Methods for Optimizing the Interface Layer and Enhancing the Stability of Perovskite Solar Cells

Abstract

This invention relates to the field of perovskite photovoltaic technology and discloses a method for optimizing and enhancing the stability of perovskite solar cell interface layers. The method first acquires real-time operating data of the cell and filters a set of interface layers to be optimized. Based on the interface material properties and the target stability threshold, an interface performance weighted network is established with the overall cell structure as the central node, the interface layers to be optimized as the starting nodes, and the target stability threshold as the ending node. The edge weights are derived from the degree of matching between material properties and the threshold. In the network, for each path from the starting node to the central node and then to the ending node, an adaptive optimization strategy is applied to search for the optimal solution, identifying multiple optimization trajectories. An interface layer optimization scheme is generated based on the optimal solution. After receiving a selection instruction, the detailed material parameter record of the interface layer corresponding to the selected scheme is presented. This method achieves interface layer optimization and stability enhancement through network construction and optimization trajectory analysis.

Description

Technical Field

[0001] This invention relates to the field of perovskite photovoltaic technology, specifically to a method for optimizing the interface layer and enhancing the stability of perovskite solar cells. Background Technology

[0002] Perovskite solar cells have attracted widespread attention in the new energy field due to their excellent photoelectric conversion characteristics. In the structural design of these cells, the interface layer, as the connecting part between different functional layers, directly affects key processes such as charge transport and interfacial recombination. Currently, the selection and matching of interface layer materials still faces many challenges. Different materials are easily affected by environmental factors such as humidity and temperature changes during long-term operation, leading to degradation of interface properties and thus affecting the overall operating status of the cell.

[0003] In existing technologies, interface layer optimization often relies on empirical material selection or single performance index control, lacking a systematic consideration of the relationship between the overall battery structure and each interface layer. In actual operation, battery performance degradation is the result of the combined effects of multiple interface layers, and local optimization of a single interface often fails to effectively improve overall stability. Furthermore, traditional optimization methods fail to fully utilize real-time operational data, making it difficult to dynamically respond to the performance degradation process. This leads to deviations between the optimization scheme and actual needs, and an inability to adapt to changes in the battery's state at different operating stages.

[0004] Furthermore, the matching between interface material properties and stability requirements lacks quantitative analysis methods, with most solutions remaining at the qualitative description level, making it difficult to accurately identify key interface factors affecting stability. In this situation, the interface layer optimization process is often arbitrary, increasing R&D costs and limiting the continued application of perovskite solar cells in practical scenarios. Summary of the Invention

[0005] The purpose of this invention is to provide a method for optimizing the interface layer and enhancing the stability of perovskite solar cells, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a method for optimizing the interface layer and enhancing the stability of perovskite solar cells, the method comprising:

[0007] The real-time operating data of the perovskite solar cell is obtained, which includes performance degradation indicators and target stability thresholds. The set of interface layers to be optimized is selected from multiple candidate interface layers of the perovskite solar cell.

[0008] Based on the interface material properties in the real-time running data and the target stability threshold, an interface performance weighted network is established by taking the overall structure of the perovskite solar cell as the central node, each interface layer in the set of interface layers to be optimized as the starting node, and the target stability threshold as the ending node. The weight values ​​of the connecting edges in the interface performance weighted network are derived by the degree of matching between the interface material properties and the target stability threshold.

[0009] In the interface performance weighted network, for each set of paths from the starting node to the ending node via the central node, an adaptive optimization strategy is applied to search for the optimal solution of the path weights, thereby identifying multiple optimization trajectories of the set of interface layers to be optimized that are associated with the performance degradation index via the central node.

[0010] Based on the optimal solution of path weights among the multiple optimized trajectories, an interface layer optimization scheme is generated for the real-time running data. When a selection instruction for the interface layer optimization scheme is received, detailed material parameter records of the set of interface layers to be optimized corresponding to the selected interface layer optimization scheme are presented.

[0011] Preferably, the step of establishing an interface performance weighted network based on the interface material properties in the real-time operating data and the target stability threshold, taking the overall structure of the perovskite solar cell as the central node, each interface layer in the set of interface layers to be optimized as the starting node, and the target stability threshold as the ending node, includes the following sub-steps:

[0012] Extract the historical aging rate of each interface layer in each set of interface layers to be optimized, and calculate the durability factor of each interface layer based on the historical aging rate.

[0013] The influence of each interface layer on the photoelectric conversion efficiency of the perovskite solar cell is analyzed. Based on the expected efficiency level in the real-time operating data, the efficiency-sensitive connection weights from each starting node to the center node in the interface performance weighted network are derived.

[0014] By combining the thermal expansion coefficient records of each interface layer, the historical fault statistics of the set of interface layers to be optimized, and the design lifetime requirements in the real-time operating data, the thermal stability sensitive connection weight of each starting node to the central node in the interface performance weighted network is calculated.

[0015] The durability factor, the efficiency-sensitive connection weight, and the thermal stability-sensitive connection weight are integrated and multiplied by the minimum matching value of the interface material property corresponding to each interface layer and the target stability threshold to determine the connection edge weight value from the starting node to the center node in the interface performance weighted network.

[0016] Using the target stability threshold in the real-time running data, the weight values ​​of the connection edges from the termination node to the center node in the interface performance weighted network are set, and the construction process of the interface performance weighted network is completed by summarizing the weight values ​​of all connection edges from the starting node to the center node and the weight values ​​of connection edges from the termination node to the center node.

[0017] Preferably, the analysis of the effect of each interface layer on the photoelectric conversion efficiency of the perovskite solar cell, combined with the expected efficiency level in the real-time operating data, derives the efficiency-sensitive connection weights from each starting node to the central node in the interface performance weighted network, including the following operations:

[0018] The difference between the maximum and minimum efficiency gains of each interface layer in a perovskite solar cell is measured, and the efficiency gain range is defined accordingly.

[0019] Collect historical efficiency response datasets for each interface layer, and calculate the average efficiency response benchmark by combining the time decay factor of each historical efficiency response dataset.

[0020] The total efficiency response offset for each interface layer is generated by summing the absolute deviations between the historical efficiency response data and the average efficiency response benchmark for each interface layer.

[0021] Divide the total efficiency response offset of each interface layer by the efficiency improvement range to obtain the deviation ratio, and map the deviation ratio to a preset influence range to obtain the corresponding weight correction coefficient.

[0022] The deviation ratio is multiplied by the weight correction coefficient to finally generate the efficiency-sensitive connection weights from the starting node to the center node in the interface performance weighted network.

[0023] Preferably, the step of collecting historical efficiency response datasets for each interface layer and calculating the average efficiency response benchmark by combining the time decay factors of each historical efficiency response dataset includes the following steps:

[0024] Based on the preset material degradation rate and the average and dispersion of the efficiency response data of each interface layer in the most recent historical period, the adaptive change rate parameter is derived.

[0025] The time decay weight of each interface layer is calculated using the timestamp offset of the efficiency response data of each interface layer and the adaptive rate of change parameter.

[0026] Multiply the time decay weight of each interface layer by the corresponding historical efficiency response data to generate the weighted efficiency response value of each interface layer.

[0027] The weighted efficiency response values ​​of all interface layers are summed to obtain the total weighted efficiency response, and the time decay weights of all interface layers are summed to obtain the total decay weight.

[0028] The average efficiency response benchmark is derived by dividing the sum of the weighted efficiency responses by the total attenuation weight.

[0029] Preferably, the step of combining the thermal expansion coefficient records of each interface layer, the historical fault statistics of the set of interface layers to be optimized, and the design lifetime requirements in the real-time operating data to calculate the thermal stability sensitive connection weight of each starting node to the central node in the interface performance weighted network includes the following process:

[0030] Extract the failure frequency of each interface layer from the historical failure statistics, and calculate the failure probability coefficient of each interface layer based on the failure frequency.

[0031] By combining the failure probability coefficient of each interface layer with the periodic difference value between the thermal expansion coefficient record and the design life requirement, the thermal stability sensitive connection weight of each starting node to the central node in the interface performance weighted network is calculated through a linear combination method.

[0032] Preferably, in the interface performance weighted network, the process of applying an adaptive optimization strategy to search for the optimal solution of path weights for each set of paths from the starting node to the ending node via the central node, and identifying multiple optimization trajectories of the set of interface layers to be optimized via the central node associated with the performance degradation index, includes the following steps:

[0033] An optimization solution space is created based on the path combinations from each starting node to the ending node in the interface performance weighted network.

[0034] Heuristic search algorithms or stochastic optimization methods are used to evaluate the gradient changes or fitness scores of all connected edges in the optimization solution space.

[0035] For the edge weights in each path, the optimal weights of a single path that independently satisfies the target stability threshold, or the optimal weights of a combination of paths that collaboratively satisfy the target stability threshold, are calculated by minimizing the path cost function.

[0036] Based on the optimal weight configuration of each single path and the optimal weight configuration of the combined path, the gradient change or fitness score in the optimization solution space is updated. When the path cost function cannot be further reduced, multiple single paths and combined paths are output.

[0037] By integrating the multiple single paths and combined paths, multiple optimization trajectories are determined for the set of interface layers to be optimized, which are associated with the performance degradation indicators via the central node.

[0038] Preferably, the step of acquiring real-time operating data of the perovskite solar cell, wherein the real-time operating data includes performance degradation indicators and target stability thresholds, and selecting a set of interface layers to be optimized from multiple candidate interface layers of the perovskite solar cell, includes the following operations:

[0039] Obtain the real-time operating data, and query the interface layer records associated with the performance degradation index in the perovskite solar cell database according to the performance degradation index;

[0040] The historical optimization frequency and number of optimizations for each interface layer are recorded in the perovskite solar cell database.

[0041] Based on the historical optimization frequency and number of optimizations, the set of interface layers to be optimized is dynamically selected from the multiple candidate interface layers.

[0042] Preferably, the method further includes: after generating the interface layer optimization scheme, feeding back the optimal solution of the path weight in the interface layer optimization scheme to the perovskite solar cell control system to adjust the material deposition parameters of the interface layer.

[0043] Preferably, adjusting the material deposition parameters of the interface layer includes modifying the thickness distribution and composition ratio of the interface layer according to the optimal solution of the path weight.

[0044] Preferably, the method further includes a verification mechanism:

[0045] After generating the interface layer optimization scheme, the simulated environment accelerated aging test process is executed to collect the performance degradation trajectory of the interface layer during the test.

[0046] The performance degradation trajectory is compared in real time with the performance improvement parameters corresponding to the interface layer optimization scheme.

[0047] When the actual performance degradation rate is detected to exceed the predicted value of the scheme and reach the preset deviation threshold, the operation of re-filtering the set of interface layers to be optimized is automatically triggered.

[0048] Compared with the prior art, the beneficial effects of the present invention are:

[0049] Starting from real-time operational data of perovskite solar cells, a set of interface layers to be optimized is selected, making the determination of optimization targets more closely aligned with the actual operating state of the cells and avoiding ineffective analysis of irrelevant interface layers. By establishing an interface performance weighted network, the overall cell structure, the interface layers to be optimized, and the target stability threshold are incorporated into a unified analysis framework, revealing the correlation between each interface layer and the overall structure and stability requirements. This breaks through the limitations of traditional local optimization, allowing the role of each interface layer to be clearly presented in the overall system.

[0050] In interface performance weighted networks, the optimal solution for path weights is searched using an adaptive optimization strategy. This approach can identify optimization trajectories related to performance degradation metrics from multiple possible optimization paths. This method fully considers the mutual influence between interface layers, making the identification of optimization trajectories more systematic and comprehensive, rather than being limited to the independent adjustment of a single interface.

[0051] The interface layer optimization scheme, based on the optimized trajectory generation, is closely linked to real-time operational data and can be dynamically adjusted according to the actual operating status of the battery, making the scheme more aligned with actual needs. When a selection command is received, the corresponding detailed material parameter records are presented, providing clear reference information for specific implementation and facilitating the accurate application of the optimization scheme in actual operation, thus making the interface layer optimization process more operable.

[0052] Overall, this method integrates real-time data, constructs network models, and analyzes and optimizes trajectories to form a coherent process for interface layer optimization and stability enhancement. It organically combines each step to form a complete closed loop from data acquisition to solution implementation, thus meeting the optimization needs of complex interface systems in perovskite solar cells. Attached Figure Description

[0053] Figure 1 This is a schematic diagram illustrating the working principle of the perovskite solar cell interface layer optimization and stability enhancement method described in this invention.

[0054] Figure 2 A flowchart for establishing an interface performance weighted network;

[0055] Figure 3 A flowchart for deriving efficiency-sensitive connection weights;

[0056] Figure 4 A flowchart for identifying optimized trajectories;

[0057] Figure 5 The flowchart is for the verification mechanism. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Please see Figure 1 This invention provides a method for optimizing the interface layer and enhancing the stability of perovskite solar cells, the method comprising:

[0060] Step 1: Acquire real-time operational data of the perovskite solar cell. This data includes performance degradation indicators and target stability thresholds. Select a set of interface layers to be optimized from multiple candidate interface layers in the perovskite solar cell. Real-time operational data can be collected in real time through a cell operation monitoring system. Performance degradation indicators include, but are not limited to, the cell output voltage decay rate and the short-circuit current decrease. The target stability threshold is pre-set according to cell design standards and actual application requirements. Candidate interface layers cover various interface layers in the cell structure, such as electron transport layers, hole transport layers, and buffer layers. Through analysis of the real-time operational data, interface layers with performance close to or below the target stability threshold are selected to form the set of interface layers to be optimized.

[0061] Step 2: Based on the interface material properties and target stability threshold from real-time operational data, an interface performance weighted network is established, using the overall structure of the perovskite solar cell as the central node, each interface layer in the set of interface layers to be optimized as the starting node, and the target stability threshold as the ending node. The weights of the connecting edges in the interface performance weighted network are derived from the degree of matching between the interface material properties and the target stability threshold. Interface material properties include the chemical composition, electrical properties, and thermal properties of the material. By analyzing the degree of fit between these properties and the target stability threshold, the weights of the connecting edges between nodes in the network are determined. Higher weight values ​​indicate a greater potential for the corresponding interface layer to improve cell stability.

[0062] Step 3: In the interface performance weighted network, for each set of paths from the starting node to the ending node via the central node, an adaptive optimization strategy is applied to search for the optimal solution of path weights. This process identifies multiple optimization trajectories for the set of interface layers to be optimized, which are associated with performance degradation indicators via the central node. The adaptive optimization strategy can dynamically adjust the search direction and step size based on feedback information during the path search process. Through continuous iterative calculation, it finds the path with the optimal weights from many possible paths. These paths correspond to the optimized trajectories that can effectively improve the battery performance degradation problem.

[0063] Step 4: Based on the optimal solution of path weights among multiple optimized trajectories, generate an interface layer optimization scheme for real-time running data. When a selection instruction for an interface layer optimization scheme is received, present detailed material parameter records of the set of interface layers to be optimized corresponding to the selected scheme. These detailed material parameter records include material purity, thickness, and preparation process parameters, providing specific parameter guidance for actual interface layer optimization operations.

[0064] Example 1: See Figure 2 The specific sub-steps for establishing the interface performance weighted network in step 2 are as follows: Extract the historical aging rate of each interface layer in the set of interface layers to be optimized. The historical aging rate can be obtained by analyzing the performance monitoring data of the battery during different operating cycles. This data includes information such as changes in the physical structure and chemical composition of the interface layer during continuous operation. Calculate the durability factor for each interface layer based on the historical aging rate. The value of the durability factor is related to the aging rate of the interface layer; the slower the aging rate, the higher the durability factor, and vice versa.

[0065] This study analyzes the impact of each interface layer on the photoelectric conversion efficiency of perovskite solar cells. Combining this with the expected efficiency level from real-time operational data, the efficiency-sensitive connection weights from each starting node to the central node in the interface performance weighted network are derived. Different types of interface layers play different roles during cell operation. For example, the hole transport layer is responsible for the effective extraction and transport of holes; its carrier mobility directly affects the hole transport efficiency, thus altering the cell's photoelectric conversion efficiency. The conductivity of the electron transport layer relates to the rapid transfer of electrons from the light absorption layer, reducing charge recombination. By quantifying these effects and combining them with the expected efficiency level set in the real-time operational data, the specific values ​​of the efficiency-sensitive connection weights are determined.

[0066] By combining the thermal expansion coefficient records of each interface layer, historical fault statistics of the set of interface layers to be optimized, and design life requirements from real-time operating data, the thermal stability sensitive connection weights from each starting node to the central node in the interface performance weighted network are calculated. Specifically, the fault occurrence frequency of each interface layer is extracted from historical fault statistics. The fault occurrence frequency is calculated based on the ratio of the number of functional failures of that interface layer to the total operating time in the battery's past operating records. A failure probability coefficient for each interface layer is calculated based on the fault occurrence frequency; the magnitude of the failure probability coefficient increases with the increase of the fault occurrence frequency. The failure probability coefficient of each interface layer is combined with the periodic difference values ​​of the thermal expansion coefficient records and design life requirements, and the thermal stability sensitive connection weights are calculated through a linear combination method. The coefficients used in the linear combination process are determined based on the actual situation, such as the position of the interface layer in the overall battery structure and the intensity of its interaction with other material layers, to reflect the proportion of different factors in the thermal stability assessment.

[0067] The durability factor, efficiency-sensitive connection weights, and thermal stability-sensitive connection weights are integrated and multiplied by the minimum matching value of the interface material properties and the target stability threshold for each interface layer. This determines the connection edge weight value from each starting node to the center node in the interface performance weighted network. The minimum matching value is obtained by comparing each parameter of the interface material properties (such as electrical performance parameters, thermal performance parameters, chemical stability parameters, etc.) with the corresponding parameter of the target stability threshold one by one, and finding the value corresponding to the parameter with the lowest matching degree. This value is the minimum matching value; the higher the matching degree, the larger the minimum matching value.

[0068] Using the target stability threshold from real-time operational data, the weight values ​​of the connection edges from the terminating node to the central node in the interface performance weighted network are set. The setting of these weight values ​​must comprehensively consider the priority of the target stability threshold in the battery performance evaluation system and its impact on the long-term reliable operation of the battery. The interface performance weighted network is constructed by aggregating the weight values ​​of all connection edges from the starting node to the central node and from the terminating node to the central node. During the aggregation process, it is necessary to ensure that the calculation units for all weight values ​​are consistent and the data format is uniform to guarantee that the constructed interface performance weighted network accurately reflects the correlation strength and performance impact relationships between each node.

[0069] In constructing the interface performance weighted network, rigorous verification of all data is necessary to ensure the accuracy and completeness of extracted historical aging rates, thermal expansion coefficient records, and historical fault statistics. For missing or abnormal data, data processing methods such as interpolation and mean substitution should be used to supplement and correct them, avoiding deviations in network construction due to data issues. Simultaneously, when calculating the weight values, the physical meaning and calculation logic of each parameter must be clearly defined to ensure that the calculation process at each step is traceable and repeatable, thereby improving the reliability and practicality of the interface performance weighted network.

[0070] Example 2: See Figure 3 This study analyzes the impact of each interface layer on the photoelectric conversion efficiency of perovskite solar cells. Based on the expected efficiency levels in real-time operational data, the efficiency-sensitive connection weights from each starting node to the central node in the interface performance weighted network are derived. The specific steps are as follows:

[0071] The difference between the maximum and minimum efficiency gains of each interface layer in a perovskite solar cell is measured to define the efficiency gain range. The measurement process must be conducted in a standardized experimental environment, maintaining consistent external conditions such as light intensity, ambient temperature, and humidity. Only the parameters of the target interface layer, such as material purity and thickness, are varied. Through repeated tests, the photoelectric conversion efficiency of the cell under different parameters for each interface layer is obtained. The maximum and minimum efficiency gains are then selected, and the difference between these two values ​​represents the efficiency gain range.

[0072] Historical efficiency response datasets for each interface layer are collected, and an average efficiency response benchmark is calculated by combining the time decay factors of each historical efficiency response dataset. The specific steps are as follows: Based on a preset material degradation rate and the average and dispersion of the efficiency response data for each interface layer within the most recent historical period, an adaptive rate of change parameter is derived. The material degradation rate is determined based on the inherent properties of the interface layer material and past experimental observations. The average value is calculated by arithmetically averaging the efficiency response data within the most recent historical period. The dispersion is measured using the standard deviation. The adaptive rate of change parameter comprehensively reflects the material degradation trend and the fluctuations in recent data. Using the timestamp offset of the efficiency response data for each interface layer and the adaptive rate of change parameter, the time decay weight for each interface layer is calculated. The timestamp offset is the difference between the data recording time and the current time; a larger difference indicates older data, and a smaller time decay weight is assigned to reflect the higher reference value of newer data in the calculation. The time decay weight of each interface layer is multiplied by the corresponding historical efficiency response data to generate a weighted efficiency response value for each interface layer. This value retains information from historical data while highlighting the influence of recent data through weight adjustments. The weighted efficiency response values ​​of all interface layers are summed to obtain the total weighted efficiency response. At the same time, the time decay weights of all interface layers are summed to obtain the total decay weight. By dividing the total weighted efficiency response by the total decay weight, the average efficiency response benchmark is derived. This benchmark reflects the average efficiency response level of all interface layers after comprehensively considering the time factor.

[0073] For each interface layer, the absolute deviation of its historical efficiency response data from the average efficiency response benchmark is summarized to generate the total efficiency response offset for each interface layer. For each interface layer, the difference between each historical efficiency response data point and the average efficiency response benchmark is calculated, and the absolute value of the difference is taken as the absolute deviation. All absolute deviations are added together to obtain the total efficiency response offset for that interface layer. This total offset reflects the degree of deviation of the efficiency response of that interface layer from the overall average level.

[0074] The deviation ratio is obtained by dividing the total efficiency response offset of each interface layer by the efficiency improvement range. The deviation ratio reflects the proportion of the total offset within the efficiency improvement range. The deviation ratio is then mapped to a preset influence range to obtain the corresponding weight correction coefficient. The influence range is divided according to different levels of efficiency impact in actual application scenarios. Different deviation ratios will fall into different ranges, and each range corresponds to a weight correction coefficient. The correction coefficient is used to adjust the degree of influence of the deviation ratio on the final weight.

[0075] The deviation ratio is multiplied by the weight correction coefficient to generate the efficiency-sensitive connection weights from each starting node to the central node in the interface performance weighted network. These weights comprehensively consider the deviation degree of the interface layer's efficiency response, its efficiency improvement potential, and its impact, quantifying the sensitivity of each interface layer to the battery's photoelectric conversion efficiency and providing key parameters for the subsequent construction of the interface performance weighted network.

[0076] Throughout the process, it is essential to ensure the completeness and accuracy of the collected data, identify and process any abnormal data to prevent it from interfering with the calculation results. Simultaneously, the settings and calculation methods for all parameters should remain consistent to guarantee the comparability of efficiency-sensitive connection weights across different interface layers, thereby providing a reliable basis for interface layer optimization.

[0077] Example 3: See Figure 4 In the UI performance weighted network, for each set of paths from the starting node to the ending node via the central node, an adaptive optimization strategy is applied to search for the optimal solution of the path weights. This process identifies multiple optimization trajectories of the set of UI layers to be optimized via the central node and associated performance degradation indicators. The specific steps are as follows:

[0078] An optimization solution space is created based on the path combinations from each starting node to the ending node in the interface performance weighted network. The path combinations include all possible paths starting from the starting node (the interface layer to be optimized), passing through the central node (the overall battery structure), and finally reaching the ending node (the target stability threshold). Each path consists of a series of connecting edges, and different paths represent different optimization directions and combinations for the interface layer. The optimization solution space covers all possible weight configurations for these paths, providing a range for subsequent optimization searches.

[0079] Heuristic search algorithms or stochastic optimization methods are employed to evaluate the gradient changes or fitness scores of all connected edges in the optimization solution space. Heuristic search algorithms efficiently explore better solutions in the solution space by simulating natural evolution or physical processes, such as through selection, crossover, and mutation operations to gradually approach the optimal solution. Stochastic optimization methods, on the other hand, randomly generate solutions and evaluate their quality, continuously adjusting the search direction. Gradient changes describe the impact of small changes in edge weights on the total path weight; positive values ​​indicate that increasing the weight will increase the total weight, while negative values ​​have the opposite effect. Fitness scores are used to score the path according to preset evaluation criteria; higher scores reflect the quality of the corresponding optimization scheme.

[0080] For the edge weights in each path, the optimal weights for a single path that independently satisfies the target stability threshold for a single interface layer, or the optimal weight configuration for a combined path that collaboratively satisfies the target stability threshold for multiple interface layers, are calculated by minimizing the path cost function. The expression for the path cost function is:

[0081]

[0082] in, This represents the path cost function value. and These are preset coefficients. Indicates the first in the path The weight of each connecting edge, This represents the total number of connecting edges in the path. This represents the standard deviation of the weights of all connecting edges in the path.

[0083] When calculating the optimal weight of a single path, only the path corresponding to a single interface layer is considered. By adjusting the weights of each connecting edge in the path, the cost function value is minimized. When calculating the optimal weight configuration of a combined path, the paths corresponding to multiple interface layers must be considered simultaneously. The weight relationships of the connecting edges in each path must be coordinated to ensure that the overall cost function value is minimized, so as to achieve the synergistic optimization effect of multiple interface layers.

[0084] Based on the optimal weights for each individual path and the optimal weights for combined paths, the gradient changes or fitness scores in the optimization solution space are updated. During the update process, the newly obtained weight configurations are substituted into the original evaluation model, and the gradient changes or fitness scores are recalculated to match the evaluation results with the current optimal weight configurations. When the path cost function no longer decreases after several consecutive iterations, i.e., a stable state is reached, multiple individual paths and combined paths are output. These paths represent the lowest-cost paths in the current optimization solution space.

[0085] By integrating multiple single and combined paths, several optimization trajectories are determined for the set of interface layers to be optimized, linked to performance degradation indicators via a central node. During the integration process, the commonalities and differences among the paths are analyzed, duplicate or conflicting paths are eliminated, and paths with practical optimization significance are retained. The resulting optimization trajectories clearly demonstrate the specific path from the current interface layer state, through adjusting relevant parameters, to gradually reach the target stability threshold, providing a direct basis for generating subsequent interface layer optimization schemes.

[0086] To acquire real-time operational data of perovskite solar cells, including performance degradation indicators and target stability thresholds, a set of interface layers to be optimized is selected from multiple candidate interface layers of the perovskite solar cells. The specific steps are as follows:

[0087] Real-time operational data is acquired through sensors installed on the cells, including parameters such as voltage, current, and temperature. Data processing extracts performance degradation indicators, such as output power attenuation rate and fill factor decrease. Target stability thresholds are also defined, such as ensuring power attenuation does not exceed a certain percentage within a given operating cycle. Based on these performance degradation indicators, the perovskite solar cell database is queried for associated interface layer records. This database stores historical data for various interface layers under different performance degradation conditions; by performing correlation queries, potentially problematic interface layers can be preliminarily located.

[0088] The historical optimization frequency and number of optimizations for each interface layer are recorded in the perovskite solar cell database. The historical optimization frequency refers to the ratio of the number of times the interface layer was included in the optimization scope within a certain period to the total monitoring time. The number of optimizations refers to the number of times optimization operations were actually performed on the interface layer. These data can be directly obtained from the operation logs in the database.

[0089] Based on historical optimization frequency and number of optimizations, a set of interface layers to be optimized is dynamically selected from multiple candidate interface layers. During the selection process, both the historical optimization frequency and the number of optimizations are considered. Generally, interface layers with high historical optimization frequency and a large number of optimizations are more prone to performance fluctuations or degradation, and these interface layers are given priority in the set of interfaces to be optimized. For interface layers that newly show signs of performance degradation, even if their historical optimization frequency and number of optimizations are low, they will be included based on the actual situation to ensure that the set of interfaces to be optimized can comprehensively cover the interface layers that need optimization.

[0090] Throughout the process, it is essential to ensure the real-time nature and accuracy of data acquisition. Records in the database should be updated promptly to avoid deviations in the screening results due to data lag or errors. Simultaneously, the identification of optimized trajectories and the screening of interface layer sets should be dynamically adjusted based on the actual operating status of the battery to adapt to optimization needs under different operating conditions.

[0091] Example 4: The method further includes: after generating the interface layer optimization scheme, feeding back the optimal solution of the path weight in the interface layer optimization scheme to the perovskite solar cell control system to adjust the material deposition parameters of the interface layer. The perovskite solar cell control system consists of hardware devices and software programs. The hardware devices include the control module of the deposition equipment, the sensor group, and the actuator. The software program is responsible for receiving the optimal solution of the path weight and converting it into specific control commands. The optimal solution of the path weight contains the performance parameters that each interface layer should achieve after optimization, such as the conductivity of the material and the interface contact resistance. These parameters are imported into the control program through the interface of the control system. The control program analyzes these parameters to determine the type and direction of the material deposition parameters that need to be adjusted. For example, when the optimal solution of the path weight shows that the conductivity of a certain electron transport layer needs to be improved, the control program will analyze the deposition conditions affecting the parameter and generate corresponding adjustment commands. The sensor group monitors various indicators in real time during the deposition process, such as the deposition rate and film thickness, and feeds back the monitoring data to the control program. The control program continuously fine-tunes the control commands based on the deviation between the feedback data and the optimal solution of the path weight to ensure that the material deposition process proceeds according to the optimization scheme.

[0092] Adjusting the material deposition parameters of the interface layer includes modifying the thickness distribution and composition ratio of the interface layer based on the optimal solution of the path weight. Taking the hole transport layer of a perovskite solar cell as an example, if the optimal solution of the path weight indicates that the hole mobility of this layer needs to be increased, and the hole mobility is related to the film thickness and the doping ratio in the composition, then the thickness distribution and composition ratio need to be adjusted. The modification of the thickness distribution needs to be combined with the structural design of the cell. The thickness of the hole transport layer should be appropriately increased in areas with strong light absorption to ensure that holes can be fully extracted, while it can be appropriately thinned in edge areas to reduce material waste. During the adjustment process, the nozzle movement speed and spray volume of the deposition equipment are precisely controlled. The faster the nozzle movement speed, the less material is deposited per unit area and the thinner the film layer, and vice versa. Through preset movement path and speed parameters, a gradient distribution of thickness is achieved. Adjusting the composition ratio involves the mixing ratio of different raw materials. For example, in the hole transport layer, the ratio of the host material to the dopant affects the hole mobility. Based on the optimal solution of the path weight, the control system will adjust the flow rate of the feed pump to change the mixing ratio of the two materials. If the doping ratio needs to be increased, the flow rate of the dopant feed pump will be increased, and the flow rate of the host material will be reduced accordingly, and vice versa.

[0093] For the electron transport layer, if the optimal solution for path weights requires enhanced electron conductivity, this can be achieved by adjusting the thickness distribution and the proportion of metal oxides in the composition. At the interface between the electron transport layer and the light absorption layer, the thickness can be appropriately increased to reduce interface resistance, while it can be thinned in areas far from the interface to reduce series resistance. Regarding composition, if a mixture of titanium dioxide and zinc oxide is used, the proportion of zinc oxide can be increased because zinc oxide has a higher electron mobility. Adjustments can be made by controlling the evaporation rates of the two materials to change the compositional proportions in the deposited film.

[0094] The adjustment of the buffer layer also follows the above logic. If the optimal solution of the path weight shows that its ability to block ion migration is insufficient, the thickness distribution needs to be adjusted to enhance the blocking effect. The thickness is increased in areas where ion aggregation is likely to occur, and the proportion of insulating material in the composition is adjusted. By increasing the proportion of insulating material, the diffusion of ions is suppressed.

[0095] Throughout the adjustment process, the control system collects film thickness and composition analysis data in real time. By comparing these data with the target values ​​in the optimal solution of the path weights, the deposition parameters are continuously corrected. For example, an ellipsomerometer is used to monitor film thickness. If the deviation between the actual thickness and the target thickness distribution exceeds the allowable range, the control system will immediately adjust the nozzle's moving speed or feed rate. X-ray photoelectron spectroscopy is used to analyze the composition ratio. If the actual dopant ratio is lower than the target value, the flow rate parameters of the feed pump will be reset to ensure that the composition ratio meets the optimization requirements.

[0096] Adjustments to different interface layers must be coordinated to avoid mutual interference. For example, when simultaneously adjusting the thickness of the electron transport layer and the perovskite absorber layer, it is necessary to ensure that the contact interface between the two layers is flat, and that changes in the thickness of the electron transport layer do not lead to a decrease in the film quality of the perovskite absorber layer. The control system achieves matching between the interface layers by synchronously controlling the parameters of multiple deposition modules, ensuring that the overall structure meets the requirements of the optimal path weight solution.

[0097] Example 5: See Figure 5After generating the interface layer optimization scheme, a simulated environment accelerated aging test process is executed. This simulated environment accelerated aging test is achieved by constructing a closed test chamber, within which parameters such as temperature, humidity, light intensity, and gas atmosphere can be precisely controlled. Based on the actual application scenarios of perovskite solar cells, accelerated aging conditions are set within the test chamber. For example, in high-temperature and high-humidity environment testing, the temperature is set to a value higher than the actual operating environment, the humidity is set to a higher relative humidity, and a constant light intensity is maintained to accelerate the aging process of the interface layer. During the test, high-precision monitoring equipment is used to collect the performance parameters of the interface layer in real time, such as conductivity, carrier mobility, and interface contact resistance. The changes of these parameters over time form a performance degradation trajectory, and the trajectory data is continuously recorded and stored in the test system's database.

[0098] The performance degradation trajectory is compared in real time with the performance improvement parameters corresponding to the interface layer optimization scheme. The performance improvement parameters are the performance change curves of the interface layer during the aging process predicted by the optimization scheme, including performance parameter thresholds corresponding to different time points. The comparison process is completed by dedicated data processing software. The software matches the real-time collected performance degradation trajectory data with the predicted data of the performance improvement parameters one by one, calculates the performance difference at each time point, such as the difference between the actual measured interface layer conductivity and the predicted conductivity threshold after 100 hours of testing, and generates a curve of the difference changing over time, intuitively showing the deviation between the actual performance and the predicted performance.

[0099] When the actual performance degradation rate exceeds the predicted value and reaches a preset deviation threshold, the system automatically triggers a re-selection of the set of interface layers to be optimized. The preset deviation threshold is set based on the importance of the interface layer and the application requirements of the battery. For example, for the electron transport layer, which directly affects photoelectric conversion efficiency, the deviation threshold can be set to a smaller value, while for the buffer layer, which provides auxiliary functions, the deviation threshold can be appropriately relaxed. The detection process is automatically executed by the software system. The system continuously monitors the deviation between the performance degradation rate and the predicted value. When the deviation exceeds the preset threshold for multiple consecutive time points, or when a single deviation reaches a certain multiple of the threshold, the system determines that the current optimization scheme has failed to achieve the expected effect and immediately issues a trigger signal. During the re-selection of the set of interface layers to be optimized, the system re-accesses historical data from the perovskite solar cell database, combines it with the latest performance degradation trajectory, and expands the selection scope. This includes not only previously optimized interface layers but also other interface layers that may affect the current performance. For example, when the performance degradation of the electron transport layer exceeds expectations, it may be necessary to consider the perovskite layer and buffer layer in contact with it, re-evaluate the performance degradation indicators of these interface layers, and select the set of interface layers that need further optimization, entering a new optimization cycle.

[0100] Throughout the entire verification mechanism operation, the environmental parameters of the test chamber remain stable to avoid performance data distortion due to environmental fluctuations. The data acquisition frequency is set according to the aging rate; the acquisition frequency is increased during the rapid performance degradation phase to capture subtle performance changes, and decreased during the stable phase to reduce data redundancy. After the re-screening operation is initiated, the original interface layer optimization schemes are marked as needing improvement. The new optimization process is re-executed based on the updated set of interface layers to be optimized until the generated optimization scheme passes the verification mechanism's test, ensuring that the performance and stability of the interface layers meet the design requirements.

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0102] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing the interface layer and enhancing the stability of perovskite solar cells, characterized in that, The method includes the following steps: The real-time operating data of the perovskite solar cell is obtained, which includes performance degradation indicators and target stability thresholds. The set of interface layers to be optimized is selected from multiple candidate interface layers of the perovskite solar cell. Based on the interface material properties in the real-time running data and the target stability threshold, an interface performance weighted network is established with the overall structure of the perovskite solar cell as the central node, each interface layer in the set of interface layers to be optimized as the starting node, and the target stability threshold as the ending node. The weight values ​​of the connecting edges in the interface performance weighted network are derived from the degree of matching between the interface material properties and the target stability threshold. In the interface performance weighted network, for each set of paths from the starting node to the ending node via the central node, an adaptive optimization strategy is applied to search for the optimal solution of the path weights, thereby identifying multiple optimization trajectories of the set of interface layers to be optimized that are associated with the performance degradation index via the central node. Based on the optimal solution of path weights among the multiple optimized trajectories, an interface layer optimization scheme is generated for the real-time running data. When a selection instruction for the interface layer optimization scheme is received, detailed material parameter records of the set of interface layers to be optimized corresponding to the selected interface layer optimization scheme are presented.

2. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 1, characterized in that, Based on the interface material properties in the real-time operating data and the target stability threshold, an interface performance weighted network is established, taking the overall structure of the perovskite solar cell as the central node, each interface layer in the set of interface layers to be optimized as the starting node, and the target stability threshold as the ending node. This includes the following sub-steps: Extract the historical aging rate of each interface layer in each set of interface layers to be optimized, and calculate the durability factor of each interface layer based on the historical aging rate. The influence of each interface layer on the photoelectric conversion efficiency of the perovskite solar cell is analyzed. Based on the expected efficiency level in the real-time operating data, the efficiency-sensitive connection weights from each starting node to the center node in the interface performance weighted network are derived. By combining the thermal expansion coefficient records of each interface layer, the historical fault statistics of the set of interface layers to be optimized, and the design lifetime requirements in the real-time operating data, the thermal stability sensitive connection weight of each starting node to the central node in the interface performance weighted network is calculated. The durability factor, the efficiency-sensitive connection weight, and the thermal stability-sensitive connection weight are integrated and multiplied by the minimum matching value of the interface material property corresponding to each interface layer and the target stability threshold to determine the connection edge weight value from the starting node to the center node in the interface performance weighted network. Using the target stability threshold in the real-time running data, the weight values ​​of the connection edges from the termination node to the center node in the interface performance weighted network are set, and the construction process of the interface performance weighted network is completed by summarizing the weight values ​​of all connection edges from the starting node to the center node and the weight values ​​of connection edges from the termination node to the center node.

3. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 2, characterized in that, The analysis examines the impact of each interface layer on the photoelectric conversion efficiency of the perovskite solar cell. Based on the expected efficiency levels in the real-time operational data, the efficiency-sensitive connection weights from each starting node to the central node in the interface performance-weighted network are derived, including the following operations: The difference between the maximum and minimum efficiency gains of each interface layer in a perovskite solar cell is measured, and the efficiency gain range is defined accordingly. Collect historical efficiency response datasets for each interface layer, and calculate the average efficiency response benchmark by combining the time decay factor of each historical efficiency response dataset. The total efficiency response offset for each interface layer is generated by summing the absolute deviations between the historical efficiency response data and the average efficiency response benchmark for each interface layer. Divide the total efficiency response offset of each interface layer by the efficiency improvement range to obtain the deviation ratio, and map the deviation ratio to a preset influence range to obtain the corresponding weight correction coefficient. The deviation ratio is multiplied by the weight correction coefficient to finally generate the efficiency-sensitive connection weights from the starting node to the center node in the interface performance weighted network.

4. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 3, characterized in that, The process of collecting historical efficiency response datasets for each interface layer and calculating the average efficiency response benchmark by combining the time decay factors of each historical efficiency response dataset includes the following steps: Based on the preset material degradation rate and the average and dispersion of the efficiency response data of each interface layer in the most recent historical period, the adaptive change rate parameter is derived. The time decay weight of each interface layer is calculated using the timestamp offset of the efficiency response data of each interface layer and the adaptive rate of change parameter. Multiply the time decay weight of each interface layer by the corresponding historical efficiency response data to generate the weighted efficiency response value of each interface layer. The weighted efficiency response values ​​of all interface layers are summed to obtain the total weighted efficiency response, and the time decay weights of all interface layers are summed to obtain the total decay weight. The average efficiency response benchmark is derived by dividing the sum of the weighted efficiency responses by the total attenuation weight.

5. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 2, characterized in that, By combining the thermal expansion coefficient records of each interface layer, the historical fault statistics of the set of interface layers to be optimized, and the design lifetime requirements in the real-time operating data, the thermal stability sensitive connection weights from each starting node to the central node in the interface performance weighted network are calculated. The process includes the following steps: Extract the failure frequency of each interface layer from the historical failure statistics, and calculate the failure probability coefficient of each interface layer based on the failure frequency. By combining the failure probability coefficient of each interface layer with the periodic difference value between the thermal expansion coefficient record and the design life requirement, the thermal stability sensitive connection weight of each starting node to the central node in the interface performance weighted network is calculated through a linear combination method.

6. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 1, characterized in that, In the interface performance weighted network, for each set of paths from the starting node to the ending node via the central node, the process of applying an adaptive optimization strategy to search for the optimal solution of path weights, and identifying multiple optimization trajectories of the set of interface layers to be optimized via the central node and associated with the performance degradation index, includes the following steps: An optimization solution space is created based on the path combinations from each starting node to the ending node in the interface performance weighted network. Heuristic search algorithms or stochastic optimization methods are used to evaluate the gradient changes or fitness scores of all connected edges in the optimization solution space. For the edge weights in each path, the optimal weights of a single path that independently satisfies the target stability threshold, or the optimal weights of a combination of paths that collaboratively satisfy the target stability threshold, are calculated by minimizing the path cost function. Based on the optimal weight configuration of each single path and the optimal weight configuration of the combined path, the gradient change or fitness score in the optimization solution space is updated. When the path cost function cannot be further reduced, multiple single paths and combined paths are output. By integrating the multiple single paths and combined paths, multiple optimization trajectories are determined for the set of interface layers to be optimized, which are associated with the performance degradation indicators via the central node.

7. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 1, characterized in that, The process of acquiring real-time operating data of perovskite solar cells, including performance degradation indicators and target stability thresholds, and selecting a set of interface layers to be optimized from multiple candidate interface layers of perovskite solar cells, includes the following operations: Obtain the real-time operating data, and query the interface layer records associated with the performance degradation index in the perovskite solar cell database according to the performance degradation index; The historical optimization frequency and number of optimizations for each interface layer are recorded in the perovskite solar cell database. Based on the historical optimization frequency and number of optimizations, the set of interface layers to be optimized is dynamically selected from the multiple candidate interface layers.

8. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 1, characterized in that, The method further includes: after generating the interface layer optimization scheme, feeding back the optimal solution of the path weight in the interface layer optimization scheme to the perovskite solar cell control system to adjust the material deposition parameters of the interface layer.

9. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 8, characterized in that, The adjustment of the material deposition parameters of the interface layer includes: modifying the thickness distribution and composition ratio of the interface layer according to the optimal solution of the path weight.

10. The method for optimizing the interface layer and enhancing the stability of perovskite solar cells according to claim 1, characterized in that, The method further includes a verification mechanism: After generating the interface layer optimization scheme, the simulated environment accelerated aging test process is executed to collect the performance degradation trajectory of the interface layer during the test. The performance degradation trajectory is compared in real time with the performance improvement parameters corresponding to the interface layer optimization scheme. When the actual performance degradation rate is detected to exceed the predicted value of the scheme and reach the preset deviation threshold, the operation of re-filtering the set of interface layers to be optimized is automatically triggered.

Images