Search

CN-121998295-A - Intelligent deployment system of intelligent farmland Internet of things equipment

CN121998295ACN 121998295 ACN121998295 ACN 121998295ACN-121998295-A

Abstract

The invention relates to the field of intelligent deployment of farmland equipment, in particular to an intelligent deployment system of intelligent farmland Internet of things equipment, which comprises a farmland environment and crop coupling module; the system comprises a crop and environment mapping relation establishment module, a monitoring module, an area and equipment preprocessing module, an optimal equipment layout scheme searching module, a deployment module and an executable equipment deployment task list, wherein the monitoring module is used for quantitatively evaluating the sensitivity degree of each area of a farmland to environment change, the area and equipment preprocessing module is used for introducing an optimal searching mechanism based on a RANSAC algorithm to search for the optimal equipment layout scheme, and the optimal equipment layout scheme is converted into the executable equipment deployment task list. According to the invention, through optimizing the sampling and iterative logic of the RANSAC algorithm and combining the accurate quantization of the device vector field on the perceived coverage, the algorithm convergence speed is accelerated, and the output layout scheme is ensured to achieve the optimal balance between the coverage effect and the resource allocation.

Inventors

  • ZHAO JUAN
  • RONG XIONG
  • LUO CHEN

Assignees

  • 内蒙古中孚明丰农业科技有限公司

Dates

Publication Date
20260508
Application Date
20251218

Claims (10)

  1. 1. An intelligent deployment system for intelligent farmland internet of things equipment, the system comprising: the system comprises a farmland environment and crop coupling module, a mapping relation establishing module, a crop environment and crop environment mapping relation establishing module; The monitoring module is used for quantitatively evaluating the sensitivity degree of each area of the farmland to environmental changes, generating microclimate sensitivity indexes and identifying sensitive areas with different grades; The system comprises a region preprocessing module, a device preprocessing module and a device layout processing module, wherein the region preprocessing module is used for introducing an optimized search mechanism based on a RANSAC algorithm, so that the micro climate sensitive index of each region is introduced into each sampling and iteration of the RANSAC algorithm, a vector field is established for each device, and an optimal device layout scheme is searched through repeated iteration of the RANSAC algorithm; And converting the optimal equipment layout scheme into an executable equipment deployment task list, collecting actual data of system operation after equipment is online, and comparing the established mapping relation to quantify the actual effect of the deployment scheme.
  2. 2. The intelligent deployment system of the intelligent farmland internet of things equipment according to claim 1, wherein the farmland environment and crop coupling module obtains historical observation data of farmland local temperature, humidity and illumination and yield data corresponding to each observation data through a third party, and obtains a yield sequence corresponding to the observation data and the yield data; calculating the linear trend of the yield sequence to obtain trend values, and subtracting the trend values from each yield data value in the yield sequence; Calculating the average value of each observation data and trending output, calculating the dispersion, covariance and standard deviation of each observation data, and dividing the covariance of each observation data by the deposition of two standard deviations to obtain a correlation coefficient for mapping the relationship between crops and the environment.
  3. 3. The intelligent deployment system of the intelligent farmland internet of things equipment according to claim 2, wherein the monitoring module divides the farmland into a plurality of grid cells according to equal distance, and extracts a temperature sequence, a humidity sequence and an illumination intensity sequence of the grid cells corresponding to each region of the farmland for at least seven days, specifically comprising: performing spatial interpolation on satellite data acquired by a third-party platform to generate a daily average temperature sequence of each grid unit; the temperature and humidity sensors distributed in each grid cell are directly mapped to the corresponding grid cell to obtain a humidity sequence; the light intensity sensors distributed in each grid cell are directly mapped to the corresponding grid cell to obtain an illumination sequence; The monitoring module calculates the probability density of each grid unit temperature sequence, then takes the negative logarithm sum to obtain temperature information entropy, calculates the average value of the absolute value of the difference value of the humidity value of each grid unit humidity sequence and the humidity value of the adjacent grid to obtain humidity space gradient, and calculates the difference value of the maximum value and the minimum value of the illumination intensity sequence of each grid unit to obtain illumination day-night difference value; Integrating temperature information entropy, humidity space gradient and illumination day and night difference values in a self-adaptive weighting mode, wherein a weight coefficient is determined by a mapping relation between crops and the environment, and a microclimate sensitivity index is generated; After microclimate sensitivity indexes of grid cells corresponding to all areas of a farmland are obtained, the microclimate sensitivity index of each grid cell is normalized to be in a range of 0 and 1, the grid cells with microclimate sensitivity indexes exceeding a first-level preset threshold value are used as sensitive areas, the grid cells with microclimate sensitivity indexes exceeding a second-level preset threshold value are used as middle sensitive areas, and the grid cells with microclimate sensitivity indexes smaller than the second-level preset threshold value are used as low sensitive areas.
  4. 4. The intelligent deployment system of intelligent farmland internet of things equipment according to claim 1, wherein when the area and equipment preprocessing module randomly selects a plurality of points from a grid unit as sampling operations of initial samples based on a RANSAC algorithm, an optimized search mechanism is introduced, and a sensitive area for key monitoring is preferentially selected, and the intelligent deployment system comprises: Assigning an initial priority value to each grid cell, the initial priority value being positively correlated with the microclimate sensitivity index of the corresponding grid cell; Each grid unit acquires selection probability according to the initial priority value, wherein the selection probability is determined by the proportion of the initial priority value of a single grid to the sum of the initial priority values of all grid units; And the high sensitive area, the medium sensitive area and the low sensitive area are classified and divided, and the sampling proportion is distributed.
  5. 5. The intelligent deployment system of intelligent farmland internet of things equipment according to claim 4, wherein after the RANSAC algorithm sampling is completed, iterating comprises: completing primary sample sampling of a RANSAC algorithm according to the determined selection probability and sampling proportion, wherein in the primary and above sample iteration of the RANSAC algorithm, the equipment layout corresponding to the grid cells can cover the grid cell area, the equipment layout is judged to be an interior point, the RANSAC algorithm increases the added value for the initial priority value of the grid cells corresponding to the interior point, and the added value is a fixed value; In each iteration, grid cells are continuously unselected, the grid cells are judged to be outer points, the RANSAC algorithm is used for superposing attenuation values for initial priority values of the grid cells corresponding to the outer points, the attenuation values are in a fixed proportion, in the twice or more sampling of the RANSAC algorithm, the selection probability is recalculated by the sum of the initial priority values of the grid cells corresponding to the inner points and the added values, the selection probability is recalculated by the product of the initial priority values of the grid cells corresponding to the outer points and the attenuation values, and the microclimate sensitivity indexes of all areas are introduced in each sampling and iteration of the RANSAC algorithm.
  6. 6. The intelligent deployment system of the intelligent farmland internet of things equipment according to claim 5, wherein the area and equipment preprocessing module integrates and establishes the perception coverage area of each equipment as a perception behavior field, and the intelligent deployment system comprises the steps of establishing a coordinate system consistent with equipment coordinate allocation, allocating unique coordinates to grid units established by the monitoring module, and constructing a vector field of a single equipment according to a vector direction and a vector modular length, wherein the vector direction is the perception coverage area of the single equipment, the vector modular length is the perception intensity of the single equipment in the corresponding grid units, and completing the vector field establishment of each equipment; When the grid unit is covered by a plurality of devices of the same type, the method further comprises the following steps that the vector direction is the vector direction of the device with the maximum perceived intensity, and the vector module length takes the arithmetic sum of the perceived intensities of all the devices in the grid unit; When the grid unit is covered by different types of equipment, the method further comprises the following steps that the vector module length is formed by superposing the perceived intensities of a plurality of pieces of equipment, and the vector direction takes the vector direction of the equipment with the maximum perceived intensity; When the vector modulo length of the single device and/or multiple devices within the grid cell exceeds the grid cell boundary, it is determined that the device layout covers the grid cell area.
  7. 7. The intelligent deployment system of intelligent farmland internet of things equipment according to claim 6, wherein after obtaining the vector field of each equipment, inputting the vector field into RANSAC algorithm, and performing layout scheme simulation on interior points obtained in the iterative process, wherein the intelligent deployment system comprises: judging whether the sensing range of the equipment can cover the target grid unit according to the vector direction of the equipment; calculating the vector modular length of the equipment, and judging whether the vector modular length exceeds the boundary of the grid unit; After integrating the vector field according to the multi-equipment coverage rule, verifying again whether the integrated vector direction and the integrated module length meet the coverage requirement; And after the verification is passed, judging the interior points as effective interior points, and after the verification is failed, judging the interior points as undetermined interior points, and recalculating the selection probability by using the initial priority values of the grid units corresponding to the interior points, so as to carry out subsequent iteration.
  8. 8. The intelligent deployment system of the intelligent farmland internet of things equipment according to claim 7, wherein after iteration of the RANSAC algorithm is completed, the area and equipment preprocessing module establishes a consistent set with all obtained effective interior points, when the number of the effective interior points in the consistent set is greater than or equal to a limiting threshold value, a deployment model output by the RANSAC after the iteration is completed is determined to meet basic monitoring requirements, when the number of the effective interior points in the consistent set is less than the limiting threshold value, the RANSAC resamples and iterates, and finally the deployment module outputs the consistent set containing all the effective interior points as an optimal equipment layout scheme.
  9. 9. The intelligent deployment system of the intelligent farmland internet of things equipment according to claim 8, wherein the deployment module simulates according to a layout scheme of all interior points in a consistent set, and unique coordinates distributed by the area and equipment preprocessing module for grid units established by the monitoring module form an executable equipment deployment task row, each task in the equipment deployment task row corresponds to equipment deployment of one grid unit, the equipment deployment system comprises equipment types, equipment quantity, unique coordinates, a minimum vector module length value and corresponding sensitive area level, all equipment intervals of the same type in the equipment deployment task row are calculated based on the vector module length, and after the equipment deployment task row is deployed, verification is performed according to a judgment rule that the equipment layout covers the grid unit area.
  10. 10. The intelligent deployment system of the intelligent farmland internet of things equipment according to claim 8, wherein after equipment deployment is completed, the farmland environment and crop coupling module obtains a new temperature sequence, a new humidity sequence and a new illumination intensity sequence of grid cells corresponding to each area of a farmland, generates new correlation coefficients, compares the correlation coefficients before deployment, and quantifies the actual effect of a deployment scheme.

Description

Intelligent deployment system of intelligent farmland Internet of things equipment Technical Field The invention relates to the field of intelligent deployment of farmland equipment, in particular to an intelligent deployment system of intelligent farmland Internet of things equipment. Background In the intelligent farmland, the deployment effect of the Internet of things equipment directly determines the environmental monitoring precision, the crop growth data acquisition quality and the resource utilization efficiency. Most of the existing farmland Internet of things equipment deployment schemes lack of deep connection of the association relation between crop growth and environmental factors, only rely on uniform distribution or artificial experience to carry out layout, and cannot combine the sensitivity difference of different farmland areas to environmental changes such as temperature, humidity, illumination and the like, so that the problem of unbalance of equipment redundancy in sensitive areas due to insufficient monitoring force and non-sensitive areas is solved. In addition, the existing scheme is not accurate enough to quantize the sensing coverage of equipment, does not fully consider the sensing characteristics of different types of equipment and the coverage effect after superposition of multiple equipment, is easy to occur monitoring blind areas or repeatedly cover, not only affects the integrity of monitoring data, but also causes the waste of deployment cost, and the problems make the existing deployment scheme difficult to meet the requirements of smart farmland fine monitoring and efficient resource allocation. Disclosure of Invention Aiming at the technical problems in the prior art, the invention provides an intelligent deployment system of intelligent farmland Internet of things equipment. The technical scheme for solving the technical problems is as follows, an intelligent deployment system of intelligent farmland internet of things equipment, the system comprises: the system comprises a farmland environment and crop coupling module, a mapping relation establishing module, a crop environment and crop environment mapping relation establishing module; The monitoring module is used for quantitatively evaluating the sensitivity degree of each area of the farmland to environmental changes, generating microclimate sensitivity indexes and identifying sensitive areas with different grades; The system comprises a region preprocessing module, a device preprocessing module and a device layout processing module, wherein the region preprocessing module is used for introducing an optimized search mechanism based on a RANSAC algorithm, so that the micro climate sensitive index of each region is introduced into each sampling and iteration of the RANSAC algorithm, a vector field is established for each device, and an optimal device layout scheme is searched through repeated iteration of the RANSAC algorithm; And converting the optimal equipment layout scheme into an executable equipment deployment task list, collecting actual data of system operation after equipment is online, and comparing the established mapping relation to quantify the actual effect of the deployment scheme. In a preferred embodiment, the farmland environment and crop coupling module obtains historical observation data of farmland local temperature, humidity and illumination and yield data corresponding to each observation data through a third party, and obtains a yield sequence corresponding to the observation data and the yield data; Calculating a linear trend of the yield sequence to obtain trend values, such as slopes, and subtracting the trend values from each yield data value in the yield sequence, such as yield increase caused by climate warming; Calculating the average value of each observation data and trending output, calculating the dispersion, covariance and standard deviation of each observation data, dividing the covariance of each observation data by the deposition of two standard deviations, and obtaining a correlation coefficient for mapping the relationship between crops and the environment; The method comprises the steps of calculating the dispersion as the difference between an environmental variable value and each observed data and the difference between a detrend yield value and a detrend yield average value, calculating the covariance as the dispersion of each observed data, multiplying the dispersion of the detrend yield by the covariance, and summing to obtain the covariance, and calculating the standard deviation of each observed data and the detrend yield, namely the square root of the average value of the square of the dispersion, and dividing the covariance by the product of the two standard deviations to obtain the correlation coefficient. In a preferred embodiment, the monitoring module divides the farmland into a plurality of grid cells at equal intervals, specifically, automatically divides grids by