Automatic alarm method and system for abnormal data of crop planting environment

Abstract

The invention relates to the technical field of intelligent alarm, and discloses an automatic alarm method and system for abnormal data of a crop planting environment, and the method comprises the steps: carrying out the data cleaning of the planting environment data of a crop, and obtaining the current environment data of the crop; extracting historical environment data in the historical log data set of the crops within a specified time range to obtain a historical data sequence of the crops; performing trend evaluation on the change mode and the fluctuation range of the historical data sequence to obtain environmental data trend characteristics of the historical data sequence; determining a dynamic threshold parameter of the crop according to the environment data trend characteristics and the growth environment parameter range of the crop; based on the dynamic threshold parameter, performing adaptive anomaly detection on the current environment data to obtain an abnormal signal of the current environment data; and performing communication protocol packaging on the abnormal signal to obtain alarm information of the current environment data. According to the invention, the efficiency of automatically alarming the abnormal data of the crop planting environment can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent alarm, in particular to a crop planting environment abnormal data automatic alarm method and system. BACKGROUND

[0002] The existing crop planting environment abnormal data automatic alarm method generally relies on preset fixed threshold to carry out abnormality judgment, and does not fully combine the differentiated needs of different growth stages of crops for environmental parameters to adjust the threshold. This fixed threshold mode is difficult to adapt to the dynamic change characteristics of environmental data, and when the environmental parameters are at the edge of the normal fluctuation interval, it is easy to appear "false alarm" or "missed alarm" phenomenon, which not only increases the invalid operation cost of planting management personnel, but also may affect the growth of crops due to the omission of real abnormalities, resulting in a substantial reduction in the accuracy and practicality of the alarm. At the same time, such methods do not consider the time sequence correlation of environmental data, and cannot capture the gradual abnormal trend of environmental parameters over time, further weakening the timeliness of abnormal warning, and it is difficult to meet the needs of fine planting management for environmental risk prediction.

[0003] In the environmental data preprocessing link, the existing technology has a relatively rough processing method for missing values and abnormal deviation points in planting environmental data, and generally uses mean filling, simple linear interpolation and other general methods, without adaptive calibration combined with the internal correlation and physiological demand characteristics of crop growth environmental parameters. This processing method is easy to cause a large deviation between the preprocessed environmental data and the actual planting environment, making the data source reliability of subsequent abnormal detection insufficient, which directly affects the effectiveness of the abnormality judgment result. In addition, most of the existing systems lack deep trend analysis of historical environmental data, and cannot optimize the abnormal detection logic based on the change mode and fluctuation range of the historical data sequence, resulting in poor adaptability of the system to complex and variable planting environments, which ultimately restricts the overall efficiency of the crop planting environment abnormal data automatic alarm, therefore, how to improve the overall efficiency of the crop planting environment abnormal data automatic alarm has become a problem to be solved. SUMMARY

[0004] The present application provides a crop planting environment abnormal data automatic alarm method and system to solve the problems raised in the background art.

[0005] To achieve the above purpose, the present application provides a crop planting environment abnormal data automatic alarm method, which comprises:

[0006] S1, data cleaning is performed on the planting environment data of crops to obtain current environment data of the crops;

[0007] S2, extracting historical environment data in the historical log data set of the crops within a specified time range to obtain a historical data sequence of the crops;

[0008] S3. Evaluate the trend of the change pattern and fluctuation range of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence;

[0009] S4. Based on the trend characteristics of the environmental data and the range of the growth environment parameters of the crop, determine the dynamic threshold parameters of the crop;

[0010] S5. Based on the dynamic threshold parameter, perform adaptive anomaly detection on the current environmental data of the crop to obtain the anomaly signal of the current environmental data;

[0011] S6. Encapsulate the abnormal signal using a communication protocol to obtain alarm information from the current environmental data.

[0012] In a preferred embodiment, the step of cleaning the crop planting environment data to obtain the current environmental data of the crop includes:

[0013] Identify missing data points in the crop growing environment data;

[0014] Interpolation processing is performed on the valid environmental data in adjacent time periods of the missing data points to obtain the environmental data sequence of the crop after filling in the missing data points;

[0015] Based on the growth environment parameters of the crop, the abnormal deviation points in the filled environmental data sequence are adaptively calibrated to obtain the regularized environmental data of the crop.

[0016] The current environmental data of the crop is obtained by performing a status assessment on the regularized environmental data.

[0017] In a preferred embodiment, the step of extracting historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop includes:

[0018] Based on the current growth stage of the crop, determine the corresponding historical reference time interval for the crop;

[0019] Based on the historical log dataset of the crop, the corresponding historical reference time interval is associated and mapped to obtain the crop's adaptive environment data record;

[0020] The adapted environment data records are time-series aligned to obtain the historical data sequence of the crop.

[0021] In a preferred embodiment, the step of evaluating the trend of the change patterns and fluctuation ranges of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence includes:

[0022] setting a sliding time window with a proper length according to the historical data sequence and the growth environment parameter of the crop;

[0023] moving the sliding time window along a time axis to obtain an environment data segment of the sliding time window;

[0024] analyzing a change direction of the environment data in the sliding time window to determine a trend direction of the environment data segment;

[0025] performing variability evaluation on a fluctuation range of the environment data in the sliding time window to obtain a fluctuation feature of the environment data segment;

[0026] performing situation synthesis on the trend direction and the fluctuation feature to obtain an environment data trend feature of the historical data sequence.

[0027] In a preferred embodiment, the determining the dynamic threshold parameter of the crop according to the environment data trend feature and the growth environment parameter range of the crop comprises:

[0028] performing niche fitting on the environment data trend feature to obtain a trend compatibility evaluation result of the crop;

[0029] constructing a dynamic threshold corridor of the crop according to the trend compatibility evaluation result and a current trend threshold floating space of the crop;

[0030] performing threshold resilience reinforcement on the dynamic threshold corridor to obtain a disturbance-resistant dynamic threshold corridor of the crop;

[0031] performing growth adaptability calibration on the dynamic threshold corridor based on growth stage characteristics of the crop to obtain the dynamic threshold parameter of the crop, wherein the dynamic threshold parameter is self-adaptively adjusted with environment trend changes.

[0032] In a preferred embodiment, the performing niche fitting on the environment data trend feature to obtain the trend compatibility evaluation result of the crop comprises:

[0033] constructing an environment parameter benchmark framework of an ideal growth environment of the crop according to the growth environment parameter of the crop;

[0034] performing coupling coordination degree evaluation on the environment data trend feature and the environment parameter benchmark framework to obtain an ecological adaptation index of the crop, wherein a calculation formula of the ecological adaptation index is as follows:

[0035] ;

[0036] wherein, The ecological adaptability index is mentioned above. This represents the total environmental parameters of the crop. The first of the growth environment parameters of the crop The weighting coefficients of each environmental parameter. It is a natural exponential function. The first of the environmental data trend features Trend characteristic values ​​of environmental parameters For the environmental parameter benchmark framework, the first The baseline value of each environmental parameter For the first Preset tolerance parameters for each environmental parameter;

[0037] Based on the ecological adaptability index and the physiological requirements of the crops, a two-way consistency diagnosis is performed on the trend characteristics of the environmental data to obtain the trend compatibility assessment results of the crops.

[0038] In a preferred embodiment, the step of adaptively detecting anomalies in the current environmental data of the crop based on the dynamic threshold parameter to obtain anomaly signals in the current environmental data includes:

[0039] The current environmental data is compared with the dynamic threshold parameter in real time to obtain abnormal data points in the current environmental data.

[0040] The abnormal data points are continuously verified to obtain a stable abnormal state of the current environmental data;

[0041] Based on the dynamic threshold parameter, the stable abnormal state is assigned to a specific interval to obtain a graded abnormal signal for the stable abnormal state.

[0042] Based on the trend characteristics of the environmental data, a trend consistency check is performed on the graded anomaly signals to obtain the anomaly signals of the current environmental data.

[0043] In a preferred embodiment, the step of determining the interval affiliation of the stable abnormal state based on the dynamic threshold parameter to obtain the hierarchical abnormal signal of the stable abnormal state includes:

[0044] Based on the floating characteristics of the dynamic threshold parameter, the state of critical data points in the boundary region is determined.

[0045] Based on the numerical range of the dynamic threshold parameter, anomaly level intervals with clearly defined boundaries are divided;

[0046] By comparing the duration and degree of deviation in the stable abnormal state with the abnormality level interval, the specific range to which the stable abnormal state belongs is obtained.

[0047] Based on the anomaly level interval attribute corresponding to the specific attribution interval, the hierarchical anomaly signal of the stable anomaly state is obtained.

[0048] In a preferred embodiment, the step of encapsulating the abnormal signal using a communication protocol to obtain alarm information from the current environmental data includes:

[0049] The abnormal signal is processed to construct a protocol, thereby obtaining the alarm data frame of the abnormal signal;

[0050] Based on a preset communication protocol, the alarm data frames are load-integrated to obtain alarm information of the current environment data.

[0051] To address the aforementioned problems, the present invention also provides an automatic alarm system for abnormal data in crop planting environments, the system comprising:

[0052] An environmental data preprocessing module is used to clean the crop planting environment data to obtain the current environmental data of the crop.

[0053] The historical data sequence construction module is used to extract historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop.

[0054] The environmental trend analysis module is used to evaluate the change patterns and fluctuation ranges of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence.

[0055] The dynamic threshold decision module is used to determine the dynamic threshold parameters of the crop based on the trend characteristics of the environmental data and the range of the growth environment parameters of the crop.

[0056] An adaptive anomaly detection module is used to perform adaptive anomaly detection on the current environmental data of the crop based on the dynamic threshold parameter, and obtain anomaly signals of the current environmental data.

[0057] The alarm information generation and encapsulation module is used to encapsulate the abnormal signal using a communication protocol to obtain alarm information from the current environmental data.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. This invention significantly improves the accuracy and reliability of identifying abnormal data in crop planting environments through multi-stage refined data processing and analysis. In the data preprocessing stage, missing data points in the planting environment data are interpolated, and adaptive calibration is performed on abnormal deviations in the filled data sequence in conjunction with crop growth environment parameters. Simultaneously, a state assessment is conducted to ensure that the obtained current environmental data accurately reflects the actual planting environment, providing a high-quality data source for subsequent anomaly detection. By extracting historical environmental data corresponding to the growth stage to construct a historical data sequence, its change patterns and fluctuation ranges are further analyzed to obtain environmental data trend characteristics. Finally, combined with the range of growth environment parameters, threshold parameters are dynamically adjusted according to environmental trends and growth stages, making the anomaly detection standards more aligned with the growth needs of crops at different stages and effectively improving the accuracy of anomaly identification.

[0060] 2. This invention significantly improves the efficiency and practicality of automatic alarm for abnormal data in crop planting environments through its sophisticated anomaly detection logic and standardized alarm information processing. When adaptively detecting anomalies in current environmental data based on dynamic threshold parameters, it not only locates abnormal data points through real-time correlation and comparison but also continuously verifies stable abnormal states. Furthermore, it divides anomaly levels into tiered anomaly signals using dynamic threshold parameters, and then performs trend consistency verification to ensure the reliability of the anomaly signals. Subsequently, the anomaly signals are encapsulated using a communication protocol to construct standardized alarm data frames and complete load integration, forming alarm information that can be efficiently transmitted. This ensures that planting managers can obtain timely and accurate information about environmental anomalies, providing support for rapid implementation of control measures and effectively guaranteeing the stability of the crop growth environment. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating an automatic alarm method for abnormal data in crop planting environment provided in an embodiment of the present invention.

[0062] Figure 2 This is a functional block diagram of an automatic alarm system for abnormal data in crop planting environment provided in an embodiment of the present invention;

[0063] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0064] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0065] This application provides an automatic alarm method for abnormal data in crop planting environment. The executing entity of this automatic alarm method for abnormal data in crop planting environment includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the automatic alarm method for abnormal data in crop planting environment can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0066] Reference Figure 1 The diagram shown is a flowchart illustrating an automatic alarm method for abnormal crop planting environment data according to an embodiment of the present invention. In this embodiment, the automatic alarm method for abnormal crop planting environment data includes:

[0067] S1. Perform data cleaning on the crop planting environment data to obtain the current environmental data of the crop;

[0068] In this embodiment of the invention, the step of cleaning the crop planting environment data to obtain the current environmental data of the crop includes:

[0069] Identify missing data points in the crop growing environment data;

[0070] Interpolation processing is performed on the valid environmental data in adjacent time periods of the missing data points to obtain the environmental data sequence of the crop after filling in the missing data points;

[0071] Based on the growth environment parameters of the crop, the abnormal deviation points in the filled environmental data sequence are adaptively calibrated to obtain the regularized environmental data of the crop.

[0072] The current environmental data of the crop is obtained by performing a status assessment on the regularized environmental data.

[0073] Specifically, the collected crop planting environment data are first arranged in chronological order to form a complete time series data list. Then, each environmental data corresponding to each time node in the time series is checked one by one. If one or more environmental data items under a certain time node are not recorded with any value, the recorded value is empty, or it is marked as invalid, then these environmental data items that are not recorded normally under that time node are identified as missing data points in the crop planting environment data, thereby completing the identification of missing data points in the crop planting environment data.

[0074] Furthermore, for each identified missing data point, the preceding and following valid time nodes of the time node where the missing data point is located are first determined. The valid time node refers to the time node where the corresponding type of environmental data is a valid record. Then, the valid environmental data of the same type as the missing data point at these two adjacent valid time nodes are extracted. Then, the arithmetic mean of these two valid environmental data is calculated. The arithmetic mean is used as the imputation value to fill the position corresponding to the missing data point. After performing the above operation on all missing data points, the original environmental data sequence with missing data is transformed into an imputed environmental data sequence with valid data at all time nodes.

[0075] Furthermore, firstly, the standard range of growth environment parameters corresponding to different growth stages of the crop is determined. Then, each data point in the filled environmental data sequence is compared with the standard range of growth environment parameters for the corresponding growth stage. If the value of a certain data point exceeds the upper limit or falls below the lower limit of the standard range, the data point is determined as an abnormal deviation point. Subsequently, a calibration value is determined according to the standard range of growth environment parameters. If the data point exceeds the upper limit, the upper limit of the standard range is taken as the calibration value. If it falls below the lower limit, the lower limit of the standard range is taken as the calibration value. The abnormal deviation points in the filled environmental data sequence are replaced with the calibration values. After replacing all abnormal deviation points, the regularized environmental data of the crop is obtained.

[0076] Furthermore, first, obtain the current growth stage of the crop and the corresponding standard environmental data. Then, compare each indicator in the standardized environmental data with the standard environmental data for the current growth stage to determine whether all indicators in the standardized environmental data meet the standard requirements for the current growth stage. If all indicators meet the standard requirements, the standardized environmental data is directly identified as the current environmental data of the crop. If some indicators slightly exceed the standard range but the deviation is within the crop's tolerance range, there is no need to modify the standardized environmental data, and it is still identified as the current environmental data of the crop. If some indicators seriously exceed the standard range and exceed the crop's tolerance range, the data filling and abnormal calibration process in the previous steps needs to be re-verified, and any possible deviations need to be corrected before comparing and judging again, until environmental data that meets the standard requirements for the current growth stage is obtained, and finally, this data is identified as the current environmental data of the crop.

[0077] In general, identifying missing data points in crop planting environment data involves organizing the collected crop planting environment-related data in chronological order, then checking each data point at each time point to find environmental data items that are not recorded with specific values, have empty values, or are marked as invalid information, thereby clarifying the missing data points in the planting environment data.

[0078] In summary, the key to interpolating valid environmental data between adjacent time points for missing data points is to first find the time points that are adjacent to each missing data point and have valid records of the corresponding type of environmental data. Then, extract the same type of valid environmental data from these two adjacent valid time points, and fill in the gaps in the missing data points by processing these valid data. Finally, a filled environmental data sequence with valid data at all time points is formed.

[0079] In summary, the core of adaptive calibration of abnormal deviations in the filled environmental data sequence based on crop growth environment parameters is to first determine the standard range of various growth environment parameters corresponding to the current growth stage of the crop, then compare each data point in the filled environmental data sequence with the standard range one by one, screen out abnormal deviations that exceed the upper limit or fall below the lower limit of the standard range, and then determine the corresponding calibration value to replace these abnormal deviations according to the standard range, thereby obtaining the regularized environmental data of the crop.

[0080] In general, the key to assessing the status of standardized environmental data is to combine the current growth stage of the crop with the corresponding standard environmental data, compare each indicator in the standardized environmental data with the standard environmental data, and determine whether each indicator meets the requirements of the current growth stage and the crop's own tolerance range. If there are deviations, the previous steps are checked and corrected, and finally the current environmental data of the crop that meets the requirements is determined.

[0081] S2. Extract historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop.

[0082] In this embodiment of the invention, the step of extracting historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop includes:

[0083] Based on the current growth stage of the crop, determine the corresponding historical reference time interval for the crop;

[0084] Based on the historical log dataset of the crop, the corresponding historical reference time interval is associated and mapped to obtain the crop's adaptive environment data record;

[0085] The adapted environment data records are time-series aligned to obtain the historical data sequence of the crop.

[0086] Specifically, first, the standards for dividing the crop growth cycle are clarified. For example, a certain corn variety is divided into sowing period, seedling stage, jointing stage, tasseling stage, grain filling stage, and maturity stage. The typical duration of each stage is based on long-term planting observations and reflects the growth process under normal climate conditions. Combining field observations of plant height, leaf quantity, organ development, and other growth characteristics with daily planting management records, the current specific growth stage is determined and the start time is recorded. Then, complete past growth cycle records are extracted from the planting archives, and past stages with the same name as the current stage are identified. Their time intervals are integrated to obtain the corresponding historical reference time interval.

[0087] Furthermore, the storage structure of the historical log dataset was analyzed: a first-level directory was set up with "year-month-day," and second-level data entries were set up with "hour:minute." Each entry fully recorded environmental data such as temperature, humidity, light intensity, soil pH, and soil moisture content at the corresponding time point, and also indicated the crop growth stage at that time point. The first-level directory covering the historical reference time interval was opened from the dataset, all second-level entries were extracted, and the growth stage marked on the entries was checked. If it matched the current stage, it was kept; otherwise, it was discarded. The retained entries were collected to form an environmentally adapted data record.

[0088] Further, the time format of the environmental data records is checked. Entries are sorted from earliest to latest by timestamp. During sorting, adjacent data timestamps are compared; if they are the same, entries with no missing environmental data are retained first. If all are complete, the entry with the earliest original order in the historical log is retained, and duplicate entries are removed. If the time interval between adjacent data exceeds the normal collection interval, no data is added; only a "normal collection interval abnormal" message is marked between them. After sorting, deduplication, and anomaly labeling, the ordered, non-duplicate records are arranged chronologically to form a coherent historical data sequence.

[0089] In general, determining the corresponding historical reference time interval based on the current growth stage of a crop is crucial. First, it's essential to clarify the crop's growth cycle division criteria and the typical characteristics of each growth stage. Then, by observing the crop's plant height, leaf development, and other growth characteristics in the field, and combining this with daily planting and management records, the specific growth stage and its starting point can be determined. Next, records of multiple complete growth cycles from the crop's planting archives should be extracted. The stages within each complete cycle that match the current growth stage should be identified, and the corresponding time intervals for these past stages should be determined. Integrating these time intervals yields the corresponding historical reference time interval for the crop.

[0090] In summary, the process of mapping historical log datasets of crops to corresponding historical reference time intervals to obtain adapted environmental data records involves several key steps. First, the storage structure of the historical log dataset is organized. This dataset uses "year-month-day" as the first-level directory and "hour:minute" as the second-level data entries. Each entry contains environmental data such as temperature and humidity at the corresponding time point, as well as labeled growth stage information. Then, all second-level data entries within the corresponding historical reference time interval are extracted from the dataset. The growth stage labeled in each entry is compared with the current growth stage of the crop. Entries with consistent growth stages are retained, while those with inconsistent stages are removed. After summarizing and organizing all retained entries, the adapted environmental data record for the crop is formed.

[0091] In summary, the historical data sequence is obtained by aligning the adaptive environment data records with time sequence. The key is to first check the time representation format of all entries in the adaptive environment data records, unify different formats into a standard format, and then sort the entries in order of timestamp from earliest to latest. If entries with the same timestamp are found during sorting, the entries with complete data and the earlier recorded order in the log are retained. If the time interval between adjacent entries exceeds the normal collection interval, an interval anomaly is marked in the corresponding position. After sorting, deduplication, and anomaly marking are completed, the ordered adaptive environment data records are arranged in chronological order to obtain the historical data sequence of the crop.

[0092] S3. Evaluate the trend of the change pattern and fluctuation range of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence;

[0093] In this embodiment of the invention, the step of evaluating the trend of the change pattern and fluctuation range of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence includes:

[0094] Based on the historical data sequence and the growth environment parameters of the crop, a sliding time window of appropriate length is set;

[0095] The sliding time window is shifted along the time axis to obtain the environmental data segment of the sliding time window;

[0096] Analyze the direction of change of environmental data within the sliding time window to determine the trend of the environmental data segment;

[0097] The variability of the fluctuation range of environmental data within the sliding time window is evaluated to obtain the fluctuation characteristics of the environmental data segment;

[0098] The trend and the fluctuation characteristics are combined to obtain the environmental data trend characteristics of the historical data sequence.

[0099] Specifically, first, determine the fixed collection interval for environmental data in the historical data sequence, and then retrieve the crop growth environment parameters. Combine the collection interval with the regular change cycle to determine the length of the sliding time window. It needs to cover the entire cycle and contain enough data points to reflect the trend, avoiding being too short to miss patterns or too long to lag. Based on this, set an appropriate window.

[0100] Furthermore, align the sliding time window with the starting point of the historical data sequence, and select environmental data starting from the starting time point with a length equal to the window length as the first window data segment. Move the window backward along the time axis according to the acquisition interval, with each move lasting the same duration as the acquisition interval, and select a new data segment as the next window data segment. Repeat this process until the end of the window reaches the end of the historical data sequence, acquiring all window data segments.

[0101] Furthermore, for each window data segment, after sorting by time, adjacent data points are compared one by one: if the later value is greater than the previous value, it is recorded as "rising"; if it is less than the previous value, it is recorded as "falling"; if they are equal, it is recorded as "stable". The total number of the three types of changes is counted. If the number of "rising" changes exceeds the sum of the number of "falling" and "stable" changes, the trend is "rising"; if the number of "falling" changes exceeds the sum of the number of "falling" changes, the trend is "falling"; if the number of "stable" changes is the highest, or if the difference between "rising" and "falling" changes is very small and less than the number of "stable" changes, the trend is "stable". This completes the trend determination for each data segment.

[0102] Furthermore, for each window data segment, the maximum and minimum values ​​are identified and the difference is calculated to obtain the actual fluctuation range. The normal and extreme fluctuation ranges for this indicator at the current stage are then reviewed in the growth environment parameters. After comparison, the fluctuation characteristics are labeled: if the actual range is less than or equal to the normal range, it is labeled "low fluctuation - normal"; if it is greater than normal but less than or equal to the extreme range, it is labeled "medium fluctuation - controllable"; if it is greater than the extreme range, it is labeled "high fluctuation - abnormal," thus completing the fluctuation characteristic assessment.

[0103] Furthermore, the trend and fluctuation characteristics of each window data segment are combined to form characteristic combination pairs such as "rising - low fluctuation - normal" and "stable - medium fluctuation - controllable". All combination pairs are arranged in the order of window progression. Consecutive combination pairs that are completely consistent are defined as stable trend characteristics. When a trend or fluctuation change occurs, the change node and the preceding and following combinations are recorded. All stable trend characteristics and change nodes are integrated to form a comprehensive description reflecting the direction of change, fluctuation status, and change nodes of historical data sequences, i.e., environmental data trend characteristics.

[0104] In summary, setting an appropriate sliding time window based on historical data sequences and crop growth environment parameters involves first defining the fixed collection interval of environmental data in the historical data sequence, then retrieving the regular change cycle of each environmental indicator in the corresponding growth stage of the crop growth environment parameters, and combining the collection interval and the regular change cycle to determine the window length that can fully cover a change cycle and contain enough data points to reflect the change trend. This avoids the window being too short to capture the complete pattern or too long to cause a lag in trend response, thus setting an appropriate sliding time window.

[0105] In summary, obtaining environmental data segments by shifting the sliding time window along the time axis involves first aligning the sliding time window with the starting time point of the historical data sequence, selecting the first segment of environmental data that fits the window length as the first environmental data segment, and then smoothly shifting the window backward along the time axis according to the collection time interval of the historical data sequence. After each shift, a new segment of environmental data is selected as the new environmental data segment. This shifting operation is repeated until the end of the window reaches the end time point of the historical data sequence, ultimately obtaining all environmental data segments corresponding to the sliding time window shifted along the time axis.

[0106] In general, analyzing the direction of change in environmental data within a sliding time window to determine the trend involves first arranging the environmental data within the window in chronological order, then comparing the values ​​of adjacent data points one by one, and recording the total number of times each of the three types of changes—"rising," "falling," and "stable"—is counted. The trend is determined by the number of these three types of changes: if "rising" occurs most frequently, the trend is "rising"; if "falling" occurs most frequently, the trend is "falling"; and if "stable" occurs most frequently or the difference between the number of "rising" and "falling" is very small, the trend is "stable." This method determines the trend of each segment of environmental data.

[0107] In general, the variability assessment of environmental data fluctuation range within a sliding time window to obtain fluctuation characteristics involves first identifying the maximum and minimum values ​​in the environmental data within the window, calculating the difference between them to obtain the actual fluctuation range, and then comparing the actual fluctuation range with the normal and extreme fluctuation ranges of the corresponding environmental indicator in the crop growth environment parameters. Based on the comparison results, different fluctuation characteristics such as "low fluctuation - normal", "medium fluctuation - controllable", and "high fluctuation - abnormal" are labeled to complete the fluctuation characteristic assessment of each environmental data segment.

[0108] In general, the process of synthesizing trend and fluctuation characteristics to obtain the environmental data trend features of historical data sequences involves first synthesizing feature pairs of the trend and fluctuation characteristics of each environmental data segment, arranging all feature pairs in the order of sliding window movement, observing and identifying continuous and consistent stable feature segments and feature change nodes, and then integrating the information of stable feature segments and change nodes to form a comprehensive description that can fully reflect the direction of change, fluctuation state, and change nodes of environmental data in historical data sequences. This comprehensive description is the environmental data trend feature of historical data sequences.

[0109] S4. Based on the trend characteristics of the environmental data and the range of the growth environment parameters of the crop, determine the dynamic threshold parameters of the crop;

[0110] In this embodiment of the invention, determining the dynamic threshold parameter of the crop based on the trend characteristics of the environmental data and the range of the crop's growth environment parameters includes:

[0111] Ecological niche fitting is performed on the trend characteristics of the environmental data to obtain the trend compatibility assessment results of the crop;

[0112] Based on the trend compatibility assessment results and the threshold fluctuation space in the current trend of the crop, a dynamic threshold corridor for the crop is constructed.

[0113] Threshold toughness enhancement is performed on the dynamic threshold corridor to obtain the disturbance-resistant dynamic threshold corridor of the crop;

[0114] Based on the growth stage characteristics of the crop, the dynamic threshold corridor is calibrated for growth adaptability to obtain the dynamic threshold parameters of the crop, wherein the dynamic threshold parameters are adaptively adjusted with changes in environmental trends.

[0115] The process of performing niche fitting on the trend characteristics of the environmental data to obtain the trend compatibility assessment results of the crops includes:

[0116] Based on the growth environment parameters of the crop, an environmental parameter benchmark framework for the ideal growth environment of the crop is constructed.

[0117] The environmental data trend characteristics are coupled with the environmental parameter benchmark framework to evaluate the degree of coordination, and the ecological adaptability index of the crop is obtained. The calculation formula of the ecological adaptability index is as follows.

[0118] ;

[0119] In the formula, The ecological adaptability index is mentioned above. This represents the total environmental parameters of the crop. The first of the growth environment parameters of the crop The weighting coefficients of each environmental parameter. It is a natural exponential function. The first of the environmental data trend features Trend characteristic values ​​of environmental parameters For the environmental parameter benchmark framework, the first The baseline value of each environmental parameter For the first Preset tolerance parameters for each environmental parameter;

[0120] Based on the ecological adaptability index and the physiological requirements of the crops, a two-way consistency diagnosis is performed on the trend characteristics of the environmental data to obtain the trend compatibility assessment results of the crops.

[0121] Specifically, the ecological niche of a crop is first defined, which is the range of suitable environmental conditions for the crop's survival and growth under various environmental indicators during its growth process. This range is derived through long-term field trials observing the crop's environmental adaptability under different growth states, combined with environmental records from the complete growth cycle. The optimal and tolerable ranges for each environmental indicator within the ecological niche are clearly delineated. Next, core information is extracted from the trend characteristics of the environmental data, including the long-term trend of each environmental indicator, the fluctuation characteristics at different time periods, and abrupt changes in environmental indicators at trend change points. Then, the environmental indicator data in the trend characteristics are matched point by point in chronological order with the tolerable range of the corresponding indicator in the ecological niche. The proportion of data falling within the tolerable range is statistically analyzed. Simultaneously, it is analyzed whether the trend of each time period is consistent with the ideal change trend of the environmental indicator at the corresponding growth stage of the crop, and whether the fluctuation characteristics are within the allowable fluctuation range of the ecological niche. If the data proportion reaches the fixed compatibility threshold set based on the crop growth tolerance rate, the trend is completely consistent with the ideal trend, and the fluctuation characteristics are within the allowable range, the trend compatibility assessment result is "fully compatible". If the data proportion does not reach the compatibility threshold but exceeds the minimum tolerance threshold, the trend deviates slightly from the ideal trend, or the fluctuation characteristics occasionally exceed the allowable range, the assessment result is "partially compatible". If the data proportion is lower than the minimum tolerance threshold, the trend is completely opposite to the ideal trend, or the fluctuation characteristics frequently exceed the allowable range, the assessment result is "incompatible". This completes the niche fitting of the environmental data trend characteristics and obtains the crop trend compatibility assessment result.

[0122] Furthermore, key values ​​for each environmental indicator are extracted from the range of crop growth environment parameters, including the upper limit, lower limit, absolute upper limit, and absolute lower limit of the suitable value. The threshold fluctuation space is defined as the range between the upper and lower limits of the suitable value for each environmental indicator, which can be flexibly adjusted according to environmental trends. Next, this threshold fluctuation space is adjusted based on the trend compatibility assessment results. If the assessment result is "fully compatible," it indicates that the current environmental trend is well-suited to crop growth, and there is no need to change the threshold fluctuation space, maintaining its consistency with the suitable value range of the growth environment parameters. If the assessment result is "partially compatible," the environmental indicator data points causing partial incompatibility are identified, the specific intervals where these data points are located are determined, and then the intervals containing these incompatible data points in the threshold fluctuation space are removed. The adjusted threshold fluctuation spaces of each environmental indicator are then concatenated and integrated in chronological order to form a continuous set of intervals covering the entire current trend cycle, with each time period corresponding to a clearly defined indicator threshold range. This set constitutes the dynamic threshold corridor for crops.

[0123] Furthermore, by reviewing historical records from the local agricultural meteorological department and field management files for the crop-growing area, common environmental disturbances that have occurred in the region over the past five years were collected. These included short-term sudden temperature rises and falls, short-term intense sunlight, sudden drops in humidity, short-term fluctuations in soil pH, and unstable sunlight caused by short-term strong winds. The maximum magnitude of change in environmental indicators and the duration of each disturbance type were also recorded in detail. Then, for each environmental indicator threshold range in the dynamic threshold corridor, the impact of each common environmental disturbance on that indicator was simulated one by one. Specifically, based on the maximum magnitude of change and duration of the disturbance, the highest, lowest, and average values ​​that the environmental indicator might reach during the disturbance were calculated. These calculated values ​​were then compared with the corresponding threshold range in the current dynamic threshold corridor to determine if the calculated values ​​exceeded the range. If the simulation results showed that a certain type of disturbance would cause the calculated values ​​of the environmental indicator to exceed the corridor range, the threshold corridor range for that environmental indicator was adjusted according to the maximum magnitude of change of the disturbance. After simulating all common environmental disturbances and adjusting the corresponding corridor intervals, the resulting dynamic threshold corridor, which can resist common environmental disturbances in the region and ensure that environmental indicators still meet the threshold requirements when disturbances occur, is the dynamic threshold corridor for crop disturbance resistance.

[0124] Furthermore, by first observing the growth morphology of crops in the field and combining this with the sowing time and early growth stage transition time recorded in the planting management file, the specific growth stage of the crop is determined. Then, based on the characteristics of this growth stage, the dynamic threshold corridors for various environmental indicators are adjusted accordingly. For air humidity characteristics, the lower limit of the air humidity threshold corridor is appropriately raised and the upper limit is appropriately lowered to ensure that humidity meets the requirements for pollen dispersal while preventing excessive humidity from causing diseases. During the adjustment process, the optimal growth cases for the same growth stage in the crop's historical planting records are consulted, and the actual range of each environmental indicator in the cases is referenced to ensure that the adjusted threshold corridor is consistent with the indicator range of the historical optimal growth environment, guaranteeing optimal growth conditions for the crop at the current stage. After completing the growth stage adaptation adjustment of the dynamic threshold corridors for all environmental indicators, the adjusted threshold ranges for each indicator are organized into a clear numerical range corresponding to the current growth stage. This numerical range is the dynamic threshold parameter of the crop, and this parameter will be adaptively adjusted again based on the characteristics of the growth stage as environmental trends change, achieving the function of adaptive adjustment with changes in environmental trends.

[0125] Specifically, the first step is to analyze the environmental parameters for crop growth. These parameters include the suitable numerical ranges and variation patterns of core environmental indicators such as temperature, humidity, light intensity, soil pH, and soil moisture content, as well as the differentiated requirements of each indicator at different growth stages of the crop. Next, these environmental parameters are categorized and organized according to the crop's growth stages. For each growth stage, the ideal numerical ranges and ideal variation trends of each environmental indicator are defined. These stage-specific ideal environmental indicator standards are then structurally integrated to form a framework covering different stages of the entire crop growth cycle, with clear ideal standards for each environmental indicator. This framework is the environmental parameter benchmark framework for the ideal growth environment of crops.

[0126] Furthermore, three core dimensions for evaluating the coupling coordination degree are first determined: the matching degree between the actual values ​​of each environmental indicator in the environmental data trend characteristics and the ideal range of the corresponding growth stage indicators in the environmental parameter benchmark framework; the consistency between the overall change trend of the environmental data trend characteristics and the ideal change trend of the corresponding stage in the benchmark framework; and the adaptability of the fluctuation amplitude of the environmental data trend characteristics to the allowable fluctuation range of the benchmark framework. Next, the environmental data trend characteristics are broken down into the actual value sequence, actual change trend, and actual fluctuation of each environmental indicator, and compared one by one with the ideal range, ideal change trend, and allowable fluctuation range of the indicators in the current crop growth stage in the environmental parameter benchmark framework: the percentage of time periods in which the actual values ​​of each environmental indicator fall within the ideal range is statistically analyzed to determine whether the actual change trend is the same as the ideal trend, and to verify whether the actual fluctuation amplitude is within the allowable fluctuation range. The degree of coupling and coordination is judged based on the comparison results of these three dimensions. If the numerical matching degree is high, the trend is completely consistent, and the fluctuation is completely adapted, an ecological adaptation index of "high adaptation" is assigned. If the numerical matching degree is medium, the trend is basically consistent, and the fluctuation occasionally exceeds the allowable range, an ecological adaptation index of "medium adaptation" is assigned. If the numerical matching degree is low, the trend is completely opposite, and the fluctuation frequently exceeds the allowable range, an ecological adaptation index of "low adaptation" is assigned. Thus, the ecological adaptation index of crops is obtained.

[0127] Furthermore, the physiological requirements of crops are first systematically analyzed, encompassing the core environmental requirements of key physiological processes at different growth stages. Then, a two-way consistency diagnosis is conducted using the obtained ecological adaptability index: If the ecological adaptability index is "high adaptability," the environmental conditions corresponding to the trend characteristics of the environmental data are further checked to see if they fully cover the key physiological requirements of the current growth stage. For example, can the temperature trend of "high adaptability" during the grain-filling stage meet the physiological requirements for dry matter accumulation? If fully covered, the final trend compatibility assessment result is determined to be "fully compatible." If the ecological adaptability index is "medium adaptability," it is analyzed whether the incompletely adapted environmental factors only affect non-critical physiological processes. For example, the light trend during the seedling stage is slightly lower than the ideal standard but does not affect the initiation of photosynthesis. If only affecting non-critical processes, the assessment result is determined to be "partially compatible." If the ecological adaptability index is "low adaptability," it is confirmed whether there are environmental factors that seriously violate key physiological requirements. For example, the temperature trend during the flowering stage is too low, leading to loss of pollen viability. If so, the assessment result is determined to be "incompatible." Through this two-way verification of the ecological adaptability index and physiological requirements, the final trend compatibility assessment result of the crops is obtained.

[0128] Furthermore, As an ecological adaptation index, its value comes from the process of evaluating the coupling and coordination between the trend characteristics of crop environmental data and the benchmark framework of environmental parameters. It is obtained by weighting and normalizing the adaptation of each environmental parameter. As the total environmental parameters for crops, its value comes from the sorting out of crop growth environment parameters. Specifically, it is to count the number of all core environmental indicators included in the growth environment parameters. These core indicators include temperature, humidity, light intensity, soil pH, soil moisture content, etc. When making the statistics, it is necessary to ensure that all key environmental parameters required for crop growth are covered, without omitting any core indicators that affect growth. As the first The weight coefficients of each environmental parameter are derived from the determination of the degree of influence of each environmental parameter on crop growth. The determination process needs to combine the physiological needs of crops and refer to the degree of influence of abnormal environmental parameters on crop growth in historical planting records. Based on the difference in the degree of influence, a fixed weight coefficient is assigned to each environmental parameter.

[0129] Furthermore, As the first The trend characteristic value of the first environmental parameter is derived from the trend characteristics of environmental data. Specifically, the value of the first parameter is extracted from the trend characteristics of environmental data. The actual change data of each environmental parameter were used to select representative values ​​of that parameter within the corresponding growth stage, which were then used as... The specific value. As the first The baseline value of the environmental parameter is derived from the environmental parameter baseline framework, specifically by extracting the corresponding value of the first environmental parameter from the environmental parameter baseline framework. The ideal values ​​of environmental parameters at the current growth stage are derived from long-term crop growth experiments and ensure that the crops achieve optimal growth under these conditions. As the first The preset tolerance parameters for each environmental parameter are derived from the analysis of historical crop growth environment data and research on the crop's own tolerance capabilities. Specifically, they are based on statistical analysis of the first environmental parameter over multiple past growth cycles. The normal fluctuation range of an environmental parameter, combined with the maximum tolerance range of crops to maintain normal growth when this parameter fluctuates, is used to determine the median value of this range, or the value that covers most of the normal fluctuations, as [the value to be determined]. The specific value.

[0130] Furthermore, the core function of this formula is to quantify the degree of fit between the trend characteristics of crop environmental data and the benchmark framework of environmental parameters. Through mathematical calculations, it integrates the fit of each environmental parameter into a comprehensive ecological fit index, intuitively reflecting the degree to which current environmental trends conform to the ideal growth environment for crops. The calculation of the numerator involves assigning a corresponding weight coefficient to each environmental parameter, and then substituting the difference between the trend characteristic value and the benchmark value of each environmental parameter into the natural exponential function. The result of the natural exponential function reflects the closeness between the trend characteristic value and the benchmark value. The closer the trend characteristic value is to the benchmark value, the larger the result of the function, indicating a higher contribution of that environmental parameter to the fit. Through the weighting of the weight coefficients, the influence of key environmental parameters on the degree of fit can be highlighted.

[0131] Furthermore, the calculation of the denominator involves summing the weighting coefficients of all environmental parameters. This normalizes the weighted calculation result in the numerator, preventing the absolute values ​​of the weighting coefficients from affecting the comparability of the final ecological suitability index. It ensures that regardless of the specific values ​​of the weighting coefficients for each environmental parameter, the final ecological suitability index falls within a uniform and measurable range. After calculating the ecological suitability index using the ratio of the numerator to the denominator, the index value directly corresponds to the suitability level between the environmental data trend characteristics and the ideal growth environment. A high index value indicates that the environmental parameters in the environmental data trend characteristics are generally close to the benchmark value, indicating a high degree of suitability, corresponding to a "high suitability" level. A medium index value indicates that some environmental parameters deviate slightly but still meet the overall requirements, indicating a moderate degree of suitability, corresponding to a "medium suitability" level. A low index value indicates that most environmental parameters deviate significantly from the benchmark value, indicating a low degree of suitability, corresponding to a "low suitability" level. This provides a quantitative basis for subsequent trend compatibility assessments.

[0132] In summary, the core of obtaining crop trend compatibility assessment results by fitting environmental data trend characteristics to niche is to identify the optimal and tolerable ranges of each environmental indicator in the crop niche, extract the trend, fluctuation and abrupt changes of environmental data trend characteristics, match trend data with niche tolerable ranges point by point, count the proportion of compatible data and analyze the adaptability, and determine "fully compatible", "partially compatible" or "incompatible" according to the degree of adaptability.

[0133] In summary, constructing a dynamic threshold corridor based on the trend compatibility assessment results and the current trend threshold fluctuation space of crops involves determining the appropriate upper and lower limits of each indicator from the range of growth environment parameters to define the fluctuation space, and then adjusting it according to the assessment results: "fully compatible" remains unchanged, "partially compatible" removes the incompatible interval, and "incompatible" shrinks to the ecological niche optimal interval. Finally, the adjusted threshold fluctuation space is integrated in chronological order to form a dynamic threshold corridor.

[0134] In summary, the key to obtaining an anti-disturbance dynamic threshold corridor by strengthening the threshold resilience of the dynamic threshold corridor is to collect the common environmental disturbance types and corresponding indicator changes and durations in the planting area over the past five years. For each indicator range in the corridor, the impact of each disturbance on the indicator is simulated and the values ​​are calculated. If the range is exceeded, the range coverage is adjusted. After completing all simulations and adjustments, the anti-disturbance dynamic threshold corridor is obtained.

[0135] In summary, the dynamic threshold corridor is calibrated for growth adaptability based on the characteristics of crop growth stages to obtain dynamic threshold parameters. This involves first determining the current growth stage by combining growth morphology and planting records, clarifying the characteristic requirements of this stage for various environmental indicators, adjusting the disturbance-resistant dynamic threshold corridor accordingly, and referring to the historical optimal growth environment range for this stage to ensure adaptability. Finally, the adjusted indicator threshold range is compiled as the dynamic threshold parameter and adaptively adjusted according to subsequent environmental trends.

[0136] In summary, constructing an environmental parameter benchmark framework for an ideal growth environment based on crop growth environment parameters involves sorting out the core indicators in the growth environment parameters, clarifying the appropriate numerical range, variation law, and differentiated requirements of each indicator at different growth stages, classifying and organizing indicator standards according to growth stages, and structurally integrating the ideal indicator standards for each stage to form a benchmark framework that covers the entire growth cycle and has clear ideal standards for each indicator.

[0137] In summary, the process of coupling environmental data trend characteristics with environmental parameter benchmark framework to obtain an ecological fit index involves first determining three dimensions: numerical matching degree, trend consistency, and fluctuation adaptability. Then, the environmental data trend characteristics are broken down into the actual numerical sequence, change trend, and fluctuation of each indicator. These are compared with the ideal range, trend, and fluctuation range of the benchmark framework at the current stage. Based on the comparison results, the degree of fit is comprehensively judged, and an ecological fit index of "high fit," "medium fit," or "low fit" is assigned.

[0138] In summary, the method of conducting a two-way consistency diagnosis of environmental data trend characteristics based on the ecological adaptability index and crop physiological needs to obtain trend compatibility assessment results involves first identifying physiological needs characteristics, and then verifying them in conjunction with the ecological adaptability index: "High adaptability" requires confirmation of key physiological needs of environmental coverage and is judged as "fully compatible"; "Medium adaptability" requires analysis of whether misfit factors only affect non-key physiological processes and is judged as "partially compatible"; "Low adaptability" requires checking whether it violates key physiological needs and is judged as "incompatible", ultimately obtaining the assessment results.

[0139] S5. Based on the dynamic threshold parameter, perform adaptive anomaly detection on the current environmental data of the crop to obtain the anomaly signal of the current environmental data;

[0140] In this embodiment of the invention, the step of adaptively detecting anomalies in the current environmental data of the crop based on the dynamic threshold parameter to obtain anomaly signals in the current environmental data includes:

[0141] The current environmental data is compared with the dynamic threshold parameter in real time to obtain abnormal data points in the current environmental data.

[0142] The abnormal data points are continuously verified to obtain a stable abnormal state of the current environmental data;

[0143] Based on the dynamic threshold parameter, the stable abnormal state is assigned to a specific interval to obtain a graded abnormal signal for the stable abnormal state.

[0144] Based on the trend characteristics of the environmental data, a trend consistency check is performed on the graded anomaly signals to obtain the anomaly signals of the current environmental data.

[0145] The step of determining the interval affiliation of the stable abnormal state based on the dynamic threshold parameter to obtain the hierarchical abnormal signal of the stable abnormal state includes:

[0146] Based on the floating characteristics of the dynamic threshold parameter, the state of critical data points in the boundary region is determined.

[0147] Based on the numerical range of the dynamic threshold parameter, anomaly level intervals with clearly defined boundaries are divided;

[0148] By comparing the duration and degree of deviation in the stable abnormal state with the abnormality level interval, the specific range to which the stable abnormal state belongs is obtained.

[0149] Based on the anomaly level interval attribute corresponding to the specific attribution interval, the hierarchical anomaly signal of the stable anomaly state is obtained.

[0150] Specifically, the core environmental indicators included in the current environmental data are first identified, such as temperature, humidity, light intensity, soil pH, and soil moisture content. These indicators are real-time data collected on the actual growth environment of the crops. Simultaneously, dynamic threshold parameters are retrieved. These parameters contain threshold ranges corresponding to each indicator in the current environmental data, with the threshold range for each indicator determined based on the characteristics of the crop's growth stage and environmental trends. Then, the actual value of each indicator in the current environmental data is compared one by one with the corresponding threshold range in the dynamic threshold parameters. If the actual value of an indicator is higher than the upper limit of its threshold range or lower than the lower limit, the current actual data point for that indicator is marked as an abnormal data point; if the actual value is within the threshold range, it is considered a normal data point. Through this real-time correlation and comparison of each indicator, all the data points marked as abnormal are ultimately collected as the abnormal data points of the current environmental data.

[0151] Furthermore, the regular data collection interval for the current environmental data is first determined. This interval is a fixed time period and is the collection frequency followed in daily monitoring of crop environmental data. For each identified anomalous data point, the actual data of the corresponding environmental indicator is continuously collected at the regular collection interval. The collection process needs to continue for a fixed duration that allows for a clear determination of whether the anomaly is persistent. After each collection, the newly collected indicator value is compared with the threshold range of the corresponding indicator in the dynamic threshold parameters. If all continuously collected data points still exceed the threshold range, and the values ​​do not show a trend of returning to the threshold range, it indicates that the anomaly persists. If any data point returns to the threshold range during continuous collection, the anomaly is determined to be a temporary fluctuation and not a persistent anomaly. Through this process of continuous collection and repeated comparison, the anomaly states corresponding to all data points that continuously exceed the threshold range are finally selected as the stable anomaly states of the current environmental data.

[0152] Furthermore, the graded abnormality intervals corresponding to each environmental indicator are first extracted from the dynamic threshold parameters. These intervals are divided according to the severity of the impact of the indicator's abnormality on crop growth, specifically including mild, moderate, and severe abnormality intervals. The mild abnormality interval is where the indicator value slightly exceeds the threshold range and has only a minor impact on crop growth; the moderate abnormality interval is where the indicator value significantly exceeds the threshold range and has a significant impact on crop growth; and the severe abnormality interval is where the indicator value greatly exceeds the threshold range and may cause serious damage to crop growth. Then, the actual value of each environmental indicator corresponding to the stable abnormal state is matched one by one with its graded abnormality interval. If the actual value of the indicator falls within the mild abnormality interval, the signal corresponding to the stable abnormal state is marked as a mild graded abnormality signal; if it falls within the moderate abnormality interval, it is marked as a moderate graded abnormality signal; and if it falls within the severe abnormality interval, it is marked as a severe graded abnormality signal. Through this clear interval assignment determination, all the graded signals obtained are the graded abnormality signals of the stable abnormal state.

[0153] Furthermore, the previously generated environmental data trend characteristics are retrieved first. These characteristics include the changing trends of each environmental indicator over historical periods, as well as the normal fluctuation patterns and ranges of each indicator within these trends. For each graded abnormal signal corresponding to an environmental indicator, the current stable abnormal state of the indicator is analyzed to determine if there is a logical correlation between it and its historical trend. If the historical trend of the indicator is stable, but the current stable abnormal state is a sudden and significant deviation of the indicator value from the threshold range, and the direction of deviation is unrelated to the historical trend, then it is necessary to check whether the monitoring equipment is faulty or whether there is temporary external interference. If the abnormal state persists after ruling out equipment faults and temporary interference, then the authenticity of the abnormality is confirmed in conjunction with the graded abnormal signal. If the deviation direction of the current stable abnormal state is consistent with the historical trend, and the deviation magnitude is within the reasonable range of the historical trend extension, then the graded abnormal signal is directly determined to be consistent with the environmental trend characteristics. Through this trend consistency verification of each graded abnormal signal, false abnormal signals caused by equipment errors or temporary interference are eliminated. The signals that truly reflect the current abnormal conditions of the crop's growth environment are the current environmental data abnormal signals.

[0154] Specifically, the floating characteristics of dynamic threshold parameters are first clarified, meaning that the threshold range of each environmental indicator will dynamically change with the crop growth stage and environmental trend characteristics. Critical data points are selected from stable abnormal states; these data points have values ​​close to the floating boundary of the corresponding indicator and are in the transition zone between normal and abnormal. Historical fluctuation patterns of the corresponding indicators are retrieved, and the critical data point values ​​are compared with these patterns: if they are within the historical normal fluctuation range and the crop shows no abnormal growth, they are determined to be normal data points; if they are outside the range or, although within the range, the crop has shown slight abnormalities, they are determined to be abnormal data points, thus completing the determination of the critical data point status.

[0155] Furthermore, the complete numerical range of each environmental indicator is extracted from the dynamic threshold parameters, including the upper and lower limits of normal operation determined based on the range of crop growth environmental parameters and the characteristics of the growth stage. Abnormality levels are divided into intervals according to the degree of impact of values ​​exceeding the normal range on crop growth: the mild interval is defined as a small deviation from the normal range, which may only slightly slow crop growth without obvious abnormalities; the moderate interval is a moderate deviation from the normal range, with obvious abnormalities such as pale leaves and stagnant plant height, but no irreversible damage; the severe interval is a very large deviation from the normal range, with severe damage such as leaf wilting and root rot, and even possible death. Clear numerical boundaries are set for each interval to ensure no overlap and continuous coverage of all values ​​exceeding the normal range, forming clearly defined abnormality level intervals.

[0156] Furthermore, for each stable abnormal state, the duration is statistically analyzed using the regular sampling interval as the unit, and the number of sampling cycles in which the abnormality exists is recorded. The degree of deviation is determined by the difference between the actual value of the indicator and the normal range; the larger the difference, the more severe the deviation. The duration and the degree of deviation are combined and matched with the abnormality level interval: short duration + small deviation meets the criteria for the mild interval and is assigned to the mild interval; medium duration + moderate deviation meets the criteria for the moderate interval and is assigned to the moderate interval; long duration + extreme deviation meets the criteria for the severe interval and is assigned to the severe interval. The matching must simultaneously meet the interval criteria corresponding to both factors to avoid misjudgment based on a single factor, ultimately yielding the specific interval to which the stable abnormal state belongs.

[0157] Furthermore, a fixed correspondence is pre-established between specific attribution intervals and anomaly level attributes. This relationship is based on interval definition standards: attribution to a mild interval corresponds to the "mild anomaly" attribute, moderate interval to "moderate anomaly," and severe interval to "severe anomaly," ensuring consistent attribute matching. The specific attribution interval for each stable anomaly state is associated with this relationship: if attribution to a mild interval, the attribute is determined as "mild anomaly" according to the correspondence, generating a mild-level anomaly signal; if attribution to a moderate interval, the corresponding attribute is determined. In general, the current environmental data is compared with dynamic thresholds in real time to obtain anomaly data points. This involves first identifying the core indicators of the current environmental data and the threshold ranges of the corresponding indicators in the dynamic thresholds, then comparing the actual values ​​of each indicator with the corresponding thresholds one by one. Values ​​exceeding the upper or lower limits of the threshold are marked, and finally, all anomaly data points are collected.

[0158] In general, continuous verification of abnormal data points to obtain stable abnormal states involves continuously collecting corresponding indicator data at regular collection intervals and comparing it with dynamic thresholds. If the continuous data all exceed the threshold and there is no regression trend, it is determined to be a stable abnormal state, excluding temporary fluctuations and non-continuous abnormalities.

[0159] In summary, the method of determining the interval affiliation of stable abnormal states based on dynamic thresholds to obtain graded abnormal signals involves extracting mild, moderate, and severe graded abnormal intervals for each indicator from the dynamic thresholds, matching the indicator values ​​corresponding to stable abnormalities with the intervals, and marking the intervals as graded abnormal signals of the corresponding level.

[0160] In summary, to obtain the final abnormal signal by performing trend consistency verification on graded abnormal signals based on the trend characteristics of environmental data, we need to retrieve the historical trends and fluctuation patterns in the trend characteristics of environmental data, analyze the logical relationship between the stable abnormality corresponding to the graded abnormal signal and the historical trend, eliminate false abnormalities caused by equipment failure and temporary interference, and retain the real environmental abnormal signal.

[0161] In summary, determining the state of critical data points in the boundary region based on the dynamic threshold fluctuation characteristics involves first clarifying the fluctuation characteristics of the dynamic threshold—the dynamic characteristics of how the thresholds of various environmental indicators adjust with the crop growth stage and environmental trends. Then, critical data points whose values ​​are close to the fluctuation boundary are selected from stable anomalies. The historical fluctuation patterns of the corresponding indicators are retrieved and observed in conjunction with the current growth status of the crops. If the data is within the historical normal fluctuation range and the crops are not abnormal, it is determined to be normal; otherwise, it is determined to be abnormal, thus completing the determination of the state of the critical data points.

[0162] In general, the process of dividing anomaly level intervals with clear boundaries based on the dynamic threshold value range involves first extracting the complete value range of each environmental indicator in the dynamic threshold, including the upper and lower limits of normal operation, and then dividing the intervals according to the degree of impact of values ​​exceeding the normal range on crop growth. Clear numerical boundaries without overlap are set for each interval, thus forming anomaly level intervals with clear boundaries.

[0163] In general, matching the duration of stable anomalies with the degree of deviation and the anomaly level range to obtain a specific attribution range involves first statistically analyzing the duration of stable anomalies at regular collection intervals, calculating the difference between the actual value of the indicator and the normal range to determine the degree of deviation, and then combining both with the anomaly level range. Once the dual criteria of the corresponding range are met, the specific attribution range is obtained.

[0164] In general, obtaining graded anomalous signals based on the anomalous level attributes of specific attribution intervals involves first establishing a fixed correspondence between specific attribution intervals and anomalous level attributes, then associating the specific attribution intervals of each stable anomaly with this relationship, and generating graded anomalous signals of mild, moderate, and severe according to the attributes.

[0165] S6. Encapsulate the abnormal signal using a communication protocol to obtain alarm information from the current environmental data.

[0166] In this embodiment of the invention, the step of encapsulating the abnormal signal using a communication protocol to obtain alarm information from the current environmental data includes:

[0167] The abnormal signal is processed to construct a protocol, thereby obtaining the alarm data frame of the abnormal signal;

[0168] Based on a preset communication protocol, the alarm data frames are load-integrated to obtain alarm information of the current environment data.

[0169] Specifically, the fixed structure of the alarm data frame is first defined, which includes three parts: a frame header, an abnormal signal data field, and a frame tail. The frame header is set to a preset fixed binary sequence to identify the start of the data frame, ensuring accurate identification by the receiving end. Next, the abnormal indicator type, abnormal level, and abnormal occurrence time are extracted from the abnormal signal. This information is converted into binary data according to preset rules and field lengths, and then filled into the abnormal signal data field in sequence. Finally, the frame tail is set to another set of preset fixed binary sequences, which are then concatenated in the order of the frame header, the filled abnormal signal data field, and the frame tail to form a complete binary data block. This data block is the alarm data frame of the abnormal signal.

[0170] Furthermore, load balancing is performed based on a pre-defined Modbus protocol. First, the pre-defined Modbus protocol specification is retrieved, specifying that the load must include four elements: function code, data length, core data, and checksum. The function code is set as a preset function code identifying alarm information transmission, the data length indicates the number of bytes in the core data, and the checksum uses CRC verification to ensure transmission integrity. Next, the alarm data frame of the aforementioned abnormal signal is used as the core data, its total byte count is calculated, and it is converted into compliant binary data and filled into the data length field. Then, according to the Modbus protocol definition, the preset function code identifying alarm information transmission is converted into corresponding binary data and filled into... The function code field is used as the starting point. Then, a CRC check is performed on the entire data block consisting of the function code, data length, and alarm data frame. Specifically, each byte of the data block is XORed sequentially with a preset initial CRC value. The least significant bit of the result is checked; if it is 1, it is XORed with a preset check polynomial; if it is zero, it remains unchanged. The result is then shifted right by one bit, and this operation is repeated until all bytes have been processed. The resulting binary number is the CRC checksum and is filled into the checksum field. Finally, the function code, data length, alarm data frame, and checksum are concatenated according to the Modbus protocol to form a complete data unit conforming to the protocol specifications. This data unit represents the alarm information of the current environmental data.

[0171] In summary, the protocol construction of the abnormal signal to obtain the alarm data frame of the abnormal signal involves first defining its fixed structure, which includes a frame header, an abnormal signal data field, and a frame tail. Then, key information such as the abnormal indicator type, level, and occurrence time are extracted from the abnormal signal. This information is then converted into binary data according to preset rules and field lengths and filled into the data field. Finally, the frame header, data field, and frame tail are concatenated to form a complete binary data block, which is the alarm data frame of the abnormal signal.

[0172] In summary, the alarm information of the current environment data is obtained by load integration of the alarm data frame based on the preset communication protocol. First, the protocol specification is retrieved to clarify that the load must include a function code, data length, core data, and check code. Then, the alarm data frame is used as the core data. Its byte count is counted, converted to binary and filled into the data length field, and converted to the preset function code according to the protocol and filled into the function code field. After that, the block composed of function code, data length, and alarm data frame is subjected to CRC check to obtain the check code and filled in. Finally, the four elements are concatenated according to the protocol order. The complete data unit that conforms to the specification is the alarm information of the current environment data.

[0173] like Figure 2 The diagram shown is a functional block diagram of an automatic alarm system for abnormal data in crop planting environment provided by an embodiment of the present invention.

[0174] The automatic alarm system 100 for abnormal crop planting environment data described in this invention can be installed in an electronic device. Depending on the functions implemented, the automatic alarm system 100 for abnormal crop planting environment data may include an environmental data preprocessing module 101, a historical data sequence construction module 102, an environmental trend analysis module 103, a dynamic threshold decision module 104, an adaptive anomaly detection module 105, and an alarm information generation and encapsulation module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0175] In this embodiment, the functions of each module / unit are as follows:

[0176] The environmental data preprocessing module 101 is used to clean the crop planting environment data to obtain the current environmental data of the crop.

[0177] The historical data sequence construction module 102 is used to extract historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop.

[0178] The environmental trend analysis module 103 is used to evaluate the change patterns and fluctuation ranges of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence.

[0179] The dynamic threshold decision module 104 is used to determine the dynamic threshold parameters of the crop based on the trend characteristics of the environmental data and the range of the growth environment parameters of the crop.

[0180] The adaptive anomaly detection module 105 is used to perform adaptive anomaly detection on the current environmental data of the crop based on the dynamic threshold parameter, and obtain the anomaly signal of the current environmental data.

[0181] The alarm information generation and encapsulation module 106 is used to encapsulate the abnormal signal using a communication protocol to obtain alarm information from the current environmental data.

[0182] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0183] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0184] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0185] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0186] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0187] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for automatically alarming abnormal data in crop planting environment, characterized in that, The method includes: S1. Perform data cleaning on the crop planting environment data to obtain the current environmental data of the crop; S2. Extract historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop. S3. Evaluate the trend of the change pattern and fluctuation range of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence; S4. Based on the trend characteristics of the environmental data and the range of the growth environment parameters of the crop, determine the dynamic threshold parameters of the crop; S5. Based on the dynamic threshold parameter, perform adaptive anomaly detection on the current environmental data of the crop to obtain the anomaly signal of the current environmental data; S6. Encapsulate the abnormal signal using a communication protocol to obtain alarm information from the current environmental data.

2. The method for automatic alarm of abnormal data in crop planting environment as described in claim 1, characterized in that, The process of cleaning the crop planting environment data to obtain the current environmental data of the crop includes: Identify missing data points in the crop growing environment data; Interpolation processing is performed on the valid environmental data in adjacent time periods of the missing data points to obtain the environmental data sequence of the crop after filling in the missing data points; Based on the growth environment parameters of the crop, the abnormal deviation points in the filled environmental data sequence are adaptively calibrated to obtain the regularized environmental data of the crop. The current environmental data of the crop is obtained by performing a status assessment on the regularized environmental data.

3. The method for automatic alarm of abnormal data in crop planting environment as described in claim 1, characterized in that, The step of extracting historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop includes: Based on the current growth stage of the crop, determine the corresponding historical reference time interval for the crop; Based on the historical log dataset of the crop, the corresponding historical reference time interval is associated and mapped to obtain the crop's adaptive environment data record; The adapted environment data records are time-series aligned to obtain the historical data sequence of the crop.

4. The method for automatic alarm of abnormal data in crop planting environment as described in claim 1, characterized in that, The process of evaluating the change patterns and fluctuation ranges of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence includes: Based on the historical data sequence and the growth environment parameters of the crop, a sliding time window of appropriate length is set; The sliding time window is shifted along the time axis to obtain the environmental data segment of the sliding time window; Analyze the direction of change of environmental data within the sliding time window to determine the trend of the environmental data segment; The variability of the fluctuation range of environmental data within the sliding time window is evaluated to obtain the fluctuation characteristics of the environmental data segment; The trend and the fluctuation characteristics are combined to obtain the environmental data trend characteristics of the historical data sequence.

5. The method for automatic alarm of abnormal data in crop planting environment as described in claim 1, characterized in that, The step of determining the dynamic threshold parameters of the crop based on the trend characteristics of the environmental data and the range of the crop's growth environment parameters includes: Ecological niche fitting is performed on the trend characteristics of the environmental data to obtain the trend compatibility assessment results of the crop; Based on the trend compatibility assessment results and the threshold fluctuation space in the current trend of the crop, a dynamic threshold corridor for the crop is constructed. Threshold toughness enhancement is performed on the dynamic threshold corridor to obtain the disturbance-resistant dynamic threshold corridor of the crop; Based on the growth stage characteristics of the crop, the dynamic threshold corridor is calibrated for growth adaptability to obtain the dynamic threshold parameters of the crop, wherein the dynamic threshold parameters are adaptively adjusted with changes in environmental trends.

6. The method for automatic alarm of abnormal data in crop planting environment as described in claim 5, characterized in that, The process of performing niche fitting on the trend characteristics of the environmental data to obtain the trend compatibility assessment results of the crops includes: Based on the growth environment parameters of the crop, an environmental parameter benchmark framework for the ideal growth environment of the crop is constructed. The environmental data trend characteristics are coupled with the environmental parameter benchmark framework to evaluate the degree of coordination, and the ecological adaptability index of the crop is obtained. The calculation formula of the ecological adaptability index is as follows. ； In the formula, The ecological adaptability index is mentioned above. This represents the total environmental parameters of the crop. The first of the growth environment parameters of the crop The weighting coefficients of each environmental parameter. It is a natural exponential function. The first of the environmental data trend features Trend characteristic values ​​of environmental parameters For the environmental parameter benchmark framework, the first The baseline value of each environmental parameter For the first Preset tolerance parameters for each environmental parameter; Based on the ecological adaptability index and the physiological requirements of the crops, a two-way consistency diagnosis is performed on the trend characteristics of the environmental data to obtain the trend compatibility assessment results of the crops.

7. The method for automatic alarm of abnormal data in crop planting environment as described in claim 1, characterized in that, The step of adaptively detecting anomalies in the current environmental data of the crop based on the dynamic threshold parameter to obtain anomaly signals in the current environmental data includes: The current environmental data is compared with the dynamic threshold parameter in real time to obtain abnormal data points in the current environmental data. The abnormal data points are continuously verified to obtain a stable abnormal state of the current environmental data; Based on the dynamic threshold parameter, the stable abnormal state is assigned to a specific interval to obtain a graded abnormal signal for the stable abnormal state. Based on the trend characteristics of the environmental data, a trend consistency check is performed on the graded anomaly signals to obtain the anomaly signals of the current environmental data.

8. The method for automatic alarm of abnormal data in crop planting environment as described in claim 7, characterized in that, The step of determining the interval affiliation of the stable abnormal state based on the dynamic threshold parameter to obtain the hierarchical abnormal signal of the stable abnormal state includes: Based on the floating characteristics of the dynamic threshold parameter, the state of critical data points in the boundary region is determined. Based on the numerical range of the dynamic threshold parameter, anomaly level intervals with clearly defined boundaries are divided; By comparing the duration and degree of deviation in the stable abnormal state with the abnormality level interval, the specific range to which the stable abnormal state belongs is obtained. Based on the anomaly level interval attribute corresponding to the specific attribution interval, the hierarchical anomaly signal of the stable anomaly state is obtained.

9. The method for automatic alarm of abnormal data in crop planting environment as described in claim 1, characterized in that, The alarm information obtained by encapsulating the abnormal signal using a communication protocol to obtain the current environmental data includes: The abnormal signal is processed to construct a protocol, thereby obtaining the alarm data frame of the abnormal signal; Based on a preset communication protocol, the alarm data frames are load-integrated to obtain alarm information of the current environment data.

10. An automatic alarm system for abnormal data in crop planting environment, characterized in that, The system includes: An environmental data preprocessing module is used to clean the crop planting environment data to obtain the current environmental data of the crop. The historical data sequence construction module is used to extract historical environmental data from the historical log dataset of the crop within a specified time range to obtain the historical data sequence of the crop. The environmental trend analysis module is used to evaluate the change patterns and fluctuation ranges of the historical data sequence to obtain the environmental data trend characteristics of the historical data sequence. The dynamic threshold decision module is used to determine the dynamic threshold parameters of the crop based on the trend characteristics of the environmental data and the range of the growth environment parameters of the crop. An adaptive anomaly detection module is used to perform adaptive anomaly detection on the current environmental data of the crop based on the dynamic threshold parameter, and obtain anomaly signals of the current environmental data. The alarm information generation and encapsulation module is used to encapsulate the abnormal signal using a communication protocol to obtain alarm information from the current environmental data.

Images