Agricultural industry economic risk tracking and early warning method and system based on key indexes

Abstract

The invention discloses an agricultural industry economic risk tracking and early warning method and system based on key indexes, and relates to the technical field of agricultural economic monitoring and early warning. The method comprises the following steps: S1, data acquisition and preprocessing: S1-1, deploying an Internet of Things sensor, a market monitoring terminal and a related acquisition system, and acquiring data of a production end, a circulation end and a consumption end in real time; s1-2, eliminating the seasonal fluctuation of the time series data by adopting a CensuusX12 seasonal adjustment method; s2, key warning condition index construction: taking the agricultural product price fluctuation ratio as a core warning condition index, and taking the ratio of the agricultural product fluctuation ratio to the CPI fluctuation ratio as a warning limit; s3, dynamic alarm limit division; s4, constructing a hybrid early warning model; and S5, outputting early warning information to a user side. According to the method, multi-dimensional data such as production, consumption, price and circulation paths are integrated, the price prediction error rate is effectively reduced through a hybrid model architecture in combination with a classical statistical model and a machine learning algorithm, the adverse effect on monitoring precision due to existence of a blind area in monitoring is avoided, and the early warning precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural economic monitoring and early warning, and specifically relates to a method and system for tracking and early warning agricultural industrial economic risks based on multi-source data analysis, intelligent model integration and dynamic risk prediction, which are applicable to the monitoring of agricultural product market risks, policy formulation and decision-making support for market players. Background Art

[0002] In recent years, although the level of agricultural product market monitoring and early warning in China has been improved to a certain extent, due to the weak foundation caused by various reasons and the limitation of the social tolerance and lenient atmosphere, the release of many information lacks timeliness or accuracy, and the role of information in guiding agricultural production and sales is greatly reduced.

[0003] The prior art discloses the theoretical method and application for constructing a cluster of agricultural product monitoring and early warning models. This method creates a cluster of monitoring and early warning models with multiple spatio-temporal dimensions by constructing a method of "factor classification decoupling and parameter conversion and adaptation" for multiple varieties of agricultural products. These model clusters can analyze and predict supply and demand factors such as the production volume, consumption volume, trade volume and price of different agricultural products. This method uses historical data to establish a regression equation to solve each coefficient, assigns weights to each single property through expert manual setting and intelligent training, and uses remote sensing and monitoring data to comprehensively evaluate the supply and demand situation of agricultural products.

[0004] The prior art also discloses the research on the method for assessing agricultural catastrophe risks in China. This research obtains agricultural disaster loss data by using the affected area, damaged area and completely damaged area of crops, and uses Monte Carlo simulation technology and extreme value POT model to effectively fit the tail distribution of agricultural disaster losses, establishes a generalized Pareto distribution model for agricultural catastrophe losses, and an accurate measurement model for agricultural catastrophe risks based on the VaR method. This research collects agricultural disaster situation data, analyzes the data by using statistical and simulation techniques to evaluate agricultural catastrophe risks, and constructs a risk assessment model.

[0005] The prior art also discloses a quantitative analysis method for agricultural risk decision-making. This method introduces quantitative analysis methods for agricultural production decisions under risk, such as the mean-variance (E, V) analysis method, the mean-standard deviation (E, σ) analysis method, etc., which are based on the expected utility theory. By establishing a mathematical model to quantify risks and using statistical methods to analyze risk factors in agricultural production decisions.

[0006] The prior art also discloses the new development of the food security monitoring and early warning system under the background of big data. Big data technology provides real-time and comprehensive information support in food security monitoring and early warning, and strengthens the monitoring and response to potential risks. This system realizes the real-time monitoring and early warning of food security risks by integrating multi-source data. It uses big data technology to collect and analyze data, and predicts and evaluates food security risks by establishing models and algorithms.

[0007] The prior art also discloses risk identification and control strategies for food security from the perspective of the entire industrial chain. Starting from the perspective of the entire industrial chain, this research identifies food security risks and develops a set of risk control strategies. The strategy includes a risk identification, monitoring, and early warning system that combines early identification, normal monitoring, and emergency early warning. By building a risk assessment and early warning platform, integrating resources from relevant business departments, scientific research institutions, etc., developing an intelligent method system for food security risk assessment and early warning, and building a national (local) food security risk information platform.

[0008] In summary, the following problems still exist in the prior art:

[0009] 1. Fragmentation of monitoring and early warning: There is no unified standard for the monitoring scale and accuracy of various risks by different institutions, resulting in "blind spots" in risk monitoring and insufficient monitoring of "soft risks" such as planting willingness, policy risks, and public health events.

[0010] 2. Point-like risk assessment: Focusing on single risks while ignoring compound risks, and paying insufficient attention to risk relevance and superposition effects, resulting in insufficient comprehensive assessment of risks.

[0011] 3. Fragmentation of risk control: The links of food risk identification, analysis, assessment, early warning, and control are characterized by multiple departments, multiple institutions, and multiple levels. The risk monitoring, assessment, and early warning decision-making are fragmented, and there is a lack of a unified comprehensive coordination mechanism for food security risk management.

[0012] 4. Exposure of policy risks: The risks caused by unstable policy regulation rhythms and insufficient implementation accuracy are intensifying. There are relatively few long-term stability policies, and there are many "firefighting" policies such as short-term, emergency, and local ones.

[0013] 5. Primary stage of technology application: The agricultural product traceability system based on blockchain technology can effectively encourage all entities in the agricultural product supply chain to take measures to ensure the quality and safety of agricultural products. However, the application of this technology is still in its primary stage and faces key challenges such as high traceability costs, difficult-to-predict product premiums, and imperfect benefit distribution mechanisms.

[0014] 6. Insufficient informatization: Most of the informatization systems for cultivated land planting, livestock and poultry farming, and aquaculture are fragmented and isolated, unable to be effectively connected and share data with various agricultural resource management systems. At the same time, there are still problems such as low informatization levels, lack of analysis functions, and inability to achieve graphic integration.

[0015] 7. Insufficient risk management tools: The risk transfer and diversification functions of market-based risk management tools such as financial insurance, futures, and options need to be strengthened to improve the modernization level of risk management.

[0016] 8. Insufficient timeliness and accuracy of data: Traditional monitoring methods rely on manual collection, resulting in lagging data updates and a lack of multi-source data fusion capabilities.

[0017] 9. Scattered information dissemination channels: The lack of a collaborative early warning platform among the government, enterprises, and farmers leads to low information transmission efficiency.

[0018] In view of the above problems, the present invention proposes an agricultural industry economic risk tracking and early warning method and system based on key indicators. By integrating multi-dimensional data analysis technology and intelligent early warning models, it realizes the accurate identification and hierarchical response of agricultural product market risks.

[0019] The above information disclosed in this background technology is only used to increase the understanding of the background technology of the present invention. Therefore, it may include prior art that is not known to those of ordinary skill in the art. Summary of the Invention

[0020] The purpose of the present invention is to provide an agricultural industry economic risk tracking and early warning method and system based on key indicators to solve the problems raised in the above background technology.

[0021] To achieve the above purpose, the present invention provides the following technical solution: An agricultural industry economic risk tracking and early warning method based on key indicators, including the following steps:

[0022] S1. Data collection and preprocessing:

[0023] S1-1. Deploy Internet of Things sensors, market monitoring terminals, and related collection systems to collect data from the production end, circulation end, and consumption end in real time;

[0024] S1-2. Use the CensusX12 seasonal adjustment method to eliminate the seasonal fluctuations of time series data;

[0025] S2. Construction of key warning indicators:

[0026] Taking the price volatility of agricultural products as the core warning indicator, and using the ratio of the volatility of agricultural products to the volatility of CPI as the warning limit to predict and warn the market fluctuations of agricultural products. The price volatility R t The calculation formula is:

[0027] R t =(lnP t -lnP t-1 )×100

[0028] Wherein, P t is the current price, and P t-1 is the previous price;

[0029] S3. Dynamic warning limit division:

[0030] Based on the distribution characteristics of historical data, the three - percentile grading method is adopted to divide the risk level into three levels. When the price volatility R t ≤0.5×CPI volatility, the risk level is green, and the corresponding response measure is normal monitoring; when 0.5×CPI volatility < R t ≤CPI volatility, the risk level is yellow, and the corresponding response measure is risk warning; when R t > CPI volatility, the risk level is red, and the corresponding response measure is emergency intervention;

[0031] S4. Hybrid early - warning model architecture for risk simulation:

[0032] Extract the long - term trend term through H - P filtering;

[0033] Use the ARMA model to smooth the data and make short - term predictions;

[0034] Use the ARCH family of models for volatility clustering detection;

[0035] Mine non - linear relationships through the Long Short - Term Memory neural network;

[0036] S5. Output early - warning information to the user side.

[0037] Preferably, in S1 - 1, the production - end data includes but is not limited to the planting area and inventory, the circulation - end data includes but is not limited to logistics data and inventory, and the consumption - end data includes but is not limited to retail prices and e - commerce sales.

[0038] Preferably, in S2, the police situation indicators also include auxiliary indicators: production fluctuation index and market supply - demand deviation degree.

[0039] Preferably, in S4, the risk simulation includes production risk simulation, consumption substitution simulation, price conduction simulation, and policy effect simulation.

[0040] Preferably, in S5, the user side includes the government side, the enterprise side, and the farmer side. The government side pushes structured early - warning reports through the government affairs network, the enterprise side returns risk indexes in real - time through the API interface, and the farmer side pushes short messages through text messages or the APP.

[0041] The present invention also provides an early - warning system for the agricultural industrial economic risk tracking and early - warning method based on key indicators as described above, including:

[0042] A data acquisition module, equipped with distributed data acquisition node hardware and systems, for real - time acquisition of production - end, circulation - end, and consumption - end data;

[0043] A data pre - processing module, equipped with an edge computing gateway, for data pre - processing;

[0044] The cloud computing center runs a core early warning model and is configured with an industrial economy monitoring and early warning module, a price monitoring and early warning module, a circulation scale and path monitoring and early warning module, a traceability information monitoring and early warning module, a knowledge service and information release module, and a risk simulation module;

[0045] The industrial economy monitoring and early warning module is used for production layout analysis and disaster impact assessment;

[0046] The price monitoring and early warning module is used for short-term price forecasting and volatility early warning;

[0047] The circulation scale and path monitoring and early warning module is used for tracking and monitoring the circulation scale and path;

[0048] The traceability information monitoring and early warning module is used for product traceability tracking and early warning;

[0049] The knowledge service and information release module automatically generates multi-version early warning reports through natural language generation technology, which is applicable to the government side, the enterprise side, and the farmer side;

[0050] The risk simulation module is used for visual deduction of the effects of policy interventions and simulation of supply chain disruptions.

[0051] Preferably, the data includes statistical databases publicly available by government departments, self-built industrial economy and circulation information databases, agricultural credit collection, and massive network data.

[0052] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0053] 1. Improvement in early warning accuracy: The present invention integrates multi-source data, consolidates multi-dimensional data such as production, consumption, price, and circulation path, avoids blind spots in monitoring that may have an adverse impact on the monitoring accuracy, and through a hybrid model architecture, combines classical statistical models with machine learning algorithms to effectively reduce the price prediction error rate and improve the early warning accuracy;

[0054] 2. Enhancement in response timeliness: Through an intelligent early warning model, the time from data collection to early warning release is shortened, the response timeliness is enhanced, and at the same time, a hierarchical response service is established to provide differentiated early warning services for the government, enterprises, and farmers;

[0055] 3. Visualization of decision-making support: Provide an interactive tool for visual deduction of the effects of policy interventions, which helps with promotion and application.

[0056] The above summary is only for the purpose of the specification and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the present invention will become readily apparent by reference to the drawings and the following detailed description. Description of the Drawings

[0057] Figure 1 This is the system architecture diagram of the present invention;

[0058] Figure 2 This is the flowchart of the early warning model of the present invention;

[0059] Figure 3 This is the schematic diagram of the warning limit division of the present invention. Detailed Embodiments

[0060] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0061] The present invention provides an agricultural industry economic risk tracking and early warning method based on key indicators, including the following steps:

[0062] S1. Data collection and preprocessing:

[0063] S1-1. Deploy Internet of Things sensors, market monitoring terminals and related collection systems to collect data from the production end, circulation end and consumption end in real time; the production end data includes, but is not limited to, planting area and inventory, the circulation end data includes, but is not limited to, logistics data and inventory, and the consumption end data includes, but is not limited to, retail price and e-commerce sales volume;

[0064] S1-2. Use the CensusX12 seasonal adjustment method to eliminate the seasonal fluctuations of time series data;

[0065] S2. Construction of key warning indicators:

[0066] Taking the agricultural product price volatility as the core warning indicator, and taking the ratio of the agricultural product volatility to the CPI volatility as the warning limit to predict and warn the agricultural product market fluctuations, the price volatility R t The calculation formula is:

[0067] R t =(lnP t -lnP t-1 )×100

[0068] Wherein, P t is the current price, and P t-1 is the previous price;

[0069] The warning indicators also include auxiliary indicators: production fluctuation index and market supply and demand deviation degree;

[0070] S3. Dynamic warning limit division:

[0071] Based on the historical data distribution characteristics, using the three - percentile grading method, the risk level is divided into three levels. When the price volatility R t ≤0.5×CPI volatility, the risk level is green, and the corresponding response measure is normal monitoring; when 0.5×CPI volatility < R t ≤CPI volatility, the risk level is yellow, and the corresponding response measure is risk warning; when R t > CPI volatility, the risk level is red, and the corresponding response measure is emergency intervention;

[0072] S4. Hybrid early - warning model architecture for risk simulation:

[0073] Extract the long - term trend term through H - P filtering;

[0074] Use the ARMA model to smooth the data and make short - term predictions;

[0075] Use the ARCH family of models for volatility clustering detection;

[0076] Mine non - linear relationships through the Long Short - Term Memory neural network;

[0077] Risk simulation includes production risk simulation, consumption substitution simulation, price conduction simulation, and policy effect simulation;

[0078] S5. Output warning information to the user side; the user side includes the government side, the enterprise side, and the farmer side. The government side pushes structured warning reports through the e - government network, such as risk level, scope of influence, and recommended measures. The enterprise side returns the risk index in real - time through the API interface, and the farmer side pushes short messages through text messages or the APP, such as planting suggestions.

[0079] The present invention also provides an agricultural industrial economic risk tracking and early - warning system based on key indicators, including a data acquisition module, a data pre - processing module, and a cloud computing center.

[0080] The data acquisition module is equipped with distributed data acquisition node hardware and systems for real - time acquisition of data from the production side, circulation side, and consumption side; the data includes statistical databases publicly disclosed by government departments, self - built industrial economic and circulation information databases, agricultural credit collections, and massive network data;

[0081] The data pre - processing module is equipped with an edge computing gateway for data pre - processing;

[0082] The core early warning model for the operation of the cloud computing center. The cloud computing center is configured with an industrial economy monitoring and early warning module, a price monitoring and early warning module, a circulation scale and path monitoring and early warning module, a traceability information monitoring and early warning module, a knowledge service and information release module, and a risk simulation module;

[0083] The industrial economy monitoring and early warning module is used for production layout analysis and disaster impact assessment;

[0084] The price monitoring and early warning module is used for short-term price forecasting and volatility early warning;

[0085] The circulation scale and path monitoring and early warning module is used for tracking and monitoring the circulation scale and path;

[0086] The traceability information monitoring and early warning module is used for product traceability tracking and early warning;

[0087] The knowledge service and information release module automatically generates multi-version early warning reports through natural language generation technology, which is applicable to the government side, enterprise side, and farmer side;

[0088] The risk simulation module is used for visual deduction of the effects of policy interventions and simulation of supply chain disruptions.

[0089] Example 1: Early warning of live pig prices

[0090] 1. Data input:

[0091] The inventory of breeding sows (production side);

[0092] The feed price index (cost side);

[0093] The operating rate of slaughtering enterprises (circulation side);

[0094] 2. Model operation:

[0095] Decompose the long-term trend term and cycle term through HP filtering:

[0096] Selection of smoothing factor λ:

[0097] For monthly data, use the Ravn-Uhlig criterion: λ = 14,400 (formula: λ = 100 * (frequency / 4)^4, monthly frequency = 12);

[0098] The breeding cycle fluctuations of 3 - 5 years can be separated (matching the change cycle of the inventory of breeding sows);

[0099] Output items:

[0100] Long-term trend term: Reflecting the structural changes in live pig production capacity (such as supply contraction caused by environmental protection policies);

[0101] Cycle term: Capturing the pig cycle fluctuations (one cycle every 2 - 3 years);

[0102] Using LSTM to predict price trends for the next 3 months:

[0103] Input features: Standardized inventory of breeding sows (year-on-year), feed price index (month-on-month), slaughter start-up rate, historical prices;

[0104] Time step: 12 months (covering a full annual cycle);

[0105] Training configuration:

[0106] Optimizer: Adam (lr = 0.001, clipnorm = 1.0);

[0107] Loss function: Huber Loss (robust to outliers);

[0108] Early stopping mechanism: Terminate if the validation set loss does not decrease for 10 consecutive epochs;

[0109] Batch Size: 32 (suitable for medium and small-scale time series data);

[0110] Calculate the volatility and divide the warning limit against the CPI benchmark value;

[0111] 3. Output results:

[0112] Trigger a red warning when the predicted volatility exceeds 150% of the CPI volatility;

[0113] Threshold setting: Live hog price volatility > 1.5 × CPI volatility (released by the National Bureau of Statistics);

[0114] Automatically generate regulatory suggestions (such as initiating the release of reserve meat).

[0115] Example 2: Early warning of rice production risks

[0116] 1. Data fusion:

[0117] Satellite remote sensing vegetation index (NDVI);

[0118] Meteorological bureau drought monitoring data;

[0119] Fertilizer purchase volume of agricultural input dealers;

[0120] 2. Risk simulation:

[0121] Predicting yield fluctuations based on the ARIMA model:

[0122] Input features:

[0123] Historical yield data (county level, last 30 years);

[0124] Moving average of NDVI two months in advance (Smoothed by SG filter);

[0125] Year-on-year change rate of chemical fertilizer purchase volume;

[0126] Residual diagnosis:

[0127] Ljung-Box Q test: p-value > 0.05 (No autocorrelation in residuals);

[0128] Shapiro-Wilk normality test: W > 0.9;

[0129] Monte Carlo simulation of the impact of extreme weather:

[0130] Risk factor parameters:

[0131]

[0132] Number of simulations: More than 5000 times are required for an error < 1%;

[0133] Daily simulation during the growing season (120-day rice growth period);

[0134] Calculate the confidence interval of regional yield:

[0135] Calculation method: Bootstrap aggregation of prediction results (α = 0.1);

[0136] Spatial weighting: Weight w_i = Proportion of rice planting area in the county × Soil fertility index (0.8 - 1.2); 3. Decision support:

[0137] When the predicted probability of yield reduction > 30%, initiate a yellow warning;

[0138] Push suggestions for planting structure adjustment to the county-level agricultural department.

[0139] The present invention effectively solves the defects of traditional methods in terms of data timeliness and service accuracy by constructing an intelligent monitoring and warning system, and provides an innovative solution for agricultural industry risk management. The present invention can be applied to multiple fields such as industrial development, agricultural planning, compilation of regional reports, agricultural investment and financing consulting, price warning, navigation of integrated industrial data resources, demonstration of agricultural Internet of Things applications, display of agricultural science and technology, and visualization services. For example, it is applied to government decision-making consulting and formulation of policy documents, regularly releasing monitoring and warning information on the supply and demand and prices of agricultural products, reasonably guiding market expectations, and providing reference for agricultural monitoring and warning research.

[0140] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0141] Any process or method description represented in a flowchart or described in other ways herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a specific logical function or process, and the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a manner other than shown or discussed, including in a substantially simultaneous manner according to the functions involved or in a reverse order, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0142] The logic and / or steps represented in a flowchart or described in other ways herein, for example, can be considered as a sequenced list of executable instructions for implementing a logical function, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in connection with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection portion having one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing as necessary, and then stored in a computer memory.

[0143] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0144] Those of ordinary skill in the art can understand that all or part of the steps carried by the methods of the above embodiments can be completed by hardware related to program instructions. The said program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0145] In addition, in each embodiment of the present invention, each functional unit can be integrated into a processing module, or each unit can exist physically alone, or two or more units can be integrated into one module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. When the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0146] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. The agricultural industry economic risk tracking and early warning method based on key indicators is characterized by: The steps include: S1. Data collection and preprocessing: S1-1. Deploy IoT sensors, market monitoring terminals and related collection systems to collect data from production, circulation and consumption ends in real time; S1-2. Use CensusX12 seasonal adjustment method to eliminate seasonal fluctuations in time series data; S2. Construction of key warning indicators: The volatility of agricultural product prices is used as the core warning indicator, and the ratio of agricultural product volatility to CPI volatility is used as the warning limit to predict and warn of agricultural product market fluctuations. t The calculation formula is: R t =(lnP t -lnP t-1 )×100 Among them, P t is the current price, P t-1 is the price of the previous period; S3. Dynamic alarm limit division: Based on the distribution characteristics of historical data, the risk level is divided into three levels using the tertiary classification method. t When the volatility is less than or equal to 0.5×CPI, the risk level is green and the corresponding response measure is normal monitoring. <R t ≤CPI volatility, the risk level is yellow, and the corresponding response measure is risk warning; when R t >CPI volatility, the risk level is red, and the corresponding response measure is emergency intervention; S4. Hybrid early warning model architecture for risk simulation: Extract long-term trend items through HP filtering; Use the ARMA model to smooth the data and make short-term forecasts; Using ARCH family models to detect volatility clustering; Nonlinear relationship mining through Long Short-Term Memory neural network; S5. Output warning information to the user end.

2. The agricultural industry economic risk tracking and early warning method and system based on key indicators according to claim 1 is characterized by: In S1-1, the production-side data includes but is not limited to the planting area and the number of livestock on hand; the circulation-side data includes but is not limited to logistics data and inventory; the consumption-side data includes but is not limited to retail prices and e-commerce sales.

3. The agricultural industry economic risk tracking and early warning method and system based on key indicators according to claim 1 is characterized by: In S2, the warning indicators also include auxiliary indicators: production fluctuation index and market supply and demand deviation.

4. The agricultural industry economic risk tracking and early warning method and system based on key indicators according to claim 1 is characterized by: In S4, risk simulation includes production risk simulation, consumption substitution simulation, price transmission simulation, and policy effect simulation.

5. The agricultural industry economic risk tracking and early warning method and system based on key indicators according to claim 1 is characterized by: In S5, the user end includes the government end, the enterprise end and the farmer end. The government end pushes a structured early warning report through the government network, the enterprise end returns the risk index in real time through the API interface, and the farmer end pushes text messages through SMS or APP.

6. The early warning system of the agricultural industry economic risk tracking and early warning method based on key indicators as claimed in any one of claims 1 to 5, characterized in that: include: The data collection module is equipped with distributed data collection node hardware and systems to collect data from the production, circulation and consumption ends in real time; The data preprocessing module is equipped with an edge computing gateway for data preprocessing; The cloud computing center runs the core early warning model. The cloud computing center is equipped with industrial economic monitoring and early warning modules, price monitoring and early warning modules, circulation scale and path monitoring and early warning modules, traceability information monitoring and early warning modules, knowledge service and information release modules, and risk simulation modules. The industrial economic monitoring and early warning module is used for production layout analysis and disaster impact assessment; The price monitoring and early warning module is used for short-term price forecasting and volatility early warning; The circulation scale and path monitoring and early warning module is used to track and monitor the circulation scale and path; The traceability information monitoring and early warning module is used for product traceability tracking and early warning; The knowledge service and information release module automatically generates multiple versions of early warning reports through natural language generation technology, which are suitable for government, enterprise and farmer ends; The risk simulation module is used to visualize the effects of policy interventions and simulate supply chain disruptions.

7. The agricultural industry economic risk tracking and early warning system based on key indicators according to claim 6 is characterized by: The data include statistical databases disclosed by government departments, self-built industrial economic and circulation information databases, rural credit procurement and massive network data.

Images