CN-122018401-A - Saline-alkali soil drainage method, device, equipment and medium based on Internet of things monitoring

CN122018401ACN 122018401 ACN122018401 ACN 122018401ACN-122018401-A

Abstract

The application relates to a saline-alkali soil drainage method, device, equipment and medium based on internet of things monitoring. The method comprises the steps of obtaining multi-source monitoring data in saline-alkali soil, carrying out space-time alignment on the multi-source monitoring data to obtain space-time pair Ji Duo source monitoring data, inputting the space-time pair Ji Duo source monitoring data to a preset water-salt cooperative regulation model to obtain salt flux pointing to a drainage layer from a root layer, obtaining a physiological stress index used for representing the current salt stress risk and a dynamic regulation threshold used for representing a leaching drainage window period according to the salt flux and a preset crop salt tolerance threshold, and matching the physiological stress index, the dynamic regulation threshold and a preset cooperative regulation rule to generate a cooperative regulation instruction set comprising an irrigation trigger instruction, an irrigation leaching quota, a drainage valve opening condition and a drainage valve closing condition. The method can effectively solve the problem of time sequence disjoint and decision basis singleness of irrigation and drainage operation, and improves salt drainage efficiency and water resource utilization rate.

Inventors

ZHAO LIHUA
LI YUEXIA
Lei Hanting
WANG HANBO
LI JIN
YANG YUQING
SUN YUTONG
LI ZEJUN
Jia Zhaobin
HAO LINGXIAO

Assignees

河北省水利科学研究院(河北省大坝安全技术中心、河北省堤防水闸技术中心)
河北林业生态建设投资有限公司

Dates

Publication Date: 20260512
Application Date: 20260210

Claims (9)

1. The saline-alkali soil drainage method based on the monitoring of the Internet of things is characterized by comprising the following steps of: S1, acquiring multi-source monitoring data in saline-alkali soil, and performing space-time alignment on the multi-source monitoring data to obtain space-time pair Ji Duo source monitoring data; S2, inputting the space-time Ji Duo source monitoring data into a preset water-salt cooperative regulation model to obtain salt flux pointing to a drainage layer from a root layer; s3, according to the salinity flux and a preset crop salt tolerance threshold, a physiological stress index for representing the current salinity stress risk and a dynamic regulation threshold for representing the leaching drainage window period are obtained; s4, matching the physiological stress index, the dynamic regulation threshold value and a preset cooperative regulation rule to generate a cooperative regulation instruction set comprising an irrigation trigger instruction, an irrigation leaching quota, a drainage valve opening condition and a drainage valve closing condition; The cooperative regulation and control rule comprises a trigger and interlocking logic of irrigation leaching operation and drainage operation, wherein the cooperative regulation and control instruction set is used for starting irrigation according to the irrigation leaching quota when the irrigation trigger instruction is met, starting drainage when the drainage valve opening condition is met, and stopping drainage when the drainage valve closing condition is met.
2. The method of claim 1, wherein said performing a spatiotemporal alignment on said multisource monitoring data results in spatiotemporal pair Ji Duo source monitoring data comprising: S11, carrying out outlier cleaning on the multisource monitoring data to obtain cleaned monitoring data; s12, performing spatial interpolation processing on the cleaned monitoring data to obtain regular-grid spatial distribution data; s13, time alignment is carried out on the space distribution data, and time sequence data with uniform time steps are obtained; and S14, carrying out weight optimization fusion on the time sequence data to obtain the time-space pair Ji Duo source monitoring data.
3. The method of claim 2, wherein inputting the spatiotemporal pair Ji Duo source monitoring data into a preset water-salt cooperative regulation model to obtain a salt flux from a root zone to a drainage zone comprises: s21, inputting the space-time-to-Ji Duo-source monitoring data into the water-salt cooperative regulation model, wherein the water-salt cooperative regulation model comprises a feature extraction network and a deep learning network; s22, carrying out feature extraction on the Ji Duo source monitoring data of the space-time pair through the feature extraction network, and extracting key influence factors through principal component analysis to obtain feature data after dimension reduction; s23, predicting the dynamic change of the soil salinity from the feature data after the dimension reduction through the deep learning network to obtain a salinity prediction result; S24, calculating the salinity flux from the root layer to the drainage layer according to the salinity prediction result, Wherein, the expression of the salt flux is: In the formula, For the salt flux, the concentration of the salt, In order to obtain the diffusion coefficient of salt, Is the salt concentration of the soil solution, For the depth of the soil, Is the water flux.
4. The method according to claim 1, wherein the obtaining a physiological stress index for representing a current salt stress risk and a dynamic regulation threshold for representing a rinse drainage window period according to the salt flux and a preset crop salt tolerance threshold comprises: s31, establishing a salinity stress response function based on the crop growth stage and the soil salinity content, and calculating to obtain the physiological stress index through the following formula: In the formula, As an index of physiological stress, In order for the stress response coefficient to be a function of the stress, As the salt content of the current soil, Is the concentration of semi-lethal salt; S32, obtaining a leaching demand coefficient according to the salt reserve of the root layer soil and the crop salt tolerance threshold value; S33, establishing a dynamic threshold model based on the salt change rate and the leaching demand coefficient, and calculating to obtain the dynamic regulation threshold through a self-adaptive algorithm.
5. The method of claim 4, wherein the matching the physiological stress indicator, the dynamic regulation threshold, and a preset cooperative regulation rule to generate a set of cooperative regulation instructions including an irrigation trigger instruction, an irrigation rinse quota, a drain valve opening condition, and a drain valve closing condition comprises: s41, calculating irrigation starting membership and drainage starting membership based on the physiological stress index and the dynamic regulation threshold to obtain an initial action decision vector; S42, inputting the initial action decision vector into a preset rule conflict resolution model, and carrying out arbitration and correction on the conflicting decisions based on the current soil water content, weather forecast data and historical action efficiency evaluation results to obtain an optimized action decision; S43, according to the certainty of irrigation starting in the optimized action decision, combining the leaching demand coefficient and real-time soil moisture data, and calculating the irrigation leaching quota through a water balance model; And S44, dynamically generating logic judgment expressions of the drainage valve opening condition and the drainage valve closing condition based on the certainty degree of drainage starting in the optimized action decision, the salinity flux direction and the drainage layer saturation condition, and obtaining the cooperative regulation instruction set.
6. The method of claim 3, wherein predicting, by the deep learning network, the dynamic change of the soil salinity for the feature data after the dimension reduction to obtain a salinity prediction result comprises: s51, inputting the feature data subjected to dimension reduction to the deep learning network, wherein the deep learning network is a space-time attention cycle neural network; S52, in the space-time attention cycle neural network, a time attention mechanism distributes weights to the characteristics of the feature data subjected to dimension reduction in different time steps to obtain a first distribution weight; S53, processing the feature data after dimension reduction through a circulation layer of the space-time attention circulation neural network based on the first distribution weight and the second distribution weight, capturing a space-time dependency relationship of soil water salt migration, and outputting a prediction sequence of soil solution salt concentration of a plurality of time steps in the future; s54, according to the predicted soil solution salinity concentration sequence and the synchronously predicted soil water content sequence, calculating to obtain the predicted root layer soil salinity storage amount as the salinity prediction result.
7. Saline-alkali soil drainage device based on thing networking monitoring, its characterized in that, the device includes: The multi-source data acquisition module is used for acquiring multi-source monitoring data in the saline-alkali soil, and carrying out space-time alignment on the multi-source monitoring data to obtain space-time pair Ji Duo source monitoring data; the water and salt analysis module is used for inputting the time-space Ji Duo source monitoring data into a preset water and salt cooperative regulation model to obtain salt flux pointing to the drainage layer from the root layer; the salt tolerance analysis and drainage threshold analysis module is used for obtaining a physiological stress index used for representing the current salt stress risk and a dynamic regulation threshold used for representing the leaching drainage window period according to the salt flux and a preset crop salt tolerance threshold; The drainage cooperative regulation and control instruction generation module is used for matching the physiological stress index, the dynamic regulation and control threshold value and a preset cooperative regulation and control rule to generate a cooperative regulation and control instruction set comprising an irrigation trigger instruction, an irrigation leaching quota, a drainage valve opening condition and a drainage valve closing condition; The cooperative regulation and control rule comprises a trigger and interlocking logic of irrigation leaching operation and drainage operation, wherein the cooperative regulation and control instruction set is used for starting irrigation according to the irrigation leaching quota when the irrigation trigger instruction is met, starting drainage when the drainage valve opening condition is met, and stopping drainage when the drainage valve closing condition is met.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the method of any one of claims 1 to 6 when executing the computer program.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of any one of claims 1 to 6.

Description

Saline-alkali soil drainage method, device, equipment and medium based on Internet of things monitoring Technical Field The invention belongs to the technical field of agricultural drainage, and particularly relates to a saline-alkali soil drainage method, device, equipment and medium based on internet of things monitoring. Background With the development of accurate agriculture and intelligent water conservancy technologies, the internet of things monitoring has become an important means for modern agriculture resource management. In the field of saline-alkali soil improvement, a sensor network is utilized to acquire soil information, and salt regulation and control are carried out by combining drainage measures, so that the method is a current main research direction. In the prior art, saline-alkali soil drainage mainly depends on engineering measures such as concealed pipe drainage, open trench drainage and the like, and salt leaching is carried out by combining periodic irrigation. In actual operation, irrigation and drainage are often managed as two independent agronomic links, namely, irrigation is usually carried out firstly to rinse salt in root areas, and then a drainage system is started according to a fixed time schedule or experience judgment. This mode is controlled depending on a preset program or a threshold value of a single soil index (e.g., moisture content). However, the current treatment methods have significant problems. The irrigation and drainage lack of dynamic coordination based on the real-time salt migration state in time sequence, so that the drainage opportunity possibly misses an optimal window for leaching salt to a drainage layer, the salt drainage efficiency is low and water resources are wasted, and the differential accurate regulation and control are difficult to implement according to the soil salt heterogeneity in space. Essentially, the existing method can not realize linkage intelligent decision of the water-salt conjugation process, and the problem of irrigation, drainage and dislocation restricts the efficient utilization of water and soil resources of the saline-alkali soil. Disclosure of Invention Based on the above, it is necessary to provide a saline-alkali soil drainage method, device, equipment and medium based on internet of things monitoring, aiming at the technical problems. In a first aspect, the application provides a saline-alkali soil drainage method based on internet of things monitoring, which comprises the following steps: S1, acquiring multi-source monitoring data in saline-alkali soil, and performing space-time alignment on the multi-source monitoring data to obtain space-time pair Ji Duo source monitoring data; s2, inputting the space-time-to-Ji Duo-source monitoring data into a preset water-salt cooperative regulation model to obtain salt flux pointing to a drainage layer from a root layer; s3, according to the salinity flux and a preset crop salt tolerance threshold, a physiological stress index for representing the current salinity stress risk and a dynamic regulation threshold for representing the leaching drainage window period are obtained; s4, matching the physiological stress index, the dynamic regulation threshold value and a preset cooperative regulation rule to generate a cooperative regulation instruction set comprising an irrigation trigger instruction, an irrigation leaching quota, a drainage valve opening condition and a drainage valve closing condition; the cooperative regulation and control instruction set is used for starting irrigation according to the irrigation leaching quota when the irrigation trigger instruction is met, starting drainage when the drainage valve opening condition is met, and stopping drainage until the drainage valve closing condition is reached. In one embodiment, performing space-time alignment on the multi-source monitoring data to obtain space-time pair Ji Duo source monitoring data includes: s11, carrying out outlier cleaning on the multisource monitoring data to obtain cleaned monitoring data; s12, performing spatial interpolation processing on the cleaned monitoring data to obtain regular-grid spatial distribution data; S13, time alignment is carried out on the space distribution data to obtain time sequence data with uniform time step length; and S14, carrying out weight optimization fusion on the time sequence data to obtain the time-space pair Ji Duo source monitoring data. In one embodiment, the method for inputting the space-time-to-Ji Duo-source monitoring data into a preset water-salt cooperative regulation model to obtain the salt flux from the root layer to the drainage layer comprises the following steps: s21, inputting the space-time pair Ji Duo source monitoring data into a water-salt cooperative regulation model, wherein the water-salt cooperative regulation model comprises a feature extraction network and a deep learning network; S22, carrying out feature extraction on Ji Duo source monit