Sugarcane planting risk analysis system based on multi-dimensional data

Abstract

The invention discloses a sugarcane planting risk analysis system based on multi-dimensional data, and relates to the field of risk early warning, and the system comprises a risk analysis management center which comprises a multi-dimensional data collection module, a multi-dimensional data processing module, a data cross analysis module, a multi-dimensional data integration module and a planting risk early warning module; the multi-dimensional data acquisition module acquires multi-dimensional risk influence data and real-time monitoring data; the multi-dimensional data processing module is used for performing causal classification processing on the multi-dimensional risk influence data and acquiring correlation data among classification processing results; a data cross analysis module performs cross statistics on classification processing results of different planting stages to obtain dynamic associated data; the multi-dimensional data integration module is used for integrating the correlation data and the dynamic association data; the planting risk early warning module carries out risk analysis and early warning processing on the real-time monitoring data according to the integration result; according to the method, the accuracy of the risk is improved by acquiring the multi-dimensional dynamics of the data analysis early warning process.

Description

Technical Field

[0001] The present invention relates to the technical field of risk early warning analysis, and in particular to a sugarcane planting risk analysis system based on multidimensional data. Background Art

[0002] Sugarcane is an important sugar crop in the world and an important source of raw materials for bioenergy. With the growth of the global population and economic development, the demand for sugar and bioenergy continues to increase, which has led to a continuous increase in the market demand for sugarcane, further promoting the development of the sugarcane planting industry. However, fluctuations in market demand also bring certain risks to sugarcane planting. In addition, the sugarcane planting process also faces various risks such as the natural environment and agricultural operations, which in turn cause economic losses to growers and enterprises.

[0003] Traditional risk warning analysis methods mainly rely on the analysis and processing of data from a limited single data source, and are difficult to handle complex multi-dimensional data. With the development of information technology, the Internet of Things and artificial intelligence technologies have been widely used, thereby improving the reliability of the risk warning analysis process.

[0004] After searching, the invention patent with Chinese patent number CN115936430A discloses an organic agricultural product asset management system based on blockchain, which relates to the field of agricultural product asset management technology. It solves the technical problem in the existing technology that organic agricultural products cannot conduct risk analysis on each link of the industrial chain and the operation of each link during production and sales. It judges the eligibility of the execution of the organic agricultural product production chain links, accurately analyzes the type of organic agricultural products, determines whether organic agricultural products are assets or liabilities, ensures the profitability of organic agricultural products, and prevents abnormal execution of the production chain links from causing organic agricultural products to fail to bring profits, so that organic agricultural product assets cannot be accurately managed; conducts risk analysis on the link operations of the analysis object, judges whether there are risks in the link operations of the analysis object, and prevents abnormal link operations from causing the profitability of the analysis object to decrease, and cannot reasonably manage the assets of current organic agricultural products, affecting the future planting planning of organic agricultural products.

[0005] Compared with the existing technology, the invention patent with Chinese patent number CN115936430A can analyze whether there are risks in the process of organic agricultural product cultivation from multiple perspectives, thereby realizing reasonable asset management of organic agricultural products and improving the income stability of organic agricultural products to a certain extent.

[0006] However, in the actual use of the above system, there may be lags in the execution of the production link in the organic agricultural product production chain due to dynamic changes, and the dynamic impact of dynamic data changes on the various links in the organic agricultural product production chain, thereby affecting the accuracy of the qualification judgment process of each production chain link. Summary of the Invention

[0007] The purpose of the present invention is to solve the shortcomings of the existing technology of insufficient accuracy and propose a sugarcane planting risk analysis system based on multidimensional data.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A sugarcane planting risk analysis system based on multidimensional data includes a risk analysis management center, wherein the risk analysis management center includes a multidimensional data acquisition module, a multidimensional data processing module, a data cross-analysis module, a multidimensional data integration module, and a planting risk early warning module;

[0010] The multi-dimensional data acquisition module is used to collect multi-dimensional risk impact data affecting the sugarcane planting process and real-time monitoring data of the current sugarcane planting process;

[0011] The multidimensional data processing module is used to respectively set corresponding impact data sets for the obtained multidimensional risk impact data according to corresponding data types, and perform causal classification on the impact data sets, obtain cause-oriented factors and result-oriented factors, perform statistical analysis on each impact data set according to the corresponding guiding factors, set corresponding planting process sequences and segmented subsequences, and obtain correlation data between each cause-oriented factor and result-oriented factor in the corresponding planting process sequence and segmented subsequence, as well as between the result-oriented factor and the risk type data;

[0012] The data cross-analysis module is used to perform cross-statistical analysis on the corresponding cause-oriented factors and result-oriented factors in each planting process sequence and segmented subsequence, obtain corresponding cross-statistical data sets, and obtain corresponding dynamic correlation data according to the corresponding cross-statistical data sets;

[0013] The multidimensional data integration module is used to analyze and process the correlation data and dynamic association data between the cause-oriented factors and result-oriented factors corresponding to the corresponding planting process sequence and segmented subsequence, and the result-oriented factors and risk type data, and to construct the corresponding sequence impact data model and cross-sequence impact model;

[0014] The planting risk warning module is used to perform risk analysis based on real-time monitoring data, input them into the sequence impact data model and the cross-sequence impact model respectively, obtain corresponding risk assessment data and cross-risk assessment data, perform comprehensive analysis based on the risk assessment data and the cross-risk assessment data, obtain comprehensive risk assessment results, and perform warning processing based on the comprehensive risk assessment results.

[0015] The above technical solution further includes: the process of collecting multi-dimensional risk impact data and real-time monitoring data includes:

[0016] Setting up a multi-dimensional data acquisition unit and a real-time monitoring acquisition unit;

[0017] Collecting multidimensional risk impact data during the sugarcane planting process in the corresponding planting area through a multidimensional data collection unit, wherein the multidimensional risk impact data includes meteorological impact data, planting impact data, market impact data, and risk type data;

[0018] The real-time monitoring data of the current sugarcane planting process in the corresponding planting area is collected through the real-time monitoring collection unit.

[0019] Furthermore, the process of causal classification of multi-dimensional risk impact data includes:

[0020] Obtain multi-dimensional risk impact data, and set corresponding impact data sets for different types of data information in the multi-dimensional risk impact data according to corresponding risk type data;

[0021] Perform feature extraction on the impact data sets respectively to obtain assessment feature data of risk type data of different types of data information, and preset causal classification assessment feature standard data;

[0022] The obtained evaluation feature data are compared and analyzed with the causal classification evaluation feature standard data respectively, and the causal type corresponding to the corresponding type of multidimensional risk impact data is obtained according to the comparative analysis results. The causal type includes two types: cause-oriented factors and result-oriented factors.

[0023] Furthermore, the process of setting the planting process sequence and segmented subsequence includes:

[0024] Obtain multidimensional risk impact data corresponding to the result-oriented factors, perform a statistical importance assessment on the multidimensional risk impact data corresponding to the risk type data of the result-oriented factors, sort them according to the statistical importance assessment results, and obtain the multidimensional risk impact data corresponding to the result-oriented factor with the greatest importance data;

[0025] The obtained multidimensional risk influencing factors of corresponding types are segmented and processed to obtain segmented periods in the sugarcane planting process. The corresponding planting process sequences are set according to the corresponding segmented periods. The standard evaluation indicators of the corresponding multidimensional risk influencing factors in the planting process sequences are obtained. The segmented processing is performed according to the standard evaluation indicators, and the segmented subsequences are set according to the segmented processing results in the planting process sequences.

[0026] In each planting process sequence and segmented subsequence, corresponding storage space is set according to the corresponding multidimensional risk impact data type, and each storage space is numbered according to the importance data, and the obtained multidimensional risk impact data is stored in the corresponding storage space.

[0027] Furthermore, the process of obtaining the correlation data between each cause-oriented factor and result-oriented factor in the corresponding planting process sequence and segmented subsequence includes:

[0028] Acquire the multidimensional risk impact data stored in the corresponding storage space in each segmented subsequence within the corresponding planting process sequence, obtain the multidimensional risk impact data corresponding to the corresponding result-oriented factor, combine the multidimensional risk impact data collected in other storage spaces during the same period, and obtain a subsequence data group; set a subsequence data group set according to the subsequence data group corresponding to the multidimensional risk impact data stored in the segmented subsequence;

[0029] Comparatively extracting the multi-dimensional risk impact data corresponding to the corresponding result-oriented factors and cause-oriented factors in the subsequence data set, as well as the result-oriented factors and risk type data, to obtain the corresponding statistical subsequence data set;

[0030] Based on the machine learning algorithm, the obtained statistical subsequence data sets are analyzed and processed respectively to obtain the corresponding result-oriented factors and cause-oriented factors, as well as the correlation data between the result-oriented factors and risk type data.

[0031] Furthermore, the process of cross-statistical analysis of the corresponding cause-oriented factors and result-oriented factors in each planting process sequence and segmented subsequence includes:

[0032] According to the planting process sequence, segmented subsequence and storage space involved in the sugarcane planting process, graph nodes, graph subnodes and segmented subnodes are set respectively, and the corresponding sequence graph is generated;

[0033] Perform multiple cross-progressive management on the corresponding graph nodes, graph sub-nodes and sub-sub-nodes in the sequence graph, and obtain the corresponding cross-statistical data set based on the cross-progressive management results;

[0034] The corresponding multidimensional risk impact data in the obtained cross-statistical data set are mapped and aligned according to the subsequence data group, and the corresponding data information is subjected to deviation analysis based on the mapping alignment results to obtain the corresponding alignment deviation data, and the various alignment deviation data in the cross-statistical data set are integrated and statistically analyzed to obtain the corresponding dynamic correlation data.

[0035] Furthermore, the process of building a sequence impact data model and a cross-sequence impact model includes:

[0036] Obtaining the correlation data between the corresponding result-oriented factors and cause-oriented factors and the result-oriented factors and risk type data, as well as the dynamic correlation data corresponding to each cross-statistical data set;

[0037] Sequence risk indicator data are set respectively according to the correlation data between the result-oriented factors and cause-oriented factors in each segmented subsequence within the corresponding planting impact sequence and the result-oriented factors and risk type data, and the obtained sequence risk indicator data are used to generate a sequence risk indicator data set;

[0038] Cross-sequence risk indicator data are set according to the dynamic correlation data corresponding to the cross-statistical data sets between the corresponding planting impact sequences and between each segmented subsequence, and the obtained cross-sequence risk indicator data are used to generate a cross-sequence risk indicator data set;

[0039] Based on the machine learning algorithm, the sequence risk indicator data sets corresponding to each segmented subsequence and the cross-sequence risk indicator data sets corresponding to the planting impact sequences and each segmented subsequence are analyzed and processed respectively to construct the corresponding sequence impact data model and cross-sequence impact model.

[0040] Furthermore, the process of obtaining comprehensive risk assessment results includes:

[0041] Obtain real-time monitoring data of the current sugarcane planting process in the corresponding planting area, analyze and process the real-time monitoring data, and obtain the current planting process sequence and segmented subsequence;

[0042] Inputting the corresponding real-time monitoring data into the corresponding sequence impact data model and the cross-sequence impact model for analysis and processing, respectively, to obtain the corresponding risk assessment data and cross-risk assessment data;

[0043] According to the risk assessment data and cross-risk assessment data, the corresponding risk type data are obtained respectively, the deviation analysis is performed on the obtained risk type data, the risk deviation data is obtained, and it is determined whether the corresponding risk deviation data meets the preset deviation standard. If it does, a comprehensive risk assessment result is generated, and risk warning information is generated according to the comprehensive risk assessment result, and warning processing is performed according to the risk warning information.

[0044] The present invention has the following beneficial effects:

[0045] In the present invention, by obtaining multidimensional risk impact data, setting causal classification between the multidimensional risk impact data, setting cause-oriented factors and result-oriented factors according to the causal classification results, and analyzing and processing different guiding factors in different stages of the sugarcane planting process, the data processing efficiency in the sugarcane planting risk early warning analysis process is improved.

[0046] In the present invention, the corresponding planting process sequence, segmented subsequence and corresponding storage space are set according to the causal classification results corresponding to the multi-dimensional risk impact data to manage the entire sugarcane planting process, and the correlation data between different types of data in the same stage, the dynamic association data between the same type of data in different stages and the same type of data in different stages are set, so as to dynamically manage the sugarcane planting process, thereby reducing the impact of the lag of risk factors caused by dynamic changes in the sugarcane planting link on the risk warning analysis process. In addition, through cross-statistical analysis, the possibility of insufficient accuracy in the risk warning analysis process caused by the dynamic impact of dynamic changes in data on various processes of sugarcane planting is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a structural diagram of a sugarcane planting risk analysis system based on multidimensional data proposed by the present invention. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention. Example

[0049] like Figure 1 As shown, the present invention proposes a sugarcane planting risk analysis system based on multidimensional data, including a risk analysis management center, which includes a multidimensional data acquisition module, a multidimensional data processing module, a data cross-analysis module, a multidimensional data integration module and a planting risk early warning module.

[0050] In this embodiment, the risk analysis management center sequentially analyzes and processes the multidimensional risk data types that affect the sugarcane planting process, determines the dynamic impact of the sugarcane planting risk corresponding to each risk data type, and performs integrated analysis based on the dynamic impact results. Ultimately, risk data for the sugarcane planting process is obtained, and early warning analysis is performed based on the corresponding risk data. The specific implementation process includes the following steps:

[0051] The multi-dimensional data acquisition module is used to collect multi-dimensional risk impact data affecting the sugarcane planting process and real-time monitoring data of the current sugarcane planting process. The specific implementation process includes:

[0052] Setting up a multi-dimensional data acquisition unit and a real-time monitoring acquisition unit;

[0053] The multi-dimensional data acquisition unit is used to collect multi-dimensional risk impact data involved in the sugarcane planting risk assessment process, and the process includes:

[0054] Acquire planting area information corresponding to the corresponding sugarcane planting area, wherein the planting area information includes corresponding geographic space information, soil basic information, infrastructure information, and surrounding environment information, perform verification and analysis on the obtained planting area information, and mark the corresponding planting area information according to the verification and analysis results;

[0055] Collect corresponding multi-dimensional risk impact data based on the labeling results of the planting area information, wherein the multi-dimensional risk impact data is the historical planting risk impact data corresponding to the corresponding planting area information, including meteorological impact data, planting impact data, market impact data, and risk type data;

[0056] The meteorological impact data include meteorological temperature data, precipitation and irrigation data, light data and other meteorological data;

[0057] The planting impact data includes growth status data, pest and disease related data, farming operation data and other planting data;

[0058] The risk type data includes the risk types existing in the corresponding sugarcane planting process and the corresponding risk occurrence data;

[0059] The market impact data includes cost data, supply and demand data, price data and other market impact data;

[0060] Sending the obtained meteorological impact data, planting impact data, market impact data and risk type data to the multidimensional data processing module respectively;

[0061] The real-time monitoring and collection unit is used to collect real-time monitoring data corresponding to various multi-dimensional risk impact data in the corresponding sugarcane planting area, and the process includes:

[0062] The real-time monitoring and collection unit is provided with a real-time monitoring terminal, which collects real-time monitoring data corresponding to different types of multi-dimensional risk impact data, wherein the real-time monitoring terminal includes a meteorological monitoring terminal, a planting monitoring terminal and a market monitoring terminal;

[0063] The meteorological monitoring terminal is used to collect meteorological monitoring data corresponding to the corresponding sugarcane planting area;

[0064] The planting monitoring terminal is used to collect planting monitoring data corresponding to the corresponding sugarcane planting area;

[0065] The market monitoring terminal is used to collect market monitoring data corresponding to the corresponding sugarcane planting area;

[0066] The obtained meteorological monitoring data, planting monitoring data and market monitoring data are respectively associated with the corresponding multi-dimensional risk impact data and marked respectively.

[0067] The multidimensional data processing module is used to set corresponding impact data sets for the obtained multidimensional risk impact data according to corresponding data types, and perform causal classification on the impact data sets, obtain cause-oriented factors and result-oriented factors, perform statistical analysis on each impact data set according to the corresponding guiding factors, set corresponding planting process sequences and segmented subsequences, and obtain correlation data between each cause-oriented factor and result-oriented factor in the corresponding planting process sequence and segmented subsequence, as well as between the result-oriented factor and risk type data. The specific implementation process includes:

[0068] Set up classification processing units, process processing units and related analysis units;

[0069] The classification processing unit is used to perform causal classification processing on the multi-dimensional risk impact data, and the process includes:

[0070] Acquire multidimensional risk impact data, set impact data sets for the acquired multidimensional risk impact data according to corresponding data types, wherein the impact data sets correspond to three types: meteorological impact data, planting impact data, market impact data, and risk occurrence data, and label the corresponding impact data sets according to the corresponding data types;

[0071] Setting corresponding impact data subsets for the corresponding impact data types in the impact data set corresponding to the corresponding data types, performing marking processing on each impact data subset, and performing marking processing according to the marking processing result;

[0072] Perform feature extraction on the impact data sets according to a machine learning algorithm, establish mapping relationships between the impact data sets corresponding to the corresponding types and the impact data sets corresponding to the risk type data, and obtain corresponding assessment feature data according to the corresponding mapping relationships;

[0073] The risk analysis management center pre-sets corresponding causal classification assessment feature standard data based on the multi-dimensional risk impact data;

[0074] Obtain the corresponding causal classification assessment feature standard data, compare and analyze the obtained assessment feature data with the causal classification assessment feature standard data, and obtain the causal type corresponding to the corresponding type of multi-dimensional risk impact data based on the comparative analysis results. The causal type includes two types: cause-oriented factors and result-oriented factors, among which:

[0075] The types of cause-oriented factors include meteorological impact data, pest and disease related data, agricultural operation data, etc.

[0076] The types of result-oriented factors corresponding to cause-oriented factors include market impact data and growth status data;

[0077] The process processing unit is used to set the planting process sequence and segmented subsequence, and the process includes:

[0078] Extracting single causal factors from the impact data subsets corresponding to the result-oriented factors and each cause-oriented factor, and obtaining a single causal factor extraction combination, wherein the single causal factor extraction combination includes the cause-oriented factors corresponding to the corresponding result-oriented factors;

[0079] Perform comparative analysis on the obtained single factor extraction combinations to obtain corresponding correlation data, and mark the corresponding correlation data as importance data corresponding to the impact data subset corresponding to the corresponding cause-oriented factor, sort according to the importance data, and obtain the multidimensional risk impact data corresponding to the result-oriented factor with the largest importance data, which corresponds to the multidimensional risk impact data corresponding to the growth status data;

[0080] The whole process of sugarcane planting is segmented according to the growth status data, and is set as five stages: germination stage, seedling stage, tillering stage, elongation stage and maturity stage;

[0081] Performing a standard evaluation process on the segmented periods corresponding to the impact data subset, respectively obtaining the standard evaluation indicators corresponding to the corresponding segmented periods, and setting the planting process sequence corresponding to the corresponding impact data subset for the corresponding segmented periods according to the corresponding standard evaluation indicators;

[0082] Obtaining the standard evaluation indicators corresponding to each planting process sequence corresponding to the impact data subset corresponding to the result-oriented factor, segmenting the standard evaluation indicators corresponding to the corresponding planting process sequence, and setting corresponding segmented subsequences respectively;

[0083] Obtain the planting process sequence and segmented subsequence corresponding to the result-oriented factor, set corresponding storage space in each planting process sequence and segmented subsequence according to the corresponding multidimensional risk impact data type, number each storage space according to importance data, and store the obtained multidimensional risk impact data in the corresponding storage space;

[0084] Marking the planting process sequence, segmented subsequence and storage space corresponding to the impact data set corresponding to the obtained multi-dimensional risk impact data, and sending the marking results to the relevant analysis unit for analysis and processing;

[0085] The correlation analysis unit is used to obtain the correlation data between each cause-oriented factor and result-oriented factor in the corresponding planting process sequence and segmented subsequence, and the process includes:

[0086] Obtain the planting process sequence, segmented subsequence, and storage space of the impact data set corresponding to the multi-dimensional risk impact data, and perform correlation analysis on the corresponding planting process sequence, segmented subsequence, and storage space;

[0087] Perform a combined analysis of influencing factors according to the planting process sequences corresponding to the result-oriented factors, sequentially number the corresponding segmented subsequences according to the sorting order of the corresponding planting process sequences to which they belong, extract each segmented subsequence within each planting process sequence according to the sequence number, obtain the planting process sequence corresponding to each multidimensional risk impact data corresponding to the corresponding sequence number, set the corresponding subsequence data group according to the corresponding sequence number for the obtained data information, and sequentially integrate the subsequence data groups corresponding to the corresponding sequence numbers to generate a subsequence data group set corresponding to the corresponding sequence number;

[0088] Comparatively extracting the multi-dimensional risk impact data corresponding to the corresponding result-oriented factors and cause-oriented factors in the subsequence data set, as well as the result-oriented factors and risk type data, and sequentially obtaining corresponding progressive statistical subsequence data sets;

[0089] Correlation statistical analysis is performed on the data types corresponding to various multidimensional risk impact data in the obtained statistical subsequence data sets, and the obtained statistical subsequence data sets are analyzed and processed based on the machine learning algorithm to obtain the risk weight coefficients and risk assessment index data of the multidimensional risk impact data corresponding to the corresponding result-oriented factors and cause-oriented factors, as well as the result-oriented factors and risk type data, and the obtained risk weight coefficients and risk assessment index data are set as the corresponding correlation data.

[0090] The data cross-analysis module is used to perform cross-statistical analysis on the corresponding cause-oriented factors and result-oriented factors in each planting process sequence and segmented subsequence, obtain corresponding cross-statistical data sets, and obtain corresponding dynamic correlation data based on the corresponding cross-statistical data sets. The specific implementation process includes:

[0091] Set up cross-analysis units and correlation analysis units;

[0092] The cross analysis unit is used to perform cross statistical analysis on various planting process sequences between the obtained planting impact process data to obtain corresponding dynamic correlation data, and the process includes:

[0093] Obtaining the sequence numbers corresponding to the corresponding planting impact sequences, segmented subsequences, and storage spaces, integrating the data corresponding to the planting impact sequences, segmented subsequences, and storage spaces corresponding to different sequence numbers, and setting a sequence map of the planting impact sequences, segmented subsequences, and storage spaces corresponding to the multidimensional risk impact data based on the integration results, wherein corresponding map nodes, map sub-nodes, and subdivision sub-nodes are respectively set in the sequence map based on the corresponding planting process sequences, segmented sub-sequences, and storage spaces, and each segmented sub-sequence is connected and managed through the map nodes;

[0094] Perform cross statistics on the obtained sequence graphs, perform multiple cross progressive management according to the position information of the sequence graphs to which the graph nodes belong, obtain the cross combinations of the corresponding graph nodes in sequence, obtain the cross combinations of the corresponding graph sub-nodes according to the cross combinations of the graph nodes, obtain the cross combinations of the corresponding sub-nodes according to the cross combinations of the graph sub-nodes, thereby obtaining the corresponding cross statistical data sets in sequence, and mark the graph nodes, graph sub-nodes and sub-nodes corresponding to the obtained cross statistical data sets;

[0095] The association analysis unit is used to analyze and process the obtained cross-statistical data set to obtain corresponding dynamic association data, and the process includes:

[0096] Obtaining a mapping alignment relationship between corresponding data in a corresponding cross-statistical dataset, performing deviation analysis on pairwise corresponding data information according to the corresponding mapping alignment relationship, obtaining corresponding alignment deviation data, integrating and statistically analyzing the alignment deviation data in the cross-statistical dataset, determining variance data and mean data corresponding to the corresponding alignment deviation data, performing dynamic correlation verification on the corresponding alignment deviation data based on the obtained variance data and mean data, and determining whether the corresponding cross-statistical dataset meets the dynamic correlation standard;

[0097] Corresponding dynamic correlation data are set for the cross-statistical data sets that meet the dynamic correlation standards. The dynamic correlation data include the dynamic change relationship corresponding to the corresponding multi-dimensional risk impact data in the corresponding cross-statistical data sets, that is, the corresponding cross-risk weight coefficient and cross-risk assessment indicator data.

[0098] The multidimensional data integration module is used to analyze and process the correlation data and dynamic association data between the cause-oriented factors and result-oriented factors corresponding to the corresponding planting process sequence and segmented subsequence, and the result-oriented factors and risk type data, and to construct the corresponding sequence impact data model and cross-sequence impact model. The specific implementation process includes:

[0099] Set up sequence processing units and cross-integration units;

[0100] The sequence processing unit is used to construct a sequence impact model based on the correlation data between the cause-oriented factors and the result-oriented factors corresponding to the corresponding planting process sequence and the segmented subsequence. The process includes:

[0101] Corresponding sequence risk indicator data are respectively set according to the correlation data between the result-oriented factors and the cause-oriented factors in each segmented subsequence within the corresponding planting impact sequence and the result-oriented factors and the risk type data, wherein the sequence risk indicator data is set by the risk weight coefficient and the risk assessment indicator data corresponding to the corresponding type of multidimensional risk impact data with respect to the corresponding risk type data, and a sequence risk indicator data set is generated from the obtained sequence risk indicator data, and the corresponding sequence risk indicator data set is divided into a training set and a validation set;

[0102] The training set corresponding to the obtained sequence impact data set is analyzed and processed based on the deep learning algorithm, and a sequence impact data model corresponding to the corresponding segmented subsequences in each planting process sequence is constructed. The obtained sequence impact data model is verified and analyzed based on the corresponding verification set until the corresponding sequence impact data model meets the verification requirements and the corresponding sequence impact data model is output;

[0103] The cross integration unit is used to construct a cross sequence impact model based on the dynamic correlation data between each result-oriented factor and risk type data corresponding to the corresponding planting process sequence and segmented subsequence, and the process includes:

[0104] According to the dynamic correlation data corresponding to each cross-statistical data set between the result-oriented factors and the cause-oriented factors between each segmented subsequence of the corresponding planting impact sequence, corresponding cross-sequence risk indicator data are respectively set, wherein the cross-sequence risk indicator data is set by the dynamically changing cross-risk weight coefficient and cross-risk assessment indicator data corresponding to the corresponding risk type data between the multidimensional risk impact data between the corresponding sequences, and a sequence risk indicator data set is generated from the obtained sequence risk indicator data, and the corresponding sequence risk indicator data set is divided into a training set and a validation set;

[0105] Based on the deep learning algorithm, the training set corresponding to the obtained cross-sequence influence data set is analyzed and processed, and a cross-sequence influence data model corresponding to the corresponding segmented subsequences in each planting process sequence is constructed. The obtained cross-sequence influence data model is verified and analyzed according to the corresponding verification set until the corresponding cross-sequence influence data model meets the verification requirements and the corresponding cross-sequence influence data model is output.

[0106] The planting risk warning module is used to perform risk analysis based on real-time monitoring data, input them into the sequence impact data model and the cross-sequence impact model respectively, obtain corresponding risk assessment data and cross-risk assessment data, perform comprehensive analysis based on the risk assessment data and the cross-risk assessment data, obtain comprehensive risk assessment results, and perform early warning processing based on the comprehensive risk assessment results. The specific implementation process includes:

[0107] Obtain real-time monitoring data of the current sugarcane planting process in the corresponding planting area, analyze and process the real-time monitoring data, and obtain the current planting process sequence and segmented subsequence;

[0108] Inputting the corresponding real-time monitoring data into the corresponding sequence impact data model and the cross-sequence impact model for analysis and processing, respectively, to obtain the corresponding risk assessment data and cross-risk assessment data;

[0109] Obtain corresponding risk type data based on risk assessment data and cross-risk assessment data, perform bias analysis on the obtained risk type data, obtain evaluation indicators corresponding to the corresponding risk type data, perform analysis and calculation based on the Spearman correlation coefficient, and obtain corresponding risk deviation data;

[0110] Analyze and process the obtained risk deviation data to determine whether the corresponding risk deviation data meets the preset deviation standard, set the deviation standard interval, and compare and analyze the risk deviation data with the deviation standard interval;

[0111] If the risk deviation data falls within the deviation standard range, a comprehensive risk assessment result is generated;

[0112] If the risk deviation data does not fall within the deviation standard range, no comprehensive risk assessment result will be generated, and risk assessment abnormality information will be generated and fed back to the risk analysis management center;

[0113] Risk warning information is generated based on the comprehensive risk assessment results obtained. The risk warning information includes corresponding risk type data, the planting process sequence and the segmented subsequence, and warning processing is performed based on the risk warning information.

[0114] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A sugarcane planting risk analysis system based on multidimensional data, including a risk analysis management center, characterized in that: The risk analysis and management center includes a multidimensional data acquisition module, a multidimensional data processing module, a data cross-analysis module, a multidimensional data integration module and a planting risk early warning module; The multi-dimensional data acquisition module is used to collect multi-dimensional risk impact data affecting the sugarcane planting process and real-time monitoring data of the current sugarcane planting process; The multidimensional data processing module is used to respectively set corresponding impact data sets for the obtained multidimensional risk impact data according to corresponding data types, and perform causal classification on the impact data sets, obtain cause-oriented factors and result-oriented factors, perform statistical analysis on each impact data set according to the corresponding guiding factors, set corresponding planting process sequences and segmented subsequences, and obtain correlation data between each cause-oriented factor and result-oriented factor in the corresponding planting process sequence and segmented subsequence, as well as between the result-oriented factor and the risk type data; The data cross-analysis module is used to perform cross-statistical analysis on the corresponding cause-oriented factors and result-oriented factors in each planting process sequence and segmented subsequence, obtain corresponding cross-statistical data sets, and obtain corresponding dynamic correlation data according to the corresponding cross-statistical data sets; The multidimensional data integration module is used to analyze and process the correlation data and dynamic association data between the cause-oriented factors and result-oriented factors corresponding to the corresponding planting process sequence and segmented subsequence, and the result-oriented factors and risk type data, and to construct the corresponding sequence impact data model and cross-sequence impact model; The planting risk warning module is used to perform risk analysis based on real-time monitoring data, input them into the sequence impact data model and the cross-sequence impact model respectively, obtain corresponding risk assessment data and cross-risk assessment data, perform comprehensive analysis based on the risk assessment data and the cross-risk assessment data, obtain comprehensive risk assessment results, and perform warning processing based on the comprehensive risk assessment results.

2. The sugarcane planting risk analysis system based on multidimensional data according to claim 1, characterized in that: The process of collecting multi-dimensional risk impact data and real-time monitoring data includes: Setting up a multi-dimensional data acquisition unit and a real-time monitoring acquisition unit; Collecting multidimensional risk impact data during the sugarcane planting process in the corresponding planting area through a multidimensional data collection unit, wherein the multidimensional risk impact data includes meteorological impact data, planting impact data, market impact data, and risk type data; The real-time monitoring data of the current sugarcane planting process in the corresponding planting area is collected through the real-time monitoring collection unit.

3. The sugarcane planting risk analysis system based on multidimensional data according to claim 2, characterized in that: The process of causal classification of multi-dimensional risk impact data includes: Obtain multi-dimensional risk impact data, and set corresponding impact data sets for different types of data information in the multi-dimensional risk impact data according to corresponding risk type data; Perform feature extraction on the impact data sets respectively to obtain assessment feature data of risk type data of different types of data information, and preset causal classification assessment feature standard data; The obtained evaluation feature data are compared and analyzed with the causal classification evaluation feature standard data respectively, and the causal type corresponding to the corresponding type of multidimensional risk impact data is obtained according to the comparative analysis results. The causal type includes two types: cause-oriented factors and result-oriented factors.

4. The sugarcane planting risk analysis system based on multidimensional data according to claim 3, characterized in that: The process of setting up the planting process sequence and segmented subsequences includes: Obtain multidimensional risk impact data corresponding to the result-oriented factors, perform a statistical importance assessment on the multidimensional risk impact data corresponding to the risk type data of the result-oriented factors, sort them according to the statistical importance assessment results, and obtain the multidimensional risk impact data corresponding to the result-oriented factor with the greatest importance data; The obtained multidimensional risk influencing factors of corresponding types are segmented and processed to obtain segmented periods in the sugarcane planting process. The corresponding planting process sequences are set according to the corresponding segmented periods. The standard evaluation indicators of the corresponding multidimensional risk influencing factors in the planting process sequences are obtained. The segmented processing is performed according to the standard evaluation indicators, and the segmented subsequences are set according to the segmented processing results in the planting process sequences. In each planting process sequence and segmented subsequence, corresponding storage space is set according to the corresponding multidimensional risk impact data type, and each storage space is numbered according to the importance data, and the obtained multidimensional risk impact data is stored in the corresponding storage space.

5. The sugarcane planting risk analysis system based on multidimensional data according to claim 4, characterized in that: The process of obtaining the correlation data between each cause-oriented factor and result-oriented factor within the corresponding planting process sequence and segmented subsequence includes: Acquire the multidimensional risk impact data stored in the corresponding storage space in each segmented subsequence within the corresponding planting process sequence, obtain the multidimensional risk impact data corresponding to the corresponding result-oriented factor, combine the multidimensional risk impact data collected in other storage spaces during the same period, and obtain a subsequence data group; set a subsequence data group set according to the subsequence data group corresponding to the multidimensional risk impact data stored in the segmented subsequence; Comparatively extracting the multi-dimensional risk impact data corresponding to the corresponding result-oriented factors and cause-oriented factors in the subsequence data set, as well as the result-oriented factors and risk type data, to obtain the corresponding statistical subsequence data set; Based on the machine learning algorithm, the obtained statistical subsequence data sets are analyzed and processed respectively to obtain the correlation data between the corresponding result-oriented factors and cause-oriented factors, as well as the result-oriented factors and risk type data.

6. The sugarcane planting risk analysis system based on multidimensional data according to claim 5, characterized in that: The process of cross-statistical analysis of the corresponding cause-oriented factors and result-oriented factors in each planting process sequence and segmented subsequence includes: According to the planting process sequence, segmented subsequence and storage space involved in the sugarcane planting process, graph nodes, graph subnodes and segmented subnodes are set respectively, and the corresponding sequence graph is generated; Perform multiple cross-progressive management on the corresponding graph nodes, graph sub-nodes and sub-sub-nodes in the sequence graph, and obtain the corresponding cross-statistical data set based on the cross-progressive management results; The corresponding multidimensional risk impact data in the obtained cross-statistical data set are mapped and aligned according to the subsequence data group, and the corresponding data information is subjected to deviation analysis based on the mapping alignment results to obtain the corresponding alignment deviation data, and the various alignment deviation data in the cross-statistical data set are integrated and statistically analyzed to obtain the corresponding dynamic correlation data.

7. The sugarcane planting risk analysis system based on multidimensional data according to claim 6, characterized in that: The process of building a sequence impact data model and a cross-sequence impact model includes: Obtaining the correlation data between the corresponding result-oriented factors and cause-oriented factors and the result-oriented factors and risk type data, as well as the dynamic correlation data corresponding to each cross-statistical data set; Sequence risk indicator data are set respectively according to the correlation data between the result-oriented factors and cause-oriented factors in each segmented subsequence within the corresponding planting impact sequence and the result-oriented factors and risk type data, and the obtained sequence risk indicator data are used to generate a sequence risk indicator data set; Cross-sequence risk indicator data are set according to the dynamic correlation data corresponding to the cross-statistical data sets between the corresponding planting impact sequences and between each segmented subsequence, and the obtained cross-sequence risk indicator data are used to generate a cross-sequence risk indicator data set; Based on the machine learning algorithm, the sequence risk indicator data sets corresponding to each segmented subsequence and the cross-sequence risk indicator data sets corresponding to the planting impact sequences and each segmented subsequence are analyzed and processed respectively to construct the corresponding sequence impact data model and cross-sequence impact model.

8. The sugarcane planting risk analysis system based on multidimensional data according to claim 7, characterized in that: The process of obtaining a comprehensive risk assessment includes: Obtain real-time monitoring data of the current sugarcane planting process in the corresponding planting area, analyze and process the real-time monitoring data, and obtain the current planting process sequence and segmented subsequence; Inputting the corresponding real-time monitoring data into the corresponding sequence impact data model and the cross-sequence impact model for analysis and processing, respectively, to obtain the corresponding risk assessment data and cross-risk assessment data; According to the risk assessment data and cross-risk assessment data, the corresponding risk type data are obtained respectively, the deviation analysis is performed on the obtained risk type data, the risk deviation data is obtained, and it is determined whether the corresponding risk deviation data meets the preset deviation standard. If it does, a comprehensive risk assessment result is generated, and risk warning information is generated according to the comprehensive risk assessment result, and warning processing is performed according to the risk warning information.

Images