Breeding test field selection method based on microclimate characteristics in mountainous and hilly areas

Abstract

The invention relates to the technical field of breeding tests, in particular to a breeding test field selection method based on microclimate characteristics in mountainous and hilly areas, which comprises the following steps: collecting microclimate data of a land parcel, constructing a microclimate data set and generating a microclimate environment factor gradient distribution map; partitioning the microclimate environment factor gradient distribution map, generating a microclimate partition feature set, and calculating a corresponding candidate variety set; according to the candidate variety set, optimizing and generating a test field site selection scheme set; according to the test field site selection scheme set, generating a test field distribution diagram including plot space coordinates, a microclimate partition feature set and matched varieties; according to the invention, fine acquisition and analysis of microclimate data in mountainous and hilly areas are realized, space intensification and precision of site selection of test fields are improved, effective technical support is provided for agricultural production layout in mountainous and hilly areas, the space range of seed production of breeding test fields and seed production fields with different environmental factors is shortened, convenience is improved, and economic benefits are increased. And the breeding and seed production cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of breeding experiments, and in particular to a method for selecting breeding experimental fields based on microclimate characteristics of mountainous and hilly areas. Background Art

[0002] The existing breeding test method is often to select a piece of flat land near the breeding test work unit as a breeding test field. The crop types and varieties bred in this way are poorly representative of the width of the environmental adaptation niche. my country has a vast territory, rich ecosystem types, and diverse microclimates and microclimates. When promoted on a large scale, they often face the phenomenon of "water and soil incompatibility". If the representativeness of the width of the environmental adaptation niche of the selected crop types and varieties is to be improved, subsequent multi-point tests are required, which leads to high costs and long test times for breeding tests, and the existing breeding test fields are mostly concentrated in open plains. It is very susceptible to hybrid contamination by different lines and varieties of crops, especially for plant types that are spread by wind pollen.

[0003] Mountainous and hilly areas, with their varying topography and altitude, possess complex and variable microclimates, offering the potential for reducing the spatial scale of breeding experiments and improving the breadth and representativeness of environmental adaptation. However, traditional site selection methods often rely on empirical judgment or extensive meteorological data, making it difficult to accurately grasp the spatial heterogeneity and temporal dynamics of microclimates, resulting in insufficient representativeness and applicability of experimental results. Accurately capturing the spatiotemporal variations of microclimates in complex mountainous and hilly terrain and precisely matching them with the growth requirements of crop varieties is a key technical issue that needs to be addressed.

[0004] It should be noted that the information disclosed in this background technology section is only intended to increase understanding of the overall background of the present invention, and should not be regarded as an admission or any form of implication that the information constitutes prior art already known to those skilled in the art. Summary of the Invention

[0005] In order to solve the related defects of the above-mentioned existing breeding test methods, the present invention provides a method for selecting breeding test fields based on the microclimate characteristics of mountainous and hilly areas, and the breeding test field selection method based on the microclimate characteristics of mountainous and hilly areas comprises the following steps: Collect microclimate data of the plot, construct a microclimate dataset and generate a gradient distribution map of microclimate environmental factors; Partitioning the microclimate environmental factor gradient distribution map, generating a microclimate partition feature set and calculating a corresponding candidate variety set; Based on the candidate variety set, optimizing and generating a set of experimental field site selection plans; According to the experimental field site selection plan set, an experimental field distribution map including plot spatial coordinates, microclimate zoning feature sets and matching varieties is generated.

[0006] Furthermore, the microclimate data includes at least one or more of temperature, humidity, air velocity, thermal radiation Furthermore, the microclimate dataset at least includes a time series and spatial coordinates.

[0007] Furthermore, the collecting of microclimate data of the plot and constructing the microclimate dataset includes the following steps: Deploy monitoring points in flat areas of mountainous and hilly regions; Collecting plot temperature, humidity, and light data through the monitoring points; generating microclimate data including a time series according to the temperature, humidity, and light data; generating microclimate data including the spatial coordinates according to the spatial coordinates of the monitoring points; Constructing a microclimate dataset according to the microclimate data including the time series and the microclimate data including the spatial coordinates; According to the microclimate dataset, the numerical distribution characteristics of temperature, humidity and light in different plots are determined.

[0008] Furthermore, generating a microclimate environmental factor gradient distribution map comprises the following steps: extracting temperature gradient data, humidity gradient data, and light gradient data from the microclimate dataset; generating a microclimate environmental factor gradient distribution map according to the temperature gradient data, the humidity gradient data, and the light gradient data; According to the microclimate environmental factor gradient distribution map, the spatial variation characteristics of temperature, humidity and light are characterized.

[0009] Furthermore, partitioning the microclimate environmental factor gradient distribution map to generate a microclimate partition feature set comprises the following steps: Extracting the temperature gradient data, the humidity gradient data, and the light gradient data from the microclimate environmental factor gradient distribution map; If the temperature gradient data, the humidity gradient data, or the light gradient data exceeds a preset threshold, a K-means clustering algorithm is used to divide the microclimate into zones; According to the microclimate zones, calculating the mean and variance of the temperature in each zone; According to the microclimate zones, the mean and variance of humidity in each zone are calculated; According to the microclimate zones, calculating the mean and variance of the light intensity in each zone; A microclimate zoning feature set is generated based on the mean and variance.

[0010] Furthermore, the candidate variety set corresponding to the calculation includes: Extracting the mean and variance of temperature, humidity, and light from the microclimate partition feature set; Extract the temperature adaptation range of varieties from the crop variety resource library; Extract the humidity adaptation range of varieties from the crop variety database; Extract the light adaptation range of varieties from the crop variety resource library; The weighted Euclidean distance algorithm is used to calculate the similarity between the variety parameters and the microclimate zoning feature set; Based on the similarity, a set of candidate varieties for each partition is generated.

[0011] Furthermore, the optimizing and generating of the experimental field site selection plan set based on the candidate variety set includes: extracting variety parameters from the candidate variety set; Genetic algorithm was used to optimize the site selection combination of experimental fields; Calculating the scores of the site combinations based on the representativeness of the microclimate zone feature set; Calculating a score for the site selection combination based on the similarity between the variety parameters and the microclimate zoning feature set; and generating a set of experimental field site selection schemes based on the score; If the score of the experimental field site selection scheme set is lower than a preset threshold, the site selection combination is adjusted.

[0012] Furthermore, the step of optimizing and generating a set of experimental field site selection plans based on the set of candidate varieties further includes: Extracting plots with scores lower than a preset threshold from the set of experimental field site selection plans; Selecting a zone with a second highest gradient value from the microclimate zone feature set; Using a weighted Euclidean distance algorithm, the similarity between the partition and the variety parameters is calculated; generating an updated set of candidate varieties based on the similarity; According to the updated candidate variety set, a simulated annealing algorithm is used to adjust the experimental field site combination; Based on the adjusted site selection combination, a set of experimental field site selection plans is generated.

[0013] Furthermore, generating a test field distribution map including plot spatial coordinates, microclimate zoning feature sets, and matching varieties includes: Extracting the site selection plan with the highest score from the set of site selection plans for the experimental field; According to the site selection plan, obtaining the spatial coordinates of the plot; According to the site selection plan, a microclimate zoning feature set is obtained; According to the site selection plan, a set of matching candidate varieties is obtained; generating a test field distribution map based on the plot spatial coordinates, the microclimate zoning feature set, and the candidate variety set; According to the experimental field distribution map, the experimental field site selection location and variety configuration results are determined.

[0014] Based on the above, the present invention provides a method for selecting breeding test plots based on the microclimate characteristics of mountainous and hilly areas. Compared with the existing technology, this method collects microclimate data from mountainous and hilly areas, constructs a microclimate dataset, and generates a gradient distribution map of microclimate environmental factors. The gradient distribution map of microclimate environmental factors is then partitioned to generate a microclimate partition feature set, and a set of candidate varieties is screened. Ultimately, a distribution map containing plot coordinates, partition features, and matching varieties is optimized and generated, achieving the refined collection and analysis of microclimate data in mountainous and hilly areas, improving the spatial intensification and accuracy of test plot site selection, and providing effective technical support for the layout of agricultural production in mountainous and hilly areas. It shortens the spatial scope of seed production in breeding test plots and seed production fields with different environmental factors, improves convenience, and reduces breeding and seed production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be derived from these drawings without inventive effort. The positional relationships described in the drawings in the following description are based on the orientation of the components in the drawings unless otherwise specified.

[0016] Figure 1 A schematic flow chart of a method for selecting a breeding test field based on microclimate characteristics in mountainous and hilly areas according to one embodiment of the present invention; Figure 2 A schematic diagram of experimental fields in different alpine climate zones provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0018] In the description of the present invention, it should be noted that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance, or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more. In addition, the term "including" and any variations thereof all mean "at least including".

[0019] See also Figure 1 , Figure 1 A schematic flow chart of a method for selecting breeding test fields based on microclimate characteristics in mountainous and hilly areas, provided in one embodiment of the present invention.

[0020] To address the aforementioned shortcomings of existing breeding trial methods, or to achieve at least one of the aforementioned advantages or other advantages, one embodiment of the present invention provides a method for selecting breeding trial plots based on the microclimate characteristics of mountainous and hilly areas. As shown in the figure, this method for selecting breeding trial plots based on the microclimate characteristics of mountainous and hilly areas includes the following steps: Collect the microclimate data of the plot, construct a microclimate dataset and generate a gradient distribution map of microclimate environmental factors.

[0021] In practice, microclimate data for corresponding plots in mountainous and hilly areas will be collected through a sensor network. IoT nodes will be deployed in mountainous and hilly areas, each integrating equipment such as air temperature and humidity sensors, soil moisture probes, and photosynthetically active radiometers.

[0022] For example, nine monitoring points are set up in an altitude range of 800-1200 meters. Data is transmitted every 15 minutes via LoRa wireless networking, forming a standardized data set containing latitude and longitude coordinates, altitude, and timestamps. This three-dimensional deployment can capture special climate phenomena such as inversion layers and cloud belts, solving the problem that traditional single-point monitoring cannot reflect vertical gradient variations.

[0023] The collected raw data was processed using a Kalman filter to reduce noise, remove outliers, and create a spatiotemporal matrix. For example, analysis of data from the 2023 rainy season for a tea tree breeding project revealed that for every 100-meter increase in altitude, the average daily temperature dropped by 0.65°C, but the frequency of foggy days increased by 18%. This nonlinear relationship requires spatial interpolation to accurately represent.

[0024] When using the Kriging algorithm, elevation is introduced as a covariate to generate a temperature distribution map with a resolution of 1 meter, which improves the accuracy by 23% compared with the ordinary inverse distance weighted method.

[0025] Of course, it's understandable that the radiation redistribution effect caused by the ridge-valley topography requires special attention. For example, plots on the eastern slope where actual radiation received is 12% lower than the theoretical value should be separately marked. At 1,100 meters above sea level, the effective accumulated temperature is 156°C / d lower than at 900 meters, but the UV intensity is 15% higher. This quantitative result provides a direct basis for the selection and breeding of cold-resistant varieties.

[0026] A fusion model for meteorological station observation data and satellite remote sensing data was established. For example, in the karst mountainous areas of Guizhou, MODIS surface temperature products were spatially scaled with ground sensor data, and all-weather soil temperature fields were generated using the Bayesian maximum entropy method. This method compensated for the inability to conduct drone aerial surveys during rainy weather. This method reduced the root mean square error of the interpolation results from 2.1°C to 0.8°C.

[0027] The resulting gradient distribution map was linked to a crop phenotypic database. A Sichuan wheat breeding case study showed that differences in heading date matched the interpolated altitude-temperature gradient by 91%, confirming the method's ability to accurately predict phenological responses. Specifically, it was found that for every 50-meter increase in altitude, the onset of ear differentiation in spring varieties was delayed by 1.2 days, providing a quantitative standard for vertical ecotyping of varieties.

[0028] The microclimate environmental factor gradient distribution map is partitioned to generate a microclimate partition feature set and calculate the corresponding candidate variety set.

[0029] In a specific implementation, a clustering algorithm is used to divide the microclimate into zones according to the gradient distribution map of the microclimate environmental factors, and a microclimate zone feature set is generated. The clustering algorithm is an unsupervised learning method that groups similar data points and is suitable for dividing microclimate zones.

[0030] For example, in the mountainous and hilly regions of southern and southwestern China, we collected data on microclimate parameters such as altitude, temperature, humidity, and light intensity to construct a gradient distribution map of microclimate environmental factors. Using these parameters as input features, we used the K-means clustering algorithm to group regions with similar microclimate conditions into the same cluster.

[0031] For example, an area with an altitude between 500 and 800 meters, an average annual temperature between 15°C and 18°C, and an annual precipitation between 1200 and 1500 mm can be classified as a microclimate zone. In this way, multiple microclimate zone feature sets are generated, each containing a range of typical climate parameters for that zone. Based on these microclimate zone feature sets, a similarity calculation algorithm is used to generate a set of candidate varieties. This similarity calculation algorithm is used to evaluate the similarities between different microclimate zones, thereby recommending suitable crop varieties for each zone.

[0032] For example, we can use the cosine similarity algorithm to calculate the similarity between the feature sets of two microclimate zones. Suppose the feature set for zone A is [altitude 600 meters, temperature 16°C, precipitation 1300 mm], and the feature set for zone B is [altitude 650 meters, temperature 17°C, precipitation 1400 mm]. By calculating their cosine similarity, we can obtain a similarity value. The closer the value is to 1, the more similar the microclimate conditions of the two zones are. Based on the similarity calculation results, we can generate a set of candidate varieties for each zone.

[0033] For example, if the similarity between Zone A and Zone B is high, then crop varieties that perform well in Zone A can also be recommended for testing in Zone B. This approach can effectively reduce breeding costs and increase the success rate of variety promotion.

[0034] Based on the candidate variety set, optimizing and generating a set of experimental field site selection plans; The implementation process begins with establishing a variety-environment compatibility matrix, encompassing at least parameters such as variety cold tolerance and water requirements. A genetic algorithm is then used to select candidate varieties at corresponding altitudes. GIS spatial analysis is then used to initially screen candidate plots in flat areas. The entropy weight method is then used to calculate the comprehensive scores of environmental factors for each plot, ultimately generating a set of site selection options.

[0035] Taking winter wheat as an example, five candidate varieties adapted to altitudes of 800-1200m were selected through genetic algorithms. The algorithm's objective function was set to maximize the matching degree between the variety and the microclimate zone (for example, the weight of temperature matching accounted for 60%, and the weight of soil pH adaptability accounted for 30%).

[0036] Based on GIS spatial analysis, 20 candidate plots were initially screened out, and then the comprehensive scores of environmental factors of each plot (such as annual average temperature of 12-16°C and soil organic matter >2%) were calculated using the entropy weight method, and finally a set of site selection plans was output.

[0037] Micro-meteorological stations were deployed at selected altitude gradients (e.g., 800m, 1000m, and 1200m), and data such as canopy temperature and photosynthetically active radiation were collected for two consecutive years. K-means clustering was used to divide the mountains into three microclimate zones, of which the mid-altitude zone (900-1100m) was identified as a high-frequency frost zone (e.g., annual occurrence >15 times), requiring the matching of cold-resistant varieties.

[0038] According to the experimental field site selection plan set, an experimental field distribution map including plot spatial coordinates, microclimate zoning feature sets and matching varieties is generated.

[0039] Through spatial overlay, selected plots (e.g., Plot 5 at 108.3°E, 33.8°N) were associated with microclimate zones and their appropriate varieties were labeled (e.g., the 1000 m plot was associated with the cold-resistant variety "Longmai No. 7"). Mapping was performed using a thermal layer to display temperature gradients and a contour layer to indicate elevation. The final output was a 1:5000 scale spatial configuration map of the breeding experimental fields.

[0040] By replacing traditional cross-regional distribution with an altitude gradient, the variety adaptability verification cycle is shortened by 40%, and the mountain isolation characteristics reduce the cross-pollination rate to below 0.3%.

[0041] In some preferred embodiments, collecting the microclimate data of the plot and constructing the microclimate dataset comprises the following steps: Deploy monitoring points in flat areas of mountainous and hilly regions; Collecting plot temperature, humidity, and light data through the monitoring points; generating microclimate data including a time series according to the temperature, humidity, and light data; generating microclimate data including the spatial coordinates according to the spatial coordinates of the monitoring points; Constructing a microclimate dataset according to the microclimate data including the time series and the microclimate data including the spatial coordinates; According to the microclimate dataset, the numerical distribution characteristics of temperature, humidity and light in different plots are determined.

[0042] When setting up monitoring points in flat areas of mountainous and hilly regions, it is first necessary to rationally plan the locations of the monitoring points based on the terrain characteristics and altitude gradient.

[0043] For example, in a typical mountainous and hilly area, all terraces with relatively flat terrain, good enclosure, and suitable for mechanized operations can be selected. Each terrace is numbered and monitoring points are assigned to facilitate subsequent microenvironmental measurements.

[0044] Specifically, the site selection method can refer to the "Site Selection Requirements and Monitoring Specifications for Farmland Information Monitoring Points" (GB / T 37802-2019), "Technology for Mechanization Transformation of Farmland in Hilly and Mountainous Areas" (NY / T 4256-2022), "Technical Specifications for Sample Collection and Preparation for Farmland Soil Environmental Quality Monitoring" (DB61 / T 1783-2023), and "Guidelines for the Construction and Evaluation of Agricultural Seed Breeding and Promotion of Cultivated Seed Breeding Bases" (GB / T 36210-2018).

[0045] Each monitoring point should be equipped with a temperature sensor, a humidity sensor, and a light sensor. These sensors can collect real-time temperature, humidity, and light data for the plot. This ensures that the collected data is representative and reflects the microclimate differences between different plots. When collecting temperature, humidity, and light data at the monitoring points, the sensors should collect data at regular intervals, for example, once an hour. The collected data includes temperature (in degrees Celsius), humidity (in percentage), and light intensity (in μmol / m² / s). This data is transmitted to the data center via wireless transmission technology to ensure real-time and accuracy.

[0046] For example, at 10:00 AM on a particular day, monitoring point A records a temperature of 20°C, humidity of 65%, and light intensity of 1200 μmol / m² / s. This data is then recorded and transmitted. To generate microclimate data containing a time series based on temperature, humidity, and light data, the collected data must be organized and stored in chronological order.

[0047] For example, the temperature data collected at monitoring point A over a single day can form a time series: from 6:00 AM to 6:00 PM, the temperature gradually rises from 15°C to 25°C, then gradually decreases. Humidity data might decrease from 70% in the morning to 50% at noon, then rise again. Light intensity data might increase from 500 μmol / m² / s in the morning to 1500 μmol / m² / s at noon, then gradually decrease. These time series data can reflect the microclimate changes in a plot of land over the course of a day. When generating microclimate data containing spatial coordinates based on the spatial coordinates of the monitoring points, it is necessary to associate the geographic location information of each monitoring point with the collected microclimate data.

[0048] For example, the coordinates of monitoring point A are 30.5 degrees north latitude, 110.2 degrees east longitude, and an altitude of 500 meters. The coordinates of monitoring point B are 30.5 degrees north latitude, 110.2 degrees east longitude, and an altitude of 600 meters. In this way, a microclimate dataset containing spatial coordinates can be generated, with each data point containing geographic location information and corresponding microclimate data. When constructing a microclimate dataset based on microclimate data containing time series and microclimate data containing spatial coordinates, the time series data and spatial coordinate data need to be integrated.

[0049] For example, the temperature, humidity, and light data collected at monitoring point A over a single day can form a time series, while the location of monitoring point A provides the spatial coordinates. By integrating this data, a complete microclimate dataset can be constructed, encompassing both temporal changes and spatial distributions. Based on this microclimate dataset, the distribution of temperature, humidity, and light values ​​across different plots can be determined. Data analysis methods can then be used to extract the microclimate characteristics of each plot.

[0050] For example, by analyzing the temperature data at monitoring points A and B, we can find that the average temperature at monitoring point A is 20°C, while the average temperature at monitoring point B is 18°C, indicating that the temperature gradually decreases with increasing altitude. Humidity data may show that the average humidity at monitoring point A is 60%, while the average humidity at monitoring point B is 65%, indicating that humidity gradually increases with increasing altitude. Light intensity data may show that the average light intensity at monitoring point A is 1200μmol / m² / s, while the average light intensity at monitoring point B is 1100μmol / m² / s, indicating that light intensity gradually decreases with increasing altitude. Through these analyses, the numerical distribution characteristics of microclimates in different plots can be determined, providing a scientific basis for crop breeding.

[0051] In some preferred embodiments, generating a microclimate environmental factor gradient distribution map comprises the following steps: extracting temperature gradient data, humidity gradient data, and light gradient data from the microclimate dataset; generating a microclimate environmental factor gradient distribution map according to the temperature gradient data, the humidity gradient data, and the light gradient data; According to the microclimate environmental factor gradient distribution map, the spatial variation characteristics of temperature, humidity and light are characterized.

[0052] When extracting temperature, humidity, and light gradient data from microclimate datasets, temperature data can be obtained using temperature sensors placed at different altitudes, humidity data can be recorded using hygrometers or weather stations, and light data can rely on light sensors or satellite remote sensing technology. These data are typically stored as time series, containing observations at different points in time. When extracting data, it is important to select data from specific time periods or locations based on research needs to ensure representativeness and completeness.

[0053] For example, when studying the microclimate at different altitudes in a mountainous area, summer and winter temperature data can be extracted to analyze the impact of seasonal changes on temperature gradients. For example, in mountainous environments, the temperature typically drops by about 0.6°C for every 100-meter increase in altitude.

[0054] On the windward slopes of mountains, humidity is usually higher due to more precipitation, while on the leeward slopes, humidity is relatively lower. For example, rice requires higher humidity, while wheat is more adapted to dry environments.

[0055] Light intensity is often affected by factors such as topography, cloud cover, and solar altitude. For example, on the southern slopes of mountains, light intensity is typically higher due to stronger solar radiation, while on the northern slopes, light intensity is relatively lower. This light gradient information is important for crop breeding, as light is a key factor influencing crop photosynthesis and growth.

[0056] For example, corn requires ample sunlight, while tea trees are more suited to semi-shaded environments. When generating a microclimate environmental factor gradient distribution map based on temperature gradients, humidity gradients, and light intensity differences, the above calculation results need to be comprehensively analyzed and visualized. Microclimate environmental factor gradient distribution maps are usually presented in the form of maps, using different colors or contour lines to represent the gradient changes in temperature, humidity, and light.

[0057] For example, in a gradient distribution map of microclimate factors in a mountainous area, red could represent high temperatures and blue could represent low temperatures; dark green could represent high humidity and light green could represent low humidity; and yellow could represent high light levels and gray could represent low light levels. This type of map can intuitively demonstrate the spatial variations in microclimate, providing a scientific basis for crop breeding. When characterizing the spatial variations in temperature, humidity, and light using a gradient distribution map of microclimate factors, analysis must be conducted in conjunction with specific topographic and climatic conditions.

[0058] For example, a distribution map of microclimatic environmental factors in a particular mountainous region shows a trend of decreasing temperature with increasing altitude, significant differences in humidity between windward and leeward slopes, and a clear contrast in sunlight between south and north slopes. These variations can help breeders select crop varieties suited to different microclimates, improving the precision and adaptability of breeding.

[0059] For example, cold-resistant crop varieties can be selected in high-altitude and low-temperature areas, moisture-resistant crop varieties can be selected in low-altitude and high-humidity areas, and light-loving crop varieties can be selected in high-light areas on the south slope.

[0060] In some preferred embodiments, partitioning the microclimate environmental factor gradient distribution map to generate a microclimate partition feature set comprises the following steps: Extracting the temperature gradient data, the humidity gradient data, and the light gradient data from the microclimate environmental factor gradient distribution map; If the temperature gradient data, the humidity gradient data, or the light gradient data exceeds a preset threshold, a K-means clustering algorithm is used to divide the microclimate into zones; According to the microclimate zones, calculating the mean and variance of the temperature in each zone; According to the microclimate zones, the mean and variance of humidity in each zone are calculated; According to the microclimate zones, calculating the mean and variance of the light intensity in each zone; A microclimate zoning feature set is generated based on the mean and variance.

[0061] Microclimate data at different altitudes in mountainous and hilly areas is collected. A sensor network collects data on light, temperature, humidity, and soil factors to generate a high-resolution environmental factor dataset. Based on this dataset, spatial interpolation is performed using a geographic information system to generate temperature, humidity, and light gradient distribution maps. If the temperature, humidity, or light gradient exceeds a preset threshold, the K-means clustering algorithm is used to partition the distribution map to obtain preliminary microclimate zones.

[0062] Based on the preliminary microclimate zoning, the mean and variance of temperature within each zone are calculated to generate a temperature characteristic parameter set. Based on the preliminary microclimate zoning, the mean and variance of humidity within each zone are calculated to generate a humidity characteristic parameter set. Based on the preliminary microclimate zoning, the mean and variance of light within each zone are calculated to generate a light characteristic parameter set.

[0063] By integrating the temperature, humidity, and light characteristic parameter sets, a weighted algorithm was used to generate a comprehensive microclimate zoning feature set. Based on this comprehensive microclimate zoning feature set, a decision tree algorithm was used to model the environment-phenotype association between the zoning regions and determine the crop phenotypic adaptability corresponding to each zoning region.

[0064] Through the environment-phenotype association model, an optimization algorithm is used to screen out crop varieties suitable for each microclimate zone, and obtain a precise and wide-adaptability breeding plan.

[0065] Specifically, by deploying sensor networks in mountainous and hilly areas, we collect data on environmental factors such as light intensity, temperature, humidity, and soil nutrients to obtain a high-resolution environmental factor dataset.

[0066] A geographic information system (GIS) was used to generate temperature, humidity, and light gradient maps from the collected data, with a resolution of 100 m x 100 m. If the temperature gradient exceeded 5°C / 100 m, the humidity gradient exceeded 10% / 100 m, or the light gradient exceeded 500 lux / 100 m, the K-means clustering algorithm was used to partition the distribution map, setting the number of clusters to 5 to obtain preliminary microclimate zones. Based on these preliminary zoning results, the mean and variance of the temperature within each zone were calculated. For example, if the mean temperature in a zone was 18°C ​​and the variance was 2.5, a set of temperature characteristic parameters was generated.

[0067] Calculate the mean and variance of humidity within each partition. For example, if the mean humidity in a certain partition is 75% and the variance is 3.2, generate a humidity feature parameter set. Calculate the mean and variance of light within each partition. For example, if the mean light in a certain partition is 12,000 lux and the variance is 800, generate a light feature parameter set.

[0068] By fusing the temperature, humidity, and light characteristic parameter sets, a weighted algorithm was used to generate a comprehensive microclimate zoning feature set with weights of 0.4, 0.3, and 0.3, respectively.

[0069] Based on the comprehensive feature set, the CART decision tree algorithm was used to model the environmental-phenotypic associations of the sub-regions, determining the phenotypic adaptability of the crops corresponding to each sub-region. For example, a sub-region was suitable for growing cold-tolerant crops. Using this environmental-phenotypic association model, a genetic optimization algorithm was used to select crop varieties suitable for each microclimate sub-region. For example, in sub-regions with an average temperature of 18°C ​​and an average humidity of 75%, rice varieties with high cold and humidity tolerance were selected, resulting in a precise breeding plan with broad adaptability.

[0070] In some preferred embodiments, the calculation of the corresponding candidate variety set includes: Extracting the mean and variance of temperature, humidity, and light from the microclimate partition feature set; Extract the temperature adaptation range of varieties from the crop variety resource library; Extract the humidity adaptation range of varieties from the crop variety database; Extract the light adaptation range of varieties from the crop variety resource library; The weighted Euclidean distance algorithm is used to calculate the similarity between the variety parameters and the microclimate zoning feature set; Based on the similarity, a set of candidate varieties for each partition is generated.

[0071] The mean and variance of temperature, humidity, and light were extracted from the microclimate zoning feature set to generate a statistical feature set for each zoning environmental factor. Based on the crop variety resource library, the temperature, humidity, and light adaptation ranges of each variety were obtained to determine the variety's environmental adaptation parameter set. A weighted Euclidean distance algorithm was used to calculate the similarity between the variety's environmental adaptation parameter set and the microclimate zoning statistical feature set, resulting in a similarity score for each variety and zoning.

[0072] If the similarity score is greater than the preset threshold, the corresponding variety will be included in the candidate variety set to generate the initial candidate variety set for each partition.

[0073] Based on elevation gradient data for vertical mountain climate zones, we obtain multidimensional environmental variables for each sub-region, including light, temperature, water, and soil, and determine the sub-region's environmental characteristic vector. Using an environment-phenotype correlation model, we calculate the predicted phenotypic performance of the initial candidate variety set under the sub-region's environmental characteristic vector, generating a phenotypic adaptability score. If the phenotypic adaptability score meets the sub-region's environmental adaptability requirements, the corresponding variety is retained, generating an optimized set of candidate varieties.

[0074] Based on the optimized set of candidate varieties, a clustering algorithm was used to group the varieties according to their environmental adaptability, resulting in a final subset of candidate varieties for each sub-area. By matching the final subset of candidate varieties with the environmental data of the mountain experimental fields, the variety breeding plan for each sub-area experimental field was determined.

[0075] Specifically, the mean and variance of temperature, humidity, and illumination are extracted from the microclimate zoning feature set. For example, in the zones at an altitude of 500 meters, 800 meters, and 1100 meters, the mean temperatures are 18°C, 15°C, and 12°C, respectively; the mean humidity is 75%, 80%, and 85%, respectively; and the mean illumination is 1200 Lux, 1000 Lux, and 800 Lux, respectively. The variance is calculated using historical meteorological data.

[0076] Using the crop variety resource library, we retrieved the environmental adaptation parameters of each variety. For example, a rice variety's temperature adaptation range is 14-22°C, humidity adaptation range is 70-90%, and light adaptation range is 800-1500 Lux. Using a weighted Euclidean distance algorithm, with a temperature weight of 0.5, a humidity weight of 0.3, and a light weight of 0.2, we calculated the similarity between the variety parameters and the partition characteristics. If the similarity score exceeded the 0.85 threshold, the variety was included in the candidate set. Based on mountain vertical climate zone data, we extracted environmental variables such as soil pH, precipitation, and wind speed along the altitude gradient to construct a partitioned environmental vector containing 10 features.

[0077] Using an environment-phenotype association model, candidate varieties were input with their genotype data and partitioned environmental vectors to predict their yield and stress resistance phenotypes. Varieties with predicted yields ≥500 kg / mu and stress resistance scores ≥8 were retained. Using a K-means clustering algorithm, the optimized candidate varieties were divided into three groups based on their environmental adaptability scores, each representing a different adaptation type. The clustering results were then matched with real-time monitoring data from the experimental fields. For example, if the soil nitrogen content in a partition was 2.5 g / kg, a subset of varieties adapted to high-nitrogen environments was selected for breeding.

[0078] In some preferred embodiments, optimizing and generating a set of experimental field site selection plans based on the set of candidate varieties includes: extracting variety parameters from the candidate variety set; Genetic algorithm was used to optimize the site selection combination of experimental fields; Calculating the scores of the site combinations based on the representativeness of the microclimate zone feature set; Calculating a score for the site selection combination based on the similarity between the variety parameters and the microclimate zoning feature set; and generating a set of experimental field site selection schemes based on the score; If the score of the experimental field site selection scheme set is lower than a preset threshold, the site selection combination is adjusted.

[0079] Variety parameters are extracted from the candidate variety set to obtain a variety parameter set. Based on the variety parameter set, a genetic algorithm is used to generate initial experimental field site combinations to obtain a site combination set. The microclimate representativeness score of each combination in the site combination set is calculated using the microclimate zoning feature set to obtain a representative score set. Based on the similarity between the variety parameter set and the microclimate zoning feature set, a matching score set is calculated for each combination in the site combination set to obtain a matching score set. By weighted fusion of the representative score set and the matching score set, a comprehensive score is calculated for each combination in the site combination set to obtain a comprehensive score set. Based on the comprehensive score set, the site combination set is sorted to obtain a sorted set of site selection options.

[0080] If the highest comprehensive score in the sorted site selection set falls below a preset threshold, the genetic algorithm parameters are adjusted and a new set of site selection combinations is generated. By iteratively optimizing the set of site selection combinations and screening for combinations with comprehensive scores above the preset threshold, an optimized set of experimental field site selection options is obtained. Based on this optimized set of experimental field site selection options, an environmental-phenotype correlation model is constructed by correlating light, temperature, water, and soil ecological factors under varying altitude gradients, resulting in a precise breeding experimental field site selection plan.

[0081] Specifically, variety parameters, such as cold tolerance, drought tolerance, and growing period, are extracted from the candidate variety set to form a variety parameter set. Based on this variety parameter set, a genetic algorithm is used to generate initial experimental field site combinations. For example, a population size of 100, a crossover probability of 0.8, and a mutation probability of 0.1 are set to generate a set of 50 site combinations.

[0082] The microclimate representativeness score of each combination in the site selection combination set is calculated using the microclimate zoning feature set. For example, the standard deviation of factors such as altitude, temperature, and humidity for each combination is calculated using GIS data to obtain a representative score set.

[0083] According to the similarity between the variety parameter set and the microclimate zoning feature set, the matching score of each combination in the site selection combination set is calculated. For example, the cosine similarity algorithm is used to calculate the matching degree between the variety cold resistance and the temperature factor to obtain the matching score set.

[0084] By weighting the representative score set and the matching score set, a comprehensive score is calculated for each combination in the site selection combination set. For example, a representative weight of 0.6 and a matching weight of 0.4 are set to calculate the comprehensive score set. Based on the comprehensive score set, the site selection combination set is sorted, for example, from high to low by comprehensive score, to obtain a sorted set of site selection options.

[0085] If the highest comprehensive score in the sorted site selection plan set is lower than a preset threshold, such as a threshold of 0.85, the genetic algorithm parameters are adjusted, such as adjusting the population size to 150, and a new site selection combination set is generated.

[0086] By iteratively optimizing the site selection combination set, we screen combinations with comprehensive scores higher than the preset threshold, for example, we screen out 10 combinations with comprehensive scores greater than 0.85, and obtain an optimized set of experimental field site selection schemes. Based on the optimized set of experimental field site selection schemes, we associate the light, temperature, water, soil and ecological factors under the altitude gradient, for example, we use multiple regression analysis to build an environmental phenotype association model, and obtain a precise breeding experimental field site selection plan. In some preferred embodiments, optimizing and generating a set of experimental field site selection plans based on the set of candidate varieties further comprises: Extracting plots with scores lower than a preset threshold from the set of experimental field site selection plans; Selecting a zone with a second highest gradient value from the microclimate zone feature set; Using a weighted Euclidean distance algorithm, the similarity between the partition and the variety parameters is calculated; generating an updated set of candidate varieties based on the similarity; According to the updated candidate variety set, a simulated annealing algorithm is used to adjust the experimental field site combination; Based on the adjusted site selection combination, a set of experimental field site selection plans is generated.

[0087] A set of candidate varieties and experimental plot site selection plans was obtained. Data cleaning techniques were used to remove duplicates and missing values ​​to obtain a standardized candidate dataset. Based on the standardized candidate dataset, a preset threshold screening rule was used to extract plots with scores below the threshold, thus determining the set of low-scoring plots.

[0088] Obtain a set of microclimate zoning characteristics, use a gradient analysis algorithm to sort the zoning gradient values, select the zoning with the next highest gradient value, and obtain the target zoning characteristics. Based on the target zoning characteristics and candidate cultivar parameters, use a weighted Euclidean distance algorithm to calculate the similarity between the zoning and cultivar parameters, and determine a similarity score set.

[0089] Based on the similarity score set, the candidate variety set was updated using sorting and screening rules to obtain an updated candidate variety set. Based on the updated candidate variety set, the simulated annealing algorithm was used to adjust the experimental field site combinations to generate an optimized site combination set. Based on the optimized site combination set, the environmental adaptability of the site combinations was evaluated using a multidimensional environment-phenotype association model to obtain a set of adaptability scores.

[0090] If any combination in the adaptability score set falls below a preset threshold, an iterative optimization algorithm is used to readjust the site selection combinations to obtain an updated set of site selection combinations. Based on this updated set of site selection combinations, data integration technology is used to generate the final set of experimental field site selection plans and determine the experimental field site selection plan.

[0091] Specifically, a set of candidate varieties and a set of experimental field site selection plans were obtained from the database of light, temperature, soil, and air environmental factors in mountain experimental fields. The Python-based Pandas library was used for data cleaning, and duplicate records and plots with more than 30% missing values ​​were eliminated to form a standardized candidate data set.

[0092] Based on the standardized dataset, a scoring threshold of 0.7 was set to screen out plots with light intensity below 1500 lux or soil pH values ​​deviating from the range of 6.0-7.5, generating a low-scoring plot set. Data for zones with an elevation gradient of 800-1200 meters were extracted from the microclimate zone feature set. K-means clustering was used to sort the zones by temperature gradient (0.6°C per 100 meters), and the second-level zone with a gradient value of 0.8 was selected as the target zone.

[0093] A weighted Euclidean distance algorithm was used to calculate the similarity between the target partition and the rice variety "Huayou No. 1," with a temperature weight of 0.4, a humidity weight of 0.3, and a soil organic matter weight of 0.3. This output was a set of similarity scores. The similarity score set was sorted in descending order, retaining the top 20% of varieties. The candidate variety set was then updated based on soil nitrogen data from low-scoring plots.

[0094] In some preferred embodiments, generating a test field distribution map including plot spatial coordinates, microclimate zone feature sets, and matching varieties includes: Extracting the site selection plan with the highest score from the set of site selection plans for the experimental field; According to the site selection plan, obtaining the spatial coordinates of the plot; According to the site selection plan, a microclimate zoning feature set is obtained; According to the site selection plan, a set of matching candidate varieties is obtained; generating a test field distribution map based on the plot spatial coordinates, the microclimate zoning feature set, and the candidate variety set; According to the experimental field distribution map, the experimental field site selection location and variety configuration results are determined.

[0095] The site selection plan with the highest score was extracted from the set of experimental field site selection plans. A scoring algorithm based on multidimensional environmental factors was used, combining the weights of factors such as photosynthetically active radiation, temperature, soil moisture content, pH value, organic matter content and available nitrogen to obtain the site selection plan with the highest score.

[0096] The spatial coordinates of the plots were obtained from the top-scoring site selection proposal. The longitude, latitude, and altitude of each experimental field plot in the site selection proposal were extracted through a geographic information system interface to determine the set of plot spatial coordinates. A microclimate zoning feature set was obtained from the top-scoring site selection proposal. Cluster analysis was used to zonate the environmental factors (light, temperature, water, soil, etc.) for the plots to generate the microclimate zoning feature set.

[0097] If the number of zones in the microclimate zoning feature set exceeds the number of candidate varieties, the environment-phenotype association model is used to screen for varieties that closely match the characteristics of each zone, resulting in a set of matching candidate varieties. A spatial visualization algorithm is used to generate a distribution map of the experimental fields based on the spatial coordinates of the plots, the microclimate zoning feature set, and the candidate variety set. A preliminary distribution map of the experimental fields is generated by overlaying the plot coordinates and the microclimate zoning feature rendering.

[0098] Based on the preliminary experimental field distribution map, the difference data of environmental factors between adjacent plots were obtained, and the environmental independence between plots was determined by calculating the spatial gradients of photosynthetically active radiation, day and night temperature difference and soil moisture content.

[0099] If the environmental independence between plots meets the preset threshold, the genetic algorithm is used to optimize the variety configuration, combining the candidate variety set and the microclimate zoning feature set to obtain the experimental field site selection location and variety configuration results.

[0100] Based on the experimental field site selection and variety configuration results, spatial variation data for mountain vertical climate zones was obtained. Interpolation analysis was used to supplement the microclimate zoning feature set to obtain an optimized experimental field distribution map. Based on this optimized experimental field distribution map, 3D modeling technology was used to generate a final experimental field distribution map containing the spatial coordinates of the plots, the microclimate zoning feature set, and the matching varieties, thus determining the final site selection and variety configuration results.

[0101] Specifically, when extracting the site selection plan with the highest score from the experimental field site selection plan set, a weighted scoring algorithm is used, setting the photosynthetic active radiation (PAR) weight to 30%, the temperature weight to 25%, the soil moisture weight to 20%, the pH value weight to 15%, the organic matter content weight to 5%, and the available nitrogen weight to 5%. The comprehensive score of each site selection plan is calculated, and the plan with a score of more than 85 points is selected as the highest-scoring site selection plan.

[0102] High-precision terrain data (resolution 10m×10m) were called through the geographic information system (GIS) interface to extract the latitude and longitude coordinates and altitude of each plot in the site selection plan (e.g., 29°15' north latitude, 107°05' east longitude, 800m above sea level) to form a set of plot spatial coordinates.

[0103] K-means cluster analysis (number of clusters k = 5) was performed on the light, temperature, soil and other data of each plot to generate a microclimate zoning feature set.

[0104] If the number of zones (e.g., 5) exceeds the number of candidate varieties (e.g., 3), the matching degree between each zone and the variety is calculated using an environment-phenotype association model to select a set of matching varieties. Using ArcGIS spatial visualization tools, the plot coordinates are overlaid with the microclimate zone data (e.g., red indicates high-temperature, low-humidity areas, blue indicates low-temperature, high-humidity areas) to generate a preliminary experimental field distribution map.

[0105] The spatial gradients of PAR differences, temperature differences, and soil moisture differences between adjacent plots were calculated. If more than 90% of the plots met the independence threshold, the NSGA-II genetic algorithm was used to optimize the variety configuration and output the Pareto optimal solution. Based on the mountain vertical climate zone data (the temperature drops by 0.6°C for every 100m increase in altitude), microclimate characteristics were supplemented through Kriging interpolation to generate an optimized distribution map. Finally, a three-dimensional model was constructed using DEM data, annotating the plot coordinates (e.g., plot A1: 850m altitude, PAR 1500μmol / m² / s), zoning characteristics (e.g., zone B: pH 6.2, organic matter 3.1%), and matching varieties (e.g., rice variety "Cold-resistant No. 1") to output the final experimental field distribution map. In summary, the present invention provides a method for selecting breeding test plots based on the microclimate characteristics of mountainous and hilly areas. Compared with the existing technology, the present invention collects microclimate data from mountainous and hilly areas, constructs a microclimate data set, generates a gradient distribution map of microclimate environmental factors, partitions the gradient distribution map of microclimate environmental factors, generates a microclimate partition feature set, and screens a set of candidate varieties. Finally, a distribution map containing plot coordinates, partition features, and matching varieties is optimized and generated, achieving the refined collection and analysis of microclimate data in mountainous and hilly areas, improving the spatial intensification and accuracy of test plot site selection, providing effective technical support for the layout of agricultural production in mountainous and hilly areas, shortening the spatial scope of seed production in breeding test plots and seed production fields with different environmental factors, improving convenience, and reducing breeding and seed production costs.

[0106] In addition, those skilled in the art should understand that, although there are many problems in the prior art, each embodiment or technical solution of the present invention may be improved in only one or several aspects, without having to simultaneously solve all the technical problems listed in the prior art or background art. Those skilled in the art should understand that any content not mentioned in a claim should not be construed as limiting the claim.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for selecting breeding experimental fields based on the microclimate characteristics of mountainous and hilly areas, characterized by: The following steps are involved: Collect microclimate data of the plot, construct a microclimate dataset and generate a gradient distribution map of microclimate environmental factors; Partitioning the microclimate environmental factor gradient distribution map, generating a microclimate partition feature set and calculating a corresponding candidate variety set; Based on the candidate variety set, optimizing and generating a set of experimental field site selection plans; According to the experimental field site selection plan set, an experimental field distribution map including plot spatial coordinates, microclimate zoning feature sets and matching varieties is generated.

2. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: The microclimate data includes at least one or more of temperature, humidity, air flow velocity, and thermal radiation.

3. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: The microclimate dataset at least includes a time series and spatial coordinates.

4. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: The microclimate data of the plots are collected and the microclimate dataset is constructed, which includes the following steps: Deploy monitoring points in flat areas of mountainous and hilly regions; Collecting plot temperature, humidity, and light data through the monitoring points; generating microclimate data including a time series according to the temperature, humidity, and light data; generating microclimate data including the spatial coordinates according to the spatial coordinates of the monitoring points; Constructing a microclimate dataset according to the microclimate data including the time series and the microclimate data including the spatial coordinates; According to the microclimate dataset, the numerical distribution characteristics of temperature, humidity and light in different plots are determined.

5. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: Generating a microclimate environmental factor gradient distribution map comprises the following steps: extracting temperature gradient data, humidity gradient data, and light gradient data from the microclimate dataset; Generating a microclimate environmental factor gradient distribution map according to the temperature gradient, the humidity gradient, and the light gradient; According to the microclimate environmental factor gradient distribution map, the spatial variation characteristics of temperature, humidity and light are characterized.

6. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: Partitioning the microclimate environmental factor gradient distribution map to generate a microclimate partition feature set comprises the following steps: Extracting the temperature gradient data, the humidity gradient data, and the light gradient data from the microclimate environmental factor gradient distribution map; If the temperature gradient data, the humidity gradient data, or the light gradient data exceeds a preset threshold, a K-means clustering algorithm is used to divide the microclimate into zones; According to the microclimate zones, calculating the mean and variance of the temperature in each zone; According to the microclimate zones, the mean and variance of humidity in each zone are calculated; According to the microclimate zones, calculating the mean and variance of the light intensity in each zone; A microclimate zoning feature set is generated based on the mean and variance.

7. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: The candidate variety set corresponding to the calculation includes: Extracting the mean and variance of temperature, humidity, and light from the microclimate partition feature set; Extract the temperature adaptation range of varieties from the crop variety resource library; Extract the humidity adaptation range of varieties from the crop variety database; Extract the light adaptation range of varieties from the crop variety resource library; The weighted Euclidean distance algorithm is used to calculate the similarity between the variety parameters and the microclimate zoning feature set; Based on the similarity, a set of candidate varieties for each partition is generated.

8. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: The optimizing and generating a set of experimental field site selection schemes based on the set of candidate varieties includes: extracting variety parameters from the candidate variety set; Genetic algorithm was used to optimize the site selection combination of experimental fields; Calculating the scores of the site combinations based on the representativeness of the microclimate zone feature set; Calculating a score for the site selection combination based on the similarity between the variety parameters and the microclimate zoning feature set; and generating a set of experimental field site selection schemes based on the score; If the score of the experimental field site selection scheme set is lower than a preset threshold, the site selection combination is adjusted.

9. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 8, characterized in that: The step of optimizing and generating a set of experimental field site selection plans based on the set of candidate varieties further comprises: Extracting plots with scores lower than a preset threshold from the set of experimental field site selection plans; Selecting a zone with a second highest gradient value from the microclimate zone feature set; Using a weighted Euclidean distance algorithm, the similarity between the partition and the variety parameters is calculated; generating an updated set of candidate varieties based on the similarity; According to the updated candidate variety set, a simulated annealing algorithm is used to adjust the experimental field site combination; Based on the adjusted site selection combination, a set of experimental field site selection plans is generated.

10. The method for selecting breeding experimental fields based on microclimate characteristics in mountainous and hilly areas according to claim 1, characterized in that: The generating of the experimental field distribution map including the plot spatial coordinates, the microclimate zoning feature set and the matching varieties includes: Extracting the site selection plan with the highest score from the set of site selection plans for the experimental field; According to the site selection plan, obtaining the spatial coordinates of the plot; According to the site selection plan, a microclimate zoning feature set is obtained; According to the site selection plan, a set of matching candidate varieties is obtained; generating a test field distribution map based on the plot spatial coordinates, the microclimate zoning feature set, and the candidate variety set; According to the experimental field distribution map, the experimental field site selection location and variety configuration results are determined.

Images