A saline-alkali land management early warning method

Abstract

The present invention provides a saline-alkali land governance early warning method, which relates to the field of saline-alkali land governance early warning. The saline-alkali land data of the successfully improved saline-alkali land are collected during the improvement period of the successfully improved saline-alkali land; the historical saline-alkali land data of the successfully improved saline-alkali land are used to fill the missing data; a standard curve is established according to the completed saline-alkali land data of the successfully improved saline-alkali land; the standard curve is corrected according to the difference in the improvement influencing factors to obtain the governance target curve; the remote sensing data of the saline-alkali land to be evaluated is collected in real time to obtain the saline-alkali land data of the saline-alkali land to be evaluated; the saline-alkali land data of the saline-alkali land to be evaluated is subtracted from the governance target curve; if the difference is within a preset range, the governance of the saline-alkali land to be evaluated does not require an early warning; if the difference exceeds the preset range, a BP neural network is established to predict the saline-alkali land data at the next collection moment of the saline-alkali land to be evaluated and to conduct governance early warning. The present invention realizes accurate and timely early warning of the saline-alkali land governance process.

Description

Technical Field

[0001] The invention relates to the field of early warning for saline-alkali land management, and in particular to an early warning method for saline-alkali land management. Background Art

[0002] Saline-alkali land is not conducive to the growth of crops due to its high soil salinity and high alkaline content. With the development of industrialization and the destruction of arable land, the area of ​​saline-alkali land in the world is increasing year by year. The management of saline-alkali land is receiving more and more attention.

[0003] At present, the evaluation of the effectiveness of saline-alkali land management has been applied in practice, but there is a lack of early warning during the management process, and it is impossible to provide early warning for areas where management is not ideal and provide timely intervention. Summary of the invention

[0004] The purpose of the present invention is to propose a saline-alkali land management early warning method, aiming to solve the problem of lack of early warning of undesirable management in the saline-alkali land management process.

[0005] The present invention provides a saline-alkali land management early warning method, comprising the following steps:

[0006] Collecting saline-alkali land data of successfully improved saline-alkali land at a preset period during the improvement period of the successfully improved saline-alkali land, wherein the saline-alkali land data includes vegetation coverage, plant growth conditions and soil data;

[0007] Determine whether the saline-alkali land data of successfully improved saline-alkali land is missing. If missing, use the historical saline-alkali land data of successfully improved saline-alkali land to complete the missing data;

[0008] A standard curve was established based on the completed saline-alkali land data of successfully improved saline-alkali land;

[0009] The standard curve is corrected according to the difference in the improvement factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land, and the corrected standard curve is used as the control target curve;

[0010] Collect remote sensing data of the saline-alkali land to be evaluated in real time according to a preset period, pre-process the collected remote sensing data, and obtain saline-alkali land data of the saline-alkali land to be evaluated;

[0011] Make a difference between the saline-alkali land data of the saline-alkali land to be evaluated and the saline-alkali land data at the corresponding time point of the treatment target curve;

[0012] If the difference is within the preset range, no early warning is required for the management of the saline-alkali land to be evaluated;

[0013] If the difference exceeds the preset range, a BP neural network is established, and the saline-alkali land data of the successfully improved saline-alkali land are corrected according to the difference in improvement influencing factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land, to obtain the corrected saline-alkali land data of the successfully improved saline-alkali land, and data are selected from the corrected saline-alkali land data of the successfully improved saline-alkali land to form a training set and a test set, and the BP neural network is trained and tested to obtain the tested BP neural network, and the saline-alkali land data of the saline-alkali land to be evaluated are input to predict the saline-alkali land data at the next collection time of the saline-alkali land to be evaluated. If the difference between the predicted saline-alkali land data and the corrected saline-alkali land data of the successfully improved saline-alkali land at the corresponding time still exceeds the preset range, an early warning for the management of the saline-alkali land to be evaluated is issued.

[0014] Preferably, if there are missing saline-alkali land data for successfully improved saline-alkali land, the saline-alkali land data within the preset periods are fitted by working backwards from the time corresponding to the missing data to obtain a fitting curve, and the data at the time corresponding to the missing data is predicted based on the fitting curve to complete the missing data.

[0015] Preferably, correcting the standard curve according to the difference in improvement influencing factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land specifically includes: comparing each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land, and correcting the standard curve based on the comparison results of each indicator.

[0016] Preferably, each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land is compared respectively, and the standard curve is corrected based on the comparison results of each indicator. Specifically, the values ​​of each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land are compared with the values ​​of the corresponding indicators of the non-salty-alkali land suitable for vegetation growth, and then the comparison results of each indicator are multiplied to obtain the improvement difference factor δ, and the improvement difference factor is multiplied by the standard curve to obtain the control target curve, which is calculated specifically using the following formula:

[0017]

[0018] in,

[0019] X i is the value of the i-th indicator in the factors affecting saline-alkali land improvement to be evaluated;

[0020] B i is the value of the corresponding index of item i for non-saline-alkali land suitable for vegetation growth;

[0021] Y i is the value of the ith indicator in the factors affecting the successful improvement of saline-alkali land;

[0022] n is the total number of indicators in the improved impact factor.

[0023] Preferably, the improvement influencing factors include initial soil salinity, initial vegetation coverage, climatic conditions, groundwater level, soil moisture and soil thickness.

[0024] Preferably, the climate conditions include temperature, humidity, sunshine intensity, percentage of sunny days, precipitation and wind speed.

[0025] Preferably, the method further includes obtaining an electronic map of the saline-alkali land to be evaluated, dynamically dividing the saline-alkali land to be evaluated into regions, and respectively obtaining saline-alkali land data of each region.

[0026] Preferably, multiple detection points are set up in each area, and the locations of the detection points where the saline-alkali land data to be evaluated obtained in the same area exceeds a preset range are removed from the area, and the area is divided into a virtual area. The number of deployed detection points is increased in the virtual area, and the virtual area is used as a key focus area for subsequent governance and early warning.

[0027] Preferably, each detection point collects a group of saline-alkali land data of the saline-alkali land to be evaluated. After the virtual area is divided, the saline-alkali land data in each area or virtual area are averaged to obtain the saline-alkali land data of the area or virtual area, and the saline-alkali land data of each area or virtual area are compared with the treatment target curve. According to the comparison result, it is determined whether to establish a BP neural network for the area or virtual area to predict the saline-alkali land data.

[0028] Preferably, it also includes adjusting improvement measures for the saline-alkali land to be evaluated for which a governance warning is given, and determining the improvement impact factor based on the time after the adjustment of the improvement measures, and continuing to judge whether a governance warning is needed.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] (1) The present invention adopts a method of fitting the saline-alkali land data within the preset period by working backward from the current time for the missing data of the successfully improved saline-alkali land to obtain a fitting curve, and predicts the data at the time corresponding to the missing data based on the fitting curve, and fills in the missing data, so that the final control target curve is more accurate, control early warning can be achieved, and control early warning is also more accurate and timely.

[0031] (2) The present invention first makes a difference between the saline-alkali land data of the saline-alkali land to be evaluated collected in real time and the data in the control target curve at the corresponding time point, and makes the first step of judging whether to issue an early warning based on the difference. Only when the difference exceeds a preset range, a BP neural network is established to predict the saline-alkali land data of the saline-alkali land to be evaluated at the next collection time. A control prediction is made based on the prediction result, and a control early warning can be realized, which not only simplifies the early warning process, but also makes the early warning more accurate.

[0032] (3) The present invention first performs preliminary regional division on the saline-alkali land to be evaluated, removes the original area for the detection points whose saline-alkali land data of the saline-alkali land to be evaluated in the same area exceeds the preset range, constructs a new virtual area, calculates the data average value for each area and the virtual area respectively, and takes the virtual area as the focus of later attention, thereby achieving more reasonable and accurate regional division of saline-alkali land, achieving governance early warning, and making the early warning of each area more accurate;

[0033] (4) When the present invention corrects the standard curve, the initial soil salinity, initial vegetation coverage, climatic conditions, groundwater level, soil moisture and soil thickness are taken into consideration, and corresponding indicators of non-salt-alkali land suitable for vegetation growth are set. The difference factor is obtained by multiplying the comparison results of each indicator in the improvement influencing factors of the successfully improved saline-alkali land, each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the corresponding indicator of the non-salt-alkali land suitable for vegetation growth. The standard curve is corrected using the difference factor to obtain the control target curve, so that the reference data is more accurate and early warning can be carried out in a timely and accurate manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0035] Figure 1 The present invention provides a saline-alkali land management early warning method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] The present invention provides a saline-alkali land management early warning method, comprising the following steps:

[0038] Collecting saline-alkali land data of successfully improved saline-alkali land at a preset period during the improvement period of the successfully improved saline-alkali land, wherein the saline-alkali land data includes vegetation coverage, plant growth conditions and soil data;

[0039] Determine whether the saline-alkali land data of successfully improved saline-alkali land is missing. If missing, use the historical saline-alkali land data of successfully improved saline-alkali land to complete the missing data;

[0040] A standard curve was established based on the completed saline-alkali land data of successfully improved saline-alkali land;

[0041] The standard curve is corrected according to the difference in the improvement factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land, and the corrected standard curve is used as the control target curve;

[0042] Collect remote sensing data of the saline-alkali land to be evaluated in real time according to a preset period, pre-process the collected remote sensing data, and obtain saline-alkali land data of the saline-alkali land to be evaluated;

[0043] Make a difference between the saline-alkali land data of the saline-alkali land to be evaluated and the saline-alkali land data at the corresponding time point of the treatment target curve;

[0044] If the difference is within the preset range, no early warning is required for the management of the saline-alkali land to be evaluated;

[0045] If the difference exceeds the preset range, a BP neural network is established, and the saline-alkali land data of the successfully improved saline-alkali land are corrected according to the difference in improvement influencing factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land, to obtain the corrected saline-alkali land data of the successfully improved saline-alkali land, and data are selected from the corrected saline-alkali land data of the successfully improved saline-alkali land to form a training set and a test set, and the BP neural network is trained and tested to obtain the tested BP neural network, and the saline-alkali land data of the saline-alkali land to be evaluated are input to predict the saline-alkali land data at the next collection time of the saline-alkali land to be evaluated. If the difference between the predicted saline-alkali land data and the corrected saline-alkali land data of the successfully improved saline-alkali land at the corresponding time still exceeds the preset range, an early warning for the management of the saline-alkali land to be evaluated is issued.

[0046] According to a specific embodiment of the present invention, if there are missing saline-alkali land data for successfully improved saline-alkali land, the saline-alkali land data within the preset periods are fitted from the time corresponding to the missing data to obtain a fitting curve, and the data at the time corresponding to the missing data is predicted based on the fitting curve to complete the missing data.

[0047] According to a specific embodiment of the present invention, correcting the standard curve according to the difference in improvement influencing factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land specifically includes: comparing each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land, and correcting the standard curve based on the comparison results of each indicator.

[0048] According to a specific implementation scheme of the present invention, each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land is compared respectively, and the standard curve is corrected by comprehensively comparing the comparison results of each indicator. Specifically, the values ​​of each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land are compared with the values ​​of the corresponding indicators of the non-saline-alkali land suitable for vegetation growth, and then the comparison results of each indicator are multiplied to obtain the improvement difference factor δ, and the improvement difference factor is multiplied by the standard curve to obtain the control target curve, which is specifically calculated using the following formula:

[0049]

[0050] in,

[0051] X i is the value of the i-th indicator in the factors affecting saline-alkali land improvement to be evaluated;

[0052] B i is the value of the corresponding index of item i for non-saline-alkali land suitable for vegetation growth;

[0053] Y i is the value of the ith indicator in the factors affecting the successful improvement of saline-alkali land;

[0054] n is the total number of indicators in the improved impact factor.

[0055] According to a specific embodiment of the present invention, the improvement influencing factors include initial soil salinity, initial vegetation coverage, climatic conditions, groundwater level, soil moisture and soil thickness.

[0056] According to a specific embodiment of the present invention, the climate conditions include temperature, humidity, sunshine intensity, percentage of sunny days, precipitation and wind speed.

[0057] According to a specific embodiment of the present invention, it also includes obtaining an electronic map of the saline-alkali land to be evaluated, dynamically dividing the saline-alkali land to be evaluated into regions, and respectively obtaining saline-alkali land data of each region.

[0058] According to a specific implementation scheme of the present invention, multiple detection points are set in each area, and the locations of the detection points where the saline-alkali land data to be evaluated obtained in the same area exceeds a preset range are removed from the area, and the area is divided into a virtual area. The number of deployed detection points in the virtual area is increased, and the virtual area is used as a key focus area for subsequent governance and early warning.

[0059] According to a specific embodiment of the present invention, each detection point collects a group of saline-alkali land data of the saline-alkali land to be evaluated. After the virtual area is divided, the saline-alkali land data in each area or the virtual area are averaged to obtain the saline-alkali land data of the area or the virtual area, and the saline-alkali land data of each area or the virtual area are compared with the control target curve. According to the comparison result, it is determined whether to establish a BP neural network for the area or the virtual area to predict the saline-alkali land data.

[0060] According to a specific embodiment of the present invention, it also includes adjusting improvement measures for the saline-alkali land to be evaluated for which a governance warning is given, and determining the improvement impact factor based on the time after the adjustment of the improvement measures, and continuing to judge whether a governance warning is needed.

[0061] According to a specific implementation scheme of the present invention, the remote sensing data preprocessing process mainly includes the following steps:

[0062] Data import: Import remote sensing data into the processing software. This can include different types of remote sensing data such as multispectral images, hyperspectral images, radar data, etc.

[0063] Atmospheric correction: Atmospheric correction is necessary for data in the visible and near-infrared bands to remove atmospheric scattering and absorption effects, making the data more accurate and comparable.

[0064] Radiation correction: Convert raw remote sensing data into reflectance or radiance to eliminate radiation differences between data at different times and locations.

[0065] Geometric correction: Aligning remote sensing data to a geographic coordinate system for geospatial analysis. This includes steps such as image registration, selection of ground control points, and projection conversion.

[0066] Denoising: Remove noise and clutter from remote sensing data to improve image quality and accuracy. Common denoising methods include filtering, wavelet transform, etc.

[0067] Data clipping and subset extraction: According to research needs, remote sensing data can be clipped and subset extracted to reduce data complexity.

[0068] Image enhancement: Improve the visual effect and information extraction ability of remote sensing images. Common enhancement methods include histogram equalization, filtering, band combination, etc.

[0069] Data verification and validation: Perform quality checks and validation on preprocessed data to ensure data accuracy and reliability.

[0070] The above descriptions are only optional embodiments of the present invention, and are not intended to limit the patent scope of the present invention. All equivalent structural changes made using the contents of the present invention's specification and drawings, or directly / indirectly applied in other related technical fields, are included in the patent protection scope of the present invention.

Claims

1. A saline-alkali land management early warning method, characterized in that: The steps include: Collecting saline-alkali land data of successfully improved saline-alkali land at a preset period during the improvement period of the successfully improved saline-alkali land, wherein the saline-alkali land data includes vegetation coverage, plant growth conditions and soil data; Determine whether the saline-alkali land data of successfully improved saline-alkali land is missing. If missing, use the historical saline-alkali land data of successfully improved saline-alkali land to complete the missing data; A standard curve was established based on the completed saline-alkali land data of successfully improved saline-alkali land; The standard curve is corrected according to the difference in the improvement factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land, and the corrected standard curve is used as the control target curve; Collect remote sensing data of the saline-alkali land to be evaluated in real time according to a preset period, pre-process the collected remote sensing data, and obtain saline-alkali land data of the saline-alkali land to be evaluated; Make a difference between the saline-alkali land data of the saline-alkali land to be evaluated and the saline-alkali land data at the corresponding time point of the treatment target curve; If the difference is within the preset range, no early warning is required for the management of the saline-alkali land to be evaluated; If the difference exceeds the preset range, a BP neural network is established, and the saline-alkali land data of the successfully improved saline-alkali land are corrected according to the difference in the improvement influencing factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land, to obtain the corrected saline-alkali land data of the successfully improved saline-alkali land, and data are selected from the corrected saline-alkali land data of the successfully improved saline-alkali land to form a training set and a test set, and the BP neural network is trained and tested to obtain the tested BP neural network, and the saline-alkali land data of the saline-alkali land to be evaluated are input to predict the saline-alkali land data at the next collection time of the saline-alkali land to be evaluated. If the difference between the predicted saline-alkali land data and the corrected saline-alkali land data of the successfully improved saline-alkali land at the corresponding time still exceeds the preset range, an early warning for the management of the saline-alkali land to be evaluated is issued; Among them, the standard curve is corrected according to the difference in the improvement influencing factors between the saline-alkali land to be evaluated and the successfully improved saline-alkali land, specifically including: comparing each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land, and correcting the standard curve based on the comparison results of each indicator, specifically: The values ​​of each indicator in the improvement influencing factors of the saline-alkali land to be evaluated and the successfully improved saline-alkali land are compared with the values ​​of the corresponding indicators of the non-saline-alkali land suitable for vegetation growth, and then the comparison results of each indicator are multiplied to obtain the improvement difference factor δ, and the improvement difference factor δ is multiplied by the standard curve to obtain the control target curve, which is calculated using the following formula: in, X i is the value of the i-th indicator in the factors affecting saline-alkali land improvement to be evaluated; B i is the value of the corresponding index of item i for non-saline-alkali land suitable for vegetation growth; Y i is the value of the ith indicator in the factors affecting the successful improvement of saline-alkali land; n is the total number of indicators in the improved impact factor.

2. The early warning method for saline-alkali land management according to claim 1, characterized in that: If there are missing saline-alkali land data for successfully improved saline-alkali land, the preset periods are counted back from the time corresponding to the missing data, and the saline-alkali land data within the preset periods are fitted to obtain a fitting curve. The data at the time corresponding to the missing data is predicted based on the fitting curve to complete the missing data.

3. The early warning method for saline-alkali land management according to any one of claims 1-2, characterized in that: Improvement influencing factors include initial soil salinity, initial vegetation coverage, climatic conditions, groundwater level, soil moisture and soil thickness.

4. The early warning method for saline-alkali land management according to claim 3 is characterized in that: Climate conditions include temperature, humidity, sunshine intensity, percentage of sunny days, precipitation and wind speed.

5. The early warning method for saline-alkali land management according to claim 1 is characterized in that: It also includes obtaining an electronic map of the saline-alkali land to be evaluated, dynamically dividing the saline-alkali land to be evaluated into regions, and obtaining saline-alkali land data for each region respectively.

6. The early warning method for saline-alkali land management according to claim 5, characterized in that: Multiple detection points are set up in each area. The locations of the detection points where the saline-alkali land data to be assessed in the same area exceeds the preset range are removed from the area, and the area is divided into a virtual area. The number of deployed detection points is increased in the virtual area, and the virtual area is used as a key focus area for subsequent governance and early warning.

7. The early warning method for saline-alkali land management according to claim 6, characterized in that: Each detection point collects a group of saline-alkali land data of the saline-alkali land to be evaluated. After the virtual area is divided, the saline-alkali land data in each area or virtual area are averaged to obtain the saline-alkali land data of the area or virtual area, and the saline-alkali land data of each area or virtual area are compared with the governance target curve. According to the comparison result, it is determined whether to establish a BP neural network for the area or virtual area to predict the saline-alkali land data.

8. The early warning method for saline-alkali land management according to claim 1, characterized in that: It also includes adjusting improvement measures for the saline-alkali land to be evaluated for which a governance warning has been given, and determining the improvement influencing factors based on the time after the adjustment of the improvement measures, and continuing to judge whether a governance warning is needed.

Images