Assessment method for the impact of centralized photovoltaic power station development on vegetation

Abstract

The invention discloses an evaluation method for the impact of the development of a centralized photovoltaic power station on vegetation, which comprises the following steps: obtaining a remote sensing image of an area where the centralized photovoltaic power station is located, and determining a threshold range of the impact of the centralized photovoltaic power station on vegetation; constructing a buffer zone whose distance from the centralized photovoltaic power station is within the threshold range, and using it as an impact zone, and then establishing a mask for the impact zone; selecting an area near the impact zone, which has substantially the same natural geography, shape and area as the impact zone, as a control zone, and establishing a mask for the control zone; extracting a vegetation index time series of all grids within the mask range according to the masks of the impact zone and the control zone, constructing a monthly / annual vegetation index time series curve of the impact zone and the control zone, and then respectively calculating the change amplitude and average change rate of the vegetation index time series curve of the impact zone and the control zone; and calculating the difference in the change amplitude and the difference in the average change rate between the impact zone and the control zone.

Description

Technical Field

[0001] The present invention relates to the field of energy and environmental technology, and in particular to a method for evaluating the impact of centralized photovoltaic power station development on vegetation. Background Art

[0002] Centralized photovoltaic power stations are generally built in deserts, Gobi and other areas. While they bring clean energy supply to the power system, due to the relatively poor natural conditions and sensitive and fragile ecological environment in these areas, they also have potential impacts on the regional water, energy, carbon and other circulation processes, thereby changing the regional meteorological, hydrological, ecological environment and even social and economic conditions.

[0003] At present, relevant work has conducted preliminary discussions on the impact of photovoltaic power stations on vegetation. The basic conclusion is that building centralized photovoltaic power stations in arid ecosystems can promote regional vegetation recovery.

[0004] However, existing research methods on the impact of photovoltaic power stations on vegetation have certain limitations, mainly manifested in the following aspects: First, most of them rely on short-term, small-scale, and local research based on manually collected ground monitoring data, and lack long-term, large-scale, and regional assessments; second, most of them explore the vegetation recovery situation from a qualitative perspective, and lack quantitative assessments; third, there is a lack of comprehensive and systematic assessments of the impact of centralized photovoltaic power station development on vegetation.

[0005] Due to the shortcomings of the existing assessment methods, the assessment results are incomplete and inaccurate, which cannot provide support for scientific planting. This may lead to blind planting of tree and shrub belts and green cash crops, which will not survive, and to a certain extent limit the restoration of the ecological environment. At the same time, it may also lead to unnecessary waste of human and material resources, affecting the progress of my country's photovoltaic power station construction and ecological restoration. Therefore, it is urgent to develop comprehensive and effective technical methods to ensure the coordinated development of photovoltaic power station construction and ecological environment, so as to achieve the goal of effective resource utilization and sustainable development. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for evaluating the impact of centralized photovoltaic power station development on vegetation, which solves the problem that the existing evaluation methods are inaccurate and limit the restoration of the ecological environment.

[0007] A method for evaluating the impact of centralized photovoltaic power station development on vegetation is provided, which comprises the following steps:

[0008] S1. Obtain remote sensing images of the area where the centralized photovoltaic power station is located, and determine the threshold range of the impact of the centralized photovoltaic power station on vegetation based on the remote sensing images and vegetation index;

[0009] S2, construct a buffer zone with a distance from the centralized photovoltaic power station within a threshold range and use it as the impact zone, and then establish a mask of the impact zone;

[0010] S3. Within a preset distance range from the impact area, select an area with substantially the same physical geography, shape and area as the impact area as a control area, and establish a mask for the control area;

[0011] S4. According to the masks of the impact area and the control area, the vegetation index time series of all grids within the mask range are extracted, and the monthly / yearly vegetation index time series curves of the impact area and the control area are constructed;

[0012] S5. Calculate the change amplitude and average change rate of the vegetation index time series curves of the impact area and the control area respectively according to the vegetation index time series curves;

[0013] S6. Calculate the difference in the magnitude of change and the difference in the average rate of change between the impact area and the control area as a means of quantitatively evaluating the impact of centralized photovoltaic power station construction on vegetation from a spatial comparison perspective.

[0014] Furthermore, the method for constructing the vegetation index time series curve includes:

[0015] S41, using the maximum value synthesis method, selecting the maximum value of the vegetation index of each month / year in the vegetation index time series as the output value;

[0016] S42. Draw monthly / yearly vegetation index time series curves of the impact area and the control area according to the output values ​​of the impact area and the control area.

[0017] Furthermore, the expressions for calculating the change amplitude and average change rate are:

[0018]

[0019]

[0020] Where CA is the change amplitude; ACR is the average change rate; i is the time series index; VI i+1 With VI i are the i+1th and ith vegetation indices respectively; m is the length of the time series; Δt i For VI i+1 With VI i The time interval between them; max is the maximum value; min is the minimum value.

[0021] The beneficial effects of the above technical scheme are as follows: the existing technical scheme for the impact of photovoltaic power station development on vegetation does not involve the determination of the threshold range of vegetation impact, but conducts impact assessment within a certain range based on experience, which will bring inaccuracy to the assessment results; this scheme makes up for the shortcomings of the existing assessment method, first determines the threshold range of the impact of centralized photovoltaic power stations on vegetation, and then on this basis, compares and evaluates the vegetation index time series curves of the impact area and the control area, so as to obtain time series curve evaluation parameters (change amplitude and average change rate) with high credibility and high precision.

[0022] By comparing the curve evaluation parameters of the impact zone obtained by this scheme with those of the control zone, we can intuitively understand the impact of photovoltaic power stations on vegetation growth. When the parameters of the impact zone are better than those of the control zone, it indicates that the construction of photovoltaic power stations has a promoting effect on vegetation growth; conversely, when they are worse than those of the control zone, it indicates that they have an inhibitory effect. The degree of difference between the parameters of the impact zone obtained by this scheme and those of the control zone can reflect the degree to which photovoltaic power stations promote the growth of surrounding vegetation. When the degree of difference is small, the scale of surrounding vegetation planting can be reduced. As the degree of difference increases, the scale of planting can be expanded accordingly.

[0023] Through the quantitative evaluation of this plan, managers can make preliminary plans for the planting of vegetation around centralized photovoltaic power stations and determine whether the area around photovoltaic power stations is suitable for planting vegetation. If suitable, the scale of planting can be further determined, providing more accurate data for subsequent technical solutions.

[0024] Furthermore, the assessment method of the impact of centralized photovoltaic power station development on vegetation also includes a time comparison quantitative assessment method of the impact on vegetation:

[0025] A1. Divide the affected area into the t1 period before the construction of the centralized photovoltaic power station and the t2 period after the construction;

[0026] A2, extract the climatic and environmental factors of the impact area at period t1, and construct a climatic-environmental factor time series with the same length as the vegetation index time series of the impact area;

[0027] A3. Use the climate-environmental factor time series and vegetation index time series of period t1 to train multiple machine learning models, and select the machine learning model with the highest accuracy;

[0028] A4. Input the time series of climate-environmental factors in the affected area at period t2 into the selected machine learning model to obtain the time series of predicted vegetation index affected by climate-environmental factors at period t2;

[0029] A5. Based on the vegetation index time series in the impact area during period t2 and the predicted vegetation index time series, the contribution ratios of climate-environmental factors and the construction of centralized photovoltaic power stations to the vegetation index change are calculated respectively.

[0030] The beneficial effect of the above technical scheme is: since the original vegetation coverage conditions of the control area are not completely consistent with those of the impact area, the evaluation results of the control area cannot fully represent the actual situation of the impact area; on this basis, this scheme conducts a time comparison evaluation of the impact area to obtain more reliable and accurate impact parameters (two contribution ratios).

[0031] The numerical value of the contribution ratio corresponding to the influence of climate-environmental factors obtained in this scheme can indicate the magnitude of climate-environmental improvement. When the value is relatively small, it is suitable to plant drought-resistant plants. As the value increases, it is suitable to select plants with relatively weak drought resistance. For the contribution ratio corresponding to the construction of centralized photovoltaic power stations, its value determines the vegetation planting area. The larger the value, the larger the planting area.

[0032] Through the two contribution ratios accurately obtained through this plan, managers can make more accurate and reasonable plans to determine the vegetation planting area and vegetation planting type to promote ecological restoration in arid areas.

[0033] Furthermore, the expression for calculating the contribution ratio is:

[0034] ΔVI clim =ΔVI-ΔVI PV

[0035]

[0036] Among them, ΔVI is the change of vegetation index in period t2 relative to period t1; ΔVI PV is the change in vegetation index caused by the construction of photovoltaic power stations; ΔVI clim is the magnitude of the change in vegetation index due to climate-environmental factors; and are the observed average values ​​of vegetation index during period t1 and period t2, respectively; is the average value of the predicted vegetation index time series; CR clim and CR PV They are the contribution ratios of climate-environmental factors and photovoltaic power station construction to the changes in vegetation index.

[0037] The beneficial effects of the above technical scheme are as follows: the existing technical scheme focuses on evaluating the impact of photovoltaic power station construction on vegetation growth from a qualitative perspective, and lacks a quantitative evaluation method; this scheme provides a quantitative calculation method for determining the contribution ratio of photovoltaic power station construction and climate-environmental factors to vegetation impact, combining actual data and model prediction data, and on the basis of spatial and temporal comparison of quantitative evaluation results, more accurately and flexibly calculates the contribution ratio of photovoltaic power station construction and climate-environmental factors to vegetation changes.

[0038] When the contribution ratio of climate-environmental factors obtained by this scheme is greater than the contribution ratio of photovoltaic power station construction, it indicates that photovoltaic power station construction will promote vegetation growth, but climate-environmental factors still dominate; if the contribution ratio of climate-environmental factors is less than the contribution ratio of photovoltaic power station construction, it indicates that photovoltaic power station construction plays a leading role in vegetation growth. Based on the contribution ratio of different factors, it is helpful to understand the main driving force behind vegetation changes, and guide managers to reasonably plan and determine the planting scale and vegetation planting type.

[0039] Furthermore, step S1 further comprises:

[0040] S11. Acquire a remote sensing image of the area where the centralized photovoltaic power station is located, and identify the spatial coverage of the centralized photovoltaic power station;

[0041] S12, taking the centralized photovoltaic power station as the center and Δd as the distance step, establish n discrete buffers 0~Δd, Δd~2·Δd, …, (n-1)·Δd~n·Δd, and use GIS tools to establish a mask for each discrete buffer;

[0042] S13, according to the mask of each discrete buffer zone, extracting the vegetation index time series of all grids within the mask range, and constructing a monthly / yearly vegetation index time series curve of each discrete buffer zone;

[0043] S14. Calculate the change size and degree of the vegetation index before and after the construction of the centralized photovoltaic power station based on the vegetation index time series curve of the discrete buffer zone;

[0044] S15. According to the change size and degree of the vegetation index of all discrete buffer zones, the relationship curves between the change size and degree of the vegetation index and the distance from the buffer zone to the centralized photovoltaic power station are drawn respectively, and the distance when it tends to be stable is selected as the threshold range of the impact of the centralized photovoltaic power station on the vegetation.

[0045] Furthermore, the expression of the change size and degree of vegetation index is:

[0046] ΔVI slp =VI slp,t2 -VIslp,t1

[0047] in, and ΔVI slp are the magnitude and degree of change of vegetation index respectively; and is the average vegetation index of the vegetation index time series curve of the discrete buffer zone before and after the construction of the centralized photovoltaic power station; VI slp,t1 and VI slp,t2 They are the slopes of the vegetation index time series curves of the discrete buffer zone before and after the construction of the centralized photovoltaic power station.

[0048] The beneficial effects of the above technical scheme are as follows: before conducting a spatial and temporal comparative evaluation of the impact of centralized photovoltaic power station development on vegetation, the threshold range of the impact of photovoltaic power stations on the surrounding environment is first determined, which makes up for the shortcomings of previous methods. In particular, this scheme obtains long-term vegetation index data of different discrete buffer zones by acquiring remote sensing images, which is easier to capture the long-term trend changes of vegetation indexes than short-term data. Long-term monitoring provides richer data, which can capture the relative differences in vegetation indices in different regions, thereby generating more representative time series curve evaluation parameters. These parameters can more finely reflect the vegetation changes in different regions, and help to efficiently and accurately determine the scope and extent of the impact of centralized photovoltaic power stations on vegetation. Managers can comprehensively plan the scale and type of planting based on the determined impact threshold range and the contribution ratio parameters of climate-environmental factors and photovoltaic power station construction to ensure the coordinated development of photovoltaic power station construction and ecological restoration.

[0049] Furthermore, the vegetation index is a normalized vegetation index, an enhanced vegetation index or a leaf area index.

[0050] Furthermore, various machine learning models include (but are not limited to) random forest RF, support vector machine SVM and artificial neural network ANN.

[0051] The beneficial effects of the present invention are as follows: when the difference in the amplitude of change and the difference in the average rate of change obtained by the present scheme are both positive values, it indicates that the construction of the photovoltaic power station has a positive impact on the vegetation; when they are both negative values, it has a negative impact on the vegetation; the present scheme can accurately obtain the two parameters through the above method, so that when the two parameters are both positive, the management personnel can accurately plan the planting area of ​​the tree and shrub belt and green economic crops based on their sizes, thereby promoting the restoration of the ecological environment; at the same time, it can also avoid unnecessary waste of manpower and material resources.

[0052] By determining the threshold range of the impact of photovoltaic power stations on vegetation, this plan can lay the foundation for the accurate calculation of the difference in the change amplitude and the average change rate between the impact area and the control area; by comparing the impact area and the control area, the impact of centralized photovoltaic power station development on vegetation can be accurately determined, providing a reference for the planning, design and operation management of large-scale development of centralized photovoltaic power stations under the new situation of ecological protection.

[0053] This scheme can provide a feasible technical method for the coordinated improvement of photovoltaic power station planning and design, operation and management, and ecological environmental protection and restoration. The two are organically combined and promote each other, achieving a balance between economic, social and ecological benefits. In addition, the direct benefits and contributions of photovoltaic power station operation to carbon emission reduction can be calculated through coal substitution, but the indirect benefits of carbon sinks caused by increased vegetation coverage are more difficult to calculate. This scheme provides a feasible technical method for evaluating the indirect benefits of carbon sinks brought by photovoltaic power stations by quantitatively evaluating the impact of photovoltaic power stations on vegetation. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 Flowchart for developing an assessment methodology for the impact of concentrated photovoltaic power plants on vegetation.

[0055] Figure 2 This is a flowchart for implementing a specific example of this solution.

[0056] Figure 3 Schematic diagram of discrete buffers in the threshold range determination process.

[0057] Figure 4 For different discrete buffers in the threshold range determination process Graph.

[0058] Figure 5 ΔNDVI for different discrete buffers in the threshold range determination process slp Graph.

[0059] Figure 6 Schematic diagram of the affected area and the control area.

[0060] Figure 7 This is a schematic diagram of the change amplitude and average change rate of the vegetation index time series curve.

[0061] Figure 8 Schematic diagram of the comparison of vegetation index time series changes before (t1) and after (t2) the construction of centralized photovoltaic power stations. DETAILED DESCRIPTION

[0062] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0063] refer to Figure 1 , Figure 1 A flow chart showing the method for assessing the impact of centralized photovoltaic power station development on vegetation; Figure 1 As shown, the method S includes steps S1 to S6.

[0064] In step S1, a remote sensing image of the area where the centralized photovoltaic power station is located is obtained, and based on the remote sensing image and the vegetation index, a threshold range of the impact of the centralized photovoltaic power station on vegetation is determined. The preferred vegetation index of this solution is preferably the normalized vegetation index, the enhanced vegetation index or the leaf area index.

[0065] In one embodiment of the present invention, step S1 further comprises:

[0066] S11. Acquire a remote sensing image of the area where the centralized photovoltaic power station is located, and identify the spatial coverage of the centralized photovoltaic power station;

[0067] S12, taking the centralized photovoltaic power station as the center and Δd as the distance step, establish n discrete buffers 0~Δd, Δd~2·Δd, …, (n-1)·Δd~n·Δd, and use GIS tools to establish a mask for each discrete buffer;

[0068] S13, according to the mask of each discrete buffer zone, extracting the vegetation index time series of all grids within the mask range, and constructing a monthly / yearly vegetation index time series curve of each discrete buffer zone;

[0069] S14. Based on the vegetation index time series curve of the discrete buffer zone, calculate the change size and degree of the vegetation index before and after the construction of the centralized photovoltaic power station:

[0070] ΔVI slp =VI slp,t2 -VI slp,t1

[0071] in, and ΔVI slp are the magnitude and degree of change of vegetation index respectively; and is the average vegetation index of the vegetation index time series curve of the discrete buffer zone before and after the construction of the centralized photovoltaic power station; VIslp,t1 and VI slp,t2 They are the slopes of the vegetation index time series curves of the discrete buffer zone before and after the construction of the centralized photovoltaic power station.

[0072] S15. According to the change size and degree of the vegetation index of all discrete buffer zones, the relationship curves between the change size and degree of the vegetation index and the distance from the buffer zone to the centralized photovoltaic power station are drawn respectively, and the distance when it tends to be stable is selected as the threshold range of the impact of the centralized photovoltaic power station on the vegetation.

[0073] In step S2, a buffer zone with a distance from the centralized photovoltaic power station within a threshold range is constructed and used as an influence zone, and then a mask of the influence zone is established;

[0074] In step S3, within a preset distance range from the impact area, an area with substantially the same physical geography, shape and area as the impact area is selected as a control area, and a mask of the control area is established; the physical geography may specifically refer to climate, topography, land use, soil type and vegetation type, etc.

[0075] In step S4, according to the masks of the impact area and the control area, the vegetation index time series of all grids within the mask range are extracted to construct the monthly / yearly vegetation index time series curves of the impact area and the control area:

[0076] S41, using the maximum value synthesis method, selecting the maximum value of the vegetation index of each month / year in the vegetation index time series as the output value;

[0077] S42. Draw monthly / yearly vegetation index time series curves of the impact area and the control area according to the output values ​​of the impact area and the control area.

[0078] In step S5, the change amplitude and average change rate of the vegetation index time series curves of the impact area and the control area are calculated respectively according to the vegetation index time series curves:

[0079]

[0080]

[0081] Where CA is the change amplitude; ACR is the average change rate; i is the time series index; VI i+1 With VI i are the i+1th and ith vegetation indices respectively; m is the length of the time series; Δt i For VI i+1 With VI i The time interval between them; max is the maximum value; min is the minimum value.

[0082] In step S6, the difference in the change amplitude and the difference in the average change rate between the impact area and the control area are calculated as a quantitative assessment of the impact of the construction of a centralized photovoltaic power station on vegetation from a spatial comparison.

[0083] In one embodiment of the present invention, the method for evaluating the impact of centralized photovoltaic power station development on vegetation also includes a time comparison quantitative evaluation method for the impact on vegetation:

[0084] A1. Divide the affected area into the t1 period before the construction of the centralized photovoltaic power station and the t2 period after the construction;

[0085] A2. Extract the climatic and environmental factors of the impact area at period t1, and construct a climatic-environmental factor time series with the same length as the vegetation index time series of the impact area; climatic factors include precipitation, temperature, etc., and environmental factors include topography, soil, etc.

[0086] A3. Use the climate-environmental factor time series and vegetation index time series of period t1 to train multiple machine learning models, and select the machine learning model with the highest accuracy;

[0087] A4. Input the time series of climate-environmental factors in the affected area at period t2 into the selected machine learning model to obtain the time series of predicted vegetation index affected by climate-environmental factors at period t2;

[0088] A5. Based on the vegetation index time series in the impact area at period t2 and the predicted vegetation index time series, calculate the contribution ratio of climate-environmental factors and centralized photovoltaic power station construction to the vegetation index change:

[0089] ΔVI clim =ΔVI-ΔVI PV

[0090]

[0091] Among them, ΔVI is the change of vegetation index in period t2 relative to period t1; ΔVI PV is the change in vegetation index caused by the construction of photovoltaic power stations; ΔVI clim is the magnitude of the change in vegetation index due to climate-environmental factors; and are the observed average values ​​of vegetation index during period t1 and period t2, respectively; is the average value of the predicted vegetation index time series; CR clim and CR PV They are the contribution ratios of climate-environmental factors and photovoltaic power station construction to the changes in vegetation index.

[0092] The detailed implementation process of step A3 of this solution is as follows:

[0093] The climate-environmental factor time series and vegetation index time series in period t1 are divided into a training set (accounting for a in period t1) and a validation set (accounting for b in period t1), where a+b=100%. For the training period, a machine learning model (such as random forest RF, support vector machine SVM, artificial neural network ANN, etc.) is selected, and the mapping relationship between climate-environmental factors and vegetation index is established through model parameter optimization.

[0094] To prove the effectiveness of the model, the climate-environmental factor time series in the validation set was used as the model input to obtain the simulated vegetation index. By introducing evaluation indicators, the simulated vegetation index was compared with the observed vegetation index to evaluate the simulation effect of the machine learning model. The evaluation indicators included mean square error (MSE), mean absolute error (MAE), and coefficient of determination (R 2 ), etc. The calculation formula is as follows:

[0095]

[0096]

[0097]

[0098] Where j is the time series index of the vegetation index during the validation period, j∈[1,J], and J is the length of the time series during the validation period; Y j is the observed value of vegetation index in period j, is the simulated value of the vegetation index in time period j, is the observed average value of vegetation index in the whole period; the closer MSE and MAE are to 0, the higher R 2 The closer it is to 1, the better the simulation accuracy of the machine learning model. The evaluation indicators can be used to compare the simulation accuracy of multiple machine learning models (such as random forest RF, support vector machine SVM, artificial neural network ANN, etc.) and select the machine learning model with the best performance.

[0099] In this scheme, steps S2 to S6 mainly determine the impact on vegetation by comparing the corresponding parameters of the impact area and the control area, which can be regarded as a spatial comparison. In this scheme, steps A1 to A5 mainly analyze the impact area in time, which can be regarded as a temporal comparison.

[0100] In order to facilitate the understanding of this scheme, the evaluation method of this scheme is described in detail below with reference to specific examples:

[0101] refer to Figure 2The flowchart of the threshold range part in the embodiment can specifically correspond to the technical means of steps S11 to S15 of this solution. A centralized photovoltaic power station is selected as the research object, and the spatial coverage of the centralized photovoltaic power station is identified based on Google ultra-high resolution images, that is, Figure 3 The black area in the middle. With the photovoltaic power station as the center and Δd = 100m as the distance step, discrete buffer zones are established one by one, and the number of discrete buffer zones n = 20.

[0102] Based on LANDSAT remote sensing images, the Normalized Difference Vegetation Index (NDVI) was selected as an indicative indicator. According to the mask of each discrete buffer, the NDVI time series of all grids within the mask range was extracted. The maximum value composite method (MVC) was applied to the extracted NDVI time series data, and the maximum value of the monthly NDVI was selected as the output value to construct the monthly NDVI time series of each discrete buffer.

[0103] For any discrete buffer zone, two vegetation change indicators are calculated, namely, the vegetation change magnitude and vegetation change degree ΔNDVI slp For all discrete buffers, two vegetation change indicators and the distance d from the buffer to the centralized photovoltaic power station are plotted respectively. Buf-PV The relationship curve is as follows: Figure 4 and Figure 5 All discrete buffers are are all positive, indicating that the vegetation coverage increased after the construction of the centralized photovoltaic power station at t2 compared with that before construction at t1, and with the increase of d Buf-PV increase, Gradually decreases, at d Buf-PV It tends to be stable at about 1000-1400m. ΔNDVI of all discrete buffer zones slp are all positive, indicating that the vegetation coverage growth rate becomes faster at t2 after the construction of the centralized photovoltaic power station compared with t1 before the construction, and as d Buf-PV The increase in ΔNDVI slp Gradually decreases, at d Buf-PV It tends to be stable at about 800-1200m. Taking the above two vegetation change indicators into consideration, it is finally determined that the threshold range d of the impact of centralized photovoltaic power stations on vegetation is about 1000m.

[0104] refer to Figure 2The flowchart of the spatial comparison part in , which specifically corresponds to the technical means of steps S2 to S6 of this solution, selects a centralized photovoltaic power station as the research object, determines the threshold range as 1000m, establishes a buffer zone with a distance of 1000m to the centralized photovoltaic power station as the impact zone S, and based on Google ultra-high resolution remote sensing images, delineates an area near the impact zone S that is basically consistent with the natural geographical conditions before the construction of the photovoltaic power station, and has the same shape and area as the impact zone S, as the control zone C, such as Figure 6 shown.

[0105] Based on LANDSAT remote sensing images, the normalized difference vegetation index NDVI was selected as an indicative indicator, and the NDVI time series of all grids within the mask range were extracted based on the masks of the affected area S and the control area C. The maximum value synthesis method MVC was applied to the extracted NDVI time series data, and the maximum value of the monthly NDVI was selected as the output value. The monthly NDVI time series curves of the affected area S and the control area C were constructed and drawn respectively.

[0106] Using the two indicators of change amplitude CA and average change rate ACR, the changes in the NDVI time series curves of the affected area S and the control area C were calculated respectively. The results are as follows: Figure 7 As shown. It can be seen that during the entire period t, CA S =0.2454, CA C =0.1022, ΔCA S-C =0.1432, the change amplitude of the NDVI time series curve in the affected area S is CA S The change range of NDVI time series curve CA is greater than that of control area C C ; ACR S =9.62×10 -6 , ACR C =5.89×10 -6 , ΔACR S-C =3.73×10 -6 , the average change rate ACR of the NDVI time series curve of the affected area S S The average change rate ACR of the NDVI time series curve is greater than that of the control area C C .

[0107] The curve changes before and after the centralized photovoltaic power station are further compared. S =0.1659, CA C =0.0848, ΔCA S-C =0.0811, the change range of NDVI time series curve in the affected area S is CA SThe change range of NDVI time series curve CA is greater than that of control area C C ; ACR S =5.67×10 -6 , ACR C =3.76×10 -6 , ΔACR S-C =1.91×10 -6 , the average change rate ACR of the NDVI time series curve of the affected area S S The average change rate ACR of the NDVI time series curve is greater than that of the control area C C .

[0108] After the construction of centralized photovoltaic power station, t2, CA S =0.2238, CA C =0.0783, ΔCA S-C =0.1455, the change amplitude of the NDVI time series curve in the affected area S is CA S The change range of NDVI time series curve CA is greater than that of control area C C ; ACR S =1.62×10 -5 , ACR C =5.72×10 -6 , ΔACR S-C =1.05×10 -5 , the average change rate ACR of the NDVI time series curve of the affected area S S The average change rate ACR of the NDVI time series curve is greater than that of the control area C C Moreover, the difference between the two indicators, whether ΔCA S-C or ΔACR S-C After the construction of the centralized photovoltaic power station, t2 becomes larger than t1 before construction.

[0109] refer to Figure 2The flowchart of the time comparison part in the figure can specifically correspond to the technical means of steps A1 to A5 of this scheme. A centralized photovoltaic power station is selected as the research object, the threshold range is determined to be 1000m, and a buffer zone with a distance of 1000m to the centralized photovoltaic power station is established as the study area S (corresponding to the impact area in steps A1 to A5). The entire study period from 2000 to 2023 is divided into the t1 period from 2000 to 2013 before the construction of the centralized photovoltaic power station and the t2 period from 2014 to 2023. Based on the LANDSAT remote sensing image, the normalized difference vegetation index NDVI is selected as an indicative indicator. Based on the mask of the study area S, the NDVI time series of all grids within the mask range is extracted. The maximum value synthesis method MVC is applied to the extracted NDVI time series data, and the maximum value of each month's NDVI is selected as the output value. The monthly NDVI time series curve is constructed and drawn, as shown in the figure. Figure 8 shown.

[0110] The data experiment of this example shows that the two factors of temperature and precipitation have a good explanatory power for the vegetation index NDVI in the study area S. Therefore, the monthly NDVI time series and the monthly temperature and precipitation time series are used as model inputs, and the random forest RF model is used to establish the mapping relationship between the three in the t1 period. The results show that the mean square error MSE is 2.14×10 -4 , the mean absolute error MAE is 0.0101, and the determination coefficient R 2 It is 0.865, and the model simulation effect is good.

[0111] The random forest RF model is used to take the monthly temperature and precipitation in the period t2 after the construction of the centralized photovoltaic power station as the model input to simulate the monthly vegetation index VI time series in the period t2. The calculated vegetation index change ΔNDVI before and after the construction of the centralized photovoltaic power station is 0.0509, of which the vegetation index change ΔNDVI caused by the climate-environmental factor Clim is clim The contribution ratio is 0.0243, CR clim The ΔNDVI caused by the construction of photovoltaic power plants is 47.7%. PV The contribution ratio is 0.0266, CR PV It is 52.3%.

[0112] The above examples show that this scheme is operational, accurate and practical in practical applications. By identifying the spatial coverage of centralized photovoltaic power stations and establishing discrete buffer zones, the operability of the threshold range determination method is verified; based on the determined threshold range, multi-angle spatial and temporal comparative evaluations are carried out to determine the impact of photovoltaic power stations on vegetation growth, revealing the degree of influence of different factors on vegetation, and verifying that this scheme has formed a more comprehensive and accurate evaluation system compared with other schemes, especially in distinguishing the contribution ratio of climate-environmental factors and photovoltaic power station construction to vegetation growth. It has higher adaptability, flexibility and practicality. The results obtained can provide data and content on planting scale, planting method, planting type, etc. for managers when planning and designing photovoltaic power stations and ecological restoration related work.

[0113] To sum up, this scheme can provide important technical support for scientifically monitoring and evaluating the impact of centralized photovoltaic power station development on vegetation, understanding the ecological and environmental effects of photovoltaic power station construction, promoting the coordinated improvement of photovoltaic power station development and ecological governance and restoration, rationally planning and constructing photovoltaic power station projects, promoting the sustainable development of the photovoltaic industry, and assisting in the construction of ecological civilization. It has extremely broad application prospects.

Claims

1. The method for assessing the impact of centralized photovoltaic power station development on vegetation is characterized by: Includes steps: S1. Obtain remote sensing images of the area where the centralized photovoltaic power station is located, and determine the threshold range of the impact of the centralized photovoltaic power station on vegetation based on the remote sensing images and vegetation index; S2, construct a buffer zone with a distance from the centralized photovoltaic power station within a threshold range and use it as the impact zone, and then establish a mask of the impact zone; S3. Within a preset distance range from the impact area, select an area with substantially the same physical geography, shape and area as the impact area as a control area, and establish a mask for the control area; S4. According to the masks of the impact area and the control area, the vegetation index time series of all grids within the mask range are extracted, and the monthly / yearly vegetation index time series curves of the impact area and the control area are constructed; S5. Calculate the change amplitude and average change rate of the vegetation index time series curves of the impact area and the control area respectively according to the vegetation index time series curves; S6. Calculate the difference in the magnitude of change and the difference in the average rate of change between the affected area and the control area as a means of quantitatively evaluating the impact of centralized photovoltaic power station construction on vegetation from a spatial comparison perspective; Step S1 further comprises: S11. Acquire a remote sensing image of the area where the centralized photovoltaic power station is located, and identify the spatial coverage of the centralized photovoltaic power station; S12, with centralized photovoltaic power station as the center, Δ d is the distance step, establish n discrete buffers 0~Δd, Δ d ~2·Δ d ,…,( n -1)·Δ d ~ n ·Δ d , and use GIS tools to create masks for each discrete buffer zone; S13, according to the mask of each discrete buffer zone, extracting the vegetation index time series of all grids within the mask range, and constructing a monthly / yearly vegetation index time series curve of each discrete buffer zone; S14. Calculate the change size and degree of the vegetation index before and after the construction of the centralized photovoltaic power station based on the vegetation index time series curve of the discrete buffer zone; S15. According to the change size and degree of the vegetation index of all discrete buffer zones, the relationship curves between the change size and degree of the vegetation index and the distance from the buffer zone to the centralized photovoltaic power station are drawn respectively, and the distance when it tends to be stable is selected as the threshold range of the impact of the centralized photovoltaic power station on the vegetation; The expression of the change size and degree of vegetation index is: ； in, and are the magnitude and degree of change of vegetation index respectively; and is the average value of the vegetation index time series curve of the discrete buffer zone before and after the construction of the centralized photovoltaic power station; and They are the slopes of the vegetation index time series curves of the discrete buffer zone before and after the construction of the centralized photovoltaic power station.

2. The method for evaluating the impact of centralized photovoltaic power station development on vegetation according to claim 1, characterized in that: The construction method of vegetation index time series curve includes: S41, using the maximum value synthesis method, selecting the maximum value of the vegetation index of each month / year in the vegetation index time series as the output value; S42. Draw monthly / yearly vegetation index time series curves of the impact area and the control area according to the output values ​​of the impact area and the control area.

3. The method for evaluating the impact of centralized photovoltaic power station development on vegetation according to claim 1, characterized in that: The expressions for calculating the change amplitude and average change rate are: in, CA is the amplitude of change; ACR is the average rate of change; i is the time series index; and Respectively i +1 and i vegetation index; m is the length of the time series; for and the time interval between max To obtain the maximum value; min To take the minimum value.

4. The method for evaluating the impact of centralized photovoltaic power station development on vegetation according to claim 1, characterized in that: It also includes methods for temporal quantitative assessment of vegetation impacts: A1. Divide the affected area into the t1 period before the construction of the centralized photovoltaic power station and the t2 period after the construction; A2, extract the climatic and environmental factors of the impact area at period t1, and construct a climatic-environmental factor time series with the same length as the vegetation index time series of the impact area; A3. Use the climate-environmental factor time series and vegetation index time series of period t1 to train multiple machine learning models, and select the machine learning model with the highest accuracy; A4. Input the time series of climate-environmental factors in the affected area at period t2 into the selected machine learning model to obtain the time series of predicted vegetation index affected by climate-environmental factors at period t2; A5. Based on the vegetation index time series in the impact area during period t2 and the predicted vegetation index time series, the contribution ratios of climate-environmental factors and the construction of centralized photovoltaic power stations to the vegetation index change are calculated respectively.

5. The method for evaluating the impact of centralized photovoltaic power station development on vegetation according to claim 1, characterized in that: The expression for calculating the contribution ratio is: ， ， ， in, is the change of vegetation index in period t2 relative to period t1; is the change in vegetation index caused by the construction of photovoltaic power stations; is the magnitude of the change in vegetation index due to climate-environmental factors; and are the observed average values ​​of vegetation index during period t1 and period t2, respectively; is the average value of the predicted vegetation index time series; and They are the contribution ratios of climate-environmental factors and photovoltaic power station construction to the changes in vegetation index.

6. The method for evaluating the impact of centralized photovoltaic power station development on vegetation according to any one of claims 1 to 5, characterized in that: The vegetation index is the normalized difference vegetation index, the enhanced vegetation index or the leaf area index.

7. The method for evaluating the impact of centralized photovoltaic power station development on vegetation according to claim 4, characterized in that: Various machine learning models include random forest RF, support vector machine SVM and artificial neural network ANN.

Images