A method and system for evaluating citrus suitable planting areas

Abstract

The invention discloses a method and system for evaluating citrus suitable planting areas, which relates to the field of citrus suitable planting area evaluation, including: obtaining the suitable planting characteristics of land quality and environmental data suitable for citrus planting based on big data; establishing the weight coefficient of the influence of different citrus suitable planting characteristics on citrus planting based on the Pearson correlation coefficient formula; judging the influence of seasonal effects on citrus growth according to an algorithm model; establishing a citrus suitable planting area evaluation model based on an LSTM neural network algorithm model, and obtaining the evaluation level of citrus suitable planting areas according to a citrus suitable planting area evaluation matrix; establishing a citrus suitable planting area evaluation platform for inputting, storing and displaying citrus suitable planting characteristic data and evaluation results. The invention has the advantages of effectively reducing seasonal influences and improving the accuracy and reliability of citrus suitable planting area evaluation.

Description

Technical Field

[0001] The invention relates to the field of evaluation of citrus suitable planting areas, and in particular to a method and system for evaluating citrus suitable planting areas. Background Art

[0002] Citrus is a perennial evergreen fruit tree with a long life cycle. The tree lifespan can reach decades or even hundreds of years. Once a planting area is selected, large-scale citrus planting is carried out in the area, but it is found that the area has a great impact on the quality of the planted citrus. Further transplanting or manual intervention is likely to lead to waste of manpower and material resources and low quality of citrus products. Therefore, in the process of citrus planting, it is necessary to give priority to the suitability of citrus planting in the area, analyze it from multiple aspects, and comprehensively evaluate the suitable citrus planting area; secondly, the growth and development of citrus and product quality are not only affected by soil quality and environmental factors, but also by seasonal factors in the planting area. Therefore, when evaluating the suitable citrus planting area, it is necessary to consider the impact of seasonal factors in the area. A reasonable algorithm model can effectively improve the accuracy and reliability of the evaluation of the suitable citrus planting area and reduce the impact of seasonal factors.

[0003] Traditional evaluation of suitable citrus planting areas mainly relies on citrus planting experts to conduct on-site inspections of the areas to be tested. They use their rich practical experience and professional knowledge, combined with soil conditions and climate data of the areas to be tested, to conduct subjective assessments of the evaluation areas based on the growth needs of citrus and the characteristics of natural conditions. This lack of scientific and systematic analysis can easily lead to differences and uncertainties in the evaluation results, thus affecting the accuracy and reliability of the evaluation of suitable citrus planting areas and easily causing a waste of manpower and material resources. Summary of the invention

[0004] In order to solve the above-mentioned technical problems, a method and system for evaluating suitable citrus planting areas are provided. The technical solution solves the problem that the analysis mentioned in the above-mentioned background technology lacks scientificity and systematicity, which easily leads to differences and uncertainties in the evaluation results, thereby affecting the accuracy and reliability of the evaluation of suitable citrus planting areas, and easily leads to waste of manpower and material resources.

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

[0006] A method for evaluating citrus suitable planting areas, comprising:

[0007] Obtain the suitable planting characteristics of land quality and environmental data suitable for citrus planting based on big data;

[0008] Based on the Pearson correlation coefficient formula, the weight coefficients of the effects of different citrus planting characteristics on citrus planting were established;

[0009] according to Algorithmic model to determine the impact of seasonal effects on citrus growth;

[0010] Based on the LSTM neural network algorithm model, a citrus suitable planting area evaluation model was established, and the evaluation level of citrus suitable planting areas was obtained according to the citrus suitable planting area evaluation matrix;

[0011] Establish a citrus suitable planting area evaluation platform for inputting, storing and displaying citrus suitable planting characteristic data and evaluation results.

[0012] Preferably, the suitable planting characteristics of the land quality and environmental data suitable for citrus planting obtained based on big data specifically include:

[0013] Based on big data and historical data analysis, obtain the suitable planting characteristics of land quality and environmental data suitable for citrus planting;

[0014] According to the suitable planting characteristics of citrus, various data intervals of the suitable planting characteristics of citrus are established, and a sample list of the suitable planting characteristics of citrus is established;

[0015] Among them, the characteristics suitable for citrus cultivation include soil pH, soil thickness, organic matter content, soil permeability and drainage performance, as well as temperature, humidity, wind speed, light intensity, rainfall and growth cycle;

[0016] Based on historical data and big data analysis, an evaluation grade of citrus suitable planting areas is established for different data intervals of citrus suitable planting characteristics in the sample list of citrus suitable planting characteristics;

[0017] Based on the evaluation levels of areas suitable for growing citrus, a classification database of evaluation levels of areas suitable for growing citrus will be established.

[0018] Preferably, the weight coefficients of the effects of different citrus planting characteristics on citrus planting are established based on the Pearson correlation coefficient formula and specifically include:

[0019] Through simulation experiments and historical data, the data on the impact of different citrus planting characteristics on citrus planting are obtained by using the control variable method;

[0020] According to the data on the impact of different citrus planting characteristics on citrus planting, the data is cleaned and standardized;

[0021] According to the Pearson correlation coefficient formula, the weight coefficients of the effects of different citrus planting characteristics on citrus planting were calculated;

[0022] According to the deviation range of the evaluation grade intervals of the citrus suitable planting areas, the deviation coefficients of each grade interval and the most suitable interval are calculated;

[0023] According to the weight coefficients of the influence of different citrus suitable planting characteristics on citrus planting and the deviation coefficients between each grade interval and the most suitable interval, the input matrix of general citrus suitable planting characteristics is established;

[0024] The weight coefficient expression for calculating the influence of different citrus suitable planting characteristics on citrus planting is:

[0025]

[0026] In the formula, is the correlation coefficient between different citrus planting characteristics and the suitability of citrus planting, that is, the weight coefficient of the influence of different citrus planting characteristics on citrus planting. For the Data on citrus suitable planting characteristics, For the The average value of the citrus planting characteristics data, For the The suitability of citrus planting corresponding to the citrus planting characteristic data is For the The average value of the suitability of citrus planting corresponding to the citrus planting characteristic data, For the The number of groups of citrus suitable planting characteristic data;

[0027] The expression for calculating the deviation coefficient between each grade interval and the most suitable interval is:

[0028]

[0029] In the formula, For the Citrus Planting Characteristics The deviation coefficient between the grade interval and the most suitable interval, is the emphasis weight of different level intervals, The most suitable interval integer value, For the The first integer value, is the number of integer values ​​in the most suitable interval, For the The number of integer values ​​in the level interval.

[0030] Preferably, the The algorithm model determines the impact of seasonal effects on citrus growth, including:

[0031] Obtain seasonal suitable planting characteristic data of the citrus growth cycle in the area to be tested, correct and standardize the data, and establish a list of seasonal suitable planting characteristic data of the citrus growth cycle;

[0032] According to the method of determining factor decomposition, the seasonal time series fitting function of citrus growth cycle is established;

[0033] according to Algorithm model, construct general autoregression operator formula and seasonal autoregression operator formula;

[0034] according to Algorithm model, construct general moving average operator formula and seasonal moving average operator formula;

[0035] According to the formulas of general autoregression and seasonal autoregression operators and general moving average operators and seasonal moving average operators, the growth cycle of citrus is established. Model;

[0036] According to the citrus growth cycle Model, extracting the quality characteristics of citrus fruits under different seasonal influences of the citrus growth cycle;

[0037] According to the citrus quality characteristics affected by the citrus growth cycle in different seasonal periods, the input matrix of seasonal citrus suitable planting characteristics was analyzed and established;

[0038] The seasonal time series fitting function expression of the citrus growth cycle is:

[0039]

[0040] in, is the observed value of the original time series, For the long-term trend, For seasonal change trends, is the cycle fluctuation value, For irregular changes;

[0041] The general autoregressive operator formula and the seasonal autoregressive operator formula are:

[0042]

[0043]

[0044] in, is a general autoregressive operator, is the seasonal autoregressive operator, is the general autoregressive order, is the seasonal autoregressive order, is the length of the seasonal cycle;

[0045] The general moving average operator formula and the seasonal moving average operator formula are:

[0046]

[0047]

[0048] in, is a general moving average operator, is the seasonal moving average operator, is the general moving average order, is the seasonal moving average order;

[0049] Said The model expression is:

[0050]

[0051] In the formula, is a constant difference, is the seasonal difference, , denote the number of non-seasonal and seasonal differences, respectively. For the Monthly seasonal suitability of citrus planting, is a constant term, is white noise and conforms to Gaussian distribution.

[0052] Preferably, the citrus suitable planting area evaluation model is established based on the LSTM neural network algorithm model, and the evaluation level of the citrus suitable planting area is obtained according to the citrus suitable planting area evaluation matrix, which specifically includes:

[0053] Based on big data, historical data or experimental data on the suitability of citrus planting in different citrus planting areas under the influence of seasonal changes in land quality and environmental factors are obtained, and a training sample set of citrus planting suitable area input is established;

[0054] According to the expected effect and standardized index of citrus planting suitability, a target training sample set of citrus planting suitable areas is established;

[0055] Standardize and normalize the data in the input training sample set of citrus suitable planting areas and the target training sample set of citrus suitable planting areas respectively;

[0056] According to the input matrix of general citrus suitable planting characteristics and the input matrix of seasonal citrus suitable planting characteristics, the evaluation matrix of citrus suitable planting areas is established;

[0057] The evaluation matrix of citrus suitable planting areas and the data in the training sample set of citrus suitable planting areas are used as the input matrix, and the data in the target training sample set of citrus suitable planting areas are used as the standard value. Based on the LSTM neural network algorithm model, the gap between the input value and the standard value is reduced through parameter iteration.

[0058] According to the learning and training results of the data, a citrus suitable planting area evaluation model is established, and according to the general citrus suitable planting characteristic data and seasonal citrus suitable planting characteristic data of the citrus suitable planting area to be tested, the citrus suitable planting level of the tested area is determined;

[0059] Determine whether the citrus suitability level of the tested area meets expectations. If so, it means that the area is suitable for citrus planting. If not, it means that the area is not suitable for citrus planting. By analyzing the parameter changes and input data in the citrus suitable planting area evaluation model, the abnormal points of citrus suitability characteristics are screened out.

[0060] Preferably, the establishment of a citrus suitable planting area evaluation platform for inputting, storing and displaying citrus suitable planting characteristic data and evaluation results specifically includes:

[0061] Establish a citrus suitable planting area evaluation platform to build the software environment for running the citrus suitable planting area evaluation model and the processing environment for data call;

[0062] Based on the citrus suitable planting area evaluation platform, general citrus suitable planting characteristic data and seasonal citrus suitable planting characteristic data are input through human-computer interaction, and the citrus suitable planting area is evaluated based on the data;

[0063] Based on the citrus suitable planting area evaluation platform, it is used to store and display citrus suitable planting characteristic data and evaluation results, and make corresponding prompts based on the evaluation results.

[0064] Furthermore, the present invention proposes a citrus suitable planting area evaluation system, which is used to implement the above-mentioned citrus suitable planting area evaluation method, including:

[0065] A feature extraction module, the feature extraction module is used to obtain the suitable planting features of the land quality and environmental data suitable for citrus planting based on big data;

[0066] Establishing an evaluation model module, the evaluation model module is used to establish a weight coefficient of the influence of different citrus suitable planting characteristics on citrus planting based on the Pearson correlation coefficient formula; Algorithm model to determine the impact of seasonal effects on citrus growth; based on the LSTM neural network algorithm model, establish a citrus suitable planting area evaluation model, and according to the citrus suitable planting area evaluation matrix, obtain the citrus suitable planting area evaluation level;

[0067] An evaluation platform module is used to establish an evaluation platform for citrus suitable planting areas, and is used to input, store and display citrus suitable planting characteristic data and evaluation results.

[0068] Preferably, the evaluation model building module specifically includes:

[0069] A general impact unit, which is used to establish a weight coefficient of the impact of different citrus planting characteristics on citrus planting based on the Pearson correlation coefficient formula;

[0070] A seasonal effect unit is used to Algorithmic model to determine the impact of seasonal effects on citrus growth;

[0071] The comprehensive evaluation unit is used to establish a citrus suitable planting area evaluation model based on the LSTM neural network algorithm model, and obtain the citrus suitable planting area evaluation level according to the citrus suitable planting area evaluation matrix.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] Based on big data, the suitable planting characteristics of citrus are obtained, and the evaluation level of citrus suitable planting areas is established to evaluate the suitability of citrus planting areas. By analyzing the influence of existing soil quality and environmental data in citrus planting areas, the input matrix of general citrus suitable planting characteristics is established according to the weight coefficients of the influence of different citrus suitable planting characteristics on citrus planting and the deviation coefficients of each grade interval and the most suitable interval. Secondly, by analyzing and processing the historical data of seasonal factors in the area to be tested, the citrus growth cycle is established. The model extracts the citrus quality characteristics affected by the citrus growth cycle in different seasonal conditions, and based on the citrus quality characteristics, analyzes and establishes the input matrix of seasonal citrus suitable planting characteristics, so as to judge the impact of seasonal effects on citrus growth. Finally, based on the LSTM neural network algorithm model, the citrus suitable planting area evaluation matrix is ​​established through the input matrix of general citrus suitable planting characteristics and the input matrix of seasonal citrus suitable planting characteristics. According to the learning and training results of the data, the citrus suitable planting area evaluation model is established. The citrus suitability level of the tested area is judged through the general citrus suitable planting characteristic data and the seasonal citrus suitable planting characteristic data of the tested citrus suitable planting area, thereby effectively reducing the seasonal impact of the citrus suitable planting area and improving the accuracy and reliability of the citrus suitable planting area evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 This is a flow chart of a method for evaluating citrus suitable planting areas of the present invention;

[0075] Figure 2 The present invention provides a flow chart of weight coefficients for establishing the effects of different citrus planting characteristics on citrus planting based on the Pearson correlation coefficient formula;

[0076] Figure 3 The basis of the present invention Algorithm model, flow chart for determining the impact of seasonal effects on citrus growth;

[0077] Figure 4 The present invention is based on the LSTM neural network algorithm model, establishes a citrus suitable planting area evaluation model, and obtains a citrus suitable planting area evaluation level flow chart according to the citrus suitable planting area evaluation matrix. DETAILED DESCRIPTION

[0078] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0079] Reference Figure 1 As shown, a method for evaluating citrus suitable planting areas comprises:

[0080] Obtain the suitable planting characteristics of land quality and environmental data suitable for citrus planting based on big data;

[0081] Based on the Pearson correlation coefficient formula, the weight coefficients of the effects of different citrus planting characteristics on citrus planting were established;

[0082] according to Algorithmic model to determine the impact of seasonal effects on citrus growth;

[0083] Based on the LSTM neural network algorithm model, a citrus suitable planting area evaluation model was established, and the evaluation level of citrus suitable planting areas was obtained according to the citrus suitable planting area evaluation matrix;

[0084] Establish a citrus suitable planting area evaluation platform for inputting, storing and displaying citrus suitable planting characteristic data and evaluation results.

[0085] It can be explained that this scheme obtains the suitable planting characteristics of citrus based on big data, and establishes the evaluation level of citrus suitable planting areas to evaluate the suitability of citrus planting areas. By analyzing the influence of existing soil quality and environmental data in citrus planting areas, according to the weight coefficients of the influence of different citrus suitable planting characteristics on citrus planting and the deviation coefficients between each grade interval and the most suitable interval, an input matrix of general citrus suitable planting characteristics is established. Secondly, by analyzing and processing the historical data of seasonal factors in the area to be tested, the citrus growth cycle is established. The model extracts the citrus quality characteristics affected by the citrus growth cycle in different seasonal conditions, and based on the citrus quality characteristics, analyzes and establishes the input matrix of seasonal citrus suitable planting characteristics, so as to judge the impact of seasonal effects on citrus growth. Finally, based on the LSTM neural network algorithm model, the citrus suitable planting area evaluation matrix is ​​established through the input matrix of general citrus suitable planting characteristics and the input matrix of seasonal citrus suitable planting characteristics. According to the learning and training results of the data, the citrus suitable planting area evaluation model is established. The citrus suitability level of the tested area is judged through the general citrus suitable planting characteristic data and the seasonal citrus suitable planting characteristic data of the tested citrus suitable planting area, thereby effectively reducing the seasonal impact of the citrus suitable planting area and improving the accuracy and reliability of the citrus suitable planting area evaluation.

[0086] Reference Figure 2As shown, the weight coefficients of the influence of different citrus suitable planting characteristics on citrus planting based on the Pearson correlation coefficient formula specifically include:

[0087] Through simulation experiments and historical data, the data on the impact of different citrus planting characteristics on citrus planting are obtained by using the control variable method;

[0088] According to the data on the impact of different citrus planting characteristics on citrus planting, the data is cleaned and standardized;

[0089] According to the Pearson correlation coefficient formula, the weight coefficients of the effects of different citrus planting characteristics on citrus planting were calculated;

[0090] According to the deviation range of the evaluation grade intervals of the citrus suitable planting areas, the deviation coefficients of each grade interval and the most suitable interval are calculated;

[0091] According to the weight coefficients of the influence of different citrus suitable planting characteristics on citrus planting and the deviation coefficients between each grade interval and the most suitable interval, the input matrix of general citrus suitable planting characteristics is established;

[0092] The weight coefficient expression for calculating the influence of different citrus suitable planting characteristics on citrus planting is:

[0093]

[0094] In the formula, is the correlation coefficient between different citrus planting characteristics and the suitability of citrus planting, that is, the weight coefficient of the influence of different citrus planting characteristics on citrus planting. For the Data on the suitable planting characteristics of citrus fruits. For the The average value of the citrus planting characteristics data, For the The suitability of citrus planting corresponding to the citrus planting characteristic data is For the The average value of the suitability of citrus planting corresponding to the citrus planting characteristic data, For the The number of groups of citrus suitable planting characteristic data;

[0095] The expression for calculating the deviation coefficient between each grade interval and the most suitable interval is:

[0096]

[0097] In the formula, For the Citrus Planting Characteristics The deviation coefficient between the grade interval and the most suitable interval, is the emphasis weight of different level intervals, The first integer value, For the The first integer value, is the number of integer values ​​in the most suitable interval, For the The number of integer values ​​in the level interval.

[0098] It can be explained that different citrus planting characteristics have different effects on citrus planting suitability. It is necessary to analyze the impact of different citrus planting characteristics on citrus planting, as well as the deviation between the grade interval and the most suitable interval of different citrus planting characteristics through big data or historical data, in order to improve the accuracy and reliability of the evaluation of citrus planting areas. This scheme uses the Pearson correlation coefficient formula to analyze the correlation coefficient between different citrus planting characteristics and the suitability of citrus planting, thereby determining the weight coefficient of the impact of different citrus planting characteristics on citrus planting, and establishing the deviation coefficient between each grade interval and the most suitable interval, and further analyzing the data of different citrus planting characteristics, so as to achieve the purpose of improving the accuracy and reliability of the evaluation of citrus planting areas.

[0099] Reference Figure 3 As shown, the basis The algorithm model determines the impact of seasonal effects on citrus growth, including:

[0100] Obtain seasonal suitable planting characteristic data of the citrus growth cycle in the area to be tested, correct and standardize the data, and establish a list of seasonal suitable planting characteristic data of the citrus growth cycle;

[0101] According to the method of determining factor decomposition, the seasonal time series fitting function of citrus growth cycle is established;

[0102] according to Algorithm model, construct general autoregression operator formula and seasonal autoregression operator formula;

[0103] according to Algorithm model, construct general moving average operator formula and seasonal moving average operator formula;

[0104] According to the formulas of general autoregression and seasonal autoregression operators and general moving average operators and seasonal moving average operators, the growth cycle of citrus is established. Model;

[0105] According to the citrus growth cycle Model, extracting the quality characteristics of citrus fruits under different seasonal influences of the citrus growth cycle;

[0106] According to the citrus quality characteristics affected by the citrus growth cycle in different seasonal periods, the input matrix of seasonal citrus suitable planting characteristics was analyzed and established;

[0107] The seasonal time series fitting function expression of the citrus growth cycle is:

[0108]

[0109] in, is the observed value of the original time series, For the long-term trend, For seasonal change trends, is the cycle fluctuation value, For irregular changes;

[0110] The general autoregressive operator formula and the seasonal autoregressive operator formula are:

[0111]

[0112]

[0113] in, is a general autoregressive operator, is the seasonal autoregressive operator, is the general autoregressive order, is the seasonal autoregressive order, is the length of the seasonal cycle;

[0114] The general moving average operator formula and the seasonal moving average operator formula are:

[0115]

[0116]

[0117] in, is a general moving average operator, is the seasonal moving average operator, is the general moving average order, is the seasonal moving average order;

[0118] Said The model expression is:

[0119]

[0120] In the formula, is a constant difference, is the seasonal difference, , denote the number of non-seasonal and seasonal differences, respectively. For the Monthly seasonal suitability of citrus planting, is a constant term, is white noise and conforms to Gaussian distribution.

[0121] It can be explained that the growth and product quality of citrus are not only affected by soil quality and environmental factors, but also by seasonal factors in the planting area. When evaluating the suitable planting area for citrus, the impact of seasonal factors in the area needs to be considered. The algorithm model has good seasonal judgment and prediction fit, and can effectively analyze the impact of seasonal factors on citrus planting. This scheme analyzes and processes the historical data of seasonal factors in the area to be tested to establish a citrus growth cycle. The model extracts the quality characteristics of citrus affected by the citrus growth cycle in different seasonal periods, and based on the citrus quality characteristics, analyzes and establishes the input matrix of seasonal citrus suitable planting characteristics, so as to achieve the purpose of judging the impact of seasonal effects on citrus growth.

[0122] Reference Figure 4 As shown, the LSTM neural network algorithm model is used to establish a citrus suitable planting area evaluation model. According to the citrus suitable planting area evaluation matrix, the evaluation levels of citrus suitable planting areas are obtained, which specifically include:

[0123] Based on big data, historical data or experimental data on the suitability of citrus planting in different citrus planting areas under the influence of seasonal changes in land quality and environmental factors are obtained, and a training sample set of citrus planting suitable area input is established;

[0124] According to the expected effect and standardized index of citrus planting suitability, a target training sample set of citrus planting suitable areas is established;

[0125] Standardize and normalize the data in the input training sample set of citrus suitable planting areas and the target training sample set of citrus suitable planting areas respectively;

[0126] According to the input matrix of general citrus suitable planting characteristics and the input matrix of seasonal citrus suitable planting characteristics, the evaluation matrix of citrus suitable planting areas is established;

[0127] The evaluation matrix of citrus suitable planting areas and the data in the training sample set of citrus suitable planting areas are used as the input matrix, and the data in the target training sample set of citrus suitable planting areas are used as the standard value. Based on the LSTM neural network algorithm model, the gap between the input value and the standard value is reduced through parameter iteration.

[0128] According to the learning and training results of the data, a citrus suitable planting area evaluation model is established, and according to the general citrus suitable planting characteristic data and seasonal citrus suitable planting characteristic data of the citrus suitable planting area to be tested, the citrus suitable planting level of the tested area is determined;

[0129] Determine whether the citrus suitability level of the tested area meets expectations. If so, it means that the area is suitable for citrus planting. If not, it means that the area is not suitable for citrus planting. By analyzing the parameter changes and input data in the citrus suitable planting area evaluation model, the abnormal points of citrus suitability characteristics are screened out.

[0130] It can be explained that this scheme establishes a citrus suitable planting area evaluation matrix through the input matrix of general citrus suitable planting characteristics and the input matrix of seasonal citrus suitable planting characteristics confirmed in the early stage. Combined with the advantages of long and short-term memory of the LSTM neural network algorithm model, the citrus suitable planting area evaluation model can effectively judge the citrus suitability level of the tested area and improve the accuracy and reliability of the citrus suitable planting area evaluation.

[0131] Further, based on the same inventive concept as the above-mentioned method for evaluating citrus suitable planting areas, this solution proposes a citrus suitable planting area evaluation system, comprising:

[0132] A feature extraction module, the feature extraction module is used to obtain the suitable planting features of the land quality and environmental data suitable for citrus planting based on big data;

[0133] Establishing an evaluation model module, the evaluation model module is used to establish a weight coefficient of the influence of different citrus suitable planting characteristics on citrus planting based on the Pearson correlation coefficient formula; Algorithm model to determine the impact of seasonal effects on citrus growth; based on the LSTM neural network algorithm model, establish a citrus suitable planting area evaluation model, and according to the citrus suitable planting area evaluation matrix, obtain the citrus suitable planting area evaluation level;

[0134] An evaluation platform module, which is used to establish an evaluation platform for citrus suitable planting areas, and is used to input, store and display citrus suitable planting characteristic data and evaluation results;

[0135] The evaluation model module includes:

[0136] A general impact unit, which is used to establish a weight coefficient of the impact of different citrus planting characteristics on citrus planting based on the Pearson correlation coefficient formula;

[0137] A seasonal effect unit is used to Algorithmic model to determine the impact of seasonal effects on citrus growth;

[0138] The comprehensive evaluation unit is used to establish a citrus suitable planting area evaluation model based on the LSTM neural network algorithm model, and obtain the citrus suitable planting area evaluation level according to the citrus suitable planting area evaluation matrix.

[0139] In summary, the advantages of the present invention are: based on big data, the citrus suitable planting characteristics are obtained, and the evaluation level of citrus suitable planting areas is established to evaluate the suitability of citrus planting areas. By analyzing the influence of existing soil quality and environmental data in citrus planting areas, according to the weight coefficients of the influence of different citrus suitable planting characteristics on citrus planting and the deviation coefficients of each grade interval and the most suitable interval, an input matrix of general citrus suitable planting characteristics is established. Secondly, by analyzing and processing the historical data of seasonal factors in the area to be detected, the citrus growth cycle is established. The model extracts the citrus quality characteristics affected by the citrus growth cycle in different seasonal conditions, and based on the citrus quality characteristics, analyzes and establishes the input matrix of seasonal citrus suitable planting characteristics, so as to judge the impact of seasonal effects on citrus growth. Finally, based on the LSTM neural network algorithm model, the citrus suitable planting area evaluation matrix is ​​established through the input matrix of general citrus suitable planting characteristics and the input matrix of seasonal citrus suitable planting characteristics. According to the learning and training results of the data, the citrus suitable planting area evaluation model is established. The citrus suitability level of the tested area is judged through the general citrus suitable planting characteristic data and the seasonal citrus suitable planting characteristic data of the tested citrus suitable planting area, thereby effectively reducing the seasonal impact of the citrus suitable planting area and improving the accuracy and reliability of the citrus suitable planting area evaluation.

[0140] 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 only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for evaluating citrus suitable planting areas, characterized in that: include: Obtain the suitable planting characteristics of land quality and environmental data suitable for citrus planting based on big data; Based on the Pearson correlation coefficient formula, the weight coefficients of the effects of different citrus planting characteristics on citrus planting were established; according to Algorithmic model to determine the impact of seasonal effects on citrus growth; Based on the LSTM neural network algorithm model, a citrus suitable planting area evaluation model was established, and the evaluation level of citrus suitable planting areas was obtained according to the citrus suitable planting area evaluation matrix; Establish a citrus suitable planting area evaluation platform for inputting, storing and displaying citrus suitable planting characteristic data and evaluation results.

2. A method for evaluating citrus suitable planting areas according to claim 1, characterized in that: The suitable planting characteristics of the land quality and environmental data suitable for citrus planting obtained based on big data specifically include: Based on big data and historical data analysis, obtain the suitable planting characteristics of land quality and environmental data suitable for citrus planting; According to the suitable planting characteristics of citrus, various data intervals of the suitable planting characteristics of citrus are established, and a sample list of the suitable planting characteristics of citrus is established; Among them, the characteristics suitable for citrus cultivation include soil pH, soil thickness, organic matter content, soil permeability and drainage performance, as well as temperature, humidity, wind speed, light intensity, rainfall and growth cycle; Based on historical data and big data analysis, an evaluation grade of citrus suitable planting areas is established for different data intervals of citrus suitable planting characteristics in the sample list of citrus suitable planting characteristics; Based on the evaluation levels of areas suitable for growing citrus, a classification database of evaluation levels of areas suitable for growing citrus will be established.

3. A method for evaluating citrus suitable planting areas according to claim 2, characterized in that: The weight coefficients of the influence of different citrus planting characteristics on citrus planting based on the Pearson correlation coefficient formula specifically include: Through simulation experiments and historical data, the data on the impact of different citrus planting characteristics on citrus planting are obtained by using the control variable method; According to the data on the impact of different citrus planting characteristics on citrus planting, the data is cleaned and standardized; According to the Pearson correlation coefficient formula, the weight coefficients of the effects of different citrus planting characteristics on citrus planting were calculated; According to the deviation range of the evaluation grade intervals of the citrus suitable planting areas, the deviation coefficients of each grade interval and the most suitable interval are calculated; According to the weight coefficients of the influence of different citrus suitable planting characteristics on citrus planting and the deviation coefficients between each grade interval and the most suitable interval, the input matrix of general citrus suitable planting characteristics is established; The weight coefficient expression for calculating the influence of different citrus suitable planting characteristics on citrus planting is: In the formula, is the correlation coefficient between different citrus planting characteristics and the suitability of citrus planting, that is, the weight coefficient of the influence of different citrus planting characteristics on citrus planting. For the Data on the suitable planting characteristics of citrus fruits. For the The average value of the citrus planting characteristics data, For the The suitability of citrus planting corresponding to the citrus planting characteristic data is For the The average value of the suitability of citrus planting corresponding to the citrus planting characteristic data, For the The number of groups of citrus suitable planting characteristic data; The expression for calculating the deviation coefficient between each grade interval and the most suitable interval is: In the formula, For the Citrus Planting Characteristics The deviation coefficient between the grade interval and the most suitable interval, is the emphasis weight of different level intervals, The most suitable interval integer value, For the The first integer value, is the number of integer values ​​in the most suitable interval, For the The number of integer values ​​in the level interval.

4. A method for evaluating citrus suitable planting areas according to claim 3, characterized in that: The basis The algorithm model determines the impact of seasonal effects on citrus growth, including: Obtain seasonal suitable planting characteristic data of the citrus growth cycle in the area to be tested, correct and standardize the data, and establish a list of seasonal suitable planting characteristic data of the citrus growth cycle; According to the method of determining factor decomposition, the seasonal time series fitting function of citrus growth cycle is established; according to Algorithm model, construct general autoregression operator formula and seasonal autoregression operator formula; according to Algorithm model, construct general moving average operator formula and seasonal moving average operator formula; According to the formulas of general autoregression and seasonal autoregression operators and general moving average operators and seasonal moving average operators, the growth cycle of citrus is established. Model; According to the citrus growth cycle Model, extracting the quality characteristics of citrus fruits under different seasonal influences of the citrus growth cycle; According to the citrus quality characteristics affected by the citrus growth cycle in different seasonal periods, the input matrix of seasonal citrus suitable planting characteristics was analyzed and established; The seasonal time series fitting function expression of the citrus growth cycle is: in, is the observed value of the original time series, For the long-term trend, For seasonal change trends, is the cycle fluctuation value, For irregular changes; The general autoregressive operator formula and the seasonal autoregressive operator formula are: in, is a general autoregressive operator, is the seasonal autoregressive operator, is the general autoregressive order, is the seasonal autoregressive order, is the length of the seasonal cycle; The general moving average operator formula and the seasonal moving average operator formula are: in, is a general moving average operator, is the seasonal moving average operator, is the general moving average order, is the seasonal moving average order; Said The model expression is: In the formula, is a constant difference, is the seasonal difference, , denote the number of non-seasonal and seasonal differences, respectively. For the Monthly seasonal suitability of citrus planting, is a constant term, is white noise and conforms to Gaussian distribution.

5. A method for evaluating citrus suitable planting areas according to claim 4, characterized in that: Based on the LSTM neural network algorithm model, a citrus suitable planting area evaluation model is established, and according to the citrus suitable planting area evaluation matrix, the evaluation levels of citrus suitable planting areas are obtained, which specifically include: Based on big data, historical data or experimental data on the suitability of citrus planting in different citrus planting areas under the influence of seasonal changes in land quality and environmental factors are obtained, and a training sample set of citrus planting suitable area input is established; According to the expected effect and standardized index of citrus planting suitability, a target training sample set of citrus planting suitable areas is established; Standardize and normalize the data in the input training sample set of citrus suitable planting areas and the target training sample set of citrus suitable planting areas respectively; According to the input matrix of general citrus suitable planting characteristics and the input matrix of seasonal citrus suitable planting characteristics, the evaluation matrix of citrus suitable planting areas is established; The evaluation matrix of citrus suitable planting areas and the data in the training sample set of citrus suitable planting areas are used as the input matrix, and the data in the target training sample set of citrus suitable planting areas are used as the standard value. Based on the LSTM neural network algorithm model, the gap between the input value and the standard value is reduced through parameter iteration. According to the learning and training results of the data, a citrus suitable planting area evaluation model is established, and according to the general citrus suitable planting characteristic data and seasonal citrus suitable planting characteristic data of the citrus suitable planting area to be tested, the citrus suitable planting level of the tested area is determined; Determine whether the citrus suitability level of the tested area meets expectations. If so, it means that the area is suitable for citrus planting. If not, it means that the area is not suitable for citrus planting. By analyzing the parameter changes and input data in the citrus suitable planting area evaluation model, the abnormal points of citrus suitability characteristics are screened out.

6. A method for evaluating citrus suitable planting areas according to claim 5, characterized in that: The establishment of a citrus suitable planting area evaluation platform for inputting, storing and displaying citrus suitable planting characteristic data and evaluation results specifically includes: Establish a citrus suitable planting area evaluation platform to build the software environment for running the citrus suitable planting area evaluation model and the processing environment for data call; Based on the citrus suitable planting area evaluation platform, general citrus suitable planting characteristic data and seasonal citrus suitable planting characteristic data are input through human-computer interaction, and the citrus suitable planting area is evaluated based on the data; Based on the citrus suitable planting area evaluation platform, it is used to store and display citrus suitable planting characteristic data and evaluation results, and make corresponding prompts based on the evaluation results.

7. A citrus suitable planting area evaluation system, characterized in that: A method for evaluating a citrus suitable planting area according to any one of claims 1 to 6, comprising: A feature extraction module, the feature extraction module is used to obtain the suitable planting features of the land quality and environmental data suitable for citrus planting based on big data; Establishing an evaluation model module, the evaluation model module is used to establish a weight coefficient of the influence of different citrus suitable planting characteristics on citrus planting based on the Pearson correlation coefficient formula; Algorithm model to determine the impact of seasonal effects on citrus growth; based on the LSTM neural network algorithm model, establish a citrus suitable planting area evaluation model, and according to the citrus suitable planting area evaluation matrix, obtain the citrus suitable planting area evaluation level; An evaluation platform module is used to establish an evaluation platform for citrus suitable planting areas, and is used to input, store and display citrus suitable planting characteristic data and evaluation results.

8. The citrus suitable planting area evaluation system according to claim 7, characterized in that: The evaluation model building module specifically includes: A general impact unit, which is used to establish a weight coefficient of the impact of different citrus planting characteristics on citrus planting based on the Pearson correlation coefficient formula; A seasonal effect unit is used to Algorithmic model to determine the impact of seasonal effects on citrus growth; The comprehensive evaluation unit is used to establish a citrus suitable planting area evaluation model based on the LSTM neural network algorithm model, and obtain the citrus suitable planting area evaluation level according to the citrus suitable planting area evaluation matrix.

Images