A big data analysis platform based on smart agriculture

Abstract

The present invention discloses a big data analysis platform based on smart agriculture, which relates to the field of data analysis technology, including: establishing a regional multi-source heterogeneous database; analyzing the matching index of seasonal cultivable crops on regional land based on the comprehensive factors affecting the cultivable crops on regional land in the regional multi-source heterogeneous database; screening positively correlated cultivable crops based on the dynamic suitability correlation coefficient of the cultivable crops on regional land, and establishing a cultivable crop data matrix A for regional land; estimating the yield of cultivable crops based on the cultivable crop data matrix of regional land and multi-source heterogeneous data, analyzing and estimating the output quality of cultivable crops, and determining the output income of cultivable crops on regional land; and generating the optimal cultivable crop plan for regional land according to the maximum output income of cultivable crops on regional land. The advantages of the present invention are: improving the total grain output and increasing farmers' income.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a big data analysis platform based on smart agriculture. Background Art

[0002] Big data analysis in smart agriculture refers to the use of modern information technology and big data technology to collect, store, process, analyze and utilize large amounts of data generated in the agricultural production process to optimize agricultural production, improve agricultural production efficiency, ensure the quality of agricultural products and promote sustainable development.

[0003] Since smart agriculture involves a large data span, existing systems mainly rely on network agricultural big data to conduct regional similarity training models to assist in agricultural task decision-making, but they cannot be applied to different regions, and the actual agricultural assistance implementation effect is poor. Summary of the Invention

[0004] In order to solve the above technical problems, a big data analysis platform based on smart agriculture is provided. This technical solution solves the above-mentioned problem that due to the large data span involved in smart agriculture, the existing systems mainly rely on network agricultural big data to conduct regional similarity training models to achieve auxiliary agricultural task decision-making, but cannot be applied to different regions, and the actual agricultural assistance implementation effect is poor.

[0005] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0006] A big data analysis platform based on smart agriculture, including:

[0007] Based on the relevant data of various departments in the region, a regional multi-source heterogeneous database is established; the relevant data of various departments in the region include: natural resources data, agricultural and rural data, and water conservancy data;

[0008] Based on the comprehensive factors affecting the crops planted on regional land in the regional multi-source heterogeneous database, the matching indicators of seasonal crops that can be planted on regional land are analyzed;

[0009] Based on the dynamic suitability correlation coefficient of the crops that can be planted on the regional land, positively correlated crops are screened and a data matrix A of the crops that can be planted on the regional land is established;

[0010]

[0011] Among them, G ik is the matching index of the kth arable crop in the i-th season

[0012] Based on the regional land's cultivable crop data matrix and multi-source heterogeneous data, the yield of cultivable crops is estimated, the output quality of cultivable crops is analyzed and estimated, and the yield income of cultivable crops on regional land is determined; external impact risks and internal response capabilities

[0013] According to the maximum yield of crops that can be planted on regional land, the optimal planting crop plan for regional land is generated.

[0014] Preferably, based on the comprehensive factors affecting the crops planted on regional land in the regional multi-source heterogeneous database, the matching indexes of seasonally plantable crops on regional land are analyzed, specifically including:

[0015] Acquire multi-source heterogeneous parameters in each season in a regional multi-source heterogeneous database to construct seasonal multi-source heterogeneous time series data; the multi-source heterogeneous time series data includes: soil fertility parameters, environmental climate parameters, and water resource parameters;

[0016] Based on the crops that can be grown in each season, determine the growth demand data of the crops that can be grown in the season;

[0017] Normalize seasonal multi-source heterogeneous time series data and seasonal crop demand data;

[0018] Based on seasonal multi-source heterogeneous time series data, the unit time is used as the time window, and the multi-source heterogeneous data of each season is used as the index attribute of the time window. By sliding the time window, the multi-source heterogeneous time series feature vectors of each season are obtained;

[0019] Using One-Hot coding, the seasonal crop demand data is discretized into the seasonal crop growth demand vector.

[0020] Data is aligned according to seasonal attributes, and the matching degree between the multi-source heterogeneous time series feature vectors of each season and the seasonal plantable crop growth demand vectors is calculated to obtain the matching index of seasonal plantable crops in the regional land.

[0021] The specific matching indicators of seasonal crops that can be planted on the land in the area are:

[0022]

[0023] Where G ik is the matching index of the kth arable crop in the i-th season, x ij is the jth multi-source heterogeneous time series feature vector in the i-th season, y iK is the kth arable crop growth demand vector in the i-th season, is the mean value of the j-th multi-source heterogeneous time series feature vector, is the mean value of the growth demand vector of the kth cultivable crop, is the standard deviation of the j-th multi-source heterogeneous time series feature vector, is the standard deviation of the kth cultivable crop growth demand vector, and n is the total number of seasons.

[0024] Preferably, the yield of cultivable crops is estimated based on the cultivable crop data matrix of regional land and multi-source heterogeneous data, the output quality of cultivable crops is analyzed and estimated, and the yield income of cultivable crops on regional land is determined, specifically including:

[0025] Based on SVR support vector regression, a prediction model for crop yield of regional land is constructed;

[0026] Based on the regional land's cultivable crop data matrix and multi-source heterogeneous data, the multi-source heterogeneous data are labeled and matched according to the demand for seasonal cultivable crops. This data is used as the influencing feature data for seasonal cultivable crops, and the yield of cultivable crops is used as label data to package the data into the original sample dataset of seasonal cultivable crops.

[0027] Substitute the original sample data set of seasonal cultivable crops into the cultivable crop yield prediction model of regional land, take the influencing characteristic data of seasonal cultivable crops as input, take the predicted cultivable crop yield label data as output, and take minimizing the error function of the model as the end goal;

[0028] Determine several hidden dangers in the growth process of crops that can be planted in the season, which are recorded as external growth hidden dangers; the external growth hidden dangers include: drought, floods, diseases and insect pests;

[0029] Based on the potential growth hazards of crops that can be planted in a given season, several response strategies for these potential growth hazards are marked in the region and recorded as internal response capabilities. The internal response capabilities include irrigation control and pest and disease control.

[0030] Assess the growth risk coefficient of crops that can be planted in a season based on external growth risks and internal response capabilities;

[0031] Using the growth risk coefficient of seasonal cultivable crops, the yield of the predicted cultivable crops is corrected to obtain the output quality index of the cultivable crops;

[0032] The crop yield prediction model for the land in the region specifically includes:

[0033]

[0034] Where P is the predicted yield of crops that can be planted, d z is the zth influencing characteristic data of seasonal crops, α z is the Lagrange multiplier of the zth influencing characteristic data, d is the influencing characteristic data of the input seasonal crops, V(dz ,d) is the kernel function of the model, b is the bias term;

[0035] Among them, the output quality indicators of cultivable crops are as follows:

[0036] S=P×R

[0037] Where S is the output quality index of cultivable crops, and R is the growth risk coefficient of cultivable crops in that season.

[0038] Preferably, based on external growth risks and internal response capabilities, the growth risk coefficients of crops that can be planted in a season are evaluated to include:

[0039] Based on various growth hazards within the growth cycle of seasonal cultivable crops, the output quality reduction value of cultivable crops caused by external growth hazards is calculated;

[0040] Calculate the added value of internal response capacity on the output quality of cultivable crops when various growth hazards occur during the growth cycle of seasonal cultivable crops;

[0041] Based on the decrease in the output quality of cultivable crops due to external growth hazards and the increase in the output quality of cultivable crops due to internal response capabilities, the growth risk coefficient of seasonal cultivable crops is calculated;

[0042] The calculation of the output quality reduction value of the cultivable crops due to the external growth hidden danger is specifically as follows:

[0043]

[0044] Where ΔA i' is the output quality reduction value of the arable crops caused by the i'th external growth hidden danger, U0 is the initial value of the output quality of the arable crops, U(t) is the actual value of the output quality of the arable crops, is the sensitivity coefficient of cultivable crops to external growth hazards, E i' (t) is the impact intensity of the i'th external growth hazard on the t growth cycles of the plantable crop, n' is the total number of external growth hazards, T is the length of the growth cycle of the plantable crop, and dt is the differential sign;

[0045] The calculation of the added value of the internal coping capacity for the output quality of cultivable crops is specifically as follows:

[0046]

[0047] Where, ΔB j' The j'th is the added value of the internal response capacity for the output quality of cultivable crops, β is the weight coefficient of cultivable crops to external growth hazards, C j'(t) is the improvement intensity of the j'th internal coping capacity for t growth cycles of the plantable crops, m' is the total number of internal coping capacities, T is the length of the growth cycle of the plantable crops, and dt is the differential sign;

[0048] The growth risk coefficient of crops that can be planted in the calculation season is specifically:

[0049]

[0050] Where R is the growth risk coefficient of crops that can be planted in the season.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention proposes a big data analysis platform based on smart agriculture. By integrating natural resources, agriculture, rural areas and water conservancy data in regional multi-source heterogeneous databases, it analyzes the comprehensive factors affecting the land planting of crops in the region, dynamically evaluates the correlation coefficient of the land's suitability for crop planting, and uses external risks and internal response capabilities to predict the crop output quality and income for the crops planted. Finally, it generates the optimal crop planting plan for the region. The beneficial effects are: increasing total grain output and increasing farmers' income. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of a big data analysis platform method based on smart agriculture;

[0054] Figure 2 Flowchart of the matching indicator method for analyzing seasonal crops that can be grown on regional land;

[0055] Figure 3 Flowchart of a method for assessing the yield of crops that can be grown on regional land;

[0056] Figure 4 Flowchart of the method for assessing the growth risk coefficient of seasonal cultivable crops. DETAILED DESCRIPTION

[0057] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0058] Reference Figure 1 As shown, a big data analysis platform based on smart agriculture includes:

[0059] Based on the relevant data of various departments in the region, a regional multi-source heterogeneous database is established; the relevant data of various departments in the region include: natural resources data, agricultural and rural data, and water conservancy data;

[0060] Based on the comprehensive factors affecting the crops planted on regional land in the regional multi-source heterogeneous database, the matching indicators of seasonal crops that can be planted on regional land are analyzed;

[0061] Based on the dynamic suitability correlation coefficient of the crops that can be planted on the regional land, positively correlated crops are screened and a data matrix A of the crops that can be planted on the regional land is established;

[0062]

[0063] Among them, G ik is the matching index of the kth arable crop in the i-th season

[0064] Based on the regional land's cultivable crop data matrix and multi-source heterogeneous data, the yield of cultivable crops is estimated, the output quality of cultivable crops is analyzed and estimated, and the yield income of cultivable crops on regional land is determined;

[0065] According to the maximum yield of crops that can be planted on regional land, the optimal planting crop plan for regional land is generated.

[0066] This plan integrates natural resources, agriculture, rural areas, and water conservancy data from regional multi-source heterogeneous databases, analyzes the comprehensive factors affecting crop planting on land in the region, dynamically evaluates the correlation coefficient of land suitability for crop planting, and uses external risks and internal response capabilities to predict crop output quality and returns for crops. Ultimately, it generates the optimal crop planting plan for the region, with the beneficial effects of increasing total grain output and farmers' income.

[0067] Reference Figure 2 As shown in the figure, based on the comprehensive factors affecting the crops planted on regional land in the regional multi-source heterogeneous database, the matching indicators for seasonal crops planted on regional land are analyzed, including:

[0068] Acquire multi-source heterogeneous parameters in each season in a regional multi-source heterogeneous database to construct seasonal multi-source heterogeneous time series data; the multi-source heterogeneous time series data includes: soil fertility parameters, environmental climate parameters, and water resource parameters;

[0069] Based on the crops that can be grown in each season, determine the growth demand data of the crops that can be grown in the season;

[0070] Normalize seasonal multi-source heterogeneous time series data and seasonal crop demand data;

[0071] Based on seasonal multi-source heterogeneous time series data, the unit time is used as the time window, and the multi-source heterogeneous data of each season is used as the index attribute of the time window. By sliding the time window, the multi-source heterogeneous time series feature vectors of each season are obtained;

[0072] Using One-Hot coding, the seasonal crop demand data is discretized into the seasonal crop growth demand vector.

[0073] Data is aligned according to seasonal attributes, and the matching degree between the multi-source heterogeneous time series feature vectors of each season and the seasonal plantable crop growth demand vectors is calculated to obtain the matching index of seasonal plantable crops in the regional land.

[0074] The specific matching indicators of seasonal crops that can be planted on the land in the area are:

[0075]

[0076] Where G ik is the matching index of the kth arable crop in the i-th season, x ij is the jth multi-source heterogeneous time series feature vector in the i-th season, y iK is the kth arable crop growth demand vector in the i-th season, is the mean value of the j-th multi-source heterogeneous time series feature vector, is the mean value of the growth demand vector of the kth cultivable crop, is the standard deviation of the j-th multi-source heterogeneous time series feature vector, is the standard deviation of the kth cultivable crop growth demand vector, and n is the total number of seasons.

[0077] It should be noted that the matching index of seasonal crops in the above-mentioned regional land is in the range of [-1 to 1], where -1 means the vector directions are opposite, 0 means there is no matching in the vector directions, and 1 means the vector directions are the same. Therefore, by calculating the matching degree between the multi-source heterogeneous time series feature vectors of each season and the seasonal crop growth demand vectors, the seasonal crops in the current regional land can be quickly screened out.

[0078] Reference Figure 3 As shown in the figure, based on the regional land cultivable crop data matrix and multi-source heterogeneous data, the yield of cultivable crops is estimated, the output quality of cultivable crops is analyzed and estimated, and the output benefits of cultivable crops on regional land are determined. Specifically, the following are included:

[0079] Based on SVR support vector regression, a prediction model for crop yield of regional land is constructed;

[0080] Based on the regional land's cultivable crop data matrix and multi-source heterogeneous data, the multi-source heterogeneous data are labeled and matched according to the demand for seasonal cultivable crops. This data is used as the influencing feature data for seasonal cultivable crops, and the yield of cultivable crops is used as label data to package the data into the original sample dataset of seasonal cultivable crops.

[0081] Substitute the original sample data set of seasonal cultivable crops into the cultivable crop yield prediction model of regional land, take the influencing characteristic data of seasonal cultivable crops as input, take the predicted cultivable crop yield label data as output, and take minimizing the error function of the model as the end goal;

[0082] Determine several hidden dangers in the growth process of crops that can be planted in the season, which are recorded as external growth hidden dangers; the external growth hidden dangers include: drought, floods, diseases and insect pests;

[0083] Based on the potential growth hazards of crops that can be planted in a given season, several response strategies for these potential growth hazards are marked in the region and recorded as internal response capabilities. The internal response capabilities include irrigation control and pest and disease control.

[0084] Assess the growth risk coefficient of crops that can be planted in a season based on external growth risks and internal response capabilities;

[0085] Using the growth risk coefficient of seasonal cultivable crops, the yield of the predicted cultivable crops is corrected to obtain the output quality index of the cultivable crops;

[0086] The crop yield prediction model for the land in the region specifically includes:

[0087]

[0088] Where P is the predicted yield of crops that can be planted, d z is the zth influencing characteristic data of seasonal crops, α z is the Lagrange multiplier of the zth influencing characteristic data, d is the influencing characteristic data of the input seasonal crops, V(d z ,d) is the kernel function of the model, b is the bias term;

[0089] Among them, the output quality indicators of cultivable crops are as follows:

[0090] S=P×R

[0091] Where S is the output quality index of cultivable crops, and R is the growth risk coefficient of cultivable crops in that season.

[0092] This solution, based on support vector regression (SVR) technology, combines a regional crop data matrix with multi-source heterogeneous data to construct a yield prediction model. By analyzing seasonal crop influencing characteristics, growth hazards, and response strategies, it assesses growth risk factors, then refines the predicted yield to derive crop output quality indicators. This approach offers the potential benefit of accurately estimating crop yield and quality by comprehensively considering multiple factors, providing a basis for regional land planting planning and benefit assessment, and improving the efficiency and profitability of agricultural production.

[0093] Reference Figure 4 As shown in the figure, based on external growth risks and internal response capabilities, the growth risk factors of crops that can be planted in the season are specifically assessed as follows:

[0094] Based on various growth hazards within the growth cycle of seasonal cultivable crops, the output quality reduction value of cultivable crops caused by external growth hazards is calculated;

[0095] Calculate the added value of internal response capacity on the output quality of cultivable crops when various growth hazards occur during the growth cycle of seasonal cultivable crops;

[0096] Based on the decrease in the output quality of cultivable crops due to external growth hazards and the increase in the output quality of cultivable crops due to internal response capabilities, the growth risk coefficient of seasonal cultivable crops is calculated;

[0097] The calculation of the output quality reduction value of the cultivable crops due to the external growth hidden danger is specifically as follows:

[0098]

[0099] Where ΔA i' is the output quality reduction value of the arable crops caused by the i'th external growth hidden danger, U0 is the initial value of the output quality of the arable crops, U(t) is the actual value of the output quality of the arable crops, is the sensitivity coefficient of cultivable crops to external growth hazards, E i' (t) is the impact intensity of the i'th external growth hazard on the t growth cycles of the plantable crop, n' is the total number of external growth hazards, T is the length of the growth cycle of the plantable crop, and dt is the differential sign;

[0100] The calculation of the added value of the internal coping capacity for the output quality of cultivable crops is specifically as follows:

[0101]

[0102] Where, ΔB j' The j'th is the added value of the internal response capacity for the output quality of cultivable crops, β is the weight coefficient of cultivable crops to external growth hazards, C j' (t) is the improvement intensity of the j'th internal coping capacity for t growth cycles of the plantable crops, m' is the total number of internal coping capacities, T is the length of the growth cycle of the plantable crops, and dt is the differential sign;

[0103] The growth risk coefficient of crops that can be planted in the calculation season is specifically:

[0104]

[0105] Where R is the growth risk coefficient of crops that can be planted in the season.

[0106] It can be understood that by quantitatively analyzing the output quality decline caused by various external growth hazards during the growth cycle of seasonal crops and the output quality increase brought about by internal coping capabilities, the growth risk coefficient of seasonal crops can be comprehensively calculated to evaluate the potential risks in the crop growth process and its impact on output quality, thereby promoting the rational use of agricultural resources and optimizing planting decisions.

[0107] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A big data analysis platform based on smart agriculture, characterized by: include: Based on the relevant data of various departments in the region, a regional multi-source heterogeneous database is established; The relevant data of various departments in the region include: natural resources data, agricultural and rural data, and water conservancy data; Based on the comprehensive factors affecting the crops planted on regional land in the regional multi-source heterogeneous database, the matching indicators of seasonal crops that can be planted on regional land are analyzed; Based on the dynamic suitability correlation coefficient of the crops that can be planted on the regional land, positively correlated crops are screened and a data matrix A of the crops that can be planted on the regional land is established; Among them, G ik is the matching index of the kth arable crop in the i-th season Based on the regional land's cultivable crop data matrix and multi-source heterogeneous data, the yield of cultivable crops is estimated, the output quality of cultivable crops is analyzed and estimated, and the yield income of cultivable crops on regional land is determined; Generate the optimal planting plan for regional land according to the maximum yield of crops that can be planted on regional land; Among them, the yield of cultivable crops is estimated based on the cultivable crop data matrix of regional land and multi-source heterogeneous data, the output quality of cultivable crops is analyzed and estimated, and the output income of cultivable crops on regional land is determined. Specifically, it includes: Based on SVR support vector regression, a prediction model for crop yield of regional land is constructed; Based on the regional land's cultivable crop data matrix and multi-source heterogeneous data, the multi-source heterogeneous data are labeled and matched according to the demand for seasonal cultivable crops. This data is used as the influencing feature data for seasonal cultivable crops, and the yield of cultivable crops is used as label data to package the data into the original sample dataset of seasonal cultivable crops. Substitute the original sample data set of seasonal cultivable crops into the cultivable crop yield prediction model of regional land, take the influencing characteristic data of seasonal cultivable crops as input, take the predicted cultivable crop yield label data as output, and take minimizing the error function of the model as the end goal; Determine several hidden dangers in the growth process of crops that can be planted in the season, which are recorded as external growth hidden dangers; the external growth hidden dangers include: drought, floods, diseases and insect pests; Based on the potential growth risks of crops that can be planted in a season, several response strategies for these potential growth risks are marked in the region and recorded as internal response capabilities. The internal response capabilities include irrigation control, pest and disease control, and Assess the growth risk coefficient of crops that can be planted in a season based on external growth risks and internal response capabilities; Using the growth risk coefficient of seasonal cultivable crops, the yield of the predicted cultivable crops is corrected to obtain the output quality index of the cultivable crops; The crop yield prediction model for the land in the region specifically includes: Where P is the predicted yield of crops that can be planted, d z is the zth influencing characteristic data of seasonal crops, α z is the Lagrange multiplier of the zth influencing characteristic data, d is the influencing characteristic data of the input seasonal crops, V(d z ,d) is the kernel function of the model, b is the bias term; Among them, the output quality indicators of cultivable crops are as follows: S=P×R Where S is the output quality index of cultivable crops, and R is the growth risk coefficient of cultivable crops in the season; Among them, based on external growth risks and internal response capabilities, the growth risk factors of crops that can be planted in the season are specifically assessed as follows: Based on various growth hazards within the growth cycle of seasonal cultivable crops, the output quality reduction value of cultivable crops caused by external growth hazards is calculated; Calculate the added value of internal response capacity on the output quality of cultivable crops when various growth hazards occur during the growth cycle of seasonal cultivable crops; The growth risk coefficient of seasonal cultivable crops is calculated based on the decrease in output quality of cultivable crops due to external growth hazards and the increase in output quality of cultivable crops due to internal response capabilities. The calculation of the output quality reduction value of the cultivable crops due to the external growth hidden danger is specifically as follows: Where ΔA i' For the i ' = The output quality reduction value of the cultivable crops due to the external growth hidden danger, U0 is the initial value of the output quality of the cultivable crops, U(t) is the actual value of the output quality of the cultivable crops, φ is the sensitivity coefficient of the cultivable crops to the external growth hidden danger, E i' (t) is the i-th ' The impact intensity of external growth hazards on t growth cycles of arable crops, n ' is the total number of external growth hazards, T is the length of the growing period of crops that can be planted, and dt is the differential sign; The calculation of the added value of the internal coping capacity for the output quality of cultivable crops is specifically as follows: Where ΔB j' For the jth ' is the added value of internal coping capacity for the output quality of cultivable crops, β is the weight coefficient of cultivable crops to external growth hazards, C j' (t) is the jth ' is the improvement intensity of internal coping capacity for t growth cycles of arable crops, m ' is the total amount of internal coping capacity, T is the length of the growing period of crops that can be grown, and dt is the differential sign; The growth risk coefficient of crops that can be planted in the calculation season is specifically: Where R is the growth risk coefficient of crops that can be planted in the season.

2. A big data analysis platform based on smart agriculture according to claim 1, characterized in that: Based on the comprehensive factors affecting the crops planted on regional land in the regional multi-source heterogeneous database, the matching indicators of seasonal crops that can be planted on regional land are analyzed, including: Acquire multi-source heterogeneous parameters in each season in a regional multi-source heterogeneous database to construct seasonal multi-source heterogeneous time series data; the multi-source heterogeneous time series data includes: soil fertility parameters, environmental climate parameters, and water resource parameters; Based on the crops that can be grown in each season, determine the growth demand data of the crops that can be grown in the season; Normalize seasonal multi-source heterogeneous time series data and seasonal crop demand data; Based on seasonal multi-source heterogeneous time series data, the unit time is used as the time window, and the multi-source heterogeneous data of each season is used as the index attribute of the time window. By sliding the time window, the multi-source heterogeneous time series feature vectors of each season are obtained; Using One-Hot coding, the seasonal crop demand data is discretized into the seasonal crop growth demand vector. Data alignment is performed according to seasonal attributes, and the matching degree between the multi-source heterogeneous time series feature vectors of each season and the seasonal cultivable crop growth demand vectors is calculated to obtain the matching index of seasonal cultivable crops on regional land.

3. A big data analysis platform based on smart agriculture according to claim 2, characterized in that: The matching indicators of seasonal crops that can be planted on the land in the area are as follows: Where G ik is the matching index of the kth arable crop in the i-th season, x ij is the jth multi-source heterogeneous time series feature vector in the i-th season, y iK is the kth arable crop growth demand vector in the i-th season, is the mean value of the j-th multi-source heterogeneous time series feature vector, is the mean value of the growth demand vector of the kth cultivable crop, is the standard deviation of the j-th multi-source heterogeneous time series feature vector, is the standard deviation of the kth cultivable crop growth demand vector, and n is the total number of seasons.

Images