Soil acidification intelligent early warning system and method based on big data

Abstract

The invention relates to the technical field of soil data processing, in particular to a soil acidification intelligent early warning system and method based on big data, and the system comprises a dynamic feature mapping module, a potential feature capturing module, an anomaly detection module, an acidification risk assessment module and a hot spot region recognition module. According to the method, through dynamic correlation analysis of multi-dimensional data, time-space mapping is carried out on soil acidity and alkalinity, temperature and humidity, fertilizer types, application amounts and other data, time indexes, space coordinates and dynamic characteristics are extracted, abnormal points of acidity and alkalinity changes are identified based on dynamic characteristic parameters, and acidification abnormal areas and risk distribution are positioned in combination with spatial density analysis. By extracting the pH value range, the dynamic change rate and the time distribution frequency, the acidification risk trend and hot spot area distribution are obtained, the comprehensiveness and prediction accuracy of data are improved through multi-dimensional dynamic processing and accurate risk assessment, and monitoring, recognition and trend early warning of soil acidification are effectively achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of soil data processing, and in particular to a soil acidification intelligent early warning system and method based on big data. Background Art

[0002] The field of soil data processing technology refers to a technical field that uses advanced information technology to collect, store, analyze and process soil-related data. This field usually relies on modern technologies such as big data, artificial intelligence, sensor technology and cloud computing to monitor and analyze multiple indicators such as soil quality, composition, temperature and humidity, and pH. Its main uses include the assessment of soil health status, precise fertilization and irrigation in agricultural production, monitoring and control of soil pollution, and providing data support for the rational use of land resources. Through efficient and accurate data processing methods, the field of soil data processing technology provides important support for the sustainable development of agriculture and the protection of the ecological environment.

[0003] Among them, the soil acidification intelligent early warning system refers to a system that uses soil data processing technology to monitor soil pH trends, assess the risk of soil acidification, and provide early warning information through real-time analysis and modeling of big data. The system is widely used in the fields of agricultural production and ecological protection, aiming to help farmers and relevant departments take scientific and reasonable prevention and control measures in advance to prevent further aggravation of soil acidification and ensure the stability of agricultural production and the sustainability of the ecological environment.

[0004] Existing technologies lack dynamic analysis of the relationship between time and space in soil data processing, and are unable to accurately capture the changing trends of pH over time and space, resulting in insufficient data correlation. In addition, in terms of anomaly detection, existing technologies are mostly based on single threshold judgments, without considering the spatial density and dynamic evolution of data, which can easily lead to inaccurate identification of abnormal areas, and prominent omissions or misjudgments. In risk assessment, existing technologies rely on static indicators and lack assessment of dynamic change rates and trend distributions, making it difficult to achieve forward-looking early warnings of acidification trends. These deficiencies affect the scientificity and timeliness of soil acidification prevention and control, and it is difficult to provide effective decision-making basis for agricultural production and ecological protection. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a soil acidification intelligent early warning system and method based on big data.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A soil acidification intelligent early warning system based on big data comprises:

[0007] The dynamic feature mapping module obtains soil pH, temperature and humidity, fertilizer type and application amount data, extracts time index and spatial coordinate value, matches the time relationship between pH and temperature and humidity, generates a spatiotemporal mapping table and calculates dynamic correlation features to generate soil spatiotemporal feature mapping values;

[0008] The potential feature capture module decomposes the spatial and temporal characteristics of the pH data based on the soil spatiotemporal feature mapping value, extracts the characteristic values ​​and distribution parameters of the associated changes, screens the characteristic change values ​​in the dynamic evolution matrix, and obtains the dynamic characteristic parameters of soil acidification;

[0009] The anomaly detection module calculates the pH change deviation value and time index data based on the soil acidification dynamic characteristic parameters, extracts abnormal points that are out of range and locates the coordinate values, identifies outliers with sparse spatial density distribution, and generates acidification abnormal area distribution parameters;

[0010] The acidification risk assessment module extracts the value range and dynamic change rate of the abnormal pH area based on the acidification abnormal area distribution parameters, calculates the frequency and spatial correlation characteristics of the time distribution, screens the dynamic characteristic distribution of the risk point, and obtains the acidification risk trend parameters;

[0011] The hotspot area identification module calculates the pH area density value and spatial distance based on the acidification risk trend parameters, extracts density peak points and screens hotspot areas, sets distribution thresholds to mark distribution ranges, and generates soil acidification hotspot areas.

[0012] The soil spatiotemporal characteristic mapping value includes time index, spatial coordinate value, time relationship between pH and temperature and humidity, and dynamic correlation characteristics. The soil acidification dynamic characteristic parameters include characteristic values ​​of correlation changes, distribution parameters, and characteristic change values. The acidification abnormal area distribution parameters include pH change deviation values, time index data, abnormal point coordinate values, and outliers with sparse spatial density distribution. The acidification risk trend parameters include the numerical range of pH abnormal areas, dynamic change rate, time distribution frequency, spatial correlation characteristics, and dynamic characteristic distribution of risk points. The soil acidification hotspot area includes pH area density value, spatial distance, density peak point, distribution threshold mark, and distribution range.

[0013] As a further solution of the present invention, the step of obtaining the soil spatiotemporal characteristic mapping value is specifically as follows:

[0014] Extract soil pH, temperature and humidity, fertilizer type and application amount data, match multiple parameters with time index and spatial coordinate values, generate a time series data table through the time dimension of the parameters, and obtain the matching results of time index and spatial coordinates;

[0015] Based on the matching results of the time index and the spatial coordinates, the dynamic correlation between pH and temperature and humidity under multiple time indexes is calculated, and by constructing a dynamic time relationship, the change trend between the two is matched and a time dynamic feature matrix is ​​established;

[0016] According to the fertilizer type and application amount parameters, the spatial coordinate values ​​are compared with the time dynamic feature matrix to form a time-space correspondence and perform feature weighting, using the formula:

[0017]

[0018] Calculate and generate soil spatiotemporal characteristic mapping values;

[0019] Among them, M st represents the soil spatiotemporal characteristic mapping value, p i Indicates the time dynamic correlation value between pH and temperature and humidity, w ai represents the spatial weighting parameter of fertilizer type, w ti represents the spatial weight parameter of the application rate, and n represents the number of spatial coordinates.

[0020] As a further solution of the present invention, the step of obtaining the soil acidification dynamic characteristic parameters is specifically as follows:

[0021] Based on the soil spatiotemporal characteristic mapping value, the spatial and temporal characteristics of the pH data are decomposed, the pH change trend of multiple locations is identified through spatial analysis, the periodic change characteristics are revealed through temporal analysis, and the spatial and temporal characteristic decomposition results are generated;

[0022] Extracting characteristic values ​​significantly associated with changes in pH from the spatial-temporal characteristic decomposition results, using statistical methods to determine their contribution to the soil acidification trend, and obtaining associated change characteristic values;

[0023] The associated change characteristic values ​​are used to screen the key characteristic change values ​​characterizing the soil acidification process, using the formula:

[0024]

[0025] The dynamic characteristic parameters of soil acidification are calculated;

[0026] Among them, P acid represents the dynamic characteristic parameter of soil acidification, V ti is the characteristic change value in time period i, α i is the adjustment coefficient associated with the soil type, β is the coefficient of curve adjustment, which controls the slope of the logistic function and affects the rate of contribution of pH change to the output, μ is the midpoint offset of the curve, which determines the position of the midpoint of the logistic function and affects the judgment threshold of the soil acidification state, and k is the number of time periods evaluated.

[0027] As a further solution of the present invention, the step of obtaining the distribution parameters of the acidification abnormal area is specifically as follows:

[0028] Based on the soil acidification dynamic characteristic parameters, the pH change deviation value is calculated, compared with the time index data, and the data points beyond the normal change range are identified through the set statistical threshold to generate abnormal deviation analysis results;

[0029] Matching the data points in the abnormal deviation analysis results with their spatial coordinates, locating the coordinates of each abnormal point using GIS technology, and obtaining an abnormal point coordinate matching result;

[0030] Analyze the coordinate matching results of the abnormal points, use the density clustering algorithm to identify outliers with sparse spatial density distribution, and use the formula:

[0031]

[0032] Calculate and generate the distribution parameters of the acidification anomaly area;

[0033] Among them, D acid represents the regional distribution parameter of acidification anomaly, N j represents the number of abnormal points in the jth spatial region, α j is the weight coefficient of the jth spatial region, which is used to adjust the contribution of each region in the total acidification distribution parameters, reflecting the difference in regional environmental sensitivity or soil reactivity, γ k is the adjustment coefficient of the kth spatial region, which is used to balance the impact of the number of outliers in the differentiated region, and m represents the total number of spatial regions analyzed, which is a constant determined according to the regional division defined in the geographic information system.

[0034] As a further solution of the present invention, the step of obtaining the acidification risk trend parameter is specifically as follows:

[0035] Based on the acidification abnormal area distribution parameters, the value range of the pH abnormal area is analyzed, the dynamic change rate of multiple areas is extracted, the trend change is identified by comparing with previous data, and the value range and change rate analysis results are generated;

[0036] Using the numerical range and change rate analysis results, calculate the time distribution frequency of pH changes in multiple regions, combine spatial data to analyze the correlation characteristics, process the data with a statistical model, and obtain the time distribution and spatial correlation characteristic results;

[0037] According to the time distribution and spatial correlation feature results, the dynamic feature distribution area with risk potential is screened, and the statistical formula is used:

[0038]

[0039] Adjust sensitivity to low temporal change rates to calculate acidification risk trend parameters;

[0040] Among them, T c Indicates the characteristic value of the target area in the current time period, T avg represents the time average of the characteristic value of the target area, S std represents the standard deviation of the spatial distribution characteristics of the target area, W sp Represents the spatial weight parameter, which assigns the weight of the target area according to the spatial distribution characteristics. The assignment range is set according to the actual distribution characteristics. t It represents the time change rate parameter, that is, the change rate of the acidification associated characteristic value of the target area in adjacent time periods, which is obtained by calculating the time distribution characteristics, and R is the acidification risk trend parameter.

[0041] As a further solution of the present invention, the step of obtaining the soil acidification hotspot area is specifically as follows:

[0042] Based on the acidification risk trend parameters, density values ​​of multi-pH areas are calculated, and spatial distances within the areas are referenced to capture density distribution and generate density and distance analysis results;

[0043] Utilizing the density and distance analysis results, extracting density peak points through a statistical algorithm to obtain density peak point extraction results;

[0044] According to the density peak point extraction results, the hotspot areas are screened and marked, and the distribution threshold is set to divide and mark the soil acidification areas, using the formula:

[0045]

[0046] Determine the hotspot areas with higher probability of abnormal pH, calculate and generate soil acidification hotspot areas;

[0047] Among them, H acid Indicates soil acidification hotspots, D i represents the density value in the i-th evaluation area, reflecting the degree of aggregation of abnormal events in the area, S i Represents the spatial distance of the region, measuring the distance between spatial data points, T i is the frequency of the time distribution of the region, reflecting the frequency of regional activities within the monitoring window, ∈ is a small positive number used to stabilize the calculation and avoid the divisor being zero, and p represents the total number of evaluation regions, which is used to iterate all associated regions in the summation process.

[0048] A soil acidification intelligent early warning method based on big data, the soil acidification intelligent early warning method based on big data is performed based on the above-mentioned soil acidification intelligent early warning system based on big data, and comprises the following steps:

[0049] S1: Based on the soil pH, temperature and humidity, fertilizer type and application amount data, extract the time index and spatial coordinate value, match the soil pH with temperature and humidity in time, calculate the spatiotemporal mapping value of multiple data points, classify and sort them by time index and spatial coordinates, and generate soil spatiotemporal feature mapping value;

[0050] S2: Based on the soil spatiotemporal characteristic mapping value, the spatial characteristics and temporal characteristics of the pH data are decomposed, the characteristic values ​​and distribution parameters associated with the pH change are extracted, the characteristic change values ​​in the time evolution matrix are screened, the change characteristic parameters under the spatial coordinates are extracted, and the dynamic characteristic parameters of the pH are generated;

[0051] S3: Based on the dynamic characteristic parameters of pH, calculate the pH change deviation value and time index data, extract abnormal points beyond the parameter setting range, filter the coordinate values ​​corresponding to the time index, identify outliers with sparse spatial density distribution, locate the distribution parameters of the abnormal area, and generate pH abnormal area parameters;

[0052] S4: Based on the pH abnormality area parameters, calculate the pH distribution density value and spatial distance in the area, extract the coordinate range corresponding to the density peak point, combine the pH dynamic change frequency and correlation characteristics, mark the threshold area of ​​the spatial distribution, and generate the acidification risk hotspot area.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are:

[0054] In the present invention, through the dynamic correlation analysis of multidimensional data, the soil pH, temperature and humidity, fertilizer type and application amount and other data are mapped in time and space, and the time index, spatial coordinates and dynamic characteristics are extracted to ensure that the time and space distribution and correlation relationship of the data are more accurate. Based on the dynamic characteristic parameters, the abnormal points of pH changes are identified, and combined with the spatial density analysis, the abnormal acidification areas and risk distribution are accurately located. By extracting the pH value range, dynamic change rate and time distribution frequency, the acidification risk trend and hot spot area distribution are obtained to form more targeted early warning information. Multi-dimensional dynamic processing and precise risk assessment improve the comprehensiveness of the data and the accuracy of prediction, effectively realize the monitoring, identification and trend warning of soil acidification, and provide data support for scientific prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a system flow chart of the present invention;

[0056] Figure 2 This is a flow chart of the steps for obtaining the soil spatiotemporal characteristic mapping value of the present invention;

[0057] Figure 3This is a flow chart of the steps for obtaining the dynamic characteristic parameters of soil acidification according to the present invention;

[0058] Figure 4 This is a flow chart of the steps for obtaining the distribution parameters of the acidification abnormal area of ​​the present invention;

[0059] Figure 5 This is a flow chart of the steps for obtaining the acidification risk trend parameters of the present invention;

[0060] Figure 6 This is a flow chart of the steps for obtaining the soil acidification hotspot area of ​​the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0063] Embodiment 1

[0064] See also Figure 1 The present invention provides a technical solution: a soil acidification intelligent early warning system based on big data includes:

[0065] The dynamic feature mapping module obtains soil pH, temperature and humidity, fertilizer type and application amount data, extracts time index and spatial coordinate value, matches the time relationship between pH and temperature and humidity, generates a spatiotemporal mapping table and calculates dynamic correlation features to generate soil spatiotemporal feature mapping values;

[0066] The potential feature capture module decomposes the spatial and temporal characteristics of the pH data based on the soil spatiotemporal feature mapping value, extracts the characteristic values ​​and distribution parameters of the associated changes, screens the characteristic change values ​​in the dynamic evolution matrix, and obtains the dynamic characteristic parameters of soil acidification;

[0067] The anomaly detection module calculates the pH change deviation value and time index data based on the soil acidification dynamic characteristic parameters, extracts outliers that are out of range and locates the coordinate values, identifies outliers with sparse spatial density distribution, and generates acidification abnormal area distribution parameters;

[0068] The acidification risk assessment module extracts the value range and dynamic change rate of the abnormal pH area based on the distribution parameters of the acidification abnormal area, calculates the frequency and spatial correlation characteristics of the time distribution, screens the dynamic characteristic distribution of the risk point, and obtains the acidification risk trend parameters;

[0069] The hotspot area identification module calculates the pH area density value and spatial distance based on the acidification risk trend parameters, extracts the density peak points and screens the hotspot areas, sets the distribution threshold to mark the distribution range, and generates soil acidification hotspot areas.

[0070] The soil spatiotemporal characteristic mapping values ​​include time index, spatial coordinate value, time relationship between pH and temperature and humidity, and dynamic correlation characteristics. The dynamic characteristic parameters of soil acidification include characteristic values ​​of correlation changes, distribution parameters, and characteristic change values. The distribution parameters of acidification abnormal area include pH change deviation value, time index data, abnormal point coordinate value, and outliers with sparse spatial density distribution. The acidification risk trend parameters include the numerical range of pH abnormal area, dynamic change rate, time distribution frequency, spatial correlation characteristics, and dynamic characteristic distribution of risk points. The soil acidification hotspot area includes pH regional density value, spatial distance, density peak point, distribution threshold mark, and distribution range.

[0071] See also Figure 2 , the specific steps for obtaining soil spatiotemporal characteristic mapping values ​​are as follows:

[0072] Extract soil pH, temperature and humidity, fertilizer type and application amount data, match multiple parameters with time index and spatial coordinate values, generate a time series data table through the time dimension of the parameters, and obtain the matching results of time index and spatial coordinates;

[0073] The data is monitored in real time through data acquisition equipment and transmitted to the database through the network. The data in the database contains the time label and location information of each detection point. These data are compared with the soil physical property database, and the soil pH and temperature and humidity values ​​of each location point are matched with the time label as the index. In this process, special attention is paid to the integrity and accuracy of the data, and abnormal data points are eliminated through quality control algorithms to ensure that each data point can reflect the actual soil conditions and obtain the time index and space coordinate matching results.

[0074] Based on the matching results of time index and spatial coordinates, the dynamic correlation between pH and temperature and humidity under multiple time indexes is calculated. By constructing a dynamic time relationship, the change trend between the two is matched and a time dynamic feature matrix is ​​established.

[0075] Data processing software is used to perform dynamic window analysis on pH and temperature and humidity data to capture changing trends at different time points. Time series analysis techniques, such as moving average or exponential smoothing, are used to predict and smooth data points to improve the accuracy of data prediction. The data is then interpreted in conjunction with soil science expertise to extract temporal dynamic features that aid in decision making. These features provide input for subsequent models to generate a temporal dynamic feature matrix.

[0076] According to the fertilizer type and application amount parameters, the spatial coordinate values ​​are compared with the time dynamic feature matrix to form a time-space correspondence and perform feature weighting using the formula:

[0077]

[0078] Calculate and generate soil spatiotemporal characteristic mapping values;

[0079] Among them, M st represents the soil spatiotemporal characteristic mapping value, p i Indicates the time dynamic correlation value between pH and temperature and humidity, w ai represents the spatial weighting parameter of fertilizer type, w ti represents the spatial weight parameter of the application rate, and n represents the number of spatial coordinates.

[0080] formula:

[0081]

[0082] The benefit of the formula is that by integrating the temporal dynamic correlation values ​​of pH with temperature and humidity and the spatial weights of fertilizer type and application amount, it can more accurately describe the spatiotemporal characteristics of soil and reflect the comprehensive quality of the soil environment, which is of great significance for the implementation of precision agriculture.

[0083] Detailed explanation of the formula and the process of formula calculation and derivation:

[0084] Assume that in the monitoring of a specific test plot, data from three spatial points are obtained, where the temporal dynamic correlation values ​​of pH and temperature and humidity are p1 = 0.8, p2 = 0.9, and p3 = 0.85, and the spatial weighting parameter w of the fertilizer type is a1 =0.5,w a2 =0.3,w a3 =0.2, the spatial weight parameter w of the application rate t1 =20,w t2 =15,w t3 =10, and then we can calculate:

[0085]

[0086] The results show that the soil spatiotemporal characteristic mapping value of the plot is 0.01795, which reflects the comprehensive fertility status of the soil at that specific time and space point and can provide a scientific basis for fertilization strategies.

[0087] See also Figure 3 , the specific steps for obtaining soil acidification dynamic characteristic parameters are as follows:

[0088] Based on the soil spatiotemporal characteristic mapping values, the spatial and temporal characteristics of the pH data are decomposed. The pH change trend of multiple locations is identified through spatial analysis. The periodic change characteristics are revealed through temporal analysis, and the spatial and temporal characteristic decomposition results are generated.

[0089] Identify the key change areas in the spatial distribution and the periodic characteristics of the time series, correspond these data points to the spatial coordinate system, and perform statistical analysis to evaluate the magnitude and speed of change of each point in the time series. This comprehensive analysis helps to understand the changing trends of soil pH in different regions, thereby providing data support for further soil management and adjustment. Through this stage of analysis, a detailed spatial and temporal feature decomposition result is obtained.

[0090] From the results of spatial and temporal feature decomposition, the eigenvalues ​​significantly associated with pH changes were extracted, and their contribution to soil acidification trends was determined using statistical methods to obtain associated change eigenvalues.

[0091] The principal component analysis and factor analysis methods were used to conduct an in-depth analysis of the pH data at each time point. Through these methods, the factors that most significantly affect the pH were identified, and then the key variables directly related to soil acidification were screened out. The identification of these variables helps to accurately predict the future trend of soil acidification and provide a basis for the formulation of prevention and control measures. Through these analyses, a set of correlated change characteristic values ​​that can represent the soil acidification trend are obtained.

[0092] Using the associated change characteristic values, the key characteristic change values ​​that characterize the soil acidification process are screened, using the formula:

[0093]

[0094] The dynamic characteristic parameters of soil acidification are calculated;

[0095] Among them, P acid represents the dynamic characteristic parameter of soil acidification, V ti is the characteristic change value in time period i, α i is the adjustment coefficient associated with the soil type, β is the coefficient of curve adjustment, which controls the slope of the logistic function and affects the rate of contribution of pH change to the output, μ is the midpoint offset of the curve, which determines the position of the midpoint of the logistic function and affects the judgment threshold of the soil acidification state, and k is the number of time periods evaluated.

[0096] formula:

[0097]

[0098] The benefit of the formula is that it is able to simulate the dynamic changes in soil acidification through nonlinear functions, where the adjustment of each parameter allows the model to accurately reflect the nonlinear characteristics of soil acidification trends based on actual data.

[0099] Detailed explanation of the formula and the process of formula calculation and derivation:

[0100] Set the actual parameter value to: V ti =5.2,α i =1.3, β=0.85, μ=3.5, k=12. The calculation process is as follows:

[0101]

[0102]

[0103] The results show that the calculated soil acidification dynamic characteristic parameter is 5.47. This result helps to understand the changes in soil acidification over time series, reflects the current state of soil acidification, and provides a quantitative basis for soil improvement and regulation measures.

[0104] See also Figure 4 , the specific steps for obtaining the regional distribution parameters of acidification anomaly are as follows:

[0105] Based on the soil acidification dynamic characteristic parameters, the pH change deviation value is calculated, compared with the time index data, and the data points beyond the normal change range are identified through the set statistical threshold to generate abnormal deviation analysis results;

[0106] Continuous soil pH data is obtained through high-precision sensors, and time series analysis technology is used to identify trends and periodic changes in the data. Then, anomaly detection algorithms in machine learning methods are used to determine abnormal points where the deviation values ​​exceed the normal range. This includes the use of density-based clustering techniques such as DBSCAN, which can effectively identify abnormal patterns in high-dimensional data. These technologies, combined with soil science expertise, can accurately assess the health status of soil pH and possible acidification risks, and generate abnormal deviation analysis results.

[0107] Match the data points in the abnormal deviation analysis results with their spatial coordinates, use GIS technology to locate the coordinates of each abnormal point, and obtain the abnormal point coordinate matching results;

[0108] This operation mainly involves the advanced application of GIS software. By importing the results of the abnormal deviation analysis into a layer in the GIS system and using spatial analysis tools such as spatial join query, the coordinates of each abnormal point are accurately matched. This process requires a detailed data processing flow, including data cleaning, spatial geocoding conversion and coordinate system alignment to ensure coordinate accuracy and data consistency. The effective execution of these steps depends on advanced computer vision algorithms and deep learning models, which work together to extract useful information from complex environmental data in order to better understand the spatial distribution characteristics of soil pH data and obtain the coordinate matching results of abnormal points.

[0109] Analyze the coordinate matching results of the outliers, and use the density clustering algorithm to identify outliers with sparse spatial density distribution, using the formula:

[0110]

[0111] Calculate and generate the distribution parameters of the acidification anomaly area;

[0112] Among them, D acid represents the regional distribution parameter of acidification anomaly, N j represents the number of abnormal points in the jth spatial region, α j is the weight coefficient of the jth spatial region, which is used to adjust the contribution of each region in the total acidification distribution parameters, reflecting the difference in regional environmental sensitivity or soil reactivity, γ k is the adjustment coefficient of the kth spatial region, which is used to balance the impact of the number of outliers in the differentiated region, and m represents the total number of spatial regions analyzed, which is a constant determined according to the regional division defined in the geographic information system.

[0113] formula:

[0114]

[0115] The benefit of the formula is that it combines the density distribution of spatial anomaly points with regional weight adjustment, which can more finely depict and identify the distribution of acidification anomaly areas, thereby providing more targeted data support for soil governance and management.

[0116] Detailed explanation of the formula and the process of formula calculation and derivation:

[0117] Assume there are three regions, each of which has N outliers. j They are 5, 10, and 15 respectively. Assume that the weighting coefficient α of each area j is 0.5, 1, 1.5, and the total number of spatial regions m is 3. In this way, the weighted value of the outliers in each region is first calculated, and then the formula is applied for calculation. The calculation process is as follows:

[0118] 1. Calculate the weighted anomaly points:

[0119] (0.5·5 2 )+(1·10 2 )+(1.5·15 2 )=12.5+100+337.5=450

[0120] 2. Calculate the sum of all outliers:

[0121] 5+10+15=30

[0122] 3. Apply the formula to calculate D acid :

[0123]

[0124] This result shows that when considering spatial distribution and regional weighting, the D acid A value of 0.5 reflects the distribution of acidification anomalies in the entire region. This value helps understand the severity and distribution of soil acidification problems in a specific area.

[0125] See also Figure 5 , the specific steps for obtaining acidification risk trend parameters are:

[0126] Based on the distribution parameters of the acidification abnormal area, the numerical range of the pH abnormal area is analyzed, the dynamic change rate of multiple areas is extracted, the trend change is identified by comparing with the previous data, and the numerical range and change rate analysis results are generated;

[0127] First, the previous data is compared with the current monitoring data to extract the dynamic change rate of pH in each area. In this process, advanced statistical methods are used to ensure the accuracy of the data and the reliability of the change rate. Then, through comparative analysis, those numerical areas that exceed the normal fluctuation range are determined. This analysis not only includes the calculation of the average value, but also covers the evaluation of the standard deviation and coefficient of variation to fully describe the fluctuation characteristics of pH. Through this meticulous analysis process, areas that may be at risk can be effectively identified, and numerical range and change rate analysis results can be generated. This result is of great significance for subsequent risk assessment and management.

[0128] The time distribution frequency of pH changes in multiple regions was calculated using the results of value range and change rate analysis. The correlation characteristics were analyzed in combination with spatial data. The data were processed with statistical models to obtain the time distribution and spatial correlation characteristics.

[0129] In this step, a spatial statistical model is used to process and analyze the correlation between time series data and spatial data. In addition, spatial autocorrelation and time dependence are also considered to ensure the accuracy and reliability of the analysis results, thereby obtaining the time distribution and spatial correlation characteristic results. This result provides a scientific basis for further screening of risk points and analysis of characteristic distribution.

[0130] According to the time distribution and spatial correlation characteristics, the dynamic characteristic distribution areas with risk potential are screened, and the statistical formula is used:

[0131]

[0132] Adjust sensitivity to low temporal change rates to calculate acidification risk trend parameters;

[0133] Among them, T c Indicates the characteristic value of the target area in the current time period, T avg represents the time average of the characteristic value of the target area, S std represents the standard deviation of the spatial distribution characteristics of the target area, W sp Represents the spatial weight parameter, which assigns the weight of the target area according to the spatial distribution characteristics. The assignment range is set according to the actual distribution characteristics. t It represents the time change rate parameter, that is, the change rate of the acidification associated characteristic value of the target area in adjacent time periods, which is obtained by calculating the time distribution characteristics, and R is the acidification risk trend parameter.

[0134] formula:

[0135]

[0136] The benefit of the formula is that by combining the temporal change rate and the spatial weight factor, the acidification risk trend in areas with high risk potential can be more accurately assessed. This method allows dynamic adjustment based on actual temporal data and spatial distribution characteristics, enhancing the applicability and accuracy of the prediction model.

[0137] Detailed explanation of the formula and the process of formula calculation and derivation:

[0138] First, through field monitoring or previous data collection, obtain the characteristic value T of the target area in the current time period c is 50, and the time average T avg is 45, so

[0139] |T c -T avg |=5

[0140] Assume that through statistical analysis, the standard deviation S of the spatial distribution is obtained std is 2, so

[0141]

[0142] The square root is about 1.58. The spatial weight W sp , assuming a combined score of 0.8 based on geological and environmental factors. Time rate of change V t It is calculated by comparing the data changes at consecutive time points, assuming it is 0.1, so

[0143]

[0144] Substituting these values ​​into the formula to calculate R yields

[0145] R=1.58·0.727≈1.15

[0146] The results show that the acidification risk trend in the target area is relatively stable, the risk increase is not significant, and continuous monitoring can be carried out to confirm whether there will be major changes in the trend.

[0147] See also Figure 6 , the specific steps for obtaining soil acidification hotspots are:

[0148] Based on the acidification risk trend parameters, the density value of the multi-pH area is calculated, and the spatial distance within the area is referred to capture the density distribution and generate density and distance analysis results;

[0149] Through GIS technology and advanced statistical algorithms, the calculation of density values ​​not only reflects the areas where abnormal events occur frequently, but also takes into account the relative distances between these areas in space. This process ensures that the differences and connections between regions can be accurately captured. At the same time, by comparing the density values ​​and distances between different regions, high-risk areas that require further attention can be effectively identified. This method improves the accuracy of identification and contributes to more effective resource allocation and risk management. The generated density and distance analysis results provide accurate basic data for subsequent steps.

[0150] Using the density and distance analysis results, the density peak point is extracted through a statistical algorithm to obtain the density peak point extraction result;

[0151] By applying advanced data processing techniques and algorithms, such as kernel density estimation and spatial data analysis techniques, the density peaks of each region are extracted. These peak points mark areas that may be at high risk of acidification. This step is crucial for accurately identifying potential acidification hotspots. The extraction of density peaks is not only based on a single density value, but also comprehensively considers the spatial correlation within the region and the dynamic characteristics of pH changes. Such comprehensive analysis helps to more accurately locate and predict possible hotspots, thereby providing a scientific basis for taking effective soil improvement measures.

[0152] According to the density peak point extraction results, the hotspot areas are screened and marked, and the distribution threshold is set to divide and mark the soil acidification areas. The formula is used:

[0153]

[0154] Determine the hotspot areas with higher probability of abnormal pH, calculate and generate soil acidification hotspot areas;

[0155] Among them, H acid Indicates soil acidification hotspots, D i represents the density value in the i-th evaluation area, reflecting the degree of aggregation of abnormal events in the area, S i Represents the spatial distance of the region, measuring the distance between spatial data points, T i is the frequency of the time distribution of the region, reflecting the frequency of regional activities within the monitoring window, ∈ is a small positive number used to stabilize the calculation and avoid the divisor being zero, and p represents the total number of evaluation regions, which is used to iterate all associated regions in the summation process.

[0156] formula:

[0157]

[0158] The benefit of the formula is that it comprehensively evaluates hotspot areas by combining different dimensions of density value, spatial distance and temporal frequency, which enhances the model's sensitivity to dynamic changes in different regions and its ability to accurately predict, and provides a more scientific method to determine key areas of soil acidification.

[0159] Detailed explanation of the formula and the process of formula calculation and derivation:

[0160] Assume that the density value of a specific area is 200 events / km2, the spatial distance is 50 km, the time frequency is 30 times / year, and the smoothing parameter ∈ is set to 0.01. Substitute it into the formula to calculate

[0161]

[0162] The results show that under the set parameter conditions, the soil acidification hotspot score in the region is 0.2387. This score can help decision makers identify and prioritize areas with more serious soil acidification, thereby rationally allocating resources and managing risks.

[0163] A soil acidification intelligent early warning method based on big data, the soil acidification intelligent early warning method based on big data is performed based on the above-mentioned soil acidification intelligent early warning system based on big data, and includes the following steps:

[0164] S1: Based on the soil pH, temperature and humidity, fertilizer type and application amount data, extract the time index and spatial coordinate value, match the soil pH with temperature and humidity in time, calculate the spatiotemporal mapping value of multiple data points, classify and sort them by time index and spatial coordinates, and generate soil spatiotemporal feature mapping value;

[0165] S2: Based on the soil spatiotemporal characteristic mapping value, the spatial and temporal characteristics of the pH data are decomposed, the characteristic values ​​and distribution parameters associated with the pH change are extracted, the characteristic change values ​​in the time evolution matrix are screened, the change characteristic parameters under the spatial coordinates are extracted, and the dynamic characteristic parameters of the pH are generated;

[0166] S3: Based on the dynamic characteristic parameters of pH, calculate the pH change deviation value and time index data, extract abnormal points that exceed the parameter setting range, filter the coordinate values ​​corresponding to the time index, identify outliers with sparse spatial density distribution, locate the distribution parameters of the abnormal area, and generate pH abnormal area parameters;

[0167] S4: Based on the pH abnormality area parameters, calculate the pH distribution density value and spatial distance in the area, extract the coordinate range corresponding to the density peak point, combine the pH dynamic change frequency and correlation characteristics, mark the threshold area of ​​spatial distribution, and generate acidification risk hotspot areas.

[0168] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A soil acidification intelligent early warning system based on big data, characterized in that: The system comprises: The dynamic feature mapping module obtains soil pH, temperature and humidity, fertilizer type and application amount data, extracts time index and spatial coordinate value, matches the time relationship between pH and temperature and humidity, generates a spatiotemporal mapping table and calculates dynamic correlation features to generate soil spatiotemporal feature mapping values; The potential feature capture module decomposes the spatial and temporal characteristics of the pH data based on the soil spatiotemporal feature mapping value, extracts the characteristic values ​​and distribution parameters of the associated changes, screens the characteristic change values ​​in the dynamic evolution matrix, and obtains the dynamic characteristic parameters of soil acidification; The anomaly detection module calculates the pH change deviation value and time index data based on the soil acidification dynamic characteristic parameters, extracts abnormal points that are out of range and locates the coordinate values, identifies outliers with sparse spatial density distribution, and generates acidification abnormal area distribution parameters; The acidification risk assessment module extracts the value range and dynamic change rate of the abnormal pH area based on the acidification abnormal area distribution parameters, calculates the frequency and spatial correlation characteristics of the time distribution, screens the dynamic characteristic distribution of the risk point, and obtains the acidification risk trend parameters; The hotspot area identification module calculates the pH area density value and spatial distance based on the acidification risk trend parameters, extracts density peak points and screens hotspot areas, sets distribution thresholds to mark distribution ranges, and generates soil acidification hotspot areas.

2. The soil acidification intelligent early warning system based on big data according to claim 1 is characterized in that: The soil spatiotemporal characteristic mapping value includes time index, spatial coordinate value, time relationship between pH and temperature and humidity, and dynamic correlation characteristics. The soil acidification dynamic characteristic parameters include characteristic values ​​of correlation changes, distribution parameters, and characteristic change values. The acidification abnormal area distribution parameters include pH change deviation values, time index data, abnormal point coordinate values, and outliers with sparse spatial density distribution. The acidification risk trend parameters include the numerical range of pH abnormal areas, dynamic change rate, time distribution frequency, spatial correlation characteristics, and dynamic characteristic distribution of risk points. The soil acidification hotspot area includes pH area density value, spatial distance, density peak point, distribution threshold mark, and distribution range.

3. The soil acidification intelligent early warning system based on big data according to claim 2 is characterized in that: The steps for obtaining the soil spatiotemporal characteristic mapping value are specifically as follows: Extract soil pH, temperature and humidity, fertilizer type and application amount data, match multiple parameters with time index and spatial coordinate values, generate a time series data table through the time dimension of the parameters, and obtain the matching results of time index and spatial coordinates; Based on the matching results of the time index and the spatial coordinates, the dynamic correlation between pH and temperature and humidity under multiple time indexes is calculated, and by constructing a dynamic time relationship, the change trend between the two is matched and a time dynamic feature matrix is ​​established; According to the fertilizer type and application amount parameters, the spatial coordinate values ​​are compared with the time dynamic feature matrix to form a time-space correspondence and perform feature weighting, using the formula: Calculate and generate soil spatiotemporal characteristic mapping values; Among them, M st represents the soil spatiotemporal characteristic mapping value, p i Indicates the time dynamic correlation value between pH and temperature and humidity, w ai represents the spatial weighting parameter of fertilizer type, w ti represents the spatial weight parameter of the application rate, and n represents the number of spatial coordinates.

4. The soil acidification intelligent early warning system based on big data according to claim 3 is characterized in that: The steps for obtaining the soil acidification dynamic characteristic parameters are specifically as follows: Based on the soil spatiotemporal characteristic mapping value, the spatial and temporal characteristics of the pH data are decomposed, the pH change trend of multiple locations is identified through spatial analysis, the periodic change characteristics are revealed through temporal analysis, and the spatial and temporal characteristic decomposition results are generated; Extracting characteristic values ​​significantly associated with changes in pH from the spatial-temporal characteristic decomposition results, using statistical methods to determine their contribution to the soil acidification trend, and obtaining associated change characteristic values; The associated change characteristic values ​​are used to screen the key characteristic change values ​​characterizing the soil acidification process, using the formula: The dynamic characteristic parameters of soil acidification are calculated; Among them, P acid represents the dynamic characteristic parameter of soil acidification, V ti is the characteristic change value in time period i, α i is the adjustment coefficient associated with the soil type, β is the coefficient of curve adjustment, which controls the slope of the logistic function and affects the rate of contribution of pH change to the output, μ is the midpoint offset of the curve, which determines the position of the midpoint of the logistic function and affects the judgment threshold of the soil acidification state, and k is the number of time periods evaluated.

5. The soil acidification intelligent early warning system based on big data according to claim 4 is characterized in that: The steps for obtaining the distribution parameters of the acidification abnormal area are specifically as follows: Based on the soil acidification dynamic characteristic parameters, the pH change deviation value is calculated, compared with the time index data, and the data points beyond the normal change range are identified through the set statistical threshold to generate abnormal deviation analysis results; Matching the data points in the abnormal deviation analysis results with their spatial coordinates, locating the coordinates of each abnormal point using GIS technology, and obtaining an abnormal point coordinate matching result; Analyze the coordinate matching results of the abnormal points, use the density clustering algorithm to identify outliers with sparse spatial density distribution, and use the formula: Calculate and generate the distribution parameters of the acidification anomaly area; Among them, D acid represents the regional distribution parameter of acidification anomaly, N j represents the number of abnormal points in the jth spatial region, α j is the weight coefficient of the jth spatial region, which is used to adjust the contribution of each region in the total acidification distribution parameters, reflecting the difference in regional environmental sensitivity or soil reactivity, γ k is the adjustment coefficient of the kth spatial region, which is used to balance the impact of the number of outliers in the differentiated region, and m represents the total number of spatial regions analyzed, which is a constant determined according to the regional division defined in the geographic information system.

6. The soil acidification intelligent early warning system based on big data according to claim 5 is characterized in that: The steps for obtaining the acidification risk trend parameter are specifically as follows: Based on the acidification abnormal area distribution parameters, the value range of the pH abnormal area is analyzed, the dynamic change rate of multiple areas is extracted, the trend change is identified by comparing with previous data, and the value range and change rate analysis results are generated; Using the numerical range and change rate analysis results, calculate the time distribution frequency of pH changes in multiple regions, combine spatial data to analyze the correlation characteristics, process the data with a statistical model, and obtain the time distribution and spatial correlation characteristic results; According to the time distribution and spatial correlation feature results, the dynamic feature distribution area with risk potential is screened, and the statistical formula is used: Adjust sensitivity to low temporal change rates to calculate acidification risk trend parameters; Among them, T c Indicates the characteristic value of the target area in the current time period, T avg represents the time average of the characteristic value of the target area, S std represents the standard deviation of the spatial distribution characteristics of the target area, W sp Represents the spatial weight parameter, which assigns the weight of the target area according to the spatial distribution characteristics. The assignment range is set according to the actual distribution characteristics. t It represents the time change rate parameter, that is, the change rate of the acidification associated characteristic value of the target area in adjacent time periods, which is obtained by calculating the time distribution characteristics, and R is the acidification risk trend parameter.

7. The soil acidification intelligent early warning system based on big data according to claim 6 is characterized in that: The steps for obtaining the soil acidification hotspot area are specifically as follows: Based on the acidification risk trend parameters, density values ​​of multi-pH areas are calculated, and spatial distances within the areas are referenced to capture density distribution and generate density and distance analysis results; Utilizing the density and distance analysis results, extracting density peak points through a statistical algorithm to obtain density peak point extraction results; According to the density peak point extraction results, the hotspot areas are screened and marked, and the distribution threshold is set to divide and mark the soil acidification areas, using the formula: Determine the hotspot areas with higher probability of abnormal pH, calculate and generate soil acidification hotspot areas; Among them, H acid Indicates soil acidification hotspots, D i represents the density value in the i-th evaluation area, reflecting the degree of aggregation of abnormal events in the area, S i Represents the spatial distance of the region, measuring the distance between spatial data points, T i is the frequency of the time distribution of the region, reflecting the frequency of regional activities within the monitoring window, ∈ is a small positive number used to stabilize the calculation and avoid the divisor being zero, and p represents the total number of evaluation regions, which is used to iterate all associated regions in the summation process.

8. A soil acidification intelligent early warning method based on big data, characterized in that: The soil acidification intelligent early warning system based on big data according to any one of claims 1 to 7 comprises the following steps: Based on the soil pH, temperature and humidity, fertilizer type and application amount data, extract the time index and spatial coordinate values, match the soil pH with temperature and humidity in time, calculate the spatiotemporal mapping values ​​of multiple data points, classify and sort them through time index and spatial coordinates, and generate soil spatiotemporal feature mapping values; Based on the soil spatiotemporal characteristic mapping value, the spatial characteristics and temporal characteristics of the pH data are decomposed, the characteristic values ​​and distribution parameters associated with the pH change are extracted, the characteristic change values ​​in the time evolution matrix are screened, the change characteristic parameters under the spatial coordinates are extracted, and the dynamic characteristic parameters of the pH are generated; Based on the dynamic characteristic parameters of pH, the pH change deviation value and time index data are calculated, abnormal points beyond the parameter setting range are extracted, the coordinate values ​​corresponding to the time index are screened, outliers with sparse spatial density distribution are identified, the distribution parameters of the abnormal area are located, and the pH abnormal area parameters are generated; Based on the pH abnormality area parameters, the pH distribution density value and spatial distance in the area are calculated, the coordinate range corresponding to the density peak point is extracted, and the threshold area of ​​the spatial distribution is marked in combination with the dynamic change frequency and correlation characteristics of the pH to generate the acidification risk hotspot area.

Images