Global maximum power point tracking method and system for three-terminal stacked device

Abstract

The application discloses a kind of three-terminal laminated device global maximum power point tracking method and system, method includes: according to the target short-circuit current parameter of device under preset spectrum, the spectrum of solar simulator is tuned;Based on electrical measurement system, obtain the initial power of device corresponding to initial voltage pair point, according to initial voltage point and initial power, construct probability agent model;According to probability agent model, based on bayesian optimization algorithm, obtain target voltage pair point in current iteration, stability verification is carried out to target voltage pair point using electrical measurement system, when passing through stability verification, obtain the steady-state power of device in target voltage pair point, determine probability agent model in next iteration according to stability verification result, until meeting iteration termination condition;According to the target voltage pair point and steady-state power that pass through stability verification, determine the global maximum power point of device.The scheme disclosed in the application can improve the efficiency and effectiveness of obtaining the global maximum power point of device.

Description

Technical Field

[0001] This application relates to the field of solar cell technology, and in particular to a global maximum power point tracking method and system for a three-terminal stacked device. Background Technology

[0002] In the electrical architecture of tandem solar cells, the three-terminal architecture uses a common electrode to achieve independent power extraction for the top and bottom sub-cells, avoiding current parameter matching limitations. The presence of the common electrode causes the two sub-cells to be electrically coupled; the operating state of one sub-cell affects the performance of the other through the series resistance of the common electrode. Therefore, the total output power of the three-terminal tandem device is equal to the top cell voltage V. top Bottom battery voltage V bot Two-dimensional function P total (V top V bot No single current-voltage scan can fully define the state of the device; the result depends heavily on the electrical state of the third terminal.

[0003] Currently, the method used to determine the true GMPP (Global Maximum Power Point) of a three-terminal multilayer device is two-dimensional mesh scanning. This method systematically scans a mesh consisting of (V... top V bot A matrix of voltage pairs is constructed, and power is measured on each voltage pair within this matrix to create a two-dimensional power surface and find the maximum value. This method requires testing a large number of voltage pairs, most of which are far from the maximum power point, resulting in extremely low data efficiency and a very long measurement process (requiring several hours).

[0004] Furthermore, when the top cell in a three-terminal stacked device uses perovskite material, the aforementioned characterization is difficult to amplify further. The inherent ion migration characteristics of perovskite materials result in significant hysteresis and transient effects in their IV curves, meaning the measurement results depend on the scan direction, rate, and pre-bias conditions. Therefore, if a rapid scan is performed in a two-dimensional grid scan to shorten the measurement time, a series of unsteady, artifact-affected power values ​​are measured, which cannot reflect the true steady-state performance of the device. To obtain physically meaningful steady-state performance, perovskite cells need to undergo a period of continuous illumination (photowetting) before measurement until their output power stabilizes, which typically takes several seconds to several minutes. If sufficient stabilization is performed at each voltage pair point, the measurement time becomes completely impractical, and prolonged bias and illumination may induce irreversible degradation of the device.

[0005] In summary, improving the efficiency and effectiveness of obtaining the global maximum power point of perovskite / crystalline silicon three-terminal stacked devices is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of this, the purpose of this application is to provide a global maximum power point tracking method and system for three-terminal stacked devices, which can improve the efficiency and effectiveness of obtaining the global maximum power point of perovskite / crystalline silicon three-terminal stacked devices.

[0007] To achieve the above objectives, this application provides the following technical solution: A method for global maximum power point tracking (MPPT) of a three-terminal stacked device includes: acquiring target short-circuit current parameters of the three-terminal stacked device under a preset spectrum; tuning the spectrum of a solar simulator according to the target short-circuit current parameters to perform spectral matching; wherein the three-terminal stacked device is a perovskite / crystalline silicon three-terminal stacked device; controlling an electrical measurement system to apply an initial voltage pair containing a voltage to the three-terminal stacked device; acquiring measurement data of the three-terminal stacked device at the initial voltage pair measured by the electrical measurement system; determining the initial power of the three-terminal stacked device at the initial voltage pair based on the measurement data; and constructing a probabilistic proxy based on the initial voltage point and the initial power. The model is as follows: Based on the probabilistic surrogate model and the Bayesian optimization algorithm, the target voltage pair in the current iteration is obtained. The stability of the target voltage pair is verified using the electrical measurement system. When the stability verification is passed, the steady-state power of the three-terminal stacked device at the target voltage pair is obtained. Based on the stability verification result, the probabilistic surrogate model in the next iteration is determined. The step of obtaining the target voltage pair in the current iteration based on the probabilistic surrogate model and the Bayesian optimization algorithm is returned to be executed until the iteration termination condition is met. Based on the target voltage pair that has passed the stability verification and the corresponding steady-state power, the global maximum power point of the three-terminal stacked device is determined.

[0008] A global maximum power point tracking system for a three-terminal multilayer device includes a solar simulator, an electrical measurement system, and a central control unit, wherein the central control unit is connected to the solar simulator and the electrical measurement system, and is used to implement the steps of the global maximum power point tracking method for any of the three-terminal multilayer devices described above.

[0009] The technical solution disclosed in this application utilizes intelligent sparse sampling of the Bayesian optimization algorithm to determine the global maximum power point, significantly reducing the number of measurement points required and achieving fast and efficient convergence to the global maximum power point. Furthermore, it evaluates the real-time stability of the target voltage pair selected by the Bayesian optimization algorithm by performing stability verification on the target voltage pair. Only physically valid target voltage pairs and steady-state power values ​​that pass stability verification are adopted by the Bayesian optimization algorithm. This minimizes measurement artifacts caused by the perovskite hysteresis effect, ensuring that the final global maximum power point found reflects the true performance of the three-terminal stacked device, thus improving the effectiveness of global maximum power point acquisition.

[0010] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0011] Figure 1 A flowchart of a global maximum power point tracking method for a three-terminal stacked device is provided for some embodiments of this application; Figure 2 A flowchart illustrating another global maximum power point tracking method for a three-terminal stacked device provided in some embodiments of this application; Figure 3 A flowchart of a global maximum power point tracking method for a three-terminal multilayer device provided in some embodiments of this application, in the case of applying a target voltage to a three-terminal multilayer device with a target voltage contained in the point; Figure 4 A flowchart of another method for global maximum power point tracking of a three-terminal multilayer device when applying a target voltage to a three-terminal multilayer device in the case of a target voltage contained in a point, as provided in some embodiments of this application; Figure 5 A flowchart of a global maximum power point tracking method for a three-terminal stacked device in the case of cyclic voltage scanning of a three-terminal stacked device based on a target voltage, provided for some embodiments of this application; Figure 6 A flowchart of another method for global maximum power point tracking of a three-terminal stacked device in the case of cyclic voltage scanning of a three-terminal stacked device based on a target voltage, provided for some embodiments of this application; Figure 7 A schematic diagram of the structure of a three-terminal stacked device global maximum power point tracking device provided in some embodiments of this application; Figure 8 This is a schematic diagram of the structure of a three-terminal stacked device global maximum power point tracking system provided in some embodiments of this application. Detailed Implementation

[0012] To overcome the Shockley-Queisser efficiency limit of single-junction solar cells, tandem solar cells composed of materials with different bandgap types, such as perovskite and crystalline silicon, have become a core development direction for next-generation photovoltaic technology. In the electrical architecture of tandem cells, there are three main structures: two-terminal (2T), four-terminal (4T), and three-terminal (3T). The 2T structure connects sub-cells in series, but due to the bottleneck of "current parameter matching," it experiences significant energy loss in real-world environments with varying spectra. The 4T structure completely solves the current parameter matching problem with four independent electrodes, but its complex structure and packaging process increase optical losses and manufacturing costs. The 3T architecture is a compromise that combines the advantages of 2T and 4T. It achieves independent power extraction for the top and bottom sub-cells through a common electrode, thus avoiding the current parameter matching limitation. However, the presence of the common electrode causes the two sub-cells to be electrically coupled. The operating state of one sub-cell affects the performance of the other sub-cell through the series resistance of the common electrode. Therefore, the total output power of a 3T device is a two-dimensional function P of the voltages of the two sub-cells. total (V top V bot Any single current-voltage scan cannot fully define the state of a device, and its results are highly dependent on the electrical state of a third terminal (e.g., open or short circuit), and are therefore misleading.

[0013] Currently, the method used in this field to determine the true global maximum power point of a three-terminal stacked device is two-dimensional mesh scanning. This method systematically scans a mesh consisting of (V... top V bot A matrix of voltage pairs is constructed, and power is measured at each point to build a two-dimensional power surface and find the maximum value. However, the two-dimensional grid scanning method requires testing a large number of voltage points, most of which are far from the maximum power point, resulting in extremely low data scanning efficiency and a very long measurement process (requiring several hours), which cannot meet the needs of high-throughput R&D and production quality control.

[0014] When perovskite materials are used in the top-side cell, the aforementioned characterization challenges are further amplified. The inherent ion migration characteristics of perovskite materials lead to significant hysteresis and transient effects in their IV curves. This means the measurement results depend on the scan direction, rate, and pre-bias conditions, making rapid measurements non-unique and unreliable. Furthermore, to obtain physically meaningful steady-state performance, perovskite cells require a period of continuous illumination (photowetting) before measurement until their output power stabilizes, typically taking several seconds to minutes. Therefore, using a two-dimensional grid scanning method to determine the global maximum power point of perovskite / crystalline silicon three-terminal stacked devices presents measurement validity issues. Specifically, there is a contradiction between the transient and hysteresis effects of perovskite cells and the time-consuming requirements of two-dimensional scanning: rapid scanning to shorten the time measures a series of non-steady-state power values ​​affected by artifacts, failing to reflect the true steady-state performance of the device; conversely, if sufficient stabilization is performed at each voltage pair point, the measurement time becomes completely infeasible, and prolonged bias and illumination may induce irreversible degradation of the device.

[0015] To this end, this application provides a global maximum power point tracking method and system for a three-terminal stacked device. First, the spectrum of the solar simulator is tuned according to the target short-circuit current parameter of the three-terminal stacked device under a preset spectrum to perform spectral matching and ensure that the test environment is physically equivalent to the preset standard solar spectrum. Then, the electrical measurement system applies the voltage contained in the initial voltage pair to the three-terminal stacked device, and obtains the measurement data of the three-terminal stacked device under the initial voltage pair. Based on this, the initial power of the three-terminal stacked device at the initial voltage pair is obtained. A probabilistic surrogate model is constructed based on the initial voltage pair and the initial power. Based on the probabilistic surrogate model, a Bayesian optimization algorithm is used to perform intelligent search to obtain the information-rich target voltage pair in each iteration. The electrical measurement system is used to perform stability verification on each target voltage pair to evaluate the real-time stability of the target voltage pair. When the target voltage pair passes the stability verification, the steady-state power of the three-terminal stacked device at the target voltage pair is obtained. The probabilistic surrogate model in the next iteration is determined based on the stability verification result of the target voltage pair until the iteration termination condition is met. After that, the global maximum power point of the three-terminal stacked device is determined based on the target voltage pair and steady-state power in each iteration corresponding to the stability verification. Therefore, this application utilizes a Bayesian optimization algorithm to guide the measurement process. Through intelligent sparse sampling, the Bayesian optimization algorithm significantly reduces the number of measurement points required, shortening the time to find the GMPP from hours to seconds to minutes, achieving an order-of-magnitude efficiency improvement and making it suitable for high-throughput applications. Furthermore, at each target voltage pair selected by the Bayesian optimization algorithm, this application does not immediately receive the measurement value. Instead, it first performs stability verification on the target voltage pair. Based on the stability verification results, it determines whether to update the probabilistic surrogate model in the current iteration. Only target voltage pairs and steady-state power that pass stability verification are used to update the probabilistic surrogate model in the Bayesian optimization algorithm. This minimizes measurement artifacts caused by perovskite hysteresis, ensuring that the finally found GMPP represents the true performance of the perovskite / crystalline silicon three-terminal stacked device, rather than an artifact caused by the measurement protocol, thus guaranteeing the accuracy and effectiveness of the measurement.

[0016] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0017] See Figure 1 The flowchart provided in some embodiments of this application illustrates a global maximum power point tracking method for a three-terminal stacked device, which may include: S11: Obtain the target short-circuit current parameters of the three-terminal stacked device under the preset spectrum, and tune the spectrum of the solar simulator according to the target short-circuit current parameters to perform spectral matching; the three-terminal stacked device is a perovskite / crystalline silicon three-terminal stacked device.

[0018] In some embodiments of this application, the execution entity of the global maximum power point tracking (MPPT) method for three-terminal stacked devices can be the central control unit in the MPT system for three-terminal stacked devices. This central control unit can be a computer, etc., and the MPT system can also include a solar simulator and an electrical measurement system. The central control unit runs the core algorithm software of this application to coordinate the solar simulator and the electrical measurement system, and to perform adaptive search and stability verification, etc. The solar simulator can be a multi-channel, spectrally tunable LED solar simulator, used to accurately simulate a preset spectrum under standard test conditions and achieve sub-cell current parameter matching. The preset spectrum can be an AM1.5G spectrum. The electrical measurement system includes at least two precisely synchronized source measurement units (SMUs), respectively connected to two independent loops of the three-terminal stacked device (i.e., the loop corresponding to the top cell and the loop corresponding to the bottom cell), used to apply voltage pairs (V... top V bot It can also measure the corresponding current parameters and / or power, wherein the source measurement unit can measure the corresponding current parameters and / or power at a high frequency (e.g., 10Hz). It should be noted that the current parameters mentioned in some embodiments of this application can specifically be current density or current, and correspondingly, the power mentioned in some embodiments of this application can specifically be power density corresponding to current density or electric power corresponding to current.

[0019] In some embodiments of this application, the three-terminal stacked device mentioned is specifically a perovskite / crystalline silicon three-terminal stacked device, in which the top cell is a perovskite cell and the bottom cell is a crystalline silicon cell.

[0020] When performing global maximum power point tracking on a three-terminal stacked device, initialization and spectral matching can be performed first.

[0021] Specifically, the target short-circuit current parameters of the three-terminal stacked device under a preset spectrum can be obtained first. These target short-circuit current parameters include the target short-circuit current parameters of the top cell and the bottom cell in the three-terminal stacked device. The aforementioned method for obtaining the target short-circuit current parameters of the three-terminal stacked device under the preset spectrum can be as follows: Measure the external quantum efficiency of the top cell and the external quantum efficiency of the bottom cell in the three-terminal stacked device; calculate the target short-circuit current parameters of the top cell under the preset spectrum based on the external quantum efficiency of the top cell; calculate the target short-circuit current parameters of the bottom cell under the preset spectrum based on the external quantum efficiency of the bottom cell; thus obtaining the target short-circuit current parameters of the three-terminal stacked device under the preset spectrum, thereby improving the accuracy and reliability of obtaining the target short-circuit current parameters. Alternatively, the correspondence between the structural information of the three-terminal stacked device and the target short-circuit current parameters of the three-terminal stacked device under the preset spectrum can be stored in the central control unit beforehand. The structural information of the three-terminal stacked device can then be obtained, and the target short-circuit current parameters of the three-terminal stacked device under the preset spectrum can be obtained based on the pre-stored correspondence and the structural information of the three-terminal stacked device.

[0022] Based on the target short-circuit current parameters of the three-terminal stacked device under the preset spectrum, the spectrum of the solar simulator can be tuned according to the target short-circuit current parameters of the three-terminal stacked device under the preset spectrum to perform spectral matching, ensuring that the test environment is physically equivalent to the standard solar spectrum, thereby ensuring the effectiveness and stability of the test.

[0023] After spectral matching, the output of the solar simulator can be kept constant while performing a light-immersion pretreatment on the three-terminal stacked device. This allows the device to continuously receive light, enabling ion migration within the perovskite to reach equilibrium until the power output stabilizes. This eliminates transient interference (hysteresis) caused by ion migration, ensuring that the measured performance is the true steady-state performance of the three-terminal stacked device and improving the effectiveness and stability of the test. For example, the power of the three-terminal stacked device (or the power of the top cell) can be acquired, and the time-varying rate of power can be calculated. If the time-varying rate of power is less than a first preset threshold, the power output is considered stable. Alternatively, the three-terminal stacked device can undergo a light-immersion pretreatment for a preset duration to stabilize ion migration. The preset duration can be determined in advance based on the ion migration characteristics of the perovskite, for example, 5-10 minutes. After this preset duration of light-immersion pretreatment, the power output of the three-terminal stacked device stabilizes.

[0024] S12: Control the electrical measurement system to apply the voltage contained in the initial voltage pair to the three-terminal stacked device, obtain the measurement data of the three-terminal stacked device under the initial voltage pair measured by the electrical measurement system, determine the initial power of the three-terminal stacked device at the initial voltage pair based on the measurement data, and construct a probabilistic surrogate model based on the initial voltage point and the initial power.

[0025] Based on the above, an adaptive GMPP search can be performed using a Bayesian optimization algorithm, and the search for the GMPP can be defined as maximizing a black-box function P. total (V top V bot The optimization problem of ).

[0026] First, the initial power of the three-terminal stacked device at the initial voltage pair can be obtained using an electrical measurement system. The initial voltage pair can be determined based on a two-dimensional voltage range formed by the adjustable range of the top battery (i.e., the actual voltage range of the top battery, for example, 0-1.2V) and the adjustable range of the bottom battery (i.e., the actual voltage range of the bottom battery, for example, 0-0.7V). For example, the initial voltage pair can be a uniformly distributed point within the two-dimensional voltage range, and the number of initial voltage pairs can be set according to requirements, for example, 5-10 pairs, such as (0.3V, 0.2V), (0.6V, 0.4V), etc. It should be noted that the adjustable ranges of the top and bottom batteries in the three-terminal stacked device can be determined based on the structure of the three-terminal stacked device.

[0027] Specifically, the electrical measurement system can be controlled to apply the initial voltage pair to the three-terminal stacked device, meaning the source measurement unit corresponding to the top and bottom batteries in the electrical measurement system can synchronously apply the corresponding initial voltage pair to the top and bottom batteries. The electrical measurement system then measures the measurement data of the three-terminal stacked device and sends the data to the central control unit. Specifically, the electrical measurement system can measure the current parameters corresponding to the top and bottom batteries in the three-terminal stacked device, or the power corresponding to the top and bottom batteries, or the power of the three-terminal stacked device (i.e., the total power of the three-terminal stacked device, which is the sum of the power of the top and bottom batteries). Taking the current parameter as an example, the electrical measurement system can be used to measure the corresponding current (I0). top ,I bot ) or the corresponding power (P) top ,P bot The power P of a three-terminal stacked device total , where V top V bot These are the voltages corresponding to the top battery and the bottom battery, respectively. top I bot P represents the current of the top battery and the current of the bottom battery, respectively. top P bot These represent the power of the top battery and the power of the bottom battery, respectively, P. total =P top +P bot P totalThis refers to the power of a three-terminal stacked device.

[0028] After acquiring the measurement data of the three-terminal stacked device measured by the electrical measurement system at the initial voltage point, the central control unit can determine the initial power of the three-terminal stacked device at the initial voltage point based on the measurement data. Specifically, if the central control unit acquires the current parameters of the three-terminal stacked device measured by the electrical measurement system, it can calculate the initial power corresponding to the initial voltage point based on the initial voltage point and the corresponding current parameters. If the central control unit acquires the power of the top battery and the power of the bottom battery of the three-terminal stacked device measured by the electrical measurement system, it can calculate the initial power of the three-terminal stacked device based on the power of the top battery and the bottom battery. If the central control unit acquires the power of the three-terminal stacked device measured by the electrical measurement system, it can directly determine the initial power of the three-terminal stacked device at the initial voltage point. It should be noted that the aforementioned initial power can specifically be the steady-state initial power. For example, after applying an initial voltage to the three-terminal stacked device, the time change rate of the current parameter or power of the three-terminal stacked device can be obtained. If the time change rate of the current parameter or power is less than a first preset threshold, it can be determined that the power output of the three-terminal stacked device is stable. At this time, the initial power of the three-terminal stacked device can be obtained.

[0029] After obtaining the initial power corresponding to the initial voltage pair of the three-terminal stacked device, a probabilistic surrogate model can be constructed based on the initial voltage pair and its corresponding initial power. The probabilistic surrogate model mentioned in some embodiments of this application can be a Gaussian process surrogate model, or a Bayesian linear regression surrogate model, etc. The constructed probabilistic surrogate model can not only predict the predicted power of any unmeasured voltage pair based on existing data—that is, it can predict the predicted power of each candidate voltage pair (each candidate voltage pair includes the voltage corresponding to the top battery and the voltage corresponding to the bottom battery)—but also output the prediction uncertainty value (i.e., the reliability of the predicted power). The candidate voltage pairs are determined based on the two-dimensional voltage range of the three-terminal stacked device. Specifically, the two-dimensional voltage range can be discretized into a grid composed of a large number of voltage pairs, each of which is a corresponding candidate voltage pair. For example, the uncertainty of the probabilistic surrogate model prediction is low for regions close to the initial data points (i.e., the initial voltage pair and the initial power); and high for regions far from the initial data points.

[0030] S13: Based on the probabilistic surrogate model and the Bayesian optimization algorithm, obtain the target voltage pair in the current iteration, use the electrical measurement system to verify the stability of the target voltage pair, and obtain the steady-state power of the three-terminal stacked device at the target voltage pair when the stability verification is passed.

[0031] Building upon the above, iterative optimization can be performed, and a specially designed real-time stability evaluation subroutine can be invoked at the target voltage pair selected in each iteration of the Bayesian optimization algorithm to verify the stability of the target voltage pair. By combining the Bayesian optimization algorithm with the real-time stability evaluation subroutine, a closed-loop, intelligent measurement and verification system can be constructed.

[0032] First, the target voltage pair in the current iteration can be obtained based on a Bayesian optimization algorithm using a probabilistic surrogate model. Specifically, in the current iteration, the predicted power and uncertainty value of each candidate voltage pair can be predicted based on the corresponding probabilistic surrogate model. A sampling function is used to calculate the sampling value of each candidate voltage pair based on its predicted power and uncertainty value. Based on the sampling values ​​of each candidate voltage pair, the next optimal candidate voltage pair is selected as the target voltage pair. That is, in each iteration, the sampling function selects the next sampling point with the "richest information." For example, the candidate voltage pair with the highest sampling value can be selected as the target voltage pair. This function balances the "utilization" of the high-power region and the "exploration" of the uncertain region. "Utilization" logic: If a certain region (such as V) top =0.8V, V bot The high predicted power and low uncertainty in the region (around 0.5V) indicate that the region is likely close to GMPP. The acquisition function will prioritize sampling points in this region to further confirm the location of the power peak. "Exploration" logic: If a certain area (such as V) top =0.2V, V bot The prediction uncertainty is high in the region around 0.6V (there may be undiscovered high power points). Even if the current predicted power is not high, the acquisition function will select points in this region for sampling to reduce model uncertainty.

[0033] It should be noted that the acquisition function mentioned in some embodiments of this application can be GP-UCB (Gaussian process-upper confidence bound). Of course, other acquisition functions can also be used to implement the above process. In addition, if the current iteration is the first iteration, the target voltage pair can be obtained according to the probabilistic surrogate model and acquisition function constructed in step S12; if the current iteration is not the first iteration, the target voltage pair can be obtained according to the probabilistic surrogate model and acquisition function obtained in the previous iteration.

[0034] After obtaining the target voltage pair in the current iteration, the stability of the target voltage pair can be verified using an electrical measurement system to determine whether the ion migration of the perovskite material has reached a stable state. If the target voltage pair passes the stability verification, it indicates that the power corresponding to the target voltage pair has reached a steady state and is not affected by perovskite-related transient artifacts. Therefore, after the target voltage pair passes the stability verification, the steady-state power of the three-terminal stacked device at the target voltage pair (specifically, the true steady-state power of the three-terminal stacked device at the target voltage pair) can be obtained.

[0035] In some embodiments of this application, when verifying the stability of a target voltage pair using an electrical measurement system, stability data corresponding to the target voltage pair can also be obtained. This allows for the acquisition of stability data corresponding to the target voltage pair in each iteration of the Bayesian optimization algorithm, facilitating the acquisition of instability information of the three-terminal stacked device at each target voltage pair. This, in turn, improves the understanding and optimization of the physical mechanism of the three-terminal stacked device. Based on the above, after satisfying the iteration termination condition, an instability map can be generated using an interpolation algorithm (such as Kriging interpolation) based on the target voltage pair and its corresponding stability data in each iteration. This provides systematic information about the device's uncertainty across the entire operating voltage range during the tracking of the global maximum power point, providing a diagnostic basis for subsequent device optimization. For example, if stability data for a certain voltage range indicates high instability, it suggests significant ion instability in that range, requiring improvement in the ion migration suppression capability of the perovskite material.

[0036] The above process can generate an instability map covering the operating range of the three-terminal stacked device while searching for the global maximum power point. This instability map provides valuable diagnostic information for device engineers, revealing the intrinsic relationship between device instability and operating bias voltage, thus elevating a simple performance testing tool into a powerful physical diagnostic platform.

[0037] Step S14: Determine the probabilistic surrogate model for the next iteration based on the stability verification results, and return to execute the step of obtaining the target voltage pair in the current iteration based on the Bayesian optimization algorithm according to the probabilistic surrogate model, until the iteration termination condition is met.

[0038] After verifying the stability of the target voltage at the point using an electrical measurement system, the probabilistic surrogate model for the next iteration can be determined based on the stability verification results obtained above. Specifically, it can be determined whether to update the probabilistic surrogate model in the current iteration based on the stability verification results, and corresponding operations can be performed based on the determination result to obtain the probabilistic surrogate model for the next iteration.

[0039] After determining the probabilistic surrogate model for the next iteration based on the stability verification results, the next iteration can be entered. That is, the next iteration is taken as the current iteration. The process returns to the step of obtaining the target voltage pair in the current iteration based on the Bayesian optimization algorithm according to the probabilistic surrogate model, until the iteration termination condition is met, and then the iteration process is terminated.

[0040] As can be seen from the above, some embodiments of this application utilize Bayesian optimization algorithms to guide the measurement process. The Bayesian optimization algorithm constructs a probabilistic proxy model of the two-dimensional power surface of the three-terminal stacked device and uses the acquisition function to intelligently select the next measurement point with the richest information. This method avoids redundant sampling of the vast invalid region far from the global maximum power point and concentrates the computing and measurement resources on the most critical voltage range, thereby achieving extremely high data efficiency.

[0041] S15: Determine the global maximum power point of the three-terminal stacked device based on the target voltage pair that has passed stability verification and the corresponding steady-state power.

[0042] After terminating the iteration process, the global maximum power point of the three-terminal stacked device can be determined based on the target voltage pair and steady-state power that have passed the stability verification. Specifically, the maximum steady-state power and the corresponding target voltage pair that have passed the stability verification can be determined as the global maximum power point of the three-terminal stacked device to improve the effectiveness and efficiency of determining the global maximum power point.

[0043] As described above, compared to the current method of determining the global maximum power point (GMPP) using two-dimensional grid scanning, some embodiments of this application replace scanning with intelligent search. Through intelligent sparse sampling using a Bayesian optimization algorithm, the number of measurement points is reduced from tens of thousands to dozens (e.g., less than 30), significantly reducing the required number of measurement points and shortening the time to find the GMPP from hours to seconds to minutes, achieving an order-of-magnitude efficiency improvement and making it suitable for high-throughput applications. Furthermore, some embodiments of this application ensure, through a real-time stability evaluation subroutine, that each data point used to construct the probabilistic proxy model is a physically valid steady-state value. This minimizes measurement artifacts caused by perovskite hysteresis, ensuring that the finally found GMPP represents the true performance of the three-terminal stacked device, rather than an artifact caused by the measurement protocol. This guarantees the accuracy and physical meaning of the measurement, improving the effectiveness and accuracy of obtaining the global maximum power point. Therefore, some embodiments of this application organically combine the two key concepts of intelligent search and high efficiency of real-time verification to achieve a completely new characterization process: it no longer passively and indiscriminately collects data, but actively and strategically selects the next measurement point and performs rigorous stability verification on each measured power value, thereby obtaining the most reliable measurement results at the fastest speed and highest efficiency, and achieving high-speed, high-precision and high-efficiency tracking and determination of the global maximum power point, so as to find the physically reliable global maximum power point that has been verified for stability in a much faster time than traditional methods.

[0044] In some embodiments of this application, when determining the global maximum power point of the three-terminal stacked device, the power contributions of the top and bottom cells in the three-terminal stacked device can also be calculated based on the global maximum power point. And / or, after determining that the target voltage pair has failed the stability verification, the power corresponding to the target voltage pair can also be obtained. Based on this, a two-dimensional power surface map of the three-terminal stacked device can be generated using the probabilistic surrogate model in the last iteration, according to the target voltage pair and its corresponding power in each iteration. Specifically, the power corresponding to the unsampled points in the two-dimensional voltage interval can be filled in using the probabilistic surrogate model in the last iteration, so as to generate a two-dimensional power surface map of the three-terminal stacked device based on the target voltage pair and its corresponding power, and the unsampled points and their corresponding power. The unsampled point pairs are the candidate voltage pairs other than the target voltage pair among the candidate voltage pairs.

[0045] Furthermore, in some embodiments of this application, the global maximum power point tracking (GMPP) method for three-terminal stacked devices can be embedded as a measurement module into long-term device stability testing protocols, such as the ISOS (International Summit on Organic Photovoltaic Stability) protocol. During the application of environmental stress (such as continuous illumination or thermal cycling) to the three-terminal stacked device, a complete GMPP tracking and characterization is performed at high temporal resolution (e.g., every 1-5 minutes). This approach allows for near real-time tracking of the dynamic evolution of the GMPP and the entire power surface morphology during the degradation process of the three-terminal stacked device. This provides unprecedented high-resolution data, revealing the specific degradation path and physical mechanisms of the three-terminal stacked device (e.g., whether it is a drop in top cell voltage or bottom cell voltage, or an increase in ionic instability under a specific bias voltage). This provides crucial experimental evidence for developing more stable materials and device structures and predicting the long-term lifetime of devices.

[0046] See Figure 2 This is a flowchart illustrating another global maximum power point tracking (GMP) method for a three-terminal multilayer device provided in some embodiments of this application. The GMP method for a three-terminal multilayer device provided in some embodiments of this application, which determines the probabilistic surrogate model for the next iteration based on stability verification results, may include: If the target voltage pair passes the stability verification, the probabilistic surrogate model in the current iteration is updated based on the target voltage pair and steady-state power to obtain the probabilistic surrogate model in the next iteration. If the target voltage pair fails the stability verification, the probabilistic surrogate model in the current iteration is maintained to obtain the probabilistic surrogate model in the next iteration.

[0047] In some embodiments of this application, when using an electrical measurement system to verify the stability of a target voltage pair, if the target voltage pair passes the stability verification (i.e., if the stability verification result is "passes stability verification"), it indicates that the target voltage pair and the obtained steady-state power of the three-terminal stacked device at the target voltage pair are valid data and steady-state data. This also indicates that the ion migration of the perovskite material has reached a stable state (without significant transient shift). In this case, the probabilistic surrogate model in the current iteration can be updated based on the target voltage pair and the corresponding steady-state power to obtain the probabilistic surrogate model in the next iteration. For example, the target voltage pair and the corresponding steady-state power can be added to a dataset containing the initial voltage pair and the corresponding initial power. The dataset can be used for retraining to obtain the probabilistic surrogate model in the next iteration. Alternatively, the probabilistic surrogate model in the current iteration can be incrementally fine-tuned or subjected to online learning based on the target voltage pair and the corresponding steady-state power to obtain the probabilistic surrogate model in the next iteration. Then, the next iteration can proceed, i.e., the steps of obtaining the target voltage pair and related data in the current iteration based on the probabilistic surrogate model using a Bayesian optimization algorithm can be executed. By updating the probabilistic surrogate model, the prediction accuracy will be continuously optimized as data accumulates, gradually focusing on the GMPP region.

[0048] If the target voltage pair fails the stability verification (i.e., the stability verification result is "failed"), it indicates that the power at the target voltage pair and the three-terminal stacked device at the target voltage pair is not valid or steady-state data. This means the ion migration of the perovskite material has not yet reached a stable state. In this case, updating the probabilistic surrogate model in the current iteration using the target voltage pair is rejected. The probabilistic surrogate model in the current iteration can be maintained, allowing it to participate in the next iteration. In other words, the probabilistic surrogate model in the next iteration will be the same as the one in the current iteration. It should be noted that even if the probabilistic surrogate model in the next iteration is the same as the one in the current iteration, the ion migration and transient drift within the perovskite will change over time. Therefore, the results of subsequent iterations will vary.

[0049] The above process ensures that only physically valid target voltage pairs and their corresponding steady-state power that have passed stability verification are used to update the probabilistic surrogate model in the current iteration. This ensures data validity, avoids the influence of perovskite-related transient artifacts, and improves the prediction accuracy of the probabilistic surrogate model, making the final global maximum power point found effective and accurate.

[0050] This application provides a global maximum power point tracking method for a three-terminal multilayer device, which obtains the target voltage pair in the current iteration based on a Bayesian optimization algorithm using a probabilistic surrogate model. The method may include: The target voltage pair with the highest acquisition value in the current iteration is obtained based on the probabilistic proxy model and acquisition function in the current iteration.

[0051] In some embodiments of this application, the target voltage pair in the current iteration is obtained based on a Bayesian optimization algorithm using a probabilistic surrogate model. Specifically, this can be achieved by using the probabilistic surrogate model and the acquisition function in the current iteration to obtain the target voltage pair with the highest acquired value. Specifically, the predicted power and uncertainty value of each candidate voltage pair are calculated using the aforementioned probabilistic surrogate model, and the acquired value of each candidate voltage pair is calculated using the acquisition function based on the predicted power and uncertainty value. The candidate voltage pair with the highest acquired value is then determined as the target voltage pair.

[0052] If the current iteration is the first iteration, the target voltage pair with the highest collected value in the first iteration can be obtained based on the probabilistic surrogate model and acquisition function constructed in step S12. If the current iteration is not the first iteration, the probabilistic surrogate model obtained in the previous iteration (i.e., the probabilistic surrogate model in the current iteration obtained in the previous iteration) can be obtained. The target voltage pair with the highest collected value can be obtained based on the probabilistic surrogate model obtained in the previous iteration and the acquisition function. The specific implementation process is similar to the above process and will not be repeated here.

[0053] The above process enables the most informative and valuable voltage pairs to be involved in the relevant process in each iteration, thereby efficiently balancing exploration and utilization, accelerating convergence, reducing the number of iterations, and achieving a balance between exploration (probing regions with high uncertainty) and utilization (fine sampling in regions with high predicted power) with the fewest samplings, thus converging to the global maximum power point quickly and efficiently.

[0054] See Figure 3 and Figure 4 ,in, Figure 3 The flowchart illustrates a global maximum power point tracking method for a three-terminal multilayer device (MTP) when a target voltage is applied to the device at a point containing the target voltage, as provided in some embodiments of this application. Figure 4 This is a flowchart illustrating another method for global maximum power point tracking (GMP) of a three-terminal multilayer device (TFT-TFT) when a target voltage pair containing a target voltage is applied to the TFT-TFT according to some embodiments of this application. One such method, which utilizes an electrical measurement system to verify the stability of the target voltage pair, may include: The electrical measurement system controls the application of a target voltage pair containing the target voltage to the three-terminal stacked device, and monitors the time-varying rate of change of the current parameter or power of the three-terminal stacked device under the target voltage pair. If the time change rate of the current parameter or power is less than the first preset threshold, the target voltage point is determined to have passed the stability verification; if the time change rate of the current parameter or power is not less than the first preset threshold, the process returns to the step of monitoring the time change rate of the current parameter or power of the three-terminal stacked device under the target voltage point. or, If the time change rate of the current parameter or power is less than the first preset threshold and the monitoring time has not reached the maximum monitoring time, the target voltage pair is determined to have passed the stability verification; if the time change rate of the current parameter or power is not less than the first preset threshold and the monitoring time has reached the maximum monitoring time, the target voltage pair is determined to have failed the stability verification.

[0055] In some embodiments of this application, the stability verification of a target voltage pair using an electrical measurement system can be achieved by controlling the electrical measurement system to apply voltage to the three-terminal stacked device according to the target voltage pair in the current iteration, obtaining the voltage application result, and determining whether the target voltage pair passes the stability verification based on the voltage application result. Specifically, the implementation process can be as follows: Step 101: Control the electrical measurement system to apply the target voltage contained in the target voltage pair to the three-terminal stacked device. Specifically, control the source measurement units corresponding to the top and bottom cells in the electrical measurement system to synchronously apply the corresponding target voltages in the target voltage pair to the top and bottom cells, respectively.

[0056] Step 102: Obtain the current parameters or power of the three-terminal stacked device at the target voltage point as measured by the electrical measurement system. The electrical measurement system can measure the aforementioned data at high frequency. Specifically, the aforementioned current parameters can be the current parameters of the top battery and the bottom battery; the aforementioned power can be the power of the top battery and the bottom battery, or it can be the power of the three-terminal stacked device (i.e., the total power of the three-terminal stacked device). If the central control unit obtains the power of the top battery and the power of the bottom battery, the power of the three-terminal stacked device can be calculated accordingly.

[0057] Step 103: Monitor the time rate of change of the current or power of the three-terminal multilayer device at the target voltage point. That is, the time rate of change of the current parameter or power of the three-terminal multilayer device at the target voltage point can be determined based on the current parameter or power of the three-terminal multilayer device at the target voltage point. For example, ΔP / Δt (time rate of change of power) can be calculated every 0.1 seconds.

[0058] Step 104: Determine whether the target voltage point has passed the stability verification based on the current parameter or the time rate of change of power.

[0059] Specifically, it can be determined whether the time-varying rate of change of the current parameter or power is less than a first preset threshold. This first preset threshold can be determined in advance through experiments, for example, 0.1% / s. If the time-varying rate of change of the current parameter or power is less than the first preset threshold, it indicates that the ion migration of the perovskite material has reached a stable state (no significant transient drift). Therefore, the target voltage point can be determined to have passed stability verification, and the power of the three-terminal stacked device at this time (i.e., steady-state power) can be recorded and acquired, and the recorded and acquired power is "steady-state valid data." If the time-varying rate of change of the current parameter or power is not less than the first preset threshold, it indicates that the ion migration of the perovskite material has not yet reached a stable state. Therefore, the target voltage point can be determined to have failed stability verification. In this case, the process can return to monitoring the time-varying rate of change of the current parameter or power of the three-terminal stacked device at the target voltage point until the time-varying rate of change of the current parameter or power is less than the first preset threshold, i.e., until the target voltage point passes stability verification.

[0060] Specifically, determining whether the time-varying rate of change of the current parameter or power is less than the first preset threshold can be achieved by determining whether the time-varying rate of change of the current parameter or power is less than the first preset threshold for multiple consecutive cycles (e.g., 3 cycles, i.e., multiple consecutive cycles). This method can offset random errors, improve the stability and reliability of the results, and enhance the accuracy of stability verification. Accordingly, if the time-varying rate of change of the current parameter or power is less than the first preset threshold for multiple consecutive cycles, the target voltage pair is determined to have passed stability verification. If the time-varying rate of change of the current parameter or power is not less than the first preset threshold for multiple consecutive cycles, the process can return to the step of monitoring the time-varying rate of change of the current or power of the three-terminal stacked device at the target voltage pair until the time-varying rate of change of the current parameter or power is less than the first preset threshold for multiple consecutive cycles.

[0061] In addition, in the above process, obtaining stability data can specifically be done by obtaining the monitoring duration, which is the monitoring duration corresponding to the time change rate of the current parameter or power of the three-terminal stacked device under the target voltage point. Specifically, the monitoring duration can be obtained from the moment the electrical measurement system applies the target voltage point to the three-terminal stacked device.

[0062] Alternatively, in some embodiments of this application, when monitoring the current parameters or power of the three-terminal stacked device, a monitoring duration is obtained. This monitoring duration is the monitoring duration corresponding to the time change rate of the current parameters or power of the three-terminal stacked device under the target voltage point. Specifically, the monitoring duration can be obtained from the moment the electrical measurement system applies the target voltage point to the three-terminal stacked device, and this monitoring duration is the obtained stability data. Based on this, determining whether the target voltage point passes the stability verification based on the time change rate of the current parameters or power can specifically be: determining whether the time change rate of the current parameters or power is less than a first preset threshold and whether the monitoring duration has reached the maximum monitoring duration. The maximum monitoring duration can be determined in advance through experiments or other methods. Setting the maximum monitoring duration can avoid excessive monitoring time, shorten the time used for a single iteration, and improve the efficiency of determining the global maximum power point. If the time-varying rate of change of the current parameter or power is less than the first preset threshold and the monitoring time has not reached the maximum monitoring time, it indicates that the ion migration of the perovskite material has reached a stable state before the maximum monitoring time is reached. In this case, it can be determined that the target voltage point has passed the stability verification, and the steady-state power of the three-terminal stacked device can be recorded and acquired. If the time-varying rate of change of the current parameter or power is not less than the first preset threshold and the monitoring time has reached the maximum monitoring time, it indicates that the ion migration of the perovskite material has not yet reached a stable state before the maximum monitoring time is reached. In this case, it can be determined that the target voltage point has not passed the stability verification. In this situation, the probabilistic surrogate model in the current iteration can be maintained to avoid excessive time consumption, thereby shortening the time used for a single iteration and improving the efficiency of determining the global maximum power point.

[0063] Specifically, determining whether the time change rate of the current parameter or power is less than the first preset threshold can be done by determining whether the time change rate of the current parameter or power is less than the first preset threshold in multiple consecutive cycles (e.g., 3 cycles, i.e., multiple consecutive cycles).

[0064] The above process can be used to monitor the changes in current or power of three-terminal stacked devices over time, thereby enabling the verification of the stability of the target voltage point.

[0065] This application provides a global maximum power point tracking method for a three-terminal multilayer device, which utilizes an electrical measurement system to verify the stability of a target voltage point, and may include: Based on the target voltage, the electrical measurement system controls the point-to-point control to perform cyclic voltage scanning on the three-terminal stacked device, and the hysteresis index is obtained based on the scanning results. The hysteresis index is used to determine whether the target voltage point has passed the stability verification.

[0066] In some embodiments of this application, the stability verification of the target voltage at a point using an electrical measurement system can also be as follows: Step 201: Control the electrical measurement system to perform cyclic voltage scanning on the three-terminal multilayer device according to the target voltage. The cyclic voltage scanning specifically refers to performing forward and reverse voltage scans on the three-terminal multilayer device. The scanning rate of the cyclic voltage scan is greater than a preset rate to achieve rapid cyclic voltage scanning and avoid device degradation induced by prolonged bias voltage. Simultaneously, the electrical measurement system can measure the scanning results of the three-terminal multilayer device while performing the cyclic voltage scan.

[0067] Step 202: Obtain the scanning results of the cyclic voltage scan of the three-terminal stacked device by the electrical measurement system, and obtain the hysteresis index based on the scanning results. The obtained hysteresis data can be used not only as the voltage application result but also as stability data.

[0068] Step 203: Determine whether the target voltage pair passes the stability verification based on the hysteresis index.

[0069] The above method enables cyclic voltage scanning of three-terminal stacked devices based on the target voltage point, and quantifies the fast hysteresis based on the scanning results, thereby verifying the stability of the target voltage point.

[0070] This application provides a global maximum power point tracking method for a three-terminal multilayer device, which uses a point-controlled electrical measurement system to perform cyclic voltage scanning on the three-terminal multilayer device based on a target voltage, and obtains the hysteresis index based on the scanning results. The method may include: Centered on the target voltage corresponding to the top cell in the target voltage pair, the electrical measurement system controls the application of forward scanning voltage and reverse scanning voltage to the top cell in the three-terminal stacked device; wherein, the scanning rate is greater than the preset rate, and during the scanning of the top cell, the voltage of the bottom cell in the three-terminal stacked device is maintained at the corresponding target voltage or fluctuates synchronously with the voltage of the top cell. Obtain the forward power of the three-terminal stacked device at the target voltage point during forward scanning, and obtain the reverse power of the three-terminal stacked device at the target voltage point during reverse scanning. Calculate the hysteresis index based on the forward and reverse power. Alternatively, obtain the forward scanning response current parameters of the top cell during forward scanning and the reverse scanning response current parameters of the top cell during reverse scanning. Obtain the forward scanning curve based on the forward scanning voltage and forward scanning response current parameters, and obtain the reverse scanning curve based on the reverse scanning voltage and reverse scanning response current parameters. Calculate the hysteresis index based on the forward and reverse scanning curves.

[0071] In some embodiments of this application, when the electrical measurement system is controlled to perform cyclic voltage scanning on the three-terminal stacked device according to the target voltage pair, specifically, the target voltage corresponding to the top battery in the target voltage pair can be used as the center, and a preset amplitude can be used as the scanning amplitude. The electrical measurement system first applies a forward scanning voltage to the top battery and obtains the forward scanning response current parameter of the top battery or the power of the top and bottom batteries at the target voltage pair or the forward power of the three-terminal stacked device at the target voltage pair, which are simultaneously measured by the electrical measurement system. Then, the electrical measurement system applies a reverse scanning voltage to the top battery and obtains the reverse scanning response current parameter of the top battery or the power of the top and bottom batteries at the target voltage pair or the reverse power of the three-terminal stacked device at the target voltage pair, which are simultaneously measured by the electrical measurement system.

[0072] During the scanning process of the top battery, the central control unit can control the electrical measurement system to apply the target voltage to the bottom battery, or it can control the voltage applied to the bottom battery by the electrical measurement system to fluctuate slightly in sync with the voltage of the top battery, with the target voltage corresponding to the bottom battery as the center.

[0073] It should be noted that the preset amplitude can be less than the amplitude threshold, meaning the preset amplitude can be relatively small. The preset amplitude allows for a balance between capturing local instability and avoiding interference with the device state—an excessively large amplitude may alter the ion distribution, while an excessively small amplitude will prevent effective calculation of the hysteresis area. The scan rate can be greater than the preset rate to freeze ion migration and ensure that the hysteresis index reflects only the inherent instability at that point, rather than the dynamic changes in ions during the scan process.

[0074] As described above, the central control unit can control the electrical measurement system to perform a rapid, small-amplitude cyclic voltage scan near the target voltage point. For example, the scan range (i.e., the preset amplitude) is ±0.5V. Accordingly, the central control unit can control the electrical measurement system to perform a voltage scan on the top battery from V... top0 -0.5V→V top0 →V top0 +0.5V→V top0 →V top0 -0.5V cyclic voltage scan, where V top0 V represents the target voltage corresponding to the top battery at the target point. top0 -0.5V→V top0 →V top0 +0.5V is the voltage for the forward scan, V top0 +0.5V→V top0 →V top0-0.5V is the voltage for reverse scanning. For example, if the target voltage corresponding to the top cell at the target voltage point is 0.7V, then the cyclic voltage scan of the top cell can be 0.65V→0.7V→0.75V→0.7V→0.65V; the scan rate is 1V / s to avoid device degradation induced by prolonged bias voltage. By performing rapid, small-amplitude cyclic voltage scans centered on the target voltage corresponding to the top cell at the target voltage point, the unique ion migration and hysteresis behavior of perovskite materials can be specifically induced and captured, thereby determining whether the power at the target voltage point has reached a physical steady state.

[0075] Based on the above, the hysteresis index can be obtained from the scanning results in either of the following two ways: The first method: deriving the hysteresis index based on power. Step 301: Obtain the forward power of the three-terminal stacked device at the target voltage point during forward scanning, and obtain the reverse power of the three-terminal stacked device at the target voltage point during reverse scanning.

[0076] The forward and reverse power can be obtained in any of the following ways: (a) directly obtaining the forward power of the three-terminal stacked device at the target voltage point measured by the electrical measurement system during the forward scan and the reverse power of the three-terminal stacked device at the target voltage point measured during the reverse scan; (b) obtaining the power of the top and bottom cells at the target voltage point measured by the electrical measurement system during the forward scan, and obtaining the forward power of the three-terminal stacked device at the target voltage point based on the power of the top and bottom cells at the target voltage point during the forward scan; obtaining the power of the top and bottom cells at the target voltage point measured by the electrical measurement system during the reverse scan, and obtaining the reverse power of the three-terminal stacked device at the target voltage point based on the power of the top and bottom cells at the target voltage point during the reverse scan. (c) Obtain the current parameters of the top and bottom cells at the target voltage point as measured by the electrical measurement system during the forward scan (exemplarily, these parameters can be obtained from the forward scan response current parameters), and obtain the forward power of the three-terminal stacked device at the target voltage point based on the current parameters of the top and bottom cells at the target voltage point and the target voltage point during the forward scan; Obtain the current parameters of the top and bottom cells at the target voltage point as measured by the electrical measurement system during the reverse scan (exemplarily, these parameters can be obtained from the reverse scan response current parameters), and obtain the reverse power of the three-terminal stacked device at the target voltage point based on the current parameters of the top and bottom cells at the target voltage point and the target voltage point during the reverse scan.

[0077] Step 302: Obtain the hysteresis index based on the forward power of the three-terminal stacked device at the target voltage point during forward scanning and the reverse power of the three-terminal stacked device at the target voltage point during reverse scanning. For example, the hysteresis index is obtained by calculating the percentage of the difference between the forward and reverse power relative to the forward power.

[0078] The second method: Obtain the hysteresis index based on the IV curves of the top cell's forward and reverse scanning. Step 401: Obtain the forward scan response current parameter of the top battery when a forward scan voltage is applied to the top battery, and obtain the reverse scan response current parameter of the bottom battery when a reverse scan voltage is applied to the bottom battery.

[0079] Step 402: Obtain the forward scan curve (specifically, the forward IV curve) based on the forward scan voltage and forward scan response current parameters, and obtain the reverse scan curve (specifically, the reverse IV curve) based on the reverse scan voltage and reverse scan response current parameters.

[0080] Step 403: Obtain the hysteresis index based on the forward and reverse scan curves. For example, the hysteresis index can be obtained by calculating the area enclosed by the forward and reverse scan curves. A larger area enclosed by the forward and reverse scan curves indicates stronger ion instability of the target voltage pair, while an area close to 0 indicates excellent stability of the target voltage pair.

[0081] The above method enables cyclic voltage scanning of three-terminal stacked devices based on the target voltage, calculation and quantification of the hysteresis index, and stability verification of the target voltage based on the hysteresis data.

[0082] See Figure 5 and Figure 6 ,in, Figure 5 The flowchart illustrates a global maximum power point tracking method for a three-terminal multilayer device (TFT-TFT) in the case of cyclic voltage scanning based on a target voltage, as provided in some embodiments of this application. Figure 6 This is a flowchart illustrating another three-terminal multilayer device characterization method provided in some embodiments of this application when performing cyclic voltage scanning on a three-terminal multilayer device based on a target voltage point. Some embodiments of this application also provide a global maximum power point tracking method for a three-terminal multilayer device, which determines whether the target voltage point passes stability verification based on the hysteresis index, and may include: If the hysteresis index is less than the second preset threshold, the target voltage pair is determined to have passed the stability verification. If the hysteresis index is not less than the second preset threshold, the next target voltage pair is obtained in descending order of the collected values ​​in the current iteration. The process is then repeated until the target voltage pair passes the stability verification. Alternatively, if the hysteresis index is less than the second preset threshold, the target voltage pair is determined to have passed the stability verification; if the hysteresis index is not less than the second preset threshold, the target voltage pair is determined to have failed the stability verification.

[0083] In some embodiments of this application, when determining whether a target voltage pair has passed stability verification based on the hysteresis index, the hysteresis index can be compared with a second preset threshold. The second preset threshold is related to the calculation method of the hysteresis index; different calculation methods result in different second preset thresholds. For example, if the hysteresis index is obtained by calculating the percentage of the difference between forward and reverse power to the forward power, the second preset threshold could be 5%. If the hysteresis index is obtained by calculating the area enclosed by the forward and reverse scan curves, the second preset threshold could be 0.01V. A / cm 2 .

[0084] If the hysteresis index is less than the second preset threshold, it indicates that the power of the three-terminal stacked device at the target voltage point is less affected by the hysteresis effect, and is physically valid data. This means that the target voltage point has passed the stability test, and the steady-state power of the three-terminal stacked device at the target voltage point can be obtained. The steady-state power of the three-terminal stacked device at the target voltage point can be obtained by: controlling the electrical measurement system to apply the target voltage point to the three-terminal stacked device and obtaining the steady-state power at the target voltage point; or, the steady-state power of the three-terminal stacked device can be obtained based on the forward power of the three-terminal stacked device at the target voltage point during forward scanning and the reverse power of the three-terminal stacked device at the target voltage point during reverse scanning.

[0085] If the hysteresis index is not less than the second threshold, it indicates that the power of the three-terminal stacked device at the target voltage pair is significantly affected by the hysteresis effect. Therefore, the target voltage pair is deemed unreliable, and a new target voltage pair can be selected. Specifically, the next target voltage pair can be obtained in descending order of the collected values ​​of each candidate voltage pair in the current iteration. The process then returns to execute the step of cyclically scanning the three-terminal stacked device based on the target voltage pair and obtaining the hysteresis index based on the scanning results, until the hysteresis index corresponding to the target voltage pair is less than the second preset threshold (i.e., until the target voltage pair passes the stability verification), which means until a reliable and effective target voltage pair is selected.

[0086] The above process enables the reselection of the target voltage pair when the hysteresis index corresponding to the target voltage pair is not less than the second preset threshold, until a reliable and effective target voltage pair is selected. This ensures that a target voltage pair that has passed stability verification can be obtained in the current iteration. Then, the steady-state power of the three-terminal stacked device at the target voltage pair can be obtained, and the probabilistic proxy model in the current iteration can be updated based on the target voltage pair and its steady-state power.

[0087] Alternatively, in some embodiments of this application, only the hysteresis index corresponding to the target voltage pair with the highest collected value can be compared with the second preset threshold to determine whether the target voltage pair has passed the validity verification. Based on the stability verification result of the target voltage pair, the probabilistic proxy model in the next iteration can be determined to shorten the duration of a single iteration.

[0088] Specifically, if the hysteresis index corresponding to the target voltage pair with the highest collected value is less than the second preset threshold, then the target voltage pair can be determined to have passed the stability verification. Then, the steady-state power of the three-terminal stacked device at that target voltage pair can be obtained to update the probabilistic surrogate model in the current iteration based on the target voltage pair and its steady-state power. If the hysteresis index corresponding to the target voltage pair with the highest collected value is not less than the second preset threshold, then the target voltage pair can be determined to have failed the stability verification. In this case, the probabilistic surrogate model in the current iteration can be maintained.

[0089] This application provides a global maximum power point tracking method for a three-terminal stacked device, which, in determining whether an iteration termination condition is met, may include: The maximum expected power gain is determined based on the predicted power calculated by the probabilistic surrogate model at each candidate voltage pair in the current iteration, the uncertainty value of the predicted power, and the maximum steady-state power in the previous iterations; the candidate voltage pairs are determined based on the two-dimensional voltage range of the three-terminal stacked device. Determine whether the maximum expected power gain is less than the third preset threshold, or determine whether the number of iterations has reached the maximum number of iterations; If the maximum expected power gain is less than the third preset threshold, or if the number of iterations reaches the maximum number of iterations, then the iteration termination condition is determined to be met. If the maximum expected power gain is not less than the third preset threshold and the number of iterations has not reached the maximum number of iterations, then the iteration termination condition is determined not to be met.

[0090] In some embodiments of this application, the central control unit can determine whether the iteration termination condition is met in the following ways: Step 501: Determine the maximum expected power gain based on the predicted power calculated by the probabilistic surrogate model at each candidate voltage pair in the current iteration, the uncertainty value of the predicted power, and the maximum steady-state power in previous iterations. Each candidate voltage pair can be determined based on the two-dimensional voltage range of the three-terminal stacked device. Specifically, the two-dimensional voltage range can be discretized into a grid composed of a large number of voltage pairs, with each voltage pair in the grid being a corresponding candidate voltage pair.

[0091] The expected power gain can be calculated as follows: Z = (μ - f) max f / σ, where μ is the predicted power calculated by the probabilistic surrogate model at the candidate voltage pair point in the current iteration, σ is the uncertainty value calculated by the probabilistic surrogate model at the candidate voltage pair point in the current iteration, and f max Z represents the maximum steady-state power in the previous iteration, and Z represents the expected power gain of the candidate voltage pair in the current iteration.

[0092] Step 502: Determine whether the maximum expected power gain is less than the third preset threshold, or determine whether the number of iterations (i.e. the number of iterations already performed) has reached the maximum number of iterations.

[0093] The third preset threshold can be set based on the accuracy requirements of photovoltaic device testing. For example, the third preset threshold can be 0.5%. The maximum number of iterations can be used as an auxiliary convergence criterion (e.g., the maximum number of sampling points is 30, i.e., the maximum number of iterations is 30; or the maximum number of iterations is 50, etc., which can be set according to requirements) to avoid infinite iterations due to special cases of the model.

[0094] Step 503: If the maximum expected power gain is less than the third preset threshold, it indicates that the model has sufficiently covered the high-power region, and further sampling contributes very little to improving GMPP accuracy. In this case, the iteration termination condition can be determined to be met. Alternatively, if the number of iterations reaches the maximum number of iterations, the iteration termination condition is determined to be met to avoid infinite iterations due to special cases of the model.

[0095] Step 504: If the maximum expected power gain is not less than the third preset threshold and the number of iterations has not reached the maximum number of iterations, then the iteration termination condition is not met.

[0096] Specifically, determining whether the maximum expected power gain is less than the third preset threshold can be achieved by checking whether the maximum expected power gain in multiple consecutive iterations is consistently less than the third preset threshold. This process aims to offset random errors, improve the stability and reliability of the results, and enhance the accuracy of the judgment.

[0097] This application provides a method for global maximum power point tracking of a three-terminal stacked device, which tunes the spectrum of a solar simulator according to the target short-circuit current parameter for spectral matching, and may include: Control the LEDs in each channel of the solar simulator to emit light according to the initial operating parameters; Obtain the current output spectrum of the solar simulator, the current short-circuit current parameters of the top cell and the bottom cell in the three-terminal stacked device, and calculate the spectral matching degree between the current output spectrum and the preset spectrum, the first deviation between the current short-circuit current parameter of the top cell and the corresponding target short-circuit current parameter, and the second deviation between the current short-circuit current parameter of the bottom cell and the corresponding target short-circuit current parameter. If at least one of the spectral matching degree, the first deviation, and the second deviation fails to meet the corresponding tolerance judgment condition, the operating parameters of the LEDs in each channel are adjusted, and the LEDs in each channel are controlled to emit light according to the adjusted operating parameters until the spectral matching degree, the first deviation, and the second deviation all meet the corresponding tolerance judgment condition.

[0098] In some embodiments of this application, the central control unit can control the solar simulator to initiate an iterative spectral tuning algorithm until the measured short-circuit current parameters of the top and bottom cells in the three-terminal stacked device match the corresponding target short-circuit current parameters within a preset tolerance.

[0099] Specifically, the central control unit can control the LEDs of each channel in the solar simulator to emit light according to the initial operating parameters. At the same time, it can collect the current output spectrum in real time through the spectrometer built into the solar simulator. Furthermore, it can use the electrical measurement system to measure the current short-circuit current parameters of the top cell and the bottom cell in the three-terminal stacked device.

[0100] Accordingly, the central control unit can acquire the current output spectrum of the solar simulator, the current short-circuit current parameters of the top cell and the bottom cell in the three-terminal stacked device. It calculates the spectral matching degree between the current output spectrum and the preset spectrum. For example, the spectral matching degree can be calculated according to the IEC 60904-9 standard, and the first deviation between the current short-circuit current parameter of the top cell and the corresponding target short-circuit current parameter of the top cell, and the second deviation between the current short-circuit current parameter of the bottom cell and the corresponding target short-circuit current parameter of the bottom cell, can be calculated. Then, it determines whether the spectral matching degree, the first deviation, and the second deviation meet the corresponding tolerance judgment conditions. The tolerance judgment conditions for spectral matching can be set according to the corresponding standards, and the tolerance judgment conditions for the first deviation and the second deviation can be set according to requirements, for example, 1%.

[0101] If at least one of the spectral matching degree, the first deviation, and the second deviation fails to meet the corresponding tolerance judgment condition, the operating parameters of the LEDs in each channel of the solar simulator are adjusted. For example, the current parameter of the red LED is increased to improve the light absorption of the bottom cell (crystalline silicon), and the current parameter of the blue LED is increased to improve the light absorption of the top cell (perovskite). The process then returns to the previous steps of obtaining the current output spectrum of the solar simulator, the current short-circuit current parameter of the top cell in the three-terminal stacked device, and the current short-circuit current parameter of the bottom cell. That is, the cycle of "emitting light, acquiring, calculating, and adjusting" is repeated until the spectral matching degree, the first deviation, and the second deviation all meet the corresponding tolerance judgment condition, thereby completing the spectral matching.

[0102] The above method ensures that for perovskite / crystalline silicon three-terminal stacked devices, not only does the spectral morphology match the preset spectrum, but also the number of photons absorbed by the top and bottom sub-cells (i.e., the short-circuit current parameter) accurately corresponds to the target value under standard conditions. This method of determining the spectrum by the current parameter can more directly guarantee the electrical accuracy of the test results than simply looking at the spectral curve.

[0103] Some embodiments of this application also provide a global maximum power point tracking device for a three-terminal multilayer device. See [link to relevant documentation] Figure 7 This is a schematic diagram of the structure of a global maximum power point tracking device for a three-terminal stacked device provided in some embodiments of this application. It may include: an acquisition matching module 71, used to acquire the target short-circuit current parameters of the three-terminal stacked device under a preset spectrum, and tune the spectrum of a solar simulator according to the target short-circuit current parameters for spectral matching; the three-terminal stacked device is a perovskite / crystalline silicon three-terminal stacked device; and an acquisition construction module 72, used to control the electrical measurement system to apply the voltage contained in the initial voltage pair to the three-terminal stacked device, acquire the measurement data of the three-terminal stacked device measured by the electrical measurement system under the initial voltage pair, determine the initial power of the three-terminal stacked device corresponding to the initial voltage pair based on the measurement data, and construct a probability vector based on the initial voltage point and initial power. The system includes: a model acquisition module 73, which obtains the target voltage pair in the current iteration based on the Bayesian optimization algorithm using the probabilistic surrogate model, verifies the stability of the target voltage pair using an electrical measurement system, and obtains the steady-state power of the three-terminal stacked device at the target voltage pair when the stability verification is passed; a return module 74, which determines the probabilistic surrogate model in the next iteration based on the stability verification result, returns to execute the step of obtaining the target voltage pair in the current iteration based on the Bayesian optimization algorithm using the probabilistic surrogate model, until the iteration termination condition is met; and a GMPP determination module 75, which determines the global maximum power point of the three-terminal stacked device based on the target voltage pair that has passed the stability verification and the corresponding steady-state power.

[0104] Some embodiments of this application also provide an electronic device, which may include: a memory for storing a computer program; and a processor for executing the computer program stored in the memory to implement the steps of any of the above-described three-terminal stacked device global maximum power point tracking methods.

[0105] One embodiment of this application also provides a global maximum power point tracking system for a three-terminal multilayer device. See [link to relevant documentation]. Figure 8 This is a schematic diagram of the structure of a global maximum power point tracking system for a three-terminal multilayer device provided in some embodiments of this application. It may include a solar simulator, an electrical measurement system and a central control unit, wherein the central control unit is connected to the solar simulator and the electrical measurement system and can be used to implement the steps of any of the above-mentioned global maximum power point tracking methods for three-terminal multilayer devices.

[0106] For descriptions of relevant parts of a three-terminal multilayer device global maximum power point tracking device, electronic device, and system provided in some embodiments of this application, please refer to the detailed description of the corresponding parts in the three-terminal multilayer device global maximum power point tracking method provided in the embodiments of this application, which will not be repeated here.

[0107] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0108] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0109] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0110] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0111] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "joining," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0112] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for global maximum power point tracking of a three-terminal multilayer device, characterized in that, include: The target short-circuit current parameters of the three-terminal stacked device under a preset spectrum are obtained, and the spectrum of the solar simulator is tuned according to the target short-circuit current parameters to perform spectral matching; the three-terminal stacked device is a perovskite / crystalline silicon three-terminal stacked device. The electrical measurement system applies an initial voltage pair to the three-terminal stacked device, acquires measurement data of the three-terminal stacked device at the initial voltage pair, determines the initial power of the three-terminal stacked device at the initial voltage pair based on the measurement data, and constructs a probabilistic surrogate model based on the initial voltage point and the initial power. The target voltage pair in the current iteration is obtained based on the Bayesian optimization algorithm according to the probabilistic proxy model. The stability of the target voltage pair is verified by the electrical measurement system. When the stability verification is passed, the steady-state power of the three-terminal stacked device at the target voltage pair is obtained. Based on the stability verification results, determine the probabilistic surrogate model for the next iteration, and return to execute the step of obtaining the target voltage pair in the current iteration based on the Bayesian optimization algorithm according to the probabilistic surrogate model, until the iteration termination condition is met. The global maximum power point of the three-terminal stacked device is determined based on the target voltage pair that has passed stability verification and the corresponding steady-state power.

2. The global maximum power point tracking method for a three-terminal stacked device according to claim 1, characterized in that, Based on the stability verification results, determine the probabilistic surrogate model for the next iteration, including: If the target voltage pair passes the stability verification, the probabilistic surrogate model in the current iteration is updated based on the target voltage pair and the steady-state power to obtain the probabilistic surrogate model in the next iteration. If the target voltage pair fails the stability verification, the probabilistic surrogate model in the current iteration is maintained to obtain the probabilistic surrogate model in the next iteration.

3. The global maximum power point tracking method for a three-terminal stacked device according to claim 1, characterized in that, The target voltage pair in the current iteration is obtained based on the Bayesian optimization algorithm according to the probabilistic surrogate model, including: The target voltage pair with the highest acquisition value in the current iteration is obtained based on the probabilistic proxy model and acquisition function in the current iteration.

4. The global maximum power point tracking method for a three-terminal stacked device according to claim 3, characterized in that, The stability verification of the target voltage pair point is performed using the electrical measurement system, including: The electrical measurement system is controlled to apply the target voltage contained in the target voltage pair to the three-terminal stacked device, and the time change rate of the current parameter or power of the three-terminal stacked device under the target voltage pair is monitored. If the time change rate of the current parameter or power is less than the first preset threshold, then the target voltage point is determined to have passed the stability verification; if the time change rate of the current parameter or power is not less than the first preset threshold, then return to the step of monitoring the time change rate of the current parameter or power of the three-terminal stacked device under the target voltage point. or, If the monitoring duration is less than the first preset threshold and the monitoring duration has not reached the maximum monitoring duration, then the target voltage pair is determined to have passed the stability verification; if the time change rate of the current parameter or power is not less than the first preset threshold and the monitoring duration has reached the maximum monitoring duration, then the target voltage pair is determined to have failed the stability verification.

5. The global maximum power point tracking method for a three-terminal stacked device according to claim 3, characterized in that, The stability verification of the target voltage pair point is performed using the electrical measurement system, including: Based on the target voltage, the electrical measurement system is controlled to perform a cyclic voltage scan on the three-terminal stacked device, and the hysteresis index is obtained based on the scan results. The stability verification of the target voltage pair is determined based on the hysteresis index.

6. The global maximum power point tracking method for a three-terminal stacked device according to claim 5, characterized in that, Based on the target voltage, the electrical measurement system is controlled to perform a cyclic voltage scan on the three-terminal stacked device, and the hysteresis index is obtained based on the scan results, including: Centered on the target voltage corresponding to the top battery in the target voltage point, the electrical measurement system is controlled to apply forward scanning voltage and reverse scanning voltage to the top battery in the three-terminal stacked device; wherein, the scanning rate is greater than the preset rate, and during the scanning of the top battery, the voltage of the bottom battery in the three-terminal stacked device is maintained at the corresponding target voltage or fluctuates synchronously with the voltage of the top battery; The hysteresis index is obtained by acquiring the forward power of the three-terminal stacked device at the target voltage point during forward scanning and the reverse power of the three-terminal stacked device at the target voltage point during reverse scanning, and by acquiring the forward scanning response current parameters of the top battery during forward scanning and the reverse scanning response current parameters of the top battery during reverse scanning. The forward scanning curve is obtained based on the forward scanning voltage and the forward scanning response current parameters, and the reverse scanning curve is obtained based on the reverse scanning voltage and the reverse scanning response current parameters. The hysteresis index is obtained based on the forward scanning curve and the reverse scanning curve.

7. The global maximum power point tracking method for a three-terminal stacked device according to claim 5 or 6, characterized in that, Determining whether the target voltage pair passes the stability verification based on the hysteresis index includes: If the hysteresis index is less than the second preset threshold, the target voltage pair is determined to have passed the stability verification; if the hysteresis index is not less than the second preset threshold, the next target voltage pair is obtained in descending order of the collected values ​​in the current iteration, and the process of controlling the electrical measurement system to perform cyclic voltage scanning on the three-terminal stacked device based on the target voltage pair and obtaining the hysteresis index based on the scanning results is repeated until the target voltage pair passes the stability verification. or, If the hysteresis index is less than the second preset threshold, the target voltage pair is determined to have passed the stability verification; if the hysteresis index is not less than the second preset threshold, the target voltage pair is determined to have failed the stability verification.

8. The global maximum power point tracking method for a three-terminal stacked device according to claim 1, characterized in that, Determine whether the iteration termination condition is met, including: The maximum expected power gain is determined based on the predicted power calculated by the probabilistic proxy model at each candidate voltage pair in the current iteration, the uncertainty value of the predicted power, and the maximum steady-state power in the historical iterations; the candidate voltage pairs are determined based on the two-dimensional voltage range of the three-terminal stacked device. Determine whether the maximum expected power gain is less than a third preset threshold, or determine whether the number of iterations has reached the maximum number of iterations; If the maximum expected power gain is less than the third preset threshold, or if the number of iterations reaches the maximum number of iterations, then the iteration termination condition is determined to be met. If the maximum expected power gain is not less than the third preset threshold and the number of iterations has not reached the maximum number of iterations, then it is determined that the iteration termination condition is not met.

9. The global maximum power point tracking method for a three-terminal stacked device according to claim 1, characterized in that, The spectrum of the solar simulator is tuned according to the target short-circuit current parameters for spectral matching, including: The LEDs in each channel of the solar simulator are controlled to emit light according to the initial operating parameters; The current output spectrum of the solar simulator, the current short-circuit current parameters of the top cell and the bottom cell in the three-terminal stacked device are obtained. The spectral matching degree between the current output spectrum and the preset spectrum, the first deviation between the current short-circuit current parameter of the top cell and the corresponding target short-circuit current parameter, and the second deviation between the current short-circuit current parameter of the bottom cell and the corresponding target short-circuit current parameter are calculated. If at least one of the spectral matching degree, the first deviation, and the second deviation fails to meet the corresponding tolerance judgment condition, the operating parameters of the LEDs in each channel are adjusted, and the LEDs in each channel are controlled to emit light according to the adjusted operating parameters until the spectral matching degree, the first deviation, and the second deviation all meet the corresponding tolerance judgment condition.

10. A global maximum power point tracking system for a three-terminal multilayer device, characterized in that, It includes a solar simulator, an electrical measurement system, and a central control unit, among which: The central control unit is connected to the solar simulator and the electrical measurement system, and is used to implement the steps of the global maximum power point tracking method for three-terminal stacked devices as described in any one of claims 1 to 9.

Images