Differential restoration method for grassland vegetation under photovoltaic panels and between panels

Abstract

The invention relates to the field of photovoltaic technology, and discloses a differential restoration method for grassland vegetation under photovoltaic panels and between the panels, and the method comprises the following steps: collecting grassland vegetation state data of a region under the photovoltaic panels and a region between the panels, including vegetation coverage, species composition, species diversity, soil parameters and illumination intensity data; under-panel and inter-panel environment difference analysis processing is carried out based on the grassland vegetation state data, environment difference data is generated, the grassland vegetation state data of an under-photovoltaic-panel area and an inter-panel area are collected, the environment difference data is generated, and a differential remediation scheme is made based on the moisture distribution difference, the illumination shielding difference and the soil nutrient difference. It is guaranteed that different vegetation type selection and maintenance strategies are adopted for the under-plate shading environment and the inter-plate open environment, the problem that the restoration strategy is mismatched with the local microenvironment can be solved, pertinence and adaptability of vegetation restoration are guaranteed, the vegetation survival rate and community stability are improved, and the community degradation risk is reduced.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic technology, specifically to a method for differentiated restoration of grassland vegetation under and between photovoltaic panels. Background Technology

[0002] Photovoltaics is a new type of power generation that uses the photovoltaic effect of solar cell semiconductor materials to directly convert solar radiation energy into electrical energy.

[0003] Currently, due to the differences in light, moisture, and temperature gradients between the panels and the substrates within a photovoltaic power plant, the vegetation selection and maintenance measures adopted for unified restoration of grassland vegetation often lack specificity. This makes it impossible to assess and respond in real time to the dynamic impact of the shaded environment under the panels and the open environment between the panels on vegetation growth. When the restoration strategy adopted is mismatched with the local microenvironment, it can lead to low vegetation survival rates, community degradation, and weakened soil and water conservation functions, failing to guarantee the ecological stability and landscape consistency of vegetation restoration.

[0004] Therefore, a differentiated restoration method for grassland vegetation under and between photovoltaic panels is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a differentiated restoration method for grassland vegetation under and between photovoltaic panels, solving the problem mentioned in the background art of being unable to assess and respond in real time to the dynamic impact of the shaded environment under the panels and the open environment between the panels on vegetation growth.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for differentiated restoration of grassland vegetation under and between photovoltaic panels, the method comprising the following steps: S1. Collect grassland vegetation status data in the area under the photovoltaic panels and between the panels, including vegetation coverage, soil parameters and light intensity data; S2. Based on the grassland vegetation status data, perform environmental difference analysis and processing on the area under and between the boards to generate environmental difference data, which includes differences in water distribution, light shading, and soil nutrients. S3. Based on the environmental difference data, formulate differentiated remediation plans and generate remediation plan data. The remediation plan data includes a selection plan for shade-tolerant vegetation in the area under the boards and a selection plan for light-loving vegetation in the area between the boards. S4. Implement vegetation restoration measures based on the restoration plan data to generate restoration implementation data, which includes seeding amount, irrigation frequency and fertilization strategy. S5. Monitor the vegetation status after restoration based on the restoration implementation data, and generate restoration effect data, including vegetation growth rate and coverage changes. S6. Adjust the repair strategy based on the repair effect data to generate adjusted repair data, which includes dynamically optimized irrigation and replanting schemes. S7. Integrate the adjusted restoration data and output a vegetation restoration report, summarizing the parameters and effects of the entire restoration process.

[0007] Preferably, the step S1 of collecting grassland vegetation status data in the area under the photovoltaic panels and the area between the panels includes the following steps: S11. Collect vegetation coverage data in the area under the photovoltaic panels by using a drone equipped with a multispectral camera, and generate vegetation coverage data under the panels; S12. Collect soil moisture, pH and nutrient content data in the inter-plate area through a ground sensor network to generate inter-plate soil parameter data; S13. Measure the light intensity distribution under and between photovoltaic panels using a light sensor to generate light intensity data.

[0008] Preferably, generating environmental difference data in step S2 includes the following steps: S21. Obtain the vegetation coverage data under the board, the soil parameter data between the boards, and the light intensity data; S22. Use a numerical comparison algorithm to analyze the difference in moisture distribution between the area under the plate and the area between the plates, and generate moisture difference data. S23. Calculate the occlusion rate of the area under the board based on the light intensity data, and generate occlusion difference data. S24. Combine soil parameter data to assess the soil nutrient balance in the areas under and between the boards, and generate soil nutrient difference data.

[0009] Preferably, generating the repair plan data in step S3 includes the following steps: S31. Based on the aforementioned moisture difference data, light shading difference data, and soil nutrient difference data, determine that the area under the board is suitable for planting shade-tolerant grass species; S32. Based on the light advantages of the inter-board area, light-loving grass species were selected as the main restoration species; S33. Optimize the selection scheme by combining historical vegetation data, and generate a selection scheme for shade-tolerant vegetation under the board and a selection scheme for light-loving vegetation between the boards.

[0010] Preferably, generating repair implementation data in step S4 includes the following steps: S41. Based on the shade-tolerant vegetation selection scheme under the boards and the light-loving vegetation selection scheme between the boards, calculate the sowing density in the area under the boards and the sowing interval in the area between the boards. S42. Develop differentiated irrigation plans based on moisture difference data. Use high-frequency, low-volume irrigation for the area under the boards and low-frequency, high-volume irrigation for the area between the boards. S43. Configure the ratio of organic fertilizer and chemical fertilizer based on soil nutrient difference data to generate fertilization strategy data.

[0011] Preferably, the step S5 of generating repair effect data includes the following steps: S51. Monitor vegetation cover changes through regular drone aerial photography and generate cover change data; S52. Continuously collect soil parameters using soil sensors to generate dynamic soil data; S53. Combine cover change data and soil dynamic data to assess vegetation growth rate and generate restoration effect data.

[0012] Preferably, generating the adjusted and repaired data in step S6 includes the following steps: S61. When the vegetation growth rate in the restoration effect data is lower than the threshold, increase the irrigation frequency in the area under the board. S62. When the coverage of the inter-board area is insufficient, the reseeding mechanism is activated to optimize the seeding amount; S63. Adjust the fertilization strategy dynamically based on real-time data and generate adjusted repair data.

[0013] Preferably, the vegetation restoration report output in step S7 includes the following steps: S71. Summarize the environmental difference data, repair plan data, repair implementation data, repair effect data, and adjusted repair data; S72. Generate a visual report, including charts showing vegetation restoration trends and environmental impact assessments.

[0014] Preferably, the method further includes step S8: establishing a vegetation restoration database, storing and updating all restoration process data, and supporting long-term ecological analysis, wherein S8 includes the following steps: S81. Import the collected data into the cloud platform and use encryption algorithms to ensure data security; S82. Regularly back up data and set up access control for researchers; S83. Generate an annual repair effect comparison report based on the database.

[0015] Preferably, the method further includes step S9: constructing an intelligent restoration decision support system to optimize real-time decision-making in the vegetation restoration process, wherein S9 includes the following steps: S91. An integrated IoT monitoring device collects real-time micro-environmental data in the area under and between photovoltaic panels, including temperature gradient changes, humidity distribution patterns, and light intensity fluctuations. S92. Establish a vegetation growth prediction model based on historical restoration data, and analyze the adaptability of different vegetation types to the environment under and between the boards using machine learning algorithms. S93. Use multi-objective optimization algorithms to balance ecological and economic benefits and generate dynamic adjustment suggestions for restoration plans; S94. Establish a decision support interactive interface to visually display the comparison results of remediation schemes and the simulation of expected effects; S95. Set up an early warning mechanism to automatically trigger the re-evaluation process of the repair plan when the monitored data deviates from the expected threshold.

[0016] Compared with existing technologies, this invention provides a method for differentiated restoration of grassland vegetation under and between photovoltaic panels, which has the following beneficial effects: 1. In this invention, by collecting grassland vegetation status data in the area under photovoltaic panels and the area between panels and generating environmental difference data, differentiated restoration plans are formulated based on differences in water distribution, light shading, and soil nutrients. This ensures that different vegetation selection and maintenance strategies are adopted for the shaded environment under the panels and the open environment between the panels, which can solve the problem of mismatch between restoration strategies and local microenvironments, ensure the pertinence and adaptability of vegetation restoration, improve vegetation survival rate and community stability, and reduce the risk of community degradation.

[0017] 2. In this invention, restoration effect data is generated by real-time monitoring of the vegetation status after restoration, and the restoration strategy is dynamically adjusted based on changes in vegetation growth rate and coverage. This enables the system to evaluate the growth status of shade-tolerant vegetation under the boards and light-loving vegetation between the boards in real time and to implement precise intervention in a timely manner, solving the problem of delayed adjustment of restoration strategies. When signs of vegetation degradation appear in local areas, the system can automatically trigger irrigation frequency optimization and replanting mechanisms to ensure dynamic optimization and continuous effectiveness of the restoration process.

[0018] 3. In this invention, by establishing a vegetation restoration database to store data throughout the entire process and constructing an intelligent restoration decision support system, machine learning algorithms are used to analyze vegetation adaptability patterns and multi-objective optimization algorithms are used to generate dynamic adjustment suggestions, thereby achieving precise regional control of water and nutrients, reducing resource waste and soil problems caused by uniform irrigation and fertilization, and enabling the system to improve water and fertilizer utilization efficiency, ensuring the overall health of grassland vegetation and the long-term ecological service function of photovoltaic power plants. Attached Figure Description

[0019] Figure 1 This is a flowchart of a method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to the present invention. Detailed Implementation

[0020] 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.

[0021] Please see Figure 1 This method for differentiated restoration of grassland vegetation under and between photovoltaic panels includes the following steps: S1. Collect grassland vegetation status data in the area under the photovoltaic panels and between the panels, including vegetation coverage, soil parameters and light intensity data; S2. Based on grassland vegetation status data, perform environmental difference analysis and processing on the underside and between the boards to generate environmental difference data, which includes differences in water distribution, light shading, and soil nutrients. S3. Develop differentiated remediation plans based on environmental difference data and generate remediation plan data. The remediation plan data includes the selection plan for shade-tolerant vegetation in the area under the boards and the selection plan for light-loving vegetation in the area between the boards. S4. Implement vegetation restoration measures based on the restoration plan data and generate restoration implementation data, which includes seeding amount, irrigation frequency and fertilization strategy. S5. Based on the restoration implementation data, monitor the vegetation status after restoration and generate restoration effect data, including vegetation growth rate and coverage changes. S6. Adjust the repair strategy based on the repair effect data and generate adjusted repair data, which includes dynamically optimized irrigation and replanting plans. S7. After integrating and adjusting the restoration data, output a vegetation restoration report, summarizing the parameters and effects of the entire restoration process.

[0022] The steps involved in collecting grassland vegetation status data in the area under and between photovoltaic panels in S1 are as follows: S11. Collect vegetation coverage data in the area under the photovoltaic panels using a drone equipped with a multispectral camera, and generate vegetation coverage data under the panels. The specific implementation steps are as follows: First, data collection planning and drone aerial photography were carried out. Drones equipped with multispectral cameras were used to plan flight routes based on the characteristics of the shaded areas under the photovoltaic panels, ensuring that the flight routes could completely cover all areas under the panels. The multispectral camera can simultaneously collect image data including multiple bands of visible and invisible light in one flight. The near-infrared band is extremely sensitive to the chlorophyll content of vegetation and is the key to distinguishing vegetation from non-vegetation ground features. Next, image data processing and vegetation index calculation are performed. The acquired raw multispectral images need to undergo stitching, geometric correction, and radiometric calibration preprocessing to eliminate image distortion and ensure data accuracy. Subsequently, the Normalized Difference Vegetation Index (NDVI) is calculated using the preprocessed data. NDVI is one of the most widely used indicators for assessing vegetation cover, and its calculation formula is as follows: ; in, Represents reflectivity in the near-infrared band. Represents reflectivity in the red light band; Healthy, lush vegetation strongly reflects near-infrared light and absorbs red light, so its NDVI value is close to 1, while the NDVI value of non-vegetated areas such as bare soil, rocks and water bodies is lower, or even negative. Finally, vegetation cover extraction and data generation are performed. Based on the calculated NDVI image, a reasonable threshold is set to distinguish between vegetation pixels and non-vegetation pixels, and the NDVI value of each pixel is converted into vegetation cover. S12. Collect soil moisture, pH and nutrient content data in the inter-plate area through a ground sensor network to generate inter-plate soil parameter data; After undergoing a series of processing steps on the cloud platform, the raw data generates usable inter-plate soil parameter data: Data cleaning and calibration: First, the raw data is quality controlled to remove outliers, and then the raw readings of electrical signals are converted into physically meaningful values ​​using the sensor's calibration curve.

[0023] Spatiotemporal interpolation and mapping: For data from discrete sensor nodes, the Kriging spatial interpolation algorithm can be used to generate a continuous spatial distribution map covering the entire inter-plate area. This can intuitively show the spatial variability of soil moisture, pH, and nutrients. For nutrient content estimation, sensor readings and soil spectral models are sometimes combined, and the general formula can be expressed as: ; in, A predicted concentration representing a certain nutrient. arrive This represents the sensor's response value at a specific wavelength. For the sample size, It is a predictive model built on a large amount of sample data; Data Integration and Output: Finally, the processed data is integrated to generate structured inter-plate soil parameter data, which typically includes time series datasets and spatial distribution maps of soil moisture, pH, nitrogen, phosphorus and potassium content for each monitoring point and grid, providing accurate input for subsequent analysis of soil nutrient differences between the inter-plate and sub-plate environments; S13. Measure the light intensity distribution under and between photovoltaic panels using a light sensor to generate light intensity data; The collected raw data is processed to generate differentially analyzable light intensity data. The core of this process is calculating the light shading rate, an indicator that quantitatively describes the shading effect of photovoltaic panels. The calculation formula is as follows: ; in, This represents the average light intensity at a monitoring point under a photovoltaic panel within a certain time period. This represents the average light intensity at the control monitoring points in the open area between the boards during the same time period. The final generated light intensity data is a dataset containing time series and spatial attributes. It not only provides the absolute value of light intensity in each area, but more importantly, it clearly quantifies the difference in light shading between the area under the slab and between the slabs through derived indicators such as light shading rate. This data is the core scientific basis for the subsequent formulation of differentiated restoration plans, ensuring a high degree of matching between vegetation selection and local microenvironment.

[0024] Generating environmental difference data in S2 includes the following steps: S21. Obtain vegetation coverage data under the board, soil parameter data between the boards, and light intensity data; S22. Use a numerical comparison algorithm to analyze the difference in moisture distribution between the area under the plate and the area between the plates, and generate moisture difference data. The preprocessed data will be input into the core numerical comparison algorithm, which mainly includes descriptive statistical comparison and spatial distribution feature analysis. Descriptive statistical comparison: Key statistics of soil moisture data from two regions were calculated and compared. These statistics include: Average value: Reflects the overall humidity level of the area; ; in, and These represent the average soil moisture in the areas beneath and between the boards, respectively. and For the sample size, and Representing the area below the board respectively The sampling point and the inter-plate area Soil moisture measurements at each sampling point and For indexing, and This indicates calculating the average of the sums; Coefficient of variation: reveals the spatial variability of moisture within a region; ; in, Represents the coefficient of variation. Standard deviation, This indicates the average soil moisture in the area; The coefficient of variation in the area under the photovoltaic panel is usually higher than that in the area between the panels due to uneven shading by the photovoltaic panels. Hypothesis testing: To determine whether there is a statistically significant difference in average soil moisture between two regions, hypothesis testing is often used. The test statistic is calculated using the following formula: ; in, express Test statistic For sample variance, and These represent the variances of soil moisture samples in the area beneath the boards and the area between the boards, respectively. This represents the observed difference in average soil moisture between and beneath the boards. and For the sample size, This is the difference. The standard error; Through calculation By comparing the value with the critical value, the level of moisture difference can be determined; Based on the above analysis, the system generates structured moisture difference data, which clearly quantifies the impact of photovoltaic arrays on the redistribution of moisture in the microenvironment. S23. Calculate the occlusion rate of the area under the board based on the light intensity data, and generate occlusion difference data. The calculation principle and formula for light shading rate: Light shading rate is a core calculation indicator used to quantitatively describe the degree of shading received at a specific point. It is defined as the ratio of the actual light intensity received at that point to the maximum light intensity in an open area during the same period. The most commonly used calculation formula is as follows: ; in, This represents the average photosynthetically active radiation intensity at a specific monitoring point under the plate during a specific time period. This represents the average photosynthetically active radiation intensity at the control point in the open area between the plates during the exact same time period. S24. Combine soil parameter data to assess the soil nutrient balance in the areas under and between the boards, and generate soil nutrient difference data; To obtain the differences in overall nutrient status, a comprehensive soil nutrient index is often constructed. A common calculation method is the Nemerow index method, the formula of which can be expressed as: ; in, The comprehensive nutrient index, For the first Measured values ​​of nutrients, Its corresponding standard value, The maximum value among the standard ratios of each nutrient. For the sample size, For indexing; Calculate the areas under and between the plates separately. The difference between these values ​​can quantify the differences in overall nutrient levels; To visually represent the spatial pattern of "soil nutrient differences," spatial interpolation techniques from Geographic Information Systems (GIS) are crucial. This method predicts nutrient levels in unsampled areas based on known sampling points, generating a continuous spatial distribution map of nutrients. Its general formula can be expressed as: ; in, It is the point to be predicted Nutritional value It is a known point The measured value, These are Kriging weighting coefficients. Geographic coordinates representing an unknown point in space. Its corresponding standard value, For the sample size, For indexing; By generating nutrient distribution maps for the areas under and between the boards respectively, and performing map algebra operations, a soil nutrient difference distribution map is directly generated, clearly marking the areas of nutrient deficiency and enrichment. The final generated soil nutrient difference data is a comprehensive product that includes quantitative indicators and spatial maps.

[0025] Generating repair plan data in S3 includes the following steps: S31. Based on data on differences in moisture content, light shading, and soil nutrients, determine which shade-tolerant grass species are suitable for planting in the area under the slab. S32. Based on the light advantages of the inter-board area, light-loving grass species were selected as the main restoration species; S33. Optimize the selection scheme by combining historical vegetation data, and generate a selection scheme for shade-tolerant vegetation under the board and a selection scheme for light-loving vegetation between the boards.

[0026] Generating repair implementation data in S4 includes the following steps: S41. Based on the selection scheme of shade-tolerant vegetation under the boards and the selection scheme of light-loving vegetation between the boards, calculate the sowing density in the area under the boards and the sowing interval in the area between the boards. Calculation of seeding density in the area under the board: The area beneath the slab receives relatively little sunlight. To ensure rapid vegetation coverage, suppress weeds, and conserve soil and water, a higher sowing density is typically required. The calculation formula is as follows: ; Calculation of sowing interval in inter-board area: The inter-board area has sufficient light and relatively intense water competition. In order to avoid overcrowding of plants, promote root development and enhance stress resistance, hole sowing and row sowing are often used, and the optimal sowing interval is calculated. The core is to determine the number of sowing holes per unit area, and then derive the row spacing and hole spacing. First, calculate the number of seed holes per unit area: ; in The number of seeds sown per hole is usually 3-5 to ensure germination; Subsequently, the sowing interval is determined based on the number of holes. Using a square sowing method, the row spacing and hole spacing are equal. The calculation formula is as follows: ; Ultimately, the calculated seeding density and spacing were integrated into the remediation implementation data to guide on-site seeding operations; S42. Develop differentiated irrigation plans based on moisture difference data. Use high-frequency, low-volume irrigation for the area under the boards and low-frequency, high-volume irrigation for the area between the boards. S43. Configure the ratio of organic fertilizer and chemical fertilizer based on soil nutrient difference data to generate fertilization strategy data.

[0027] Generating repair effect data in S5 includes the following steps: S51. Monitor vegetation cover changes through regular drone aerial photography and generate cover change data; Vegetation cover calculation and change detection: The most widely used method at present is the pixel-division model, which assumes that the spectral signal of a pixel is a linear mixture of vegetation and bare soil. The formula for calculating vegetation cover is as follows: in, It is the normalized difference vegetation index value. Representing pure vegetation pixels value, Representing pure bare earth pixels value; After calculating the FVC for each period, change detection can be performed. The most direct method for calculating the change in coverage is the interpolation method. ; in, and Representing two different monitoring time points, This indicates an increase in vegetation cover. This indicates coverage degradation; S52. Continuously collect soil parameters using soil sensors to generate dynamic soil data; S53. Combine cover change data and soil dynamic data to assess vegetation growth rate and generate restoration effect data.

[0028] Generating adjusted and repaired data in S6 includes the following steps: S61. When the vegetation growth rate in the restoration effect data is lower than the threshold, increase the irrigation frequency in the area under the board. S62. When the coverage of the inter-board area is insufficient, the reseeding mechanism is activated to optimize the seeding amount; S63. Adjust the fertilization strategy dynamically based on real-time data and generate adjusted repair data.

[0029] Outputting a vegetation restoration report in S7 includes the following steps: S71. Summarize environmental difference data, remediation plan data, remediation implementation data, remediation effect data, and adjusted remediation data; S72. Generate a visual report, including charts showing vegetation restoration trends and environmental impact assessments.

[0030] The method also includes step S8, establishing a vegetation restoration database to store and update all restoration process data and support long-term ecological analysis. Step S8 includes the following steps: S81. Import the collected data into the cloud platform and use encryption algorithms to ensure data security; S82. Regularly back up data and set up access control for researchers; S83. Generate an annual repair effect comparison report based on the database.

[0031] The method also includes step S9, constructing an intelligent restoration decision support system to optimize real-time decision-making in the vegetation restoration process, wherein S9 includes the following steps: S91. Integrated IoT monitoring equipment collects real-time micro-environmental data in the area under and between photovoltaic panels. The micro-environmental data includes temperature gradient changes, humidity distribution patterns, and light intensity fluctuations. S92. Establish a vegetation growth prediction model based on historical restoration data, and analyze the adaptability of different vegetation types to the environment under and between the boards using machine learning algorithms. For the typical supervised learning problem of vegetation adaptability prediction, algorithms that combine prediction accuracy and a certain degree of interpretability are often chosen, and random forest is an ideal choice. Model principle: Random forests construct multiple decision trees and perform ensemble learning. Its core formula is ensemble prediction; for regression problems, the final predicted value is the average of the predictions from all decision trees. ; in, This is the final predicted value. It is the number of decision trees in the forest. It is the first Each decision tree has input features The prediction results For indexing, These are input features.

[0032] Model training: The prepared dataset is divided into training and test sets according to a certain ratio. The training set data is used to "teach" the random forest model, that is, to let the model learn the mapping relationship from environmental features to fitness indicators. The hyperparameters of the model are optimized through cross-validation to prevent overfitting and ensure the model's generalization ability. S93. Use multi-objective optimization algorithms to balance ecological and economic benefits and generate dynamic adjustment suggestions for restoration plans; The first step in implementation is to transform the abstract concepts of "ecological benefits" and "economic benefits" into computable mathematical objective functions, which is the foundation upon which the algorithm can process them; Quantification of the ecological benefit objective function: Usually, the goal is to maximize ecological benefits, and a comprehensive ecological index is constructed as the objective function; ; in, Represents a vector of decision variables. It is a plan The predicted vegetation cover rate, It is a plan The predicted biodiversity index. It is a plan The predicted soil health index, These are weighting coefficients; Quantification of the economic benefit objective function: It usually pursues cost minimization and net profit maximization; ; in, Each represents a scheme The costs of seeds, water, fertilizer, and labor. Representative proposal The carbon sequestration benefits that will result; After defining the objective function, a multi-objective optimization algorithm can be applied to solve it. Since the two objectives are usually in conflict, there is no single optimal solution, but rather a set of optimal solutions. Multi-objective genetic algorithms are a commonly used method for solving this type of problem. Algorithm flow: 1. Initialization: Randomly generate a set of initial repair schemes; 2. Evaluation: Calculate the ecological benefit value corresponding to each scheme. and economic benefit value ; 3. Non-dominated ranking: Ranking individuals in a population based on dominance relationships; 4. Crowding Calculation: To maintain the diversity of solution distribution on the frontier, the density of each solution around it is calculated; 5. Selection, crossover, and mutation: Based on non-dominated sorting and crowding, select superior individuals to enter the next generation, and generate new schemes through genetic operators; 6. Iteration: Repeat steps 2-5 until the termination condition is met; S94. Establish a decision support interactive interface to visually display the comparison results of remediation schemes and the simulation of expected effects; S95. Set up an early warning mechanism to automatically trigger the re-evaluation process of the repair plan when the monitored data deviates from the expected threshold.

[0033] The operational steps of a method for differential vegetation restoration between and under photovoltaic panels are as follows: Step 1: Data Collection and Status Assessment This method begins with the comprehensive collection of grassland vegetation status data in the areas under and between photovoltaic panels, including vegetation cover, soil parameters, and light intensity data. The collection process utilizes a drone equipped with a multispectral camera to map vegetation cover, obtains soil parameters through a ground sensor network, and measures light distribution using a light sensor. The principle behind this step is to quantify the current vegetation status and environmental basis through multi-source data fusion, providing a scientific basis for subsequent differential analysis. Data collection not only focuses on spatial coverage but also emphasizes the continuity of time series to capture the dynamic changes in the microenvironment.

[0034] Step 2: Environmental Difference Analysis Based on the collected data, the system performs differential analysis on the environment under and between the panels, generating environmental difference data, mainly including differences in moisture distribution, light shading, and soil nutrients. The analysis principle adopts a numerical comparison algorithm and uses statistical methods to assess the spatial variability of moisture and nutrients. The core of this step is to transform the raw data into operable difference indicators through mathematical modeling and spatial interpolation techniques, revealing the redistribution effect of the photovoltaic array on the microenvironment, and laying the foundation for differentiated remediation.

[0035] Step 3: Development of a differentiated repair plan: Based on environmental difference data, the system formulates differentiated remediation plans and generates remediation plan data. The principle is to match vegetation characteristics with the local environment. Shade-tolerant grass species are suitable for planting in the shaded areas under the boards, while light-loving grass species are preferred in the open areas between the boards. The plan formulation combines historical vegetation data and uses optimization algorithms to determine the selection plan, ensuring that the selected vegetation not only adapts to the current environment but also improves community stability. This step reflects the ecological principle of "planting the right species in the right place" and aims to solve the pain point of mismatch between remediation strategies and microenvironment.

[0036] Step 4: Implementation of Remedial Measures Based on the restoration plan data, specific vegetation restoration measures are implemented, generating restoration implementation data, including seeding rate, irrigation frequency, and fertilization strategy. The implementation principle emphasizes precise control, with seeding density and interval calculated using formulas to ensure optimal resource allocation. This step transforms the plan into action, with the core being the use of parameterized operations to achieve zoned management of water and nutrients, reducing resource waste.

[0037] Step 5: Monitoring and Evaluation of Repair Results After implementation, the system continuously monitors the restored vegetation status and generates restoration effect data, including changes in vegetation growth rate and coverage. The monitoring principle is based on periodic data acquisition and sensor network tracking of soil dynamics. The evaluation process uses a pixel-based binary model algorithm to quantify coverage changes and combines growth rate thresholds to judge restoration effectiveness. The principle of this step is to establish a feedback loop, capture vegetation responses through real-time data, and provide a basis for dynamic adjustment.

[0038] Step 6: Dynamically adjust the repair strategy: Based on the restoration effect data, the system automatically adjusts the restoration strategy and generates adjusted restoration data. The adjustment principle relies on the early warning mechanism. When the vegetation growth rate is lower than the threshold, the irrigation frequency of the area under the board is increased. When the coverage is insufficient, the reseeding mechanism is activated and the seeding amount is optimized. The core is to achieve adaptive optimization through machine learning and rule engine to ensure that the restoration process can respond to degradation signs in a timely manner and maintain the continuous effectiveness of restoration.

[0039] Step 7: Report Generation and Knowledge Integration Finally, the system integrates all data to output a vegetation restoration report, summarizing environmental differences, restoration plans, implementation effects, and adjustment records. The report generation principle focuses on data visualization and knowledge accumulation. This step not only provides decision support but also accumulates historical data by establishing a database, providing a reference for future restoration projects.

[0040] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0041] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for differentiated restoration of grassland vegetation under and between photovoltaic panels, characterized in that, The method includes the following steps: S1. Collect grassland vegetation status data in the area under the photovoltaic panels and between the panels, including vegetation coverage, species composition, species diversity, soil parameters and light intensity data; S2. Based on the grassland vegetation status data, perform environmental difference analysis and processing on the area under and between the boards to generate environmental difference data, which includes differences in water distribution, light shading, and soil nutrients. S3. Based on the environmental difference data, formulate differentiated remediation plans and generate remediation plan data. The remediation plan data includes a selection plan for shade-tolerant vegetation in the area under the boards and a selection plan for light-loving vegetation in the area between the boards. S4. Implement vegetation restoration measures based on the restoration plan data to generate restoration implementation data, which includes seeding amount, irrigation frequency and fertilization strategy. S5. Monitor the vegetation status after restoration based on the restoration implementation data, and generate restoration effect data, including vegetation growth rate and coverage changes. S6. Adjust the repair strategy based on the repair effect data to generate adjusted repair data, which includes dynamically optimized irrigation and replanting schemes. S7. Integrate the adjusted restoration data and output a vegetation restoration report, summarizing the parameters and effects of the entire restoration process.

2. The method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 1, characterized in that, The step S1 involves collecting grassland vegetation status data in the area under and between photovoltaic panels, including the following steps: S11. Collect vegetation coverage data in the area under the photovoltaic panels by using a drone equipped with a multispectral camera, and generate vegetation coverage data under the panels; S12. Collect soil moisture, pH and nutrient content data in the inter-plate area through a ground sensor network to generate inter-plate soil parameter data; S13. Measure the light intensity distribution under and between photovoltaic panels using a light sensor to generate light intensity data.

3. The method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 2, characterized in that, The process of generating environmental difference data in S2 includes the following steps: S21. Obtain the vegetation coverage data under the board, the soil parameter data between the boards, and the light intensity data; S22. Use a numerical comparison algorithm to analyze the difference in moisture distribution between the area under the plate and the area between the plates, and generate moisture difference data. S23. Calculate the occlusion rate of the area under the board based on the light intensity data, and generate occlusion difference data. S24. Combine soil parameter data to assess the soil nutrient balance in the areas under and between the boards, and generate soil nutrient difference data.

4. The method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 3, characterized in that, The process of generating repair plan data in S3 includes the following steps: S31. Based on the aforementioned moisture difference data, light shading difference data, and soil nutrient difference data, determine that the area under the board is suitable for planting shade-tolerant grass species; S32. Based on the light advantages of the inter-board area, light-loving grass species were selected as the main restoration species; S33. Optimize the selection scheme by combining historical vegetation data, and generate a selection scheme for shade-tolerant vegetation under the board and a selection scheme for light-loving vegetation between the boards.

5. The method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 4, characterized in that, The process of generating repair implementation data in S4 includes the following steps: S41. Based on the shade-tolerant vegetation selection scheme under the boards and the light-loving vegetation selection scheme between the boards, calculate the sowing density in the area under the boards and the sowing interval in the area between the boards. S42. Develop differentiated irrigation plans based on moisture difference data. Use high-frequency, low-volume irrigation for the area under the boards and low-frequency, high-volume irrigation for the area between the boards. S43. Configure the ratio of organic fertilizer and chemical fertilizer based on soil nutrient difference data to generate fertilization strategy data.

6. The method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 5, characterized in that, The steps involved in generating the repair effect data in S5 are as follows: S51. Monitor vegetation cover changes through regular drone aerial photography and generate cover change data; S52. Continuously collect soil parameters using soil sensors to generate dynamic soil data; S53. Combine cover change data and soil dynamic data to assess vegetation growth rate and generate restoration effect data.

7. The method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 6, characterized in that, The process of generating the adjusted and repaired data in step S6 includes the following steps: S61. When the vegetation growth rate in the restoration effect data is lower than the threshold, increase the irrigation frequency in the area under the board. S62. When the coverage of the inter-board area is insufficient, the reseeding mechanism is activated to optimize the seeding amount; S63. Adjust the fertilization strategy dynamically based on real-time data and generate adjusted repair data.

8. The method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 7, characterized in that, The vegetation restoration report output in S7 includes the following steps: S71. Summarize the environmental difference data, repair plan data, repair implementation data, repair effect data, and adjusted repair data; S72. Generate a visual report, including charts showing vegetation restoration trends and environmental impact assessments.

9. A method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 1, characterized in that, The method further includes step S8: establishing a vegetation restoration database, storing and updating all restoration process data, and supporting long-term ecological analysis, wherein S8 includes the following steps: S81. Import the collected data into the cloud platform and use encryption algorithms to ensure data security; S82. Regularly back up data and set up access control for researchers; S83. Generate an annual repair effect comparison report based on the database.

10. A method for differentiated restoration of grassland vegetation under and between photovoltaic panels according to claim 1, characterized in that, The method further includes step S9: constructing an intelligent restoration decision support system to optimize real-time decision-making in the vegetation restoration process, wherein S9 includes the following steps: S91. An integrated IoT monitoring device collects real-time micro-environmental data in the area under and between photovoltaic panels, including temperature gradient changes, humidity distribution patterns, and light intensity fluctuations. S92. Establish a vegetation growth prediction model based on historical restoration data, and analyze the adaptability of different vegetation types to the environment under and between the boards using machine learning algorithms. S93. Use multi-objective optimization algorithms to balance ecological and economic benefits and generate dynamic adjustment suggestions for restoration plans; S94. Establish a decision support interactive interface to visually display the comparison results of remediation schemes and the simulation of expected effects; S95. Set up an early warning mechanism to automatically trigger the re-evaluation process of the repair plan when the monitored data deviates from the expected threshold.

Images