A method for predicting, assessing and designing the environmental impact of photovoltaic systems.

Abstract

This application discloses a method for predicting, assessing, and designing the environmental impact of photovoltaic (PV) systems. The steps include: collecting design parameters, structural parameters, and assessment parameters for both in-site and out-of-site locations of the PV system; calculating the absolute values ​​of the relative differences between the assessment parameters based on the design parameters; obtaining the final assessment parameters using the CRITIC weighting method based on the absolute values; constructing a predictive model between the design parameters, structural parameters, and assessment parameters using a machine learning algorithm; and providing optimal design parameters for the newly constructed PV system based on the final assessment parameters and the predictive model. This application can use the design and structural parameters of the system's location, utilize the predictive model to obtain the optimal assessment parameters for the proposed PV system, and then use these optimal assessment parameters to find the best structural and design parameters for the PV power station. This can assist in PV power station site selection and PV panel design.

Description

Technical Field

[0001] This application relates to the fields of energy and environment, specifically to a method for predicting, assessing, and designing the environmental impact of photovoltaic systems. Background Technology

[0002] Photovoltaic (PV) systems offer significant energy and environmental benefits and are widely used. Besides reducing carbon emissions from energy generation through renewable energy, PV systems also have a range of impacts on the surrounding ecosystem. While there is considerable research on environmental impact assessment methods for PV power plants, research on design guidelines for PV power plants is scarce. Summary of the Invention

[0003] To address the technical problems mentioned above, this application aims to provide a method that can predictively assess and find the optimal structural and design parameters for a photovoltaic power station based on the phenological environment and design scheme of the proposed photovoltaic system, thereby guiding the design of the photovoltaic power station.

[0004] To achieve the above objectives, this application provides a method for predicting, assessing, and designing the environmental impact of a photovoltaic system, comprising the following steps:

[0005] Collect design parameters, structural parameters, and evaluation parameters for both in-station and off-station photovoltaic systems;

[0006] Based on the design parameters, calculate the absolute value of the relative differences between the evaluation parameters;

[0007] Based on the absolute value, the CRITIC weighting method is used to obtain the final evaluation parameters;

[0008] A predictive model is constructed using machine learning algorithms to establish the relationship between the design parameters, the structural parameters, and the evaluation parameters.

[0009] Based on the final evaluation parameters and the prediction model, the optimal design parameters are provided for the newly built photovoltaic system.

[0010] Preferably, the design parameters include: soil electrical conductivity, atmospheric humidity at the site, atmospheric pressure, precipitation, total solar radiation, wind speed, wind direction, soil temperature at the site, soil moisture content at the site, soil evaporation at the site, initial atmospheric temperature, initial soil microorganisms, initial soil physicochemical properties, initial vegetation cover, initial surface albedo, and initial photovoltaic panel temperature; the structural parameters include: photovoltaic panel tilt angle, spacing between front and rear modules, photovoltaic panel size, photovoltaic panel height, shading percentage, operating temperature, current and voltage, operating power, photovoltaic cell material, photovoltaic backsheet material, photovoltaic frame material, and underground cable; the evaluation parameters include: soil temperature at the site, soil moisture content at the site, atmospheric temperature, atmospheric humidity at the site, soil microorganisms, soil physicochemical properties, net solar radiation, vegetation cover, surface albedo, soil evaporation at the site, and photovoltaic panel temperature.

[0011] Preferably, the method for calculating the absolute value includes:

[0012] C = (Internal value - External value) / External value

[0013] Where C represents absolute value.

[0014] Preferably, the method for obtaining the final parameters includes:

[0015] Calculate the data variability of each indicator in the evaluation parameters one by one;

[0016] Calculate the correlation coefficient between each indicator and the remaining indicators in the evaluation parameters one by one;

[0017] Based on the correlation coefficient, the conflict of indicators for each indicator in the evaluation parameters is calculated one by one;

[0018] Based on the data variability and the indicator conflict, the information content of each indicator in the evaluation parameters is calculated;

[0019] Based on the amount of information, calculate the weight of each indicator in the evaluation parameters;

[0020] Based on the weights, the weight index of each indicator in the evaluation parameters is calculated, and the absolute values ​​of all indicators are weighted and averaged to obtain the final evaluation parameters.

[0021] Preferably, the method for constructing the prediction model includes: using the evaluation parameters as model output items, determining multiple indicators with strong correlation to the output items among the design parameters and structural parameters through correlation analysis, and clarifying the model input items; constructing the machine learning model using Python, dividing the model dataset into a training set and a test set in a 3:1 ratio, wherein the training set is used for model training and the test set is used to evaluate the model simulation results; using the variance coefficient as the model evaluation coefficient, analyzing the test structure, and selecting the optimal machine learning calculation method based on the variance coefficient.

[0022] Preferably, the method for calculating the variance coefficient includes:

[0023]

[0024] Among them, y sim，i The predicted value of the i-th data set; y mea，i Refers to the experimental value of the i-th data set; The average of all experimental values.

[0025] Preferably, the method for providing the optimal design parameters includes:

[0026] S501. Initialization parameters: Select an initial temperature T0, set the annealing rate r, and set the termination condition;

[0027] S502. Generate initial solution: Randomly generate initial solution (x0, y0);

[0028] S503. Select a new solution: From the current solution (x n ,y n Starting from ), a new solution (x′, y′) is randomly generated;

[0029] S504. Calculate the acceptance probability using the following formula:

[0030] α = min1, exp(-ΔE / T) n )

[0031] Where, ΔE=f(x′,y′)-f(x n ,y n ) represents the energy difference; T n Indicates the current temperature;

[0032] S505. Update Solution:

[0033] Update the solution with the acceptance probability α. If accepted, the state becomes (x′, y′); otherwise, the state remains (x′, y′). n ,y n );

[0034] S506. Cooling: Reduce the temperature T according to the specified annealing rate r. n And check whether the termination conditions are met;

[0035] S507. Repeat steps S503 to S507 until the termination condition is met.

[0036] Preferably, the structural parameters are designed using the optimal evaluation parameters of the photovoltaic system as the objective function and the design parameters as boundary conditions, and the optimal combination of design parameters is searched to maximize the value of the optimal evaluation parameters.

[0037] Compared with the prior art, the beneficial effects of this application are as follows:

[0038] This application can use the design and structural parameters of the photovoltaic power station site to obtain the optimal evaluation parameters of the proposed photovoltaic system through a predictive model, and then use the optimal evaluation parameters to find the best structural and design parameters of the photovoltaic power station, which can help with the site selection and design of photovoltaic panels for photovoltaic power stations. Attached Figure Description

[0039] To more clearly illustrate the technical solutions of this application, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of the method flow of an embodiment of this application. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] like Figure 1 The diagram shown is a schematic representation of the method flow in this embodiment, and the steps include:

[0044] S1. Collect design parameters, structural parameters, and evaluation parameters for both in-station and out-of-station photovoltaic systems.

[0045] Extensive data was collected on the design, structural, and evaluation parameters of existing photovoltaic (PV) systems, both on-site and off-site. Design parameters (for off-site PV systems) include: soil conductivity, atmospheric humidity, atmospheric pressure, precipitation, total solar radiation, wind speed, wind direction, soil temperature, soil moisture content, soil evaporation, initial atmospheric temperature, initial soil microorganisms, initial soil physicochemical properties, initial vegetation cover, initial surface albedo, and initial PV panel temperature. Structural parameters include: PV panel tilt angle, spacing between front and rear modules, PV panel size, PV panel height, shading percentage, operating temperature, current, voltage, operating power, PV cell material, PV backsheet material, PV frame material, and underground cables. Evaluation parameters (for on-site PV systems) include: on-site soil temperature, on-site soil moisture content, atmospheric temperature, on-site atmospheric humidity, soil microorganisms, soil physicochemical properties, net solar radiation, vegetation cover, surface albedo, on-site soil evaporation, and PV panel temperature.

[0046] S2. Based on the design parameters, calculate the absolute value of the relative differences between the evaluation parameters.

[0047] The absolute value C of the relative difference between the evaluation parameters of in-station and out-of-station points in the existing site data (in this embodiment, five photovoltaic power stations are selected: Yunxi Photovoltaic Power Station in Hubei, Wuwei Photovoltaic Power Station in Gansu, Hami Photovoltaic Power Station in Xinjiang, Dali Photovoltaic Power Station in Yunnan, and Dongfang Photovoltaic Power Station in Hainan) is calculated using the following formula:

[0048] C = (Internal value - External value) / External value

[0049] When the numerical change of the point within the station relative to the point outside the station (the numerical change refers to the increase or decrease in the measured data of the point within the station compared to the measured data of the point outside the station) is beneficial to the surrounding ecological environment (an improved ecological environment refers to a series of situations such as vegetation restoration, suitable climate, and increased biomass, not limited to the three mentioned above), C is set to a positive value; conversely, C is set to a negative value. It should be noted that in this embodiment, the positive or negative sign is used to indicate whether it is positive or negative, and it is not a specific numerical sign but rather a designation where positive indicates an improved ecological environment and negative indicates a worsened ecological environment.

[0050] S3. Based on absolute values, the CRITIC weighting method is used to obtain the final evaluation parameters. The steps include:

[0051] S301. Calculate the data variability of each indicator in the evaluation parameters one by one.

[0052] Using one day as a time period and one hour as a data point, a set of 24 data points is used to calculate the standard deviation, which represents the data variability. The formula for the standard deviation is:

[0053]

[0054] Where σ represents the standard deviation, i.e., variability; x i denoted as each data point, μ represents the average value.

[0055] S302. Calculate the correlation coefficient between each indicator and the remaining indicators in the evaluation parameters one by one. The formula for the correlation coefficient is:

[0056]

[0057] This formula calculates the correlation between two sets of data, specifically the correlation coefficient between the first set of data (denoted as j) and the second set of data (denoted as k). Here, j represents the first set of data, k represents the second set of data, and x... ij This represents one data point from the first set of data. x represents the average of j sets of data; ik This represents one data point from the second set of data. This represents the average of k sets of data.

[0058] S303. Based on the correlation coefficient, calculate the conflict of indices for each indicator in the evaluation parameters one by one; the formula for the conflict of indices is:

[0059]

[0060] Among them, A j Indicates data metric conflict, r ij This represents the correlation coefficient for each data point. In this embodiment, r ij That is, the correlation coefficient of each data point calculated by the correlation coefficient calculation formula in S302.

[0061] S304. Based on data variability and indicator conflict, calculate the information content of each indicator in the evaluation parameters. The formula for the information content is:

[0062] Information content = indicator variability * indicator conflict.

[0063] S305. Based on the amount of information, calculate the weight of each indicator in the evaluation parameters; the weight formula is:

[0064]

[0065] Among them, z i For information content.

[0066] S306. Based on the weights, calculate the weight index of each indicator in the evaluation parameters, and take the weighted average of the absolute values ​​of all indicators to obtain the final evaluation parameters.

[0067] The weighting index is the sum of the weights of each index value, and then the weighted average of the C values ​​of all evaluation parameters is taken to obtain the final evaluation parameter E value.

[0068] S4. Employ machine learning algorithms to construct a predictive model between design parameters, structural parameters, and evaluation parameters.

[0069] Using machine learning algorithms including but not limited to artificial neural networks, random forests, and support vector machines, and based on operational data from existing photovoltaic system sites, a predictive model is constructed to connect design parameters, structural parameters, and evaluation parameters. The specific steps are as follows: Using evaluation parameters as model outputs, correlation analysis is used to identify multiple parameters among the design and structural parameters that are strongly correlated with the outputs, thus defining the model inputs; the machine learning model is constructed using Python, with the model dataset divided into training and testing sets in a 3:1 ratio. The training set is used for model training, and the testing set is used to evaluate the model simulation results; the coefficient of variance (R²) is used... 2 ) are used as model evaluation coefficients to analyze the test structure, with R... 2 Choose the optimal machine learning computation method. R 2 The specific calculation method is as follows (the characters in the formula represent the test set data content):

[0070]

[0071] Among them, y sim，i The predicted value of the i-th data set; y mea，i Refers to the experimental value of the i-th data set; The average of all experimental values.

[0072] S5. Based on the weights, calculate the weight index of each indicator in the evaluation parameters, and take the weighted average of the absolute values ​​of all indicators to obtain the final evaluation parameters.

[0073] When constructing a new photovoltaic system, the design parameters for the system's location are first determined, and the system is initially designed and its structural parameters are determined. Then, a predictive model is used to obtain the evaluation parameters for the proposed photovoltaic system. Next, using the methods in steps S2 and S3, the E-value of the system is obtained, along with the environmental impact assessment results for the system under this design scheme. (The assessment results are based on the E-value, i.e., the weight of each parameter and the magnitude of changes inside and outside the site, to determine whether the parameters with a large weight have significant changes, thus judging the degree of environmental impact, i.e., the environmental impact assessment result.)

[0074] Subsequently, simulated annealing was employed, with the system's E-value as the objective function and the design parameters as boundary conditions, to design the structural parameters and search for the optimal combination of design parameters to maximize the E-value. Specific methods included:

[0075] S501. Initialization parameters: Select an initial temperature T0, set the annealing rate r, and set the termination condition (the termination condition is the combination that maximizes the E value).

[0076] S502. Generate initial solution: Randomly generate initial solution (x0, y0).

[0077] S503. Select a new solution: From the current solution (x n ,y n Starting from , a new solution (x′, y′) is randomly generated.

[0078] S504. Calculate the acceptance probability using the following formula:

[0079] α = min1, exp(-ΔE / T) n )

[0080] Where, ΔE=f(x′,y′)-f(x n ,y n ) represents the energy difference; T n This indicates the current temperature.

[0081] S505. Update Solution:

[0082] Update the solution with the acceptance probability α. If accepted, the state becomes (x′, y′); otherwise, the state remains (x′, y′). n ,y n ).

[0083] S506. Cooling: Reduce the temperature T according to the specified annealing rate r. n And check whether the termination conditions are met.

[0084] S507. Repeat steps S503 to S507 until the termination condition is met.

[0085] The embodiments described above are merely preferred embodiments of this application and are not intended to limit the scope of this application. Any modifications and improvements made to the technical solutions of this application by those skilled in the art without departing from the spirit of this application shall fall within the protection scope defined by the claims of this application.

Claims

1. A method for predicting, assessing, and designing the environmental impact of a photovoltaic system, characterized by the following steps: include: Obtain the design parameters, structural parameters, and evaluation parameters for both in-station and off-station points of the photovoltaic system; Based on the design parameters, calculate the absolute value of the relative differences between the evaluation parameters; Based on the absolute value, the CRITIC weighting method is used to obtain the final evaluation parameters; A predictive model is constructed using machine learning algorithms to establish the relationship between the design parameters, the structural parameters, and the evaluation parameters. Based on the final evaluation parameters and the prediction model, the optimal design parameters for the new photovoltaic system are obtained. The methods for providing these optimal design parameters include: S501. Initialization parameters: Select an initial temperature T0, set the annealing rate r, and set the termination condition; S502. Generate initial solution: Randomly generate initial solution (x0, y0); S503. Select a new solution: From the current solution (x n, y n Starting from ), a new solution (x′, y′) is randomly generated; S504. Calculate the acceptance probability using the following formula: ɑ=min[1,exp(−ΔE / T n )] Where, ΔE=f(x′,y′)−f(x n ,y n ) represents the energy difference; T n Indicates the current temperature; S505. Update Solution: Update the solution with the acceptance probability α. If accepted, the state becomes (x′, y′); otherwise, the state remains (x′, y′). n ,y n ); S506. Cooling: Reduce the temperature T according to the specified annealing rate r. n And check whether the termination condition is met; S507. Repeat steps S503 to S507 until the termination condition is met.

2. The method for predicting, assessing, and designing the environmental impact of a photovoltaic system according to claim 1, characterized in that, The design parameters include: soil electrical conductivity, atmospheric humidity at the site, atmospheric pressure, precipitation, total solar radiation, wind speed, wind direction, soil temperature at the site, soil moisture content at the site, soil evaporation at the site, initial atmospheric temperature, initial soil microorganisms, initial soil physicochemical properties, initial vegetation cover, initial surface albedo, and initial photovoltaic panel temperature; the structural parameters include: photovoltaic panel tilt angle, spacing between front and rear modules, photovoltaic panel size, photovoltaic panel height, shading percentage, operating temperature, current and voltage, operating power, photovoltaic cell material, photovoltaic backsheet material, photovoltaic frame material, and underground cable; the evaluation parameters include: soil temperature at the site, soil moisture content at the site, atmospheric temperature, atmospheric humidity at the site, soil microorganisms, soil physicochemical properties, net solar radiation, vegetation cover, surface albedo, soil evaporation at the site, and photovoltaic panel temperature.

3. The method for predicting, assessing, and designing the environmental impact of a photovoltaic system according to claim 1, characterized in that, The method for calculating the absolute value includes: Where C represents absolute value.

4. The method for predicting, assessing, and designing the environmental impact of a photovoltaic system according to claim 1, characterized in that, The methods for obtaining the final evaluation parameters include: Calculate the data variability of each indicator in the evaluation parameters one by one; Calculate the correlation coefficient between each indicator and the remaining indicators in the evaluation parameters one by one; Based on the correlation coefficient, the conflict of indicators for each indicator in the evaluation parameters is calculated one by one; Based on the data variability and the indicator conflict, the information content of each indicator in the evaluation parameters is calculated; Based on the amount of information, calculate the weight of each indicator in the evaluation parameters; Based on the weights, the weight index of each indicator in the evaluation parameters is calculated, and the absolute values ​​of all indicators are weighted and averaged to obtain the final evaluation parameters.

5. The method for predicting, assessing, and designing the environmental impact of a photovoltaic system according to claim 4, characterized in that, The method for constructing the prediction model includes: using the evaluation parameters as model output items, determining multiple indicators with strong correlation to the output items among the design parameters and structural parameters through correlation analysis, and clarifying the model input items; constructing the machine learning model using Python, dividing the model dataset into a training set and a test set in a 3:1 ratio, where the training set is used for model training and the test set is used to evaluate the model simulation results; using the variance coefficient as the model evaluation coefficient, analyzing the test structure, and selecting the optimal machine learning calculation method based on the variance coefficient.

6. The method for predicting, assessing, and designing the environmental impact of a photovoltaic system according to claim 5, characterized in that, The method for calculating the variance coefficient includes: Wherein, represents the predicted value of the i-th group of data; represents the experimental value of the i-th group of data; and represents the average value of all experimental values.

7. The method for predicting, assessing, and designing the environmental impact of a photovoltaic system according to claim 1, characterized in that, Using the final evaluation parameters of the photovoltaic system as the objective function and the design parameters as the boundary conditions, the structural parameters are designed, and the optimal combination of design parameters is searched to maximize the final evaluation parameter value.

Images