equipment failure or environmental changes. The above comparative data fully demonstrates that the adaptive clustering algorithm of the present invention has strong real-time processing capabilities, flexibility and accuracy, and effectively solves the limitations of traditional methods in agricultural production data processing.

[0164] The present invention adopts an improved adaptive clustering algorithm that can dynamically adjust clustering model parameters according to changes in the time, space and production stages of agricultural production data. By dividing multidimensional agricultural data into short-term, medium-term and long-term data, and adaptively classifying and trending the data in each stage, it can more accurately handle data changes at different time scales, overcoming the limitation that traditional clustering algorithms are difficult to adapt to data variability. In the agricultural production process, the clustering model can be updated in real time in scenarios where the meteorological environment, soil conditions and crop growth data change rapidly, ensuring the accuracy and sensitivity of data classification, and effectively avoiding production risks caused by data lags or failure to identify abnormal points in time in traditional algorithms.

[0165] The present invention automatically identifies abnormal nodes in agricultural production through a multi-level clustering model, including abnormalities in crop growth, soil conditions, meteorological environment and equipment operating status. Unlike the monitoring methods in the prior art that rely on manual or fixed rules, the improved adaptive clustering algorithm can identify abnormalities by combining the clustering results of short-term, medium-term and long-term data in real time with the data dynamics of spatial location and production stage. It can not only improve the recognition accuracy of abnormal nodes, but also classify, quantify and provide real-time feedback on abnormal situations by calculating the degree of deviation between abnormalities and normal states, generate automated alarm information, reduce manual intervention, and improve the efficiency of responding to emergencies in the agricultural production process.

[0166] The present invention proposes to classify and dynamically cluster data according to different links of agricultural production (seeding, fertilizing, watering, weeding and harvesting). The data characteristics of each production stage are different, and traditional methods are difficult to cope with the differences between stages. By adjusting the clustering model parameters in real time at different production stages to adapt to the data characteristics of each stage, flexible management of the agricultural production process is achieved. The dynamic adjustment mechanism can ensure accurate data classification at each production stage, and production management is more targeted and timely, thereby improving the overall efficiency and quality of agricultural production.

[0167] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An improved agricultural production process traceability method based on artificial intelligence, characterized in that: The steps include: S1. Real-time collection of multi-dimensional agricultural production data on crop growth, soil conditions, meteorological environment and equipment operation status through equipment deployed in agricultural production environments; S2, perform data cleaning, noise removal, outlier detection and format conversion on multi-dimensional agricultural production data; S3. According to the time attribute of multidimensional agricultural production data, the multidimensional agricultural production data is divided into short-term data, medium-term data and long-term data, and the data characteristics of different time periods are analyzed by using an adaptive clustering model; S4. According to the spatial distribution characteristics of multidimensional agricultural production data, the multidimensional agricultural production data is divided according to the spatial locations of plots, regions and crop types, and adaptive clustering analysis is performed on the data of each spatial location, and the parameters of the adaptive clustering model are dynamically adjusted so that the classification results of the multidimensional agricultural production data reflect the crop growth, environmental conditions and production equipment status of different regions; S5. According to the specific links of agricultural production, the multi-dimensional agricultural production data is classified according to the multiple production stages of crops from sowing, fertilization, watering, weeding and harvesting, and the data of each production stage is dynamically clustered to identify the key features in each stage, and the adaptive clustering model parameters are adjusted in real time to cope with the data changes in the agricultural production links; S6. Automatically identify abnormal nodes in the agricultural production process based on the multi-level clustering results of steps S3 to S5, including abnormal crop growth, abnormal soil conditions, sudden changes in the meteorological environment, and equipment failures, and issue abnormal alarms in real time, and feed back information on abnormal agricultural production nodes to the management system; S7. When an abnormality occurs in the agricultural production process, based on the multi-level clustering results of multi-dimensional agricultural production data, automatic traceability analysis is performed to determine the time, location and cause of the abnormality by tracing the data source and historical changes of the abnormal node, and generate a traceability report; S8. Feedback abnormal nodes and their causes in traceability management to the agricultural production management system in real time, and dynamically adjust the irrigation frequency, fertilizer application amount or equipment maintenance plan of crops based on the feedback information.

2. The method for improving agricultural production process traceability based on artificial intelligence according to claim 1, characterized in that: The S1 includes: S11. Real-time collection of multi-dimensional data on crop growth through sensor networks deployed in agricultural production environments c (t), including leaf area, stem length, photosynthesis rate and nutrient content, D c is the multidimensional vector of crop growth parameters, and t is the collection time point; S12. Use sensor equipment deployed in the soil to collect multi-dimensional data of soil conditions in real time s (t), including soil temperature, moisture, pH value and nutrient content, D s is the multidimensional vector of soil conditions; S13. Collect meteorological environment data in real time through weather stations w (t), including air temperature, humidity, wind speed, light intensity and rainfall, D w is the multidimensional vector of meteorological environment; S14. Real-time collection of equipment operation status data related to agricultural production equipment through the equipment operation status monitoring system e (t), including the equipment's operating current, operating voltage, temperature and vibration, D e is a multidimensional vector of the equipment operation status; S15, integrating the multidimensional agricultural production data collected in steps S11 to S14 into multidimensional agricultural production data D(t): D(t)={D c (t),D s (t),D w (t),D e (t)}。 3. The method for improving agricultural production process traceability based on artificial intelligence according to claim 1 is characterized in that: The S2 includes: S21. Clean the collected multidimensional agricultural production data D(t) to remove invalid data and duplicate data, and obtain the cleaned multidimensional agricultural production data D clean (t); S22. Cleaned multidimensional agricultural production data D clean (t) Perform noise removal to eliminate the noise generated by sensor equipment and environmental interference, and obtain the multi-dimensional agricultural production data D after noise removal denoise (t); S23, multidimensional agricultural production data after noise removal D denoise (t) performs outlier detection, using a distribution-based detection method to identify abnormal data points in multidimensional agricultural production data. Abnormal data points include crop growth parameters, soil conditions, and meteorological environment data that exceed a reasonable range. The outlier detection result is expressed by the function E(t): E(t)={e1(t),e2(t),…,e n (t)}; Among them, e n (t) is an abnormal data point; S24, analyze the abnormal data in E(t), automatically correct or delete the abnormal data, and obtain the multi-dimensional agricultural production data D after abnormal processing corrected (t); S25, the multi-dimensional agricultural production data D after abnormal processing corrected (t) Perform format conversion, convert the multidimensional agricultural production data into a unified format according to the requirements of data analysis and clustering models, and obtain the converted multidimensional agricultural production data D formatted (t).

4. The method for improving agricultural production process traceability based on artificial intelligence according to claim 1 is characterized in that: The S3 includes: S31. According to the time attribute of the multidimensional agricultural production data, the multidimensional agricultural production data D formatted (t) is divided into short-term data D short (t), mid-term data D mid (t) and long-term data D long (t): D short (t) represents agricultural production data collected within a short time frame for real-time monitoring and feedback; D mid (t) represents agricultural production data collected over a medium-term period for trend identification; D long (t) represents agricultural production data collected over a long period of time for periodic pattern analysis; S32. For short-term data D short (t) Adopting adaptive clustering model for classification analysis to identify the changing patterns of crop growth, soil conditions, meteorological environment and equipment operation status in agricultural production in the short term: Among them, C short (t) is the short-term clustering result, represents the i-th short-term data point, μ j (t) is the cluster center, the mean vector of the jth class, and w ij (t) is the sample The membership function belonging to the jth class represents the distance weight between the short-term data point and the cluster center; S33. Interim data D mid (t) Trend identification: Use the adaptive clustering model to analyze the data change trend. By integrating the trend in the time interval [t0, t1], the data change trend over time is captured and used for the analysis of crop growth trend, soil nutrient fluctuation trend and meteorological change trend: Among them, T mid (t) is the mid-term trend identification result, represents the i-th mid-term data point, μ j (t) is the changing trend of cluster center, w ij (t) is the membership function; S34. For long-term data D long (t) Analyze the periodicity law and use the adaptive clustering model to detect and analyze the periodic fluctuations of long-term data in agricultural production. j (t-t0)) captures cyclical fluctuations in long-term data: Among them, P long (t) is the result of long-term periodic law analysis, is the i-th long-term data point, μ j (t) is the cluster center, f j represents the periodic frequency, w ij (t) is the membership function.

5. The method for improving agricultural production process traceability based on artificial intelligence according to claim 1 is characterized in that: The S4 includes: S41. Based on multidimensional agricultural production data formatted The spatial distribution characteristics of (t) are used to divide the multidimensional agricultural production data into spatial distributions according to plots, regions and crop types, and obtain agricultural production data D at different spatial locations. loc (t): D loc1 (t) represents the dataset of plot 1, D loc2 (t) represents the dataset of plot 2, D locn (t) represents the dataset of plot n; S42, agricultural production data D for each spatial location locn (t) Adopting adaptive clustering model for analysis, the clustering process dynamically adjusts the cluster center μ according to the different spatial positions. j (t) and membership function w ij (t), so that the adaptive clustering model reflects the crop growth, environmental conditions and production equipment status of the corresponding spatial location; S43. Based on the differences in spatial locations, agricultural production data at different spatial locations are analyzed. loc (t) Dynamic parameter adjustment is performed to make the clustering results reflect the agricultural production characteristics of each region. The dynamic adjustment of the adaptive clustering model is achieved by minimizing the function L loc (t) Implement: in, represents the multidimensional agricultural production data point at the i-th spatial location, w ij (t) is a multi-dimensional agricultural production data point The membership function of cluster center j, λ is the regularization parameter, Represents the dynamic adjustment of the cluster center over time, which adapts to the changes of agricultural production data in different time periods, and captures the acceleration changes of production data at different spatial locations through the second-order derivative term; S44. Based on the results of cluster analysis, the crop growth conditions, soil conditions, meteorological environment and production equipment status in different regions are reflected in real time, and the spatial location classification results C are generated. spatial (t): in, Producing data points for multi-dimensional agriculture and cluster center μ j (t) reflects the spatial difference of agricultural production data at time t. Used to describe dynamic adjustments in the time dimension, involving multi-dimensional agricultural production data points The time derivative of Reflects the changing speed of crop growth, β is the adjustment coefficient, and the angle parameter θ ij (t) is used to capture the relative angle changes between different cluster centers. It reflects the influence of the second-order time derivative on the cluster center and captures the acceleration change of crop growth or equipment status. δ is the spatial attenuation coefficient. Indicates that with the spatial distance |x i -x j |Increasing and decaying acceleration effects.

6. The method for improving agricultural production process traceability based on artificial intelligence according to claim 1, characterized in that: The S5 includes: S51. According to the specific links of agricultural production, multi-dimensional agricultural production data D formatted (t) Classify crops according to their production stages, including sowing, fertilizing, watering, weeding and harvesting, and obtain the production stage dataset D stage (t), where D seed (t) represents the data of the sowing stage, D fertil (t) represents the data of fertilization stage, D water (t) represents the data of the watering stage, D weed (t) represents the data of weeding stage, D harvest (t) represents the data of the harvesting stage; S52. For each production stage data set D stage (t) Adopting adaptive clustering model to conduct dynamic clustering analysis, identify key features in each production stage, and dynamically adjust the parameters of adaptive clustering model according to the changes of data in agricultural production process; S53. Identify key features of the agricultural production process based on the clustering results of each production stage, including soil conditions in the sowing stage, nutrient requirements in the fertilization stage, humidity changes in the watering stage, weed coverage in the weeding stage, and crop maturity in the harvesting stage. j (t) and w ij (t) respond to changes in data during the production phase; S54, real-time adjustment of the adaptive clustering model parameters at each production stage, by minimizing the objective function L stage (t) Optimize the clustering process: Among them, λ is the adjustment parameter, Represents the dynamic changes of cluster centers over time.

7. The method for improving agricultural production process traceability based on artificial intelligence according to claim 1, characterized in that: The S6 includes: S61, based on the multi-level clustering results of steps S3 to S5, automatically identify abnormal nodes in the agricultural production process, including abnormal crop growth, abnormal soil conditions, sudden meteorological environment changes and equipment failures, and short (t), mid-term trend identification data T mid (t) and long-term periodic data P long (t), different spatial position data C spatial (t) and data C of each production stage stage The clustering results of (t) are compared to detect deviations from the normal production mode. The result of abnormal node detection is expressed by function A(t): A(t)={a1(t),a2(t),…,a m (t)}; Among them, aa m (t) represents the mth abnormal node. The detected abnormal nodes include abnormalities in crop growth, soil conditions, meteorological environment, and equipment status; S62: for each abnormal node a m (t) Classify and combine various clustering results to determine the type of anomaly, including crop growth anomaly, soil condition anomaly, meteorological environment anomaly and equipment failure. According to different types of anomalies, automatically generate classification labels L m (t): Among them, L m (t) is the classification label of the abnormal node, They represent the variance of the cluster data at each stage, respectively. The classification process depends on the difference from the normal cluster center at each stage; S63. Based on the classification results of abnormal nodes, combined with short-term, medium-term and long-term clustering data and clustering results of different spatial locations and production stages, the abnormality degree E is automatically calculated. m (t) and assess the degree of deviation from normal production conditions; S64: detect abnormal node a m (t), exception type L m (t) and abnormality degree E m (t) Real-time feedback is provided to the agricultural production management system. The system generates an alarm based on the feedback information and, combined with the data clustering characteristics of different time periods, spatial locations and production stages, prompts managers to deal with abnormal situations and provides corresponding treatment suggestions based on the severity of the abnormality.

Images