An intelligent species configuration and space layout optimization method and system for landscaping

Abstract

The application relates to the technical field of garden ecological intelligent configuration, and particularly discloses a method and system for intelligent species configuration and space layout optimization for landscaping, wherein the method comprises the following steps: constructing a soil sensor node network to collect soil state data in real time; determining a species suitability score; adjusting light adaptability in combination with a light and shadow evolution curve collected by a drone; and outputting a space layout optimization result by a multi-constraint graph optimization algorithm. Compared with the prior art which relies on static experience configuration and lacks real-time environmental response capability, especially under the condition of a complex ecological industrial park with strong soil heterogeneity and dynamic light distribution, the technical problem that scientific species configuration and spatial intelligent cooperation are difficult to realize is solved. Since the intelligent control method integrating multi-modal perception, semantic model construction and graph optimization mechanism is introduced, dynamic adaptation of species configuration and multi-objective optimization of space layout are realized, and the ecological stability of the park greening system is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent garden ecological configuration, and in particular to an intelligent species configuration and space layout optimization method and system for garden greening. BACKGROUND

[0002] Currently, the garden greening planning and plant configuration in the ecological industrial park still mainly rely on manual experience and traditional map annotation method for deployment, which is usually based on regional soil type, sunshine condition or vegetation historical data for simple classification and manual matching. This traditional method has many technical deficiencies. For example, the existing technology lacks dynamic acquisition and high-resolution spatial modeling means for soil physical and chemical properties, especially in areas where key growth parameters such as pH value, water content, organic matter content and salinity change frequently, it is difficult to achieve accurate plant configuration matching. At the same time, due to the characteristics of the ecological industrial park, such as pollution risk residue, variable regional microtopography, significant heterogeneity of soil structure after land reclamation, etc., the traditional method of selecting plants according to "zoning and large categories" often leads to low plant survival rate, poor growth state and even frequent maintenance.

[0003] In addition, the existing plant configuration technology often lacks quantitative "plant-soil-light" coupling analysis mechanism. In actual application, the response of plants to micro-light environment has significant heterogeneity, especially in industrial parks with high buildings or large structures, the spatial and temporal characteristics such as sunshine duration, peak illuminance and shadow duration cannot be effectively modeled and matched to the physiological needs of plants, resulting in a disconnection between plant configuration and the real environment, and even light stress phenomenon, affecting the greening effect. Although some current researches based on remote sensing or satellite images attempt to obtain light distribution information, they have problems such as insufficient accuracy, large spatial delay and difficulty in depicting local dynamic shadow evolution, which cannot meet the needs of sub-meter level park greening design.

[0004] Furthermore, when the species space layout optimization is performed after the configuration is completed, the existing methods generally use rule-based planting or point placement based on horticultural aesthetic experience, without systematically introducing multi-objective optimization factors such as spatial resource constraints (such as road traffic boundaries, water supply pipe network access capacity, operation and maintenance accessibility), ecological complementarity between species, pest control isolation strategies, etc. for systematic reasoning. At the same time, traditional algorithms are difficult to balance "local optimum" and "global spatial layout efficiency", lack of mathematical models based on plant growth suitability and environmental load capacity trade-off, resulting in low utilization efficiency of greening space and poor overall ecological function. Therefore, an intelligent garden greening method that can integrate real-time monitoring of soil multi-factors, light timing modeling and plant adaptability learning mechanism, and perform optimization configuration decision under multiple constraints is urgently needed to improve the scientificity, stability and ecological restoration ability of the ecological industrial park greening project. SUMMARY

[0005] In view of the technical problems existing in the prior art, the present application aims to provide an intelligent species configuration and space layout optimization method for landscaping, which aims to solve the technical problems in the prior art that the configuration relies on static experience and lacks real-time environmental response capability, especially in the complex ecological industrial park conditions with strong soil heterogeneity and dynamic changes in light distribution, it is difficult to realize scientific configuration of species and intelligent cooperation of space.

[0006] To solve the above technical problems, the present application adopts the following technical scheme: the present application provides an intelligent species configuration and space layout optimization method for landscaping,

[0007] The intelligent species configuration and space layout optimization method for landscaping comprises:

[0008] Step S10: arranging a soil sensor node network in the ecological industrial park according to a preset grid strategy, collecting soil state data in a sampling period T in real time through the soil sensor node network, and uploading the soil state data to a preset central database; wherein the soil state data comprises soil pH value data, water content data, organic matter content data and salinity data;

[0009] Step S20: performing grid coordinate processing, adaptive interpolation processing and dynamic weighted fusion processing based on the soil state data, and outputting a multi-dimensional soil suitability index at a location ; wherein x represents the horizontal coordinate and y represents the vertical coordinate;

[0010] Step S30: obtaining species soil tolerance model data from a preset plant species database, determining a plant species at a location based on the multi-dimensional soil suitability index and the species soil tolerance model data, and calculating a species suitability score ;

[0011] Step S40: collecting park shadow distribution images at a preset typical daily time period using a UAV, generating a light shadow evolution curve at a location , performing light matching analysis in combination with the species suitability score , and outputting an optimized species suitability score after light adaptability adjustment;

[0012] Step S50: performing space layout optimization based on the optimized species suitability score in combination with a graph optimization algorithm based on multiple constraint conditions.

[0013] Preferably, in step S10, the step of deploying the soil sensor node network in the eco-industrial park according to the preset grid strategy specifically comprises: obtaining park geographical information in the eco-industrial park, dividing the green area in the park geographical information into a plurality of regular grid units; deploying soil sensor nodes at the center positions of the plurality of regular grid units; the soil sensor nodes are used for monitoring soil state data, the soil state data including soil pH value data, water content data, organic matter content data and salinity data; the soil sensor nodes include a pH sensor for monitoring soil pH value, a soil moisture sensor for monitoring soil water content, an organic matter sensor for monitoring soil organic matter content, and a conductivity sensor for monitoring soil salinity; wherein the pH sensor and the soil moisture sensor are installed at a depth of 20 cm underground; the organic matter sensor and the conductivity sensor are installed at a depth of 10 cm underground; each sensor node collects corresponding parameter values at a sampling period T and packs them into a data packet with timestamp and coordinate information, and the data packet is sent to a preset central database via a preset wireless communication module.

[0014] Preferably, in step S10, the step of obtaining park geographical information in the eco-industrial park and dividing the green area in the park geographical information into a plurality of regular grid units specifically comprises:

[0015] When the shape of the green area in the park geographical information is a regular rectangle or square, an equidistant regular grid division mechanism is adopted to uniformly divide the green area into a plurality of regular grid units with the same side length;

[0016] When the shape of the green area in the park geographical information is an irregular polygon, a region division mechanism based on Voronoi partition is adopted, taking the center position of the soil sensor node to be deployed as the generator of the region division mechanism based on Voronoi partition, to automatically divide the green area into a plurality of polygonal sub-regions as regular grid units, and the area of each polygonal sub-region is close and the geometric distance of the boundary to the center position of the adjacent soil sensor node to be deployed is equivalent;

[0017] When the shape of the green area in the park geographical information is a strip or a long and narrow shape, a strip division mechanism based on adaptive aspect ratio adjustment is adopted to divide the green area along the long side direction into a plurality of strip units, and further divide the strip units into approximately rectangular small grids as regular grid units.

[0018] Preferably, in step S20, the step of performing grid coordinate processing, adaptive interpolation processing and dynamic weighted fusion processing based on the soil state data to output a multi-dimensional soil suitability index specifically comprises:

[0019] A two-dimensional rectangular coordinate system is established based on the deployed soil sensor node network. For soil state data, an adaptive interpolation mechanism based on dynamic quantile normalization is used for multidimensional data standardization to obtain the pH sub-fitness at the location with x-axis and y-axis. Water quantum suitability Organic proton fitness and salinity and alkalinity suitability ;

[0020] And utilize a sensitivity factor weighting mechanism to assess pH suitability. Water quantum suitability Organic proton fitness and salinity and alkalinity suitability Perform dynamic weighted fusion processing to output a multidimensional soil suitability index. .

[0021] Preferably, in step S20, the soil condition data is standardized using an adaptive interpolation mechanism based on dynamic quantile normalization to obtain the pH sub-fitness at the position with x as the horizontal axis and y as the vertical axis. Water quantum suitability Organic proton fitness and salinity and alkalinity suitability The steps specifically include:

[0022] Data distribution acquisition: For soil condition data, including soil pH, water content, organic matter content, and salinity, calculate the corresponding empirical distribution function and extract the quantiles of the soil condition data corresponding to the empirical distribution function. quantiles of soil condition data ;

[0023] Outlier truncation correction: When soil condition data is higher than the upper quantile of soil condition data Or below the quantile of soil condition data At that time, they were corrected to the quantiles of the soil state data respectively. quantiles of soil condition data Output optimized soil condition data;

[0024] Quantile normalization: The optimized soil state data is mapped to the [0,1] interval to obtain the sub-fitness of each parameter. The sub-fitness of each parameter includes the pH sub-fitness at the position with x-axis and y-axis. Water quantum suitability Organic proton fitness and salinity and alkalinity suitability .

[0025] Preferably, in step S20, further comprising: for the missing coordinate points without the sensor deployed, using an adaptive spatial estimation mechanism based on Kriging interpolation, calculating the correlation and spatial semi-variance of the soil state data collected by the adjacent sensor nodes, and estimating the sub-fitness of each parameter of the missing coordinate points according to the correlation and spatial semi-variance of the soil state data collected by the adjacent sensor nodes, so as to ensure that each coordinate point in the two-dimensional rectangular coordinate system has complete four sub-fitness indicators.

[0026] Preferably, in step S30, the species soil tolerance model data is obtained from a preset plant species database, and the plant species is determined based on the multi-dimensional soil fitness index and the species soil tolerance model data. The species fitness score at the position .

[0027] The species soil tolerance model data is obtained from a preset plant species database, and the species soil tolerance model data includes: target growth parameter data corresponding to the species, pH value suitable interval parameter data, organic matter demand level parameter data, water content tolerance interval parameter data, and salinity sensitivity parameter data; a corresponding soil tolerance semantic vector is constructed based on the species soil tolerance model data ;

[0028] A support vector machine (SVM) algorithm is introduced, and a splicing vector of the soil tolerance semantic vector and the soil fitness semantic vector is taken as the model input to construct a plant species suitability evaluation model.

[0029] The plant species suitability evaluation model outputs the species fitness score of the plant species at the position .

[0030] The plant species suitability evaluation model outputs the species fitness score of the plant species at the position .

[0031] The plant species suitability evaluation model in the model training stage specifically includes: obtaining historical jth multi-dimensional soil fitness index sample , and historical species soil tolerance model sample .and corresponding plant species growth suitability experimental data , construct training sample pairs , input , output , train the species suitability evaluation model by a support vector regression training method combining a minimum structural risk criterion with a kernel function mapping mechanism, and obtain a converged suitability evaluation function model;

[0032] In the model application stage of the plant species suitability evaluation model, the splicing vector obtained by splicing the of the corresponding plant species at the target location and is input into the converged suitability evaluation function model as the model input, and the output of the species suitability score is obtained.

[0033] Preferably, in step S40, the UAV is used to collect the shadow distribution image of the park at a preset typical daily time period to generate the light and shadow evolution curve of the location , and the light matching analysis is performed in combination with the species suitability score , and the step of outputting the optimized species suitability score adjusted for light adaptability includes:

[0034] The UAV is used to collect the shadow distribution image of the park at a preset typical daily time period to generate the light and shadow evolution curve of the location ;

[0035] The light overlap duration, the number of continuous sunshine time periods, and the sunshine interruption frequency are determined based on the light and shadow evolution curve , and the light adaptability adjustment coefficient is constructed based on the light overlap duration, the number of continuous sunshine time periods, and the sunshine interruption frequency;

[0036] The species suitability score is corrected based on the light adaptability adjustment coefficient, and the optimized species suitability score adjusted for light adaptability is output.

[0037] Preferably, in step S50, the multiple constraint conditions include:

[0038] Location unique configuration constraint: only one plant species is allowed to be configured at each location , which is used to prevent multiple plant species from being repeatedly configured at the same coordinate point;

[0039] Species proportion control constraint: the configuration proportion of each plant species in the entire park should not exceed a preset upper limit proportion value, which is used to control the overall balance of the species distribution;​​​

[0040] Ecological connectivity constraint: the plant species configured between adjacent locations need to have relevance or consistency for enhancing the ecological stability of plant community structure.

[0041] The application also provides an intelligent species configuration and space layout optimization system for landscaping, comprising:

[0042] A soil monitoring and collecting module is configured to deploy a soil sensor node network according to a preset grid strategy in the ecological industrial park, collect soil state data in real time within a sampling period T through the soil sensor node network, and upload the soil state data to a preset central database; wherein the soil state data includes soil pH value data, water content data, organic matter content data and salinity data.

[0043] A soil suitability analysis module is configured to perform grid coordinate processing, adaptive interpolation processing and dynamic weighted fusion processing based on the soil state data, and output a multi-dimensional soil suitability index of the location ; wherein x represents the horizontal coordinate and y represents the vertical coordinate.

[0044] A species suitability evaluation module is configured to obtain species soil tolerance model data from a preset plant species database, determine a species suitability score of a plant species at the location based on the multi-dimensional soil suitability index and the species soil tolerance model data, and output the species suitability score.

[0045] A light adaptability correction module is configured to use a UAV to collect park shadow distribution images at a preset typical daily time period, generate a light and shadow evolution curve of the location , perform light matching analysis in combination with the species suitability score , and output an optimized species suitability score after light adaptability adjustment. A space layout optimization module is configured to perform space layout optimization based on the optimized species suitability score in combination with a graph optimization algorithm based on multiple constraints.

[0046] The application also provides an intelligent species configuration and space layout optimization device for landscaping, comprising a memory, a processor, and an intelligent species configuration and space layout optimization program for landscaping stored on the memory and executable on the processor, wherein the intelligent species configuration and space layout optimization program for landscaping, when executed by the processor, implements the intelligent species configuration and space layout optimization method for landscaping.

[0047]

[0048] ​​​The application further provides a computer program product comprising an intelligent species configuration and space layout optimization program for landscaping, which realizes the intelligent species configuration and space layout optimization method for landscaping when executed by a processor.

[0049] The application has the beneficial effect that the application realizes accurate matching of multi-dimensional soil characteristics and plant species suitability by constructing a soil sensor network and combining adaptive interpolation and a support vector machine model, and improves the scientificity and landing adaptability of species configuration.

[0050] The application introduces illumination shadow evolution analysis and multi-constraint graph optimization mechanism, effectively solves the problem that traditional landscaping schemes are difficult to dynamically adapt to illumination conditions and space limitations, and improves the overall stability and space utilization efficiency of the ecological park greening system. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0052] Figure 1 The flowchart of the first embodiment of the intelligent species configuration and space layout optimization method for landscaping of the application.

[0053] Figure 2 The RMSE decline rate distribution diagram of the traditional interpolation and dynamic weighted fusion of the intelligent species configuration and space layout optimization method for landscaping of the application.

[0054] Figure 3 The water content prediction accuracy comparison diagram of the intelligent species configuration and space layout optimization method for landscaping of the application.

[0055] Figure 4 The multi-dimensional soil suitability thermal map diagram of the intelligent species configuration and space layout optimization method for landscaping of the application.

[0056] Figure 5 The illumination shadow evolution curve diagram of the typical position of the intelligent species configuration and space layout optimization method for landscaping of the application.

[0057] Figure 6 The illumination matching score comparison diagram of the sun-loving type and shade-tolerant type species of the intelligent species configuration and space layout optimization method for landscaping of the application.

[0058] Figure 7 This is a schematic diagram of an equipment for an intelligent species configuration and spatial layout optimization method for landscaping according to the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] Example 1: As Figure 1 The diagram shown is a flowchart of the first embodiment of the intelligent species configuration and spatial layout optimization method for landscaping of the present invention, which presents the first embodiment of the intelligent species configuration and spatial layout optimization method for landscaping of the present invention.

[0061] In the first embodiment, the intelligent species configuration and spatial layout optimization method for landscaping includes:

[0062] Step S10: Deploy a soil sensor node network in the eco-industrial park according to a preset gridding strategy. Collect soil condition data in real time within the sampling period T through the soil sensor node network and upload the soil condition data to a preset central database. The soil condition data includes soil pH data, water content data, organic matter content data, and salinity data.

[0063] It should be noted that the "soil sensor node network" mentioned in this step refers to a physical network structure composed of multi-point distributed intelligent soil sensor units. Each sensor unit has independent data acquisition and communication functions, and can automatically monitor the physicochemical properties of the soil area it is located in and encode and upload the data. Its deployment is based on a preset gridding strategy, which divides the park into several adjacent grid units, each equipped with at least one sensor node, thereby achieving spatial coverage of soil condition parameters. The collected "soil condition data" mainly includes four items: soil pH, soil volumetric water content, organic matter content, and soil salinity. This data is periodically uploaded to a central database for unified storage and management via wired or wireless means.

[0064] Understandably, through the deployment and automated data collection of the aforementioned sensor network, the time-varying characteristics of the soil throughout the entire park can be continuously and frequently perceived, enabling digital modeling of soil conditions in spatial grid units. This provides accurate data support for subsequent plant species suitability analysis and spatial layout optimization, avoiding the lag and sampling blind spots inherent in traditional manual soil surveys.

[0065] It should be understood that, compared with the previous soil monitoring methods that relied on manual sampling and testing, sample delivery and low-frequency updates, this invention, by embedding a network of soil sensor nodes into the spatial layout of the park, forms a continuous and high-resolution soil data sensing capability, which shortens the soil condition update cycle from several weeks or even months to several hours or minutes, greatly improving the timeliness and accuracy of species configuration.

[0066] For example, in a pilot implementation in an industrial eco-park, 150 soil monitoring nodes were installed at a density of one sensor per 100 square meters. These nodes were set to automatically collect and upload data every ten minutes, allowing for real-time updates of soil condition information for each grid unit within the coverage area. Verification showed that this achieved 100% soil data coverage and an update latency of less than 30 minutes, significantly improving data support efficiency in plant species selection and configuration.

[0067] Step S20: Perform grid coordinate transformation, adaptive interpolation, and dynamic weighted fusion processing based on soil condition data, and output the location data. Multidimensional Soil Suitability Index Where x represents the x-coordinate and y represents the y-coordinate;

[0068] It should be noted that the "grid coordinate processing" in this step refers to mapping the collected soil condition data to a preset two-dimensional spatial grid coordinate system, giving each data point a clear spatial location identifier, facilitating subsequent interpolation and fusion processing; "adaptive interpolation processing" refers to dynamically selecting an interpolation method (such as inverse distance weighted or spline interpolation) based on the spatial distribution characteristics and data gradient change trends of multiple adjacent known nodes to fill in data gaps between grids, achieving high-reliability soil condition data filling; "dynamic weighted fusion processing" refers to comprehensively fusing soil condition data collected from different sources and at different times according to time weights, data volatility, and node acquisition stability to construct a soil condition map with high temporal consistency and spatial continuity. The dynamic weighted fusion processing steps are expressed by the following formula:

[0069]

[0070] in, Indicates the target location At any moment Robust fusion estimates for the k-th type of soil state variables (pH, water content, salinity); The timestamp of the observation data corresponding to node n; For target location The set of observation nodes in the surrounding neighborhood; For node n, the target position The dynamic comprehensive weight of the k-th type of soil state variable at time t; For target location At any moment The observed soil state value of type k; where, dynamic comprehensive weight The factor is determined by a combination of pre-defined spatial proximity factor, time freshness factor, node acquisition stability factor, data volatility factor, and data source credibility factor.

[0071] Understandably, through the aforementioned three-stage processing, a multidimensional soil suitability index set with consistent accuracy and complete distribution can be formed in the two-dimensional coordinate space (x, y) of the park. This index set comprehensively reflects indicators such as soil pH suitability, organic matter matching degree, water adequacy, and salinity tolerance at each location. It serves as the basic data input for subsequent plant species suitability assessment and light adaptability analysis, and features high spatial resolution, clear physical meaning, and stable change trends.

[0072] For example, such as Figure 2 As shown, in 5000 test locations selected within a 1000m × 1000m area of ​​the eco-industrial park, compared with the traditional inverse distance weighted interpolation method, the "dynamic weighted fusion processing" described in this invention significantly converges the overall prediction error (RMSE) globally: the error distribution shifts towards a lower value range, and the proportion of high-error tail samples decreases significantly; statistical results show that the average decrease in RMSE is approximately 35%–40%, with over 80% of samples showing a decrease of ≥30%, indicating that this method has a stable global accuracy improvement effect in heterogeneous soil scenarios. Figure 3 As shown, using water content as a representative indicator, a comparison of the fitting relationship between traditional interpolation predictions and the dynamically weighted fusion predictions of this invention at 500 sampling locations reveals that the dynamically weighted fusion prediction point cloud is closer to the ideal prediction line (y=x), with a significantly reduced average deviation; 95% of the samples have an absolute deviation controlled within ±2.5%, while the deviation of the traditional method can reach ±7%. This result demonstrates that by introducing a time freshness factor, a node acquisition stability factor, and a data volatility factor, this invention can maintain a high degree of fit to the true trend even under local missing data and high volatility interference, thus effectively supporting high-reliability filling of missing grid points. Figure 4As shown, the "multi-dimensional soil suitability thermal map" constructed based on the dynamic weighted fusion result presents a more smooth and continuous spatial gradient structure in the two-dimensional coordinate space (x, y), and the suitability belt and patch boundary after the comprehensive mapping of soil pH, water content and salinity are clear and the noise is significantly reduced, avoiding the "block pseudo-gradient" caused by the local abnormal value in the traditional method. The spatial base map can be directly used as the input of the subsequent steps: multiplication with the light satisfaction degree in step S30 to form the location-level comprehensive suitability, and joint entry with the ecological function target into the multi-objective optimization model in step S40, so as to realize the linkage optimization of species recommendation and spatial layout. The above results jointly prove that: through the cascade design of grid coordinate processing, adaptive interpolation processing and dynamic weighted fusion processing, the present application can obtain a set of multi-dimensional soil suitability indexes with high time consistency and high spatial continuity on the park scale, and provides a quantifiable and reusable data basis and empirical support for subsequent intelligent species configuration and spatial layout optimization.

[0073] Step S30: obtaining species soil tolerance model data from a preset plant species database, determining plant species based on multi-dimensional soil suitability index and species soil tolerance model data to obtain the species suitability score at the location ;

[0074] It should be noted that the "plant species database" in this step refers to a structured data set that is constructed and stored in advance and contains ecological adaptability parameters of various greening plants, mainly including plant species name, growth suitable pH interval, organic matter demand threshold, water tolerance range and salt stress threshold, etc. The "species soil tolerance model data" refers to a response model in the form of a multi-parameter threshold matrix or a function, which is trained according to historical samples or calibrated by expert experience, and is used to represent the response characteristics of each plant to different soil factors. The "species suitability score" refers to a quantitative evaluation index calculated according to the matching degree between the soil suitability index and the tolerance model of the plant species at a certain location (x, y), which reflects the growth adaptability of the species at the current location.

[0075] It can be understood that by taking the multi-dimensional soil suitability index as the input and combining the corresponding tolerance model of different plant species in the database for item-by-item matching evaluation, the comprehensive suitability score of the species at the location can be calculated. The scoring process can adopt methods such as multi-dimensional membership function matching, minimum distance penalty function or regression prediction model based on ensemble learning to realize quantitative adaptation analysis of each candidate plant species at each location in the spatial coordinate system.

[0076] It should be understood that traditional landscape design often relies on human experience or single factors (such as pH value) for species selection, neglecting the interactions between multidimensional soil factors and the multifactorial adaptability of plants themselves. This can easily lead to problems such as plant mismatch and reduced survival rates. This invention, by introducing a species tolerance model and a matching mechanism of multifactorial soil suitability, achieves personalized and precise correspondence between plant selection and soil conditions, improving the scientific nature and sustainability of plant design.

[0077] For example, at a certain location in the park (x=12, y=8), the pH suitability is 0.82, organic matter suitability is 0.74, water content suitability is 0.88, and salinity suitability is 0.65. A legume hedge plant A is selected, with tolerance model parameters of: pH range [6.0, 7.5], organic matter ≥2%, moderate water content, and moderate salinity tolerance. Its comprehensive matching score, calculated using a gradient kernel function regression model, is 0.79, classifying it as "good" suitable for planting. Another tree plant, B, scores only 0.42, is marked as "low suitability," and is automatically removed from the candidate set. This method quantifies the adaptability differences of different plant species in different spatial locations, significantly improving the spatial fit and biological adaptability of planting schemes.

[0078] Step S40: Use a drone to cruise and collect images of the park's shadow distribution during a preset typical daytime period, and generate location data. Light and shadow evolution curve Combined with species suitability score Perform light matching analysis and output the optimized species suitability score after light adaptability adjustment;

[0079] It should be noted that the "light and shadow evolution curve" refers to the process of extracting the temporal information of light and shadow changes experienced by each grid location throughout the day using image analysis algorithms based on aerial images of the park collected at different times on a typical day (with sufficient sunshine and representing seasonal characteristics), forming a quantitative description of the light exposure level at that location; "light adaptability adjustment" refers to the process of adjusting and correcting the original species suitability score according to the ecological response characteristics of different plant species to light intensity, light duration and shadow frequency, so as to more realistically reflect the actual growth adaptability of plants at that location.

[0080] Among them, for species In position Original suitability score Define its light adaptation adjustment factor :

[0081]

[0082] in, For species The baseline duration of sunshine demand; For species The optimal average light intensity; For species The maximum tolerable shadow alternation frequency; : These are the sensitivity weights of species to light duration, light intensity, and shadow frequency, respectively; This serves as the baseline duration for the measured sunshine demand; The measured average light intensity; The measured shadow alternation frequency; the adjusted optimized species suitability score. ,in, Use a sigmoid or normalization function to ensure that the result falls within the [0,1] interval.

[0083] Understandably, by introducing drones for timed aerial imagery collection and combining it with computer vision technology to obtain high spatial and temporal resolution data on light evolution, it is possible to efficiently quantify the distribution characteristics of light resources at a plant's location under natural conditions. Furthermore, by combining the light adaptation ranges (such as full sun, partial shade, and shade tolerance) registered by each plant species in the database with their tolerance to insufficient or excessive sunlight, a location-species light matching matrix can be constructed. This allows for multi-factor correction of the initial species suitability score, improving the accuracy of spatial adaptability.

[0084] It should be understood that traditional plant configuration methods often overlook the dynamic shadow effects caused by buildings, structures, or large trees in the park, which can easily lead to misjudgments of plant light requirements, resulting in problems such as slow plant growth, yellowing leaves, and even death. This invention, by introducing light evolution curves and light adaptability analysis mechanisms, can significantly improve the ecological adaptability of plant configurations and enhance the stability and durability of greening.

[0085] For example, such as Figure 5 As shown, a typical location near the building belt was selected within the eco-industrial park. Aerial images were collected by drone during a typical day (6:00 to 18:00, sampling every 5 minutes). Illumination intensity was calculated based on image analysis algorithms, resulting in the illumination-shadow evolution curve for this location. The image shows that illumination intensity decreases significantly before 9:00 AM and after 4:00 PM, remaining near full illumination during midday. Statistical results indicate that the average illumination intensity at this location is 0.73, the cumulative effective illumination duration is approximately 7.2 hours, and the shadow frequency factor is 0.18, indicating that the area is affected by building shading, with stable illumination during the middle of the day and frequent transitions between morning and evening. Figure 6As shown, based on the light shadow evolution curve and the plant light demand parameters, the light matching scores of sun-loving plants (class A) and shade-tolerant plants (class B) are calculated respectively. The results show that the suitability of class A plants is higher in the central open area, while the suitability is lower in the building shadow frequent area; on the contrary, the suitability of class B plants is higher in the area with high shadow frequency and insufficient cumulative sunshine. The results prove that the "light adaptability adjustment" step can adjust the original species suitability score according to local conditions, so that the species recommendation is more in line with the actual ecological response. Through the generation and analysis of the above light shadow evolution curve, the present application not only provides a quantitative light exposure index, but also directly relates it to the ecological needs of the species, dynamically adjusts the original species suitability score, and avoids the problem of relying only on the average light value in the traditional scheme while ignoring the timing and frequency characteristics, thereby significantly improving the rationality and scientificity of the species configuration.

[0086] Step S50: performing space layout optimization based on the optimized species suitability score combined with a graph optimization algorithm based on multiple constraint conditions.

[0087] It should be noted that the "space layout optimization" in this step refers to, after considering the light-soil suitability scores of each location and each plant, generating a plant space configuration scheme that meets the constraint conditions based on the graph structure representing the space grid and the distribution of candidate species in the park, according to multiple actual constraints such as ecological compatibility between species, visual combination, and maintenance convenience; the "multiple constraint conditions" include symbiotic / antagonistic relationships between species, limitations on the overlap of plant canopies in adjacent locations, path accessibility limitations, local species diversity index requirements, and shrub / tree / ground cover level balance requirements.

[0088] It can be understood that by taking the optimized species suitability score as the initial evaluation basis for the layout weight, a space graph model is reconstructed, where each node represents a grid cell in the park, and each edge represents the plant configuration relationship between adjacent grids. By using a heuristic graph optimization algorithm (such as multi-objective A* search, restricted genetic algorithm, or graph coloring method), the optimal configuration path that meets multiple ecological and aesthetic constraints can be quickly searched in a global range, thereby effectively improving the overall ecological adaptability and landscape performance of the space configuration.

[0089] It should be understood that in traditional garden design, plant species and distribution locations are configured relying on human experience, it is difficult to evaluate the long-term impact of micro ecological condition differences on plant growth, and the dynamic interaction between adjacent plants is often ignored in complex ecology, leading to problems such as uneven plant growth, increased maintenance costs, etc. The present application effectively avoids the above defects by introducing a graph optimization model and a multi-constraint mechanism, thereby enhancing the adaptability, long-term stability, and spatial diversity of the garden greening system.

[0090] For example, in an eco-industrial park, the suitability scores of plants A, B, and C at different positions (x, y) in a certain area are 0.89, 0.83, and 0.76 respectively after pre-analysis; in the graph optimization process, the following constraints are set: adjacent positions cannot be configured with arbor plants with a crown radius > 1.5 meters; the same area must contain at least 3 different morphological level plants (trees, shrubs, ground cover); species B and species C cannot be configured adjacent to each other due to root antagonism. Each position is regarded as a graph node, and an edge weight function is constructed based on the above restrictions, and finally a spatial layout scheme that meets the conditions is generated, which improves the local diversity index by 12% and reduces the later maintenance cost by 18%. This kind of layout mechanism significantly improves the sustainability and ecological stability of the greening system in actual deployment.

[0091] Embodiment two: In addition, the present application provides an intelligent species configuration and spatial layout optimization system for landscaping, which adopts the intelligent species configuration and spatial layout optimization method for landscaping in the above embodiment, and can solve the technical problem of intelligent species configuration and spatial layout optimization for landscaping. Compared with the prior art, the intelligent species configuration and spatial layout optimization system for landscaping provided by the present application has the same beneficial effects as the intelligent species configuration and spatial layout optimization method for landscaping provided by the above embodiment, and other technical features in the intelligent species configuration and spatial layout optimization system for landscaping are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0092] Embodiment three: The present application provides an intelligent species configuration and spatial layout optimization device for landscaping, please refer to Figure 7An intelligent species configuration and space layout optimization device for landscaping includes at least one processor and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor to enable the at least one processor to perform an intelligent species configuration and space layout optimization method for landscaping according to an embodiment. The intelligent species configuration and space layout optimization device for landscaping according to an embodiment can include, but is not limited to, a mobile terminal such as a mobile phone, a notebook, a digital broadcasting receiver, a PDA (Personal Digital Assistant), a PAD (Portable Application Description), a PMP (Portable Media Player), a vehicle terminal such as a car navigation terminal, and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The intelligent species configuration and space layout optimization device for landscaping is merely an example and should not impose any limitation on the function and use range of the embodiments. The intelligent species configuration and space layout optimization device for landscaping can include a processing device 1001 such as a central processing unit, a graphic processing unit, and the like, which can perform various appropriate actions and processes according to programs stored in a read-only memory 1002 or loaded into a random access memory 1004 from a storage device 1003. The random access memory 1004 also stores various programs and data required for the operation of the intelligent species configuration and space layout optimization device for landscaping. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other through a bus 1005. An I / O interface 1006 is also connected to the bus. In general, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, and the like; an output device 1008 including, for example, an LCD (Liquid Crystal Display), a speaker, a vibrator, and the like; the storage device 1003 including, for example, a magnetic tape, a hard disk, and the like; and a communication device 1009. The communication device 1009 can allow the intelligent species configuration and space layout optimization device for landscaping to communicate with other devices wirelessly or by wire to exchange data. Although the intelligent species configuration and space layout optimization device for landscaping having various systems is illustrated in the drawing, it should be understood that all of the illustrated systems are not required to be implemented or provided. More or less systems can be alternatively implemented or provided.

[0093] Embodiment four: the application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the intelligent species configuration and spatial layout optimization method for landscaping described above. The computer program product provided by the application can solve the technical problem of intelligent species configuration and spatial layout optimization for landscaping. Compared with the prior art, the beneficial effects of the computer program product provided by the application are the same as those of the intelligent species configuration and spatial layout optimization method for landscaping provided by the above-mentioned embodiments, and are not described here.

[0094] In particular, according to the embodiments disclosed by the application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, the embodiments disclosed by the application include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by a processing device 1001, the above-mentioned functions defined in the method of the embodiments disclosed by the application are executed.

[0095] It should be understood that various parts of the application disclosed can be realized by hardware, software, firmware or a combination thereof. In the description of the above-mentioned embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0096] Obviously, those skilled in the art can make various modifications and variations to the application without departing from the spirit and scope of the application. Thus, if these modifications and variations of the application fall within the scope of the claims of the application and their equivalent technologies, the application also intends to include these modifications and variations.

Claims

1. A method for intelligent species configuration and space layout optimization for landscaping, characterized in that, The method comprises: Step S10: arranging a soil sensor node network in the eco-industrial park according to a preset grid strategy, collecting soil state data in a sampling period T through the soil sensor node network, and uploading the soil state data to a preset central database; wherein the soil state data includes soil pH value data, water content data, organic matter content data, and salinity data; Step S20: based on the soil state data, performing grid coordinate processing, adaptive interpolation processing and dynamic weighted fusion processing, and outputting a multi-dimensional soil suitability index in the position ; wherein, x represents the horizontal coordinate, and y represents the vertical coordinate;​ Step S30: obtaining species soil tolerance model data from a pre-set plant species database, based on a multi-dimensional soil suitability index and the species soil tolerance model data to determine the plant species at the location a species suitability score ; Wherein, the species soil tolerance model data is obtained from a preset plant species database, and the plant species is determined based on the multi-dimensional soil suitability index and the species soil tolerance model data The plant species is determined based on the multi-dimensional soil suitability index The plant species is determined based on the multi-dimensional soil suitability index The plant species is determined based on the multi-dimensional soil suitability index Soil tolerance model data for each species is obtained from a pre-defined plant species database. This data includes: target growth parameters, pH range parameters, organic matter requirement level parameters, water content tolerance range parameters, and salinity sensitivity parameters for each species. Based on this soil tolerance model data, corresponding soil tolerance semantic vectors are constructed. Based on multidimensional soil suitability index Construct the corresponding soil suitability semantic vector ; Introducing the Support Vector Machine (SVM) algorithm to use soil tolerance semantic vectors Soil suitability semantic vector spliced ​​vector As input to the model, a plant species suitability evaluation model is constructed; Plant species suitability assessment model outputs plant species At location Species suitability score ; wherein the plant species suitability evaluation model outputs a plant species At the location of the species suitability score The steps include a model training phase and a model application phase; The plant species suitability evaluation model comprises the following steps of: obtaining historical first term multidimensional soil suitability index samples , historical species soil tolerance model samples and corresponding plant species growth suitability experimental data , constructing a training sample pair , taking as input, taking as output, training the species suitability evaluation model by a support vector regression training method combining a minimum structure risk criterion and a kernel function mapping mechanism, and obtaining a converged suitability evaluation function model. In the model application phase, the plant species suitability assessment model will target the location. Corresponding plant species of and The concatenated vector obtained by concatenation As input to the model, the species suitability score is obtained by feeding it into the converged suitability evaluation function model. The output; Step S40: Use a drone to cruise and collect images of the park's shadow distribution during a preset typical daytime period, and generate location data. Light and shadow evolution curve Combined with species suitability score Perform light matching analysis and output the optimized species suitability score after light adaptability adjustment; Step S50: based on the optimized species suitability score, a spatial layout optimization is performed using a graph optimization algorithm based on multiple constraint conditions. 2.The intelligent species configuration and space layout optimization method for landscaping of claim 1, wherein, In step S10, the step of arranging a soil sensor node network in the eco-industrial park according to a preset grid strategy comprises: obtaining park geographical information in the eco-industrial park, dividing the green area based on the park geographical information into a plurality of regular grid cells, arranging soil sensor nodes at the center positions of the plurality of regular grid cells, and monitoring soil state data including soil pH value data, water content data, organic matter content data, and salinity data by the soil sensor nodes. The soil sensor nodes include a pH sensor for monitoring soil pH value, a soil moisture sensor for monitoring soil water content, an organic matter sensor for monitoring soil organic matter content, and a conductivity sensor for monitoring soil salinity. The pH sensor and the soil moisture sensor are installed at a depth of 20 cm underground, and the organic matter sensor and the conductivity sensor are installed at a depth of 10 cm underground. Each sensor node collects corresponding parameter values at a sampling period T and packs them into a data packet with timestamp and coordinate information, which is sent to a preset central database via a preset wireless communication module.

3. The intelligent species configuration and space layout optimization method for landscaping of claim 2, wherein, In step S10, the step of obtaining park geographical information in the eco-industrial park and dividing the green area based on the park geographical information into a plurality of regular grid cells comprises: When the shape of the green area in the park geographical information is a regular rectangle or square, an equidistant regular grid division mechanism is used to uniformly divide the green area into a plurality of regular grid cells with the same side length; When the shape of the green area in the park geographical information is an irregular polygon, a region division mechanism based on Voronoi partitioning is used, taking the center position of the soil sensor node as the generator of the region division mechanism of Voronoi partitioning, to automatically divide the green area into a plurality of polygonal sub-regions as regular grid cells, each polygonal sub-region having a similar area and a geometric distance equivalent to the center position of the adjacent soil sensor node; When the shape of the green area in the park geographical information is a strip or a long and narrow shape, a strip division mechanism based on adaptive aspect ratio adjustment is used to divide the green area along the long side direction into a plurality of strip cells, and further divide the strip cells into approximately rectangular small cells as regular grid cells.

4. The intelligent species configuration and space layout optimization method for landscaping of claim 1, wherein, In step S20, the step of performing grid coordinate processing, adaptive interpolation processing, and dynamic weighted fusion processing based on the soil state data to output a multi-dimensional soil suitability index comprises: A two-dimensional rectangular coordinate system is established based on a laid soil sensor node network, and a multi-dimensional data standardization processing is performed on soil state data by using an adaptive interpolation mechanism based on dynamic quantile normalization to obtain a pH sub-suitability degree at a position with an x horizontal coordinate and a y vertical coordinate , a water content sub-suitability degree , an organic matter sub-suitability degree , and a salinity sub-suitability degree . And use the sensitive factor weight mechanism based on pH sub-optimal degree , the amount of water sub-optimal degree , organic matter sub-optimal degree and salinity sub-optimal degree Dynamic weighted fusion processing is performed, and a multi-dimensional soil suitability index is output .

5. The intelligent species configuration and space layout optimization method for landscaping of claim 4, wherein, In step S20, the multi-dimensional data standardization processing is performed on the soil state data by using the adaptive interpolation mechanism based on dynamic quantile normalization, to obtain the pH sub-suitability at the position with the horizontal coordinate x and the vertical coordinate y , the combined water sub-suitability , the organic matter sub-suitability , and the salinity sub-suitability , specifically including: Data distribution acquisition: for the soil state data including soil pH data, water content data, organic matter content data and salinity data, the corresponding empirical distribution function is calculated, and the soil state data upper quantile point of the corresponding empirical distribution function is extracted and the soil state data lower quantile point ; Outlier Truncation Correction: when the soil status data is higher than the upper quantile of the soil status data or lower than the lower quantile of the soil status data , it is respectively corrected to the upper quantile of the soil status data and the lower quantile of the soil status data , and the optimized soil status data is output; Quantile normalization: mapping the optimized soil status data to the interval [0, 1] to obtain the sub-fitness of each parameter, the sub-fitness of each parameter including the pH sub-fitness at the position with the horizontal coordinate x and the vertical coordinate y , the water content sub-fitness , the organic matter sub-fitness , and the salinity sub-fitness .

6. The intelligent species configuration and space layout optimization method for landscaping of claim 5, wherein, In step S20, further comprising: for the missing coordinate points without deployed sensors, using an adaptive spatial estimation mechanism based on Kriging interpolation, calculating the correlation and spatial semi-variance of the soil state data collected by the adjacent sensor nodes, and estimating the sub-fitness of each parameter of the missing coordinate points according to the correlation and spatial semi-variance of the soil state data collected by the adjacent sensor nodes, so as to ensure that each coordinate point in the two-dimensional rectangular coordinate system has complete four sub-fitness indexes.

7. The intelligent species configuration and space layout optimization method for landscaping of claim 1, wherein, In step S40, a drone is used to cruise and collect images of the park's shadow distribution during a preset typical daytime period, generating location data. Light and shadow evolution curve Combined with species suitability score The steps for performing light matching analysis and outputting the optimized species fitness score after light adaptability adjustment include: The UAV is used to collect images of shadow distribution in a preset typical day period, to generate a light shadow evolution curve of the location . ; Evolution curve based on illumination shadow determining a light overlap duration, a number of consecutive sunlight periods, and a sunlight interruption frequency; and constructing a light adaptability adjustment coefficient based on the light overlap duration, the number of consecutive sunlight periods, and the sunlight interruption frequency; adjusting the species suitability score based on the light adaptability adjustment factor The modified, light-adapted optimized species suitability score is output. 8.The intelligent species configuration and space layout optimization method for landscaping of claim 1, wherein, In step S50, the multiple constraint conditions include: Location unique configuration constraint: in each location Only one plant species is allowed to be configured, to prevent multiple plant species from being configured at the same coordinate point; Species proportion control constraint: the configuration proportion of each plant species in the entire park cannot exceed a preset upper limit proportion value, for controlling the overall balance of the species distribution; Ecological connectivity constraint: the plant species configured between adjacent positions need to have relevance or consistency, for enhancing the ecological stability of the plant community structure.

9. A system for intelligent species configuration and spatial layout optimization for landscaping, applied to the method for intelligent species configuration and spatial layout optimization for landscaping in any one of claims 1 to 6, characterized in that, The intelligent species configuration and spatial layout optimization system for landscaping includes: A soil monitoring and collecting module is configured to deploy a soil sensor node network in the ecological industrial park according to a preset grid strategy, collect soil state data in real time through the soil sensor node network within a sampling period T, and upload the soil state data to a preset central database; wherein the soil state data includes soil pH value data, water content data, organic matter content data, and salinity data; The soil suitability analysis module is used for performing grid coordinate processing, adaptive interpolation processing and dynamic weighted fusion processing based on the soil state data, and outputting a multi-dimensional soil suitability index at a location ; wherein x represents a horizontal coordinate and y represents a vertical coordinate.​ a species suitability assessment module for obtaining species soil tolerance model data from a pre-set plant species database, based on a multi-dimensional soil suitability index and the species soil tolerance model data to determine a plant species a species suitability score at a location ;​ In the step of obtaining the species soil tolerance model data from the preset plant species database, the plant species is determined based on the multi-dimensional soil suitability index and the species soil tolerance model data In the step of obtaining the species soil tolerance model data from the preset plant species database, the plant species is determined based on the multi-dimensional soil suitability index and the species soil tolerance model data In the step of obtaining the species soil tolerance model data from the preset plant species database, the plant species is determined based on the multi-dimensional soil suitability index and the species soil tolerance model data​​ Soil tolerance model data for each species is obtained from a pre-defined plant species database. This data includes: target growth parameters, pH range parameters, organic matter requirement level parameters, water content tolerance range parameters, and salinity sensitivity parameters for each species. Based on this soil tolerance model data, corresponding soil tolerance semantic vectors are constructed. Based on multidimensional soil suitability index Construct the corresponding soil suitability semantic vector ; A support vector machine (SVM) algorithm is introduced to concatenate the soil tolerance semantic vector and the soil suitability semantic vector as model input to construct a plant species suitability evaluation model;​ Plant species suitability assessment model outputs plant species At location Species suitability score ; wherein the plant species suitability evaluation model outputs a plant species At the location of the species suitability score The steps include a model training phase and a model application phase; The plant species suitability evaluation model comprises the following steps of: obtaining historical first Item multi-dimensional soil suitability index sample , historical species soil tolerance model sample and corresponding plant species growth suitability experimental data , constructing a training sample pair , taking as input, taking as output, training the species suitability evaluation model by a support vector regression training method combining a minimum structure risk criterion and a kernel function mapping mechanism, and obtaining a converged suitability evaluation function model. In the model application phase, the plant species suitability assessment model will target the location. Corresponding plant species of and The concatenated vector obtained by concatenation As input to the model, the species suitability score is obtained by feeding it into the converged suitability evaluation function model. The output; The illumination adaptation correction module is used to collect images of the park's shadow distribution by using a drone during a preset typical daytime period, and to generate location data. Light and shadow evolution curve Combined with species suitability score Perform light matching analysis and output the optimized species suitability score after light adaptability adjustment; A spatial layout optimization module is configured to perform spatial layout optimization based on the optimized species suitability score and a graph optimization algorithm based on multiple constraint conditions.

Images