Method and device for acquiring microscopic parameter aging curve of perovskite solar cell and medium
By constructing a perovskite solar cell model and using stochastic algorithms and machine learning reverse reversal, the acquisition of the microparameter change law of perovskite solar cell is simplified, the cumbersome and time-consuming problems in the existing technology are solved, and the aging process of perovskite solar cells is achieved quickly detecting and optimizing the aging process of perovskite solar cells is realized.
Patent Information
- Application Number
- CN202510544357.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-28
- Publication Date
- 2025-08-19
AI Technical Summary
The existing methods for obtaining microparameter variation patterns of perovskite solar cells require aging experiments and combined with a large number of parameter characterization techniques. The process is cumbersome and time-consuming, resulting in inefficient efficiency.
By obtaining the performance parameter aging curve of perovskite solar cells, building a model based on its structure and process, using a random algorithm to generate microparameters and train performance parameter prediction models, reverse inversion and obtaining the microparameter aging curve, combining simulation modeling and machine learning to simplify the acquisition process of microparameter change law.
Without the need for a large number of sample aging experiments and characterization technology measurements, the microparameter changes in the aging process of perovskite solar cells are quickly detected, providing a basis for optimization of structure and preparation process, and simplifying the acquisition process of microparameter changes.
Smart Images
Figure CN120509162A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of perovskite solar cells, and in particular to a method and device for obtaining a microscopic parameter aging curve of a perovskite solar cell, and a computer-readable storage medium. Background Art
[0002] Over the past fifteen years, perovskite solar cells (PSCs) have experienced rapid development, with certified power conversion efficiencies of laboratory-scale devices increasing from 3.8% to 27%, making them one of the most promising technologies in the photovoltaic field. Despite these achievements, the stability of PSCs remains inferior to that of conventional silicon solar cells, with lifetimes of the former measured in the thousands of hours, while the latter typically exceed 170,000 hours. The stability of PSCs is influenced by microscopic parameters, which, in turn, are determined by the structure and fabrication process. A key limitation to the stability of existing PSCs is the time-consuming process required to measure these parameters. This makes it difficult to detect the changes in these parameters (such as interface defects and ion mobility) during aging, hindering the identification of the specific causes of PSC performance degradation. Consequently, it is difficult to optimize the structure and fabrication process based on these changes, thereby hindering the ability to optimize PSC performance.
[0003] In the existing technology, in order to measure the change law of microscopic parameters during the aging process of perovskite solar cells, it is often necessary to combine multiple characterization techniques, which specifically include: 1. Prepare multiple perovskite solar cell samples with the same structure and process, and measure the photoelectric conversion efficiency, open circuit voltage, short circuit current and other performance parameters of each sample; 2. Place each sample in a light aging test system for different degrees of light aging experiments, and then place each sample in a wet heat aging box for different degrees of wet heat aging experiments, and measure the performance parameters of each sample after the aging experiment; 3. Place the aged sample on the sample stage of a photoluminescence spectrometer, and use a laser or light source of a specific wavelength to excite the sample, so as to obtain the photoluminescence spectrum of the sample to reflect the interface defects or bulk defects of the sample after the aging experiment; 4. Connect the sample to an electrochemical workstation and measure the sample at different The impedance response under the frequency AC signal is used to obtain the electrochemical impedance spectrum of the sample to reflect the interface resistance and charge transfer of the sample after the aging experiment; 5. The sample is placed in a secondary ion mass spectrometer for secondary ion mass spectrometry testing to obtain the ion migration of the sample; 6. The sample is placed in an X-ray photoelectron spectrometer for X-ray photoelectron spectroscopy testing to obtain an XPS spectrum to analyze the surface chemical state and chemical bond changes of the sample after the aging experiment; 7. Comprehensively analyze the performance parameters, photoluminescence spectra, electrochemical impedance spectra, ion migration data and XPS spectra of each sample to establish the relationship between the microscopic parameters of the perovskite solar cell with the current structure and process and the battery stability. This method requires a large number of characterization technologies, and the process is cumbersome and time-consuming, which makes the process of obtaining the change law of the microscopic parameters of the perovskite solar cell very inefficient.
[0004] In summary, the existing methods for obtaining the variation laws of the microscopic parameters of perovskite solar cells require aging experiments and the combination of a large number of parameter characterization technologies. The process is cumbersome and time-consuming, resulting in low efficiency in obtaining the variation laws of the microscopic parameters of perovskite solar cells. Summary of the Invention
[0005] To this end, the technical problem to be solved by the present invention is to overcome the problem that the existing method for obtaining the variation law of the microscopic parameters of perovskite solar cells requires aging experiments and the combination of a large number of parameter characterization technologies. The process is cumbersome and time-consuming, resulting in low efficiency in obtaining the variation law of the microscopic parameters of perovskite solar cells.
[0006] To solve the above technical problems, the present invention provides a method for obtaining a microscopic parameter aging curve of a perovskite solar cell, comprising: S10: Obtaining a performance parameter aging curve of the perovskite solar cell, and constructing a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and obtaining performance parameters of each perovskite solar cell model; S20: Using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, a performance parameter prediction model of the perovskite solar cell is trained and initialized to t=1; S30: generating microscopic parameters at time t using a random algorithm, inputting the microscopic parameters at time t into the perovskite solar cell performance parameter prediction model, and outputting the predicted performance parameters at time t; S40: Calculate a first error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve. If the first error is greater than or equal to a first preset error threshold, return to step S30 and execute until the first error is less than the first preset error threshold, thereby obtaining the microscopic parameter at time t. S50: Update t=t+1, and return to step S30 until t=T, and obtain the micro parameter aging curve of the perovskite solar cell based on the micro parameters at time 1~T; wherein T represents the end time of the performance parameter aging curve.
[0007] Preferably, after obtaining the microscopic parameter aging curve of the perovskite solar cell, the method further comprises: S60: inputting the microscopic parameters at time 1 to T in the microscopic parameter aging curve of the perovskite solar cell into the perovskite solar cell performance parameter prediction model, outputting the predicted performance parameters at time 1 to T, thereby obtaining the predicted performance parameter aging curve of the perovskite solar cell; S70: Calculate a second error between the predicted performance parameter aging curve of the perovskite solar cell and the performance parameter aging curve. If the second error is greater than or equal to a second preset error threshold, return to step S30 and execute until the second error is less than the second preset error threshold to obtain the target microscopic parameter aging curve of the perovskite solar cell.
[0008] Preferably, the microscopic parameters of each perovskite solar cell model are used as input and the performance parameters are used as output, and the perovskite solar cell performance parameter prediction model is trained to obtain the following: Using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, multiple machine learning models are trained to obtain multiple trained machine learning models; The performance parameter prediction error of each machine learning model is calculated respectively, and the machine learning model with the smallest performance parameter prediction error is used as the performance parameter prediction model of the perovskite solar cell.
[0009] Preferably, the first error is a root mean square error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve.
[0010] Preferably, the first preset error threshold is less than or equal to 10%.
[0011] Preferably, calculating the second error between the predicted performance parameter aging curve and the performance parameter aging curve of the perovskite solar cell comprises: Calculate the mean square error, mean absolute error, or root mean square error between the predicted performance parameter and the performance parameter at the same time in the predicted performance parameter aging curve and the performance parameter aging curve; Based on the sum of the mean square error, mean absolute error or root mean square error between the predicted performance parameters and the performance parameters at all moments, a second error between the predicted performance parameter aging curve and the performance parameter aging curve is obtained.
[0012] Preferably, the second preset error threshold is less than or equal to 10%.
[0013] Preferably, the microscopic parameters are body defects, interface defects, electron lifetime, hole lifetime, electron mobility, hole mobility, and doping concentration; The performance parameters are open circuit voltage, short circuit current density, fill factor, and photoelectric conversion efficiency.
[0014] The present invention also provides a device for obtaining a microscopic parameter aging curve of a perovskite solar cell, comprising: A data acquisition module is used to obtain the performance parameter aging curve of the perovskite solar cell, and to construct a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and to obtain the performance parameters of each perovskite solar cell model; The model training module is used to train the perovskite solar cell performance parameter prediction model using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, and initialize t=1; A performance parameter prediction module is used to generate microscopic parameters at time t using a random algorithm, input the microscopic parameters at time t into the perovskite solar cell performance parameter prediction model, and output the predicted performance parameters at time t; a micro-parameter generation module, configured to calculate a first error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve; if the first error is greater than or equal to a first preset error threshold, returning to the steps of the performance parameter prediction module until the first error is less than the first preset error threshold, thereby obtaining the micro-parameter at time t; The micro-parameter aging curve acquisition module is used to update t=t+1 and return to the steps of executing the performance parameter prediction module until t=T, and obtain the micro-parameter aging curve of the perovskite solar cell based on the micro-parameters at time 1~T; wherein T represents the cutoff time of the performance parameter aging curve.
[0015] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for obtaining the microscopic parameter aging curve of the perovskite solar cell according to any one of claims 1 to 8 are implemented.
[0016] The method for obtaining the microscopic variable aging curve of perovskite solar cells provided in this application has the following beneficial effects: By obtaining the performance parameter aging curve of the perovskite solar cell, and simulating and modeling the perovskite solar cell based on its structure and process, the perovskite solar cell model under different microscopic parameters is obtained, and then the performance parameters of different perovskite solar cells are simulated to obtain their performance parameters, so that a perovskite solar cell performance parameter prediction model with microscopic parameters as input and performance parameters as output can be trained; then the genetic algorithm is used to reversely infer the microscopic parameters of the perovskite solar cell. Specifically, initialize t=1, use a random algorithm to generate the microscopic parameters at time t, and input the microscopic parameters into the perovskite solar cell performance parameter prediction model. Energy parameter prediction model, obtain the predicted performance parameters at time t, compare the predicted performance parameters with the performance parameters at time t in the performance parameter aging curve, if the error between the two is small, it indicates that the generated microscopic parameters at time t are consistent with the actual microscopic parameters at time t, if the error between the two is large, it indicates that the generated microscopic parameters do not conform to the actual values, so it is necessary to regenerate the microscopic parameters at time t, and use this method to generate microscopic parameters at time 1~T that conform to the performance parameter aging law, and then obtain the aging curve of microscopic parameters over time, so as to deeply study the change law of microscopic parameters during the aging process of perovskite solar cells. The method provided in this application combines simulation modeling, machine learning and genetic algorithm inverse theory to obtain the change law of microscopic parameters of perovskite solar cells. There is no need to conduct aging experiments on a large number of samples and use characterization technology to measure the battery parameters after the experiment, which greatly simplifies the process of obtaining the change law of microscopic parameters, so that the change of microscopic parameters during the aging process of perovskite solar cells can be quickly detected, providing a theoretical basis for the optimization of perovskite solar cell structure and preparation process; After fitting the micro-parameter aging curve, in order to verify the accuracy of the curve, this application again inputs the micro-parameters at time 1~T used to construct the micro-parameter aging model into the perovskite solar cell performance parameter prediction model, and constructs a predicted performance parameter aging curve based on the output of the model. By comparing the error between the predicted curve and the actual curve, it is detected whether the performance parameter change law under the condition of micro-parameter changes at time 1~T conforms to the actual law, thereby verifying the accuracy of the micro-parameter aging curve. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings, wherein: Figure 1 Flowchart of the method for obtaining the microscopic parameter aging curve of perovskite solar cells provided in this application; Figure 2 This is a schematic diagram comparing different predicted performance parameter aging curves obtained by inverse deduction of the microscopic parameter aging curve provided in this application with the actual aging curve; wherein, Figure 2 (a) is a schematic diagram comparing the open circuit voltage predicted aging curve and the open circuit voltage actual aging curve. Figure 2 (b) is a schematic diagram comparing the short-circuit current density predicted aging curve and the short-circuit current density actual aging curve. Figure 2 (c) is a comparison diagram of the fill factor predicted aging curve and the fill factor actual aging curve. Figure 2 (d) is a schematic diagram comparing the predicted aging curve of photoelectric conversion efficiency and the actual aging curve of photoelectric conversion efficiency; Figure 3 A curve showing a change in the body defect density of a perovskite solar cell during aging obtained using the method provided in this application; Figure 4 Schematic diagram of the structure of the device for obtaining the microscopic parameter aging curve of the perovskite solar cell provided in this application. DETAILED DESCRIPTION
[0018] The present invention will be further described below with reference to the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. However, the embodiments are not intended to limit the present invention.
[0019] See also Figure 1 , Figure 1 The figure shows a flow chart of a method for obtaining a microscopic parameter aging curve of a perovskite solar cell provided in this application. The method specifically includes: S10: Obtaining a performance parameter aging curve of the perovskite solar cell, and constructing a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and obtaining performance parameters of each perovskite solar cell model.
[0020] Furthermore, the microscopic parameters include one or more of body defects, interface defects, electron / hole lifetime, electron / hole mobility, and doping concentration.
[0021] Performance parameters include open circuit voltage , short-circuit current density , fill factor , photoelectric conversion efficiency wait.
[0022] Specifically, TCAD or COMSOL simulation software can be used to model perovskite solar cells. By changing the input parameters of the model (for example, one or more of the microscopic parameters such as body defects, interface defects, electron / hole lifetime, electron / hole mobility, and doping concentration), perovskite solar cell models corresponding to different microscopic parameters can be obtained. By simulating different perovskite solar cell models, the performance parameters of the model can be obtained.
[0023] In some embodiments of the present application, an aging test can be performed on the perovskite solar cell to obtain its performance parameter aging curve, or existing aging data of perovskite solar cells with the same cell structure and preparation process (data in existing research or literature) can be directly used. In addition, the perovskite solar cell model can be directly simulated by using simulation software. The aging simulation program is run in the simulation software according to the set aging mechanism and conditions to simulate the performance changes of the perovskite solar cell under different aging times, thereby obtaining a performance parameter aging curve. It is worth noting that even if an aging test is performed on the perovskite solar cell to obtain its performance parameter aging curve, it is only necessary to measure the performance parameters such as the open circuit voltage and short circuit current density of the perovskite solar cell under different aging conditions, and multiple characterization techniques cannot be used.
[0024] S20: Using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, a performance parameter prediction model of the perovskite solar cell is trained and initialized to t=1.
[0025] S30: Generate microscopic parameters at time t using a random algorithm, input the microscopic parameters at time t into the perovskite solar cell performance parameter prediction model, and output the predicted performance parameters at time t.
[0026] S40: Calculate the first error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve. If the first error is greater than or equal to the first preset error threshold, return to step S30 and execute until the first error is less than the first preset error threshold to obtain the microscopic parameter at time t.
[0027] Furthermore, the first preset error threshold is less than or equal to 10%. It is worth noting that, in a specific embodiment of the present application, the first preset error threshold is set below 5%, thereby further improving the accuracy of the generated microscopic parameters.
[0028] Specifically, the first error is a root mean square error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve.
[0029] If the first error is small, it indicates that the generated microscopic parameters at time t are consistent with the actual microscopic parameters at time t, that is, the performance parameters of the perovskite solar cell under these microscopic parameters are consistent with the actual performance parameters; if the first error is large, it indicates that the generated microscopic parameters do not conform to the actual values, indicating that the performance parameters of the perovskite solar cell under these microscopic parameters do not conform to the actual values, and therefore the microscopic parameters at time t need to be regenerated.
[0030] S50: Update t=t+1, and return to step S30 until t=T, and obtain the micro parameter aging curve of the perovskite solar cell based on the micro parameters at time 1~T; wherein T represents the end time of the performance parameter aging curve.
[0031] By repeating steps S30 to S50, micro-parameter time series data that conforms to the aging patterns of perovskite solar cell performance parameters can be generated, thereby obtaining a micro-parameter aging curve for the perovskite solar cell. By combining simulation modeling, machine learning, and genetic algorithm inverse theory, the micro-parameter variation patterns of perovskite solar cells can be obtained. This eliminates the need to conduct aging experiments on a large number of samples and measure the cell parameters after the experiments using characterization techniques. This greatly simplifies the process of measuring micro-parameter variation patterns, allowing for rapid detection of micro-parameter variations during the aging process of perovskite solar cells.
[0032] Furthermore, after obtaining the micro-parameter aging curve of the perovskite solar cell, the micro-parameters that have the main influence on the performance degradation of the perovskite solar cell can be determined, and then the preparation process and cell structure can be optimized by combining the change trend of the micro-parameters and the structural and process characteristics of the perovskite solar cell. For example, by comparing the micro-parameter aging curves, it is found that the defect density gradually increases with time, resulting in a decrease in the performance parameters of the perovskite solar cell. This may be due to insufficient annealing during the preparation process. Based on this, the annealing parameters during the preparation of the perovskite solar cell can be increased to obtain an optimized perovskite solar cell; by comparing the micro-parameter aging curves, it is found that the electron lifetime gradually decreases with time, resulting in a decrease in the performance parameters of the perovskite solar cell. In this case, a buffer layer can be set between the perovskite layer and the electron transport layer or the surface of the perovskite layer can be passivated to optimize the performance of the perovskite solar cell.
[0033] Specifically, in step S20, the microscopic parameters of each perovskite solar cell model are used as input, and the performance parameters are used as output. The perovskite solar cell performance parameter prediction model is trained to obtain the following: Using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, multiple machine learning models are trained to obtain multiple trained machine learning models; The performance parameter prediction error of each machine learning model is calculated respectively, and the machine learning model with the smallest performance parameter prediction error is used as the performance parameter prediction model of the perovskite solar cell.
[0034] Specifically, the machine learning model can be a linear regression model, support vector product, convolutional neural network, random forest model, etc.; the performance parameter prediction error can be the mean square error, mean absolute error, etc. of the prediction results of the machine learning model on the validation set.
[0035] Furthermore, if Figure 1 As shown, in order to ensure the accuracy of the generated micro-variable time series data and the accuracy of the micro-parameter aging curve, the present application further includes the following steps after obtaining the micro-parameter aging curve: S60: Inputting the microscopic parameters at time 1 to T in the microscopic parameter aging curve of the perovskite solar cell into the perovskite solar cell performance parameter prediction model, outputting the predicted performance parameters at time 1 to T, thereby obtaining the predicted performance parameter aging curve of the perovskite solar cell.
[0036] S70: Calculate a second error between the predicted performance parameter aging curve of the perovskite solar cell and the performance parameter aging curve. If the second error is greater than or equal to a second preset error threshold, return to step S30 and execute until the second error is less than the second preset error threshold to obtain the target microscopic parameter aging curve of the perovskite solar cell.
[0037] Optionally, in some embodiments of the present application, the performance parameter aging curve used to calculate the second error may be the performance parameter aging curve obtained in step S10, or may be a performance parameter aging curve of other perovskite solar cells with the same structure and process.
[0038] Specifically, calculating the second error between the predicted performance parameter aging curve and the performance parameter aging curve includes: Calculate the mean square error, mean absolute error, or root mean square error between the predicted performance parameter and the performance parameter at the same time in the predicted performance parameter aging curve and the performance parameter aging curve; Based on the sum of the mean square error, mean absolute error or root mean square error between the predicted performance parameters and the performance parameters at all moments, a second error between the predicted performance parameter aging curve and the performance parameter aging curve is obtained.
[0039] Furthermore, the second preset error threshold is less than or equal to 10%. In a specific example of the present application, the second error between the predicted performance parameter aging curve obtained based on the micro-parameter aging curve and the actual performance parameter aging curve can be less than 5%, which also shows that the method provided by the present application can obtain a micro-parameter aging curve with high accuracy.
[0040] The technical solution of the present application is described in more detail below in conjunction with a plurality of embodiments. However, it should be understood that the following embodiments are only for explaining and illustrating the technical solution and do not limit the scope of the present application.
[0041] Example 1 of the present application provides a method for obtaining a microscopic parameter aging curve of a perovskite solar cell, comprising the following steps: S100: Obtaining an open-circuit voltage aging curve of the perovskite solar cell, and constructing a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and obtaining the open-circuit voltage of each perovskite solar cell model.
[0042] S200: using the microscopic parameters of each perovskite solar cell model as input and the open circuit voltage as output, training is performed to obtain the open circuit voltage prediction model of the perovskite solar cell, and initializing t=1; S300: generating microscopic parameters at time t using a random algorithm, inputting the microscopic parameters at time t into the open-circuit voltage prediction model of the perovskite solar cell, and outputting the predicted open-circuit voltage at time t; S400: Calculating a first error between the predicted open circuit voltage at time t and the open circuit voltage at time t in the open circuit voltage aging curve. If the first error is greater than or equal to a first preset error threshold, returning to step S300 and executing until the first error is less than the first preset error threshold, thereby obtaining a microscopic parameter at time t. S500: Update t=t+1, and return to step S300 until t=T, and obtain the micro parameter aging curve of the perovskite solar cell based on the micro parameters at time 1-T; wherein T represents the cutoff time of the open circuit voltage aging curve.
[0043] Example 2 of the present application provides a method for obtaining a microscopic parameter aging curve of a perovskite solar cell, comprising the following steps: S100: Obtain a short-circuit current density aging curve of the perovskite solar cell, and construct a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell to obtain the short-circuit current density of each perovskite solar cell model.
[0044] S200: using the microscopic parameters of each perovskite solar cell model as input and the short-circuit current density as output, training is performed to obtain a prediction model for the short-circuit current density of the perovskite solar cell, and initializing t=1; S300: generating microscopic parameters at time t using a random algorithm, inputting the microscopic parameters at time t into the perovskite solar cell short-circuit current density prediction model, and outputting the predicted short-circuit current density at time t; S400: Calculating a first error between the predicted short-circuit current density at time t and the short-circuit current density at time t in the short-circuit current density aging curve. If the first error is greater than or equal to a first preset error threshold, returning to step S300 and executing until the first error is less than the first preset error threshold, thereby obtaining a microscopic parameter at time t. S500: Update t=t+1, and return to step S300 until t=T, and obtain the micro parameter aging curve of the perovskite solar cell based on the micro parameters at time 1-T; wherein T represents the cutoff time of the open circuit voltage aging curve.
[0045] Example 3 of the present application provides a method for obtaining a microscopic parameter aging curve of a perovskite solar cell, comprising the following steps: S100: Obtaining a fill factor aging curve of a perovskite solar cell, and constructing a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and obtaining a fill factor of each perovskite solar cell model.
[0046] S200: using the microscopic parameters of each perovskite solar cell model as input and the fill factor as output, training is performed to obtain a fill factor prediction model for the perovskite solar cell, and initializing t=1; S300: generating microscopic parameters at time t using a random algorithm, inputting the microscopic parameters at time t into the perovskite solar cell fill factor prediction model, and outputting the predicted fill factor at time t; S400: Calculate a first error between the predicted fill factor at time t and the fill factor at time t in the fill factor aging curve. If the first error is greater than or equal to a first preset error threshold, return to step S300 and execute until the first error is less than the first preset error threshold, thereby obtaining the microscopic parameter at time t. S500: Update t=t+1, and return to step S300 until t=T, and obtain the micro parameter aging curve of the perovskite solar cell based on the micro parameters at time 1-T; wherein T represents the cutoff time of the open circuit voltage aging curve.
[0047] Example 4 of the present application provides a method for obtaining a microscopic parameter aging curve of a perovskite solar cell, comprising the following steps: S100: Obtaining an aging curve of the photoelectric conversion efficiency of the perovskite solar cell, and constructing a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and obtaining the photoelectric conversion efficiency of each perovskite solar cell model.
[0048] S200: using the microscopic parameters of each perovskite solar cell model as input and the photoelectric conversion efficiency as output, training is performed to obtain a prediction model for the photoelectric conversion efficiency of the perovskite solar cell, and initializing t=1; S300: generating microscopic parameters at time t using a random algorithm, inputting the microscopic parameters at time t into the perovskite solar cell photoelectric conversion efficiency prediction model, and outputting the predicted photoelectric conversion efficiency at time t; S400: Calculating a first error between the predicted photoelectric conversion efficiency at time t and the fill factor at time t in the photoelectric conversion efficiency aging curve. If the first error is greater than or equal to a first preset error threshold, returning to step S300 and executing until the first error is less than the first preset error threshold, thereby obtaining a microscopic parameter at time t. S500: Update t=t+1, and return to step S300 until t=T, and obtain the micro parameter aging curve of the perovskite solar cell based on the micro parameters at time 1-T; wherein T represents the cutoff time of the open circuit voltage aging curve.
[0049] like Figure 2 The figure shows a comparison diagram of different predicted performance parameter aging curves obtained by reverse deduction based on the microscopic parameter aging curves obtained in Examples 1 to 4 and the actual aging curves; wherein, Figure 2 (a) is a schematic diagram comparing the open circuit voltage predicted aging curve and the open circuit voltage actual aging curve. Figure 2 (b) is a schematic diagram comparing the short-circuit current density predicted aging curve and the short-circuit current density actual aging curve. Figure 2 (c) is a comparison diagram of the fill factor predicted aging curve and the fill factor actual aging curve. Figure 2 (d) is a schematic diagram comparing the predicted aging curve of photoelectric conversion efficiency and the actual aging curve of photoelectric conversion efficiency.
[0050] pass Figure 2 It can be seen that the predicted aging curves of different performance parameters are basically consistent with the actual aging curves, and the errors of the four curves are basically controlled within 5%, indicating that the micro-parameter aging curves obtained based on the method provided in this application are highly accurate.
[0051] like Figure 3 The figure shows a curve of the change in body defect density of a perovskite solar cell during the aging process obtained using the method provided in the present application. It can be seen from the figure that as the perovskite solar cell ages, the rate of increase of the body defect density also accelerates. This is because as the use time of the perovskite solar cell increases, its internal temperature increases and ion migration intensifies, so the rate of increase of its body defects also becomes faster. It can be seen that the body defect density aging curve obtained in the present application conforms to the actual physical laws.
[0052] Based on the method for obtaining the microscopic parameter aging curve of the perovskite solar cell provided in the above embodiment, the present application also provides a device for obtaining the microscopic parameter aging curve of the perovskite solar cell, such as Figure 4 As shown, the device specifically includes: The data acquisition module 10 is used to obtain the performance parameter aging curve of the perovskite solar cell, and to construct a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and to obtain the performance parameters of each perovskite solar cell model.
[0053] The model training module 20 is used to take the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, train to obtain the perovskite solar cell performance parameter prediction model, and initialize t=1.
[0054] The performance parameter prediction module 30 is used to generate microscopic parameters at time t using a random algorithm, input the microscopic parameters at time t into the perovskite solar cell performance parameter prediction model, and output the predicted performance parameters at time t.
[0055] The micro-parameter generation module 40 is used to calculate the first error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve. If the first error is greater than or equal to the first preset error threshold, the steps of the performance parameter prediction module are returned to execute until the first error is less than the first preset error threshold to obtain the micro-parameter at time t.
[0056] The micro-parameter aging curve acquisition module 50 is used to update t=t+1 and return to the steps of executing the performance parameter prediction module until t=T, and obtain the micro-parameter aging curve of the perovskite solar cell based on the micro-parameters at time 1~T; wherein T represents the cutoff time of the performance parameter aging curve.
[0057] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned method for measuring microscopic parameters of perovskite solar cells are implemented.
[0058] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0059] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0060] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0061] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.
[0062] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.
Claims
1. A method for obtaining microscopic parameter aging curves of perovskite solar cells, characterized in that: include: S10: Obtaining a performance parameter aging curve of the perovskite solar cell, and constructing a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and obtaining performance parameters of each perovskite solar cell model; S20: Using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, a performance parameter prediction model of the perovskite solar cell is trained and initialized to t=1; S30: generating microscopic parameters at time t using a random algorithm, inputting the microscopic parameters at time t into the perovskite solar cell performance parameter prediction model, and outputting the predicted performance parameters at time t; S40: Calculate a first error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve. If the first error is greater than or equal to a first preset error threshold, return to step S30 and execute until the first error is less than the first preset error threshold, thereby obtaining the microscopic parameter at time t. S50: Update t=t+1, and return to step S30 until t=T, and obtain the micro parameter aging curve of the perovskite solar cell based on the micro parameters at time 1~T; wherein T represents the end time of the performance parameter aging curve.
2. The method for obtaining microscopic parameter aging curve of perovskite solar cell according to claim 1, characterized in that: After obtaining the microscopic parameter aging curve of the perovskite solar cell, the following is also included: S60: inputting the microscopic parameters at time 1 to T in the microscopic parameter aging curve of the perovskite solar cell into the perovskite solar cell performance parameter prediction model, outputting the predicted performance parameters at time 1 to T, thereby obtaining the predicted performance parameter aging curve of the perovskite solar cell; S70: Calculate a second error between the predicted performance parameter aging curve of the perovskite solar cell and the performance parameter aging curve. If the second error is greater than or equal to a second preset error threshold, return to step S30 and execute until the second error is less than the second preset error threshold to obtain the target microscopic parameter aging curve of the perovskite solar cell.
3. The method for obtaining microscopic parameter aging curve of perovskite solar cells according to claim 1, characterized in that: Taking the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, the perovskite solar cell performance parameter prediction model obtained by training includes: Using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, multiple machine learning models are trained to obtain multiple trained machine learning models; The performance parameter prediction error of each machine learning model is calculated respectively, and the machine learning model with the smallest performance parameter prediction error is used as the performance parameter prediction model of the perovskite solar cell.
4. The method for obtaining microscopic parameter aging curve of perovskite solar cell according to claim 1, characterized in that: The first error is the root mean square error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve.
5. The method for obtaining microscopic parameter aging curve of perovskite solar cell according to claim 1, characterized in that: The first preset error threshold is less than or equal to 10%.
6. The method for obtaining microscopic parameter aging curve of perovskite solar cells according to claim 2, characterized in that: The second error between the predicted performance parameter aging curve and the performance parameter aging curve of the perovskite solar cell is calculated as follows: Calculate the mean square error, mean absolute error, or root mean square error between the predicted performance parameter and the performance parameter at the same time in the predicted performance parameter aging curve and the performance parameter aging curve; Based on the sum of the mean square error, mean absolute error or root mean square error between the predicted performance parameters and the performance parameters at all moments, a second error between the predicted performance parameter aging curve and the performance parameter aging curve is obtained.
7. The method for obtaining microscopic parameter aging curve of perovskite solar cells according to claim 2, characterized in that: The second preset error threshold is less than or equal to 10%.
8. The method for obtaining microscopic parameter aging curve of perovskite solar cells according to claim 1, characterized in that: The microscopic parameters are body defects, interface defects, electron lifetime, hole lifetime, electron mobility, hole mobility, and doping concentration; The performance parameters are open circuit voltage, short circuit current density, fill factor, and photoelectric conversion efficiency.
9. A device for obtaining microscopic parameter aging curves of perovskite solar cells, characterized in that: include: A data acquisition module is used to obtain the performance parameter aging curve of the perovskite solar cell, and to construct a perovskite solar cell model corresponding to different microscopic parameters based on the structure and process of the perovskite solar cell, and to obtain the performance parameters of each perovskite solar cell model; The model training module is used to train the perovskite solar cell performance parameter prediction model using the microscopic parameters of each perovskite solar cell model as input and the performance parameters as output, and initialize t=1; A performance parameter prediction module is used to generate microscopic parameters at time t using a random algorithm, input the microscopic parameters at time t into the perovskite solar cell performance parameter prediction model, and output the predicted performance parameters at time t; a micro-parameter generation module, configured to calculate a first error between the predicted performance parameter at time t and the performance parameter at time t in the performance parameter aging curve; if the first error is greater than or equal to a first preset error threshold, returning to the steps of the performance parameter prediction module until the first error is less than the first preset error threshold, thereby obtaining the micro-parameter at time t; The micro-parameter aging curve acquisition module is used to update t=t+1 and return to the steps of executing the performance parameter prediction module until t=T, and obtain the micro-parameter aging curve of the perovskite solar cell based on the micro-parameters at time 1~T; wherein T represents the cutoff time of the performance parameter aging curve.
10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the method for obtaining the microscopic parameter aging curve of the perovskite solar cell according to any one of claims 1 to 8 are implemented.