A method of laser enhanced contact repair for a photovoltaic cell and related apparatus

Abstract

The application provides a laser enhanced contact repair method of a photovoltaic cell and a related device, which is applied to a controller in a photovoltaic cell repair system, and the method comprises the following steps: obtaining first resistance detection data of an electrode area of the photovoltaic cell and manufacturing process data of an electrode of the photovoltaic cell; inputting the first resistance detection data and the manufacturing process data into a preset laser enhanced contact optimization model to obtain target repair data corresponding to the electrode area; controlling a laser enhanced contact optimization repair device to repair the electrode area according to the target repair data, and obtaining second resistance detection data of the electrode area after the repair; and performing repair analysis on the electrode area according to the first resistance detection data and the second resistance detection data to obtain a repair analysis result. In this way, the situation that some battery pieces are insufficiently repaired and some battery pieces are over-repaired in the manufacturing process of the photovoltaic cell can be solved, and the consistency of the battery conversion efficiency and the product yield can be improved.

Description

Technical Field

[0001] This application relates to the field of photovoltaic cell technology, and in particular to a laser-enhanced contact repair method and related apparatus for photovoltaic cells. Background Technology

[0002] In photovoltaic cell manufacturing, Laser Enhanced Contact Optimization (LECO) technology is applied before or independently of the passivation process to improve the contact performance between the electrode and the semiconductor. Currently, to repair the contact performance degradation caused by the passivation process, the main method is to use a secondary LECO repair process with fixed or empirical parameters. This involves applying the same or a limited number of preset repair parameters to all cells for LECO treatment to improve electrode performance.

[0003] However, due to inherent fluctuations in the preceding processes (such as batch-to-batch differences in slurry, sintering process fluctuations, and uneven passivation film thickness), different solar cells exhibit varying degrees of contact degradation after passivation. Furthermore, existing repair processes cannot detect changes in upstream processes and still employ fixed repair parameters, making it difficult to match the repair parameters with the actual degree of degradation. This can easily lead to situations where some solar cells are under-repaired while others are over-repaired, thereby reducing the consistency of solar cell repair efficiency and affecting product yield.

[0004] Therefore, it is urgent to solve the problem of how to improve the efficiency of electrode area repair and the yield rate of photovoltaic cells during the manufacturing process using laser contact optimization technology. Summary of the Invention

[0005] The purpose of this application is to provide a laser-enhanced contact repair method and related apparatus for photovoltaic cells, in order to solve the problem that some cells are under-repaired and some cells are over-repaired after the passivation process in the manufacturing process of photovoltaic cells, which leads to the inconsistency of cell conversion efficiency and low product yield.

[0006] To achieve the objectives of this application, the following solution is provided: In a first aspect, embodiments of this application provide a laser-enhanced contact repair method for photovoltaic cells, applied to a controller in a photovoltaic cell repair system, the method comprising: Acquire first resistance detection data of the electrode region of the photovoltaic cell; and manufacturing process data of the photovoltaic cell electrode; the resistance detection data includes: contact resistance data and contact resistance distribution data; The first resistance detection data and the manufacturing process data are input into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode area; The laser-enhanced contact optimization repair equipment is controlled to repair the electrode area according to the target repair data, and the second resistance detection data of the repaired electrode area is obtained. The electrode region is repaired and analyzed based on the first resistance detection data and the second resistance detection data to obtain the repair analysis results.

[0007] Secondly, this application provides a laser-enhanced contact repair device for photovoltaic cells, applied to a controller in a photovoltaic cell repair system, the device comprising: The detection unit is used to acquire first resistance detection data of the electrode region of the photovoltaic cell; and manufacturing process data of the photovoltaic cell electrode; the resistance detection data includes: contact resistance data and contact resistance distribution data; The calculation unit is used to input the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode area; The control unit is used to control the laser-enhanced contact optimization repair equipment to repair the electrode area according to the target repair data; The detection unit is also used to acquire second resistance detection data of the electrode area after repair; The repair unit is used to perform repair analysis on the electrode area based on the first resistance detection data and the second resistance detection data, and obtain the repair analysis results.

[0008] Thirdly, this application provides a controller, including a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for performing steps in any of the methods of the first aspect of the embodiments of this application.

[0009] Fourthly, this application provides a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program causes a computer to perform some or all of the steps described in any of the methods of the first aspect of the embodiments of this application.

[0010] Fifthly, this application provides a laser-enhanced contact repair system for photovoltaic cells, which can perform some or all of the steps described in any of the methods in the first aspect of the embodiments of this application.

[0011] By implementing the embodiments of this application, the following beneficial effects can be achieved; This application provides a laser-enhanced contact repair method and related apparatus for photovoltaic cells, applied to a controller in a photovoltaic cell repair system. First, it acquires first resistance detection data of the photovoltaic cell electrode region and manufacturing process data of the photovoltaic cell electrode. Then, it inputs the first resistance detection data and manufacturing process data into a preset laser-enhanced contact optimization model to obtain target repair parameters corresponding to the electrode region. Next, it controls the laser-enhanced contact repair equipment to repair the electrode region according to the target repair parameters and acquires second resistance detection data after repair. Finally, it performs repair analysis on the electrode region based on the first and second resistance detection data to obtain repair analysis results.

[0012] Thus, by inputting the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model, target repair data corresponding to the electrode region is obtained. The first resistance detection data, which characterizes the actual contact state of the photovoltaic cell, and the manufacturing process data, which characterizes the preceding manufacturing process, are used together as the basis for generating target repair parameters. This allows the generated target repair parameters to match the actual contact degradation degree formed under different manufacturing process conditions, reducing under-repair or over-repair phenomena caused by fixed repair parameters. At the same time, by performing repair analysis on the electrode region based on the first resistance detection data and the second resistance detection data, and obtaining repair analysis results, the repair effect corresponding to the target repair parameters can be evaluated by acquiring the repaired second resistance detection data and performing repair analysis. This helps to verify whether the repair results meet expectations, thereby improving the consistency of photovoltaic cell contact repair and enhancing the consistency of photovoltaic cell conversion efficiency and product yield. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a hardware architecture diagram of a photovoltaic cell repair system provided in an embodiment of this application; Figure 2 This is a software architecture diagram of a photovoltaic cell repair system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a controller provided in an embodiment of this application; Figure 4 This is a schematic flowchart of a laser-enhanced contact repair method for photovoltaic cells provided in an embodiment of this application; Figure 5 This is a schematic flowchart illustrating the electrode region repair analysis of a photovoltaic cell provided in an embodiment of this application; Figure 6 This is a flowchart illustrating an incremental learning method for a photovoltaic cell laser-enhanced contact repair model provided in an embodiment of this application. Figure 7 This is a schematic diagram of a closed-loop learning and optimization process for photovoltaic cell repair based on laser-enhanced contact optimization, provided in an embodiment of this application. Figure 8 This is a schematic diagram of a repair scenario for laser-enhanced contact optimization of photovoltaic cells provided in an embodiment of this application; Figure 9 This is a functional module block diagram of a laser-enhanced contact repair device for photovoltaic cells provided in an embodiment of this application. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0016] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0017] It should be understood that the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document indicates that the preceding and following related objects are in an "or" relationship. In the embodiments of this application, "multiple" refers to two or more.

[0018] In the embodiments of this application, "at least one item" or its similar expression refers to any combination of these items, including any combination of a single item or a plurality of items. "One or more" means one or more, while "multiple" means two or more. For example, "at least one item" of a, b, or c can represent the following seven cases: a, b, c; a and b; a and c; b and c; a, b, and c. Each of a, b, and c can be an element or a set containing one or more elements.

[0019] In this application, the term "connection" refers to various connection methods, such as direct connection or indirect connection, to achieve communication between devices. This application does not impose any limitations on this.

[0020] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] The following is an explanation of the relevant terms used in this application: Laser Enhanced Contact Optimization (LECO) is a technology that uses the synergistic effect of laser and bias electric field to induce the formation of three-dimensional conductive microstructures at the metal-semiconductor contact interface of photovoltaic cells, thereby achieving ultra-low loss ohmic contacts.

[0022] Contact degradation after passivation: Specifically refers to the thin film deposition and high-temperature annealing (usually >300°C) processes performed after the first LECO process in photovoltaic cell manufacturing, which cause thermal degradation of the high-quality contact microstructure formed in the early stage, manifested as a significant increase in contact resistance.

[0023] The following is combined Figure 1 and Figure 2 The system architecture of a laser-enhanced contact repair method for photovoltaic cells, as described in the embodiments of this application, is explained. Figure 1 This is a hardware architecture diagram of a photovoltaic cell repair system provided in an embodiment of this application. The photovoltaic cell repair system 100 includes a contact resistance detection device 110, a controller 120, a repairable LECO device 130, and a photovoltaic cell 140.

[0024] The contact resistance detection device 110 is used to detect the electrode contact state of the photovoltaic cell 140, acquire quantitative data of degradation state characterizing the electrode contact performance, and send the quantitative data of degradation state to the controller 120. In some embodiments, the quantitative data of degradation state may include contact resistance data before repair, contact resistance data after repair, resistance distribution data, local contact resistance change data, or other information that can characterize the electrode contact state, so as to reflect the degree of contact degradation in the current electrode area of ​​the photovoltaic cell 140. In one possible embodiment, the contact resistance detection device 110 may be an online contact resistance measurement (OCRM) device, which scans the contact resistance of multiple detection areas of the photovoltaic cell 140 to obtain the contact resistance detection results corresponding to each detection area and generate corresponding quantitative data of degradation state, so that the controller 120 can perform repair analysis and repair strategy calculation.

[0025] The controller 120 receives degradation state quantification data sent by the contact resistance detection device 110, analyzes and processes the degradation state quantification data, generates corresponding dynamic repair parameters based on the analysis results, and controls the restorative LECO device 130 to perform laser-enhanced contact repair operations. The controller 120 may include a model calculation module 121, an incremental learning module 122, and a system control module 123. The model calculation module 121 calculates the target repair data corresponding to the current photovoltaic cell 140 based on the degradation state quantification data and generates dynamic repair parameters. The incremental learning module 122 continuously updates the laser-enhanced contact optimization model based on feedback data after repair completion, improving the model's repair prediction capability under different manufacturing process conditions and different degradation states. In one possible embodiment, the incremental learning module 122 continuously collects repair feedback data after repair completion, constructs repair experience samples based on manufacturing process data, repair parameters, repair analysis results, and repaired contact performance data, writes the repair experience samples into a repair strategy sample library, and performs incremental learning on the laser-enhanced contact optimization model. As the number of repair experience samples increases, the incremental learning module 122 continuously updates the mapping relationship between parameters and performance within the model. This enables the model to learn the repair patterns corresponding to different manufacturing process backgrounds, different contact degradation states, and different repair strategies, thereby continuously improving the accuracy of model parameter predictions and the repair effect. Furthermore, the incremental learning module 122 can also analyze the difference between the model's predicted results and the actual repair effect by combining the IV test efficiency, EL test results, or contact resistance recovery after repair, and use the analysis results to continuously optimize the model parameters, realizing the model's self-learning and continuous optimization capabilities.

[0026] The restorative LECO device 130 receives dynamic repair parameters output by the controller 120 and performs laser-enhanced contact repair on the photovoltaic cell 140 according to these parameters. In one possible embodiment, the restorative LECO device 130 controls process parameters such as laser output power, scanning speed, bias voltage, bias waveform, and spot distribution according to the dynamic repair parameters to perform localized laser-enhanced contact repair on the electrode area of ​​the photovoltaic cell 140. This improves the ohmic contact performance between the metal electrode and the silicon substrate, reduces electrode contact resistance, and increases the photovoltaic cell conversion efficiency.

[0027] As can be seen, the contact resistance detection device 110 is responsible for acquiring quantitative data on the degradation state of the photovoltaic cell 140. The controller 120 completes the repair strategy calculation and continuous model optimization based on the quantitative data on the degradation state. The repairable LECO device 130 completes laser-enhanced contact repair based on dynamic repair parameters. A closed-loop working mechanism of "detection, analysis, repair, feedback, learning, and re-optimization" is formed among the various components, enabling the photovoltaic cell repair system 100 to adaptively generate the optimal laser-enhanced contact repair parameters based on different manufacturing process backgrounds and different electrode contact degradation states. By continuously accumulating repair experience, the model performance is continuously optimized, thereby effectively improving the photovoltaic cell repair efficiency, repair consistency, and conversion efficiency, and further enhancing the automation level of the photovoltaic cell intelligent repair system and the product yield in the mass production process.

[0028] Please continue reading. Figure 2 , Figure 2This is a software architecture diagram of a photovoltaic cell repair system provided in an embodiment of this application. As can be seen, the online sensing layer 210 includes a contact resistance detection module, which acquires contact resistance detection data of the photovoltaic cell electrode area online and generates quantitative data of degradation state characterizing the electrode contact state based on the detection results. The quantitative data of degradation state may include contact resistance magnitude, contact resistance distribution, local contact performance changes, and other detection information that can reflect the degree of electrode contact degradation. The intelligent decision layer 220 includes an adaptive repair algorithm and a parameter-performance mapping model. The adaptive repair algorithm receives the quantitative data of degradation state sent by the online sensing layer 210 and, combined with the manufacturing process background corresponding to the current photovoltaic cell, performs a comprehensive analysis of the degradation state of the current electrode area. The parameter-performance mapping model establishes a mapping relationship between the electrode contact degradation state, the manufacturing process background, and the repair performance, and dynamically calculates the corresponding repair parameters based on the optimization objectives. The manufacturing process background may include process information such as slurry type, sintering peak temperature, and passivation process number. The optimization objectives may include at least one of maximizing the conversion efficiency after repair, minimizing the efficiency dispersion, achieving optimal repair performance while meeting preset EL yield requirements, or maximizing the repair success rate. The adaptive repair algorithm outputs the optimal dynamic repair parameters for the current photovoltaic cell based on the parameter-performance mapping model. The precise execution layer 230 receives the dynamic repair parameters output by the intelligent decision layer 220 and generates corresponding dynamic repair parameter instructions to drive the restorative LECO device to complete laser-enhanced contact repair. In this embodiment, the dynamic repair parameter instructions may include one or more repair control parameters such as laser power, scanning speed, bias voltage, bias waveform, and spot shape modulation parameters, enabling the restorative LECO device to perform differentiated repairs for electrode areas with different degrees of degradation, improving electrode contact recovery and repair consistency. The closed-loop learning and optimization layer 240 includes an incremental learning module. The incremental learning module collects repair feedback data after repair completion and constructs repair experience samples based on the repair feedback data, continuously updating the parameter-performance mapping model. The repair feedback data may include contact resistance detection results, repair analysis results, IV test data, EL test data, and corresponding target repair parameters. The incremental learning module continuously optimizes the parameter-performance mapping model based on newly added repair experience samples, enabling the model to continuously learn the repair rules corresponding to different manufacturing process backgrounds, different contact degradation states, and different repair strategies. The updated model is then fed back to the intelligent decision layer 220 to achieve continuous optimization of model parameters.

[0029] In one optional embodiment, after the online sensing layer 210 acquires the quantified degradation state data of the current photovoltaic cell electrode area, it inputs the quantified degradation state data into the intelligent decision layer 220. The intelligent decision layer 220 calls the parameter-performance mapping model through an adaptive repair algorithm, using the quantified degradation state data and manufacturing process background as joint inputs, and calculates the dynamic repair parameters corresponding to the current photovoltaic cell according to the preset optimization target. The precise execution layer 230 generates dynamic repair parameter instructions based on the dynamic repair parameters and controls the repairable LECO device to complete laser-enhanced contact repair. After the repair is completed, the online sensing layer 210 collects the contact resistance detection data after repair again, and together with the repair parameters and repair performance, it constitutes repair feedback data, which is sent to the closed-loop learning and optimization layer 240. The closed-loop learning and optimization layer 240 uses the newly added repair experience samples to perform incremental learning, updates the parameter-performance mapping model, and feeds back the updated model to the intelligent decision layer 220 for subsequent photovoltaic cell repair tasks to call, thereby forming a complete software closed-loop optimization process. Compared to traditional methods that use fixed LECO repair parameters or rely on manual experience for parameter configuration, this embodiment can dynamically generate optimal repair parameters based on different manufacturing process backgrounds and different electrode degradation states. By continuously optimizing the model's predictive ability through repair feedback, it can improve the accuracy, stability, and consistency of laser-enhanced contact repair, as well as the consistency of mass production, thereby further improving the repair efficiency and conversion efficiency of photovoltaic cells.

[0030] The following is combined Figure 3 The controller in the embodiments of this application will be described. Figure 3 This is a schematic diagram of the structure of a controller provided in an embodiment of this application, such as... Figure 3 As shown, the controller 120 includes a processor 310, a memory 320, a communication interface 330, and one or more programs 321. The processor 310 is communicatively connected to the memory 320 and the communication interface 330 via an internal communication bus.

[0031] The one or more programs 321 are stored in the memory 320 and configured to be executed by the processor 310. The one or more programs 321 include instructions for performing any step in the above method embodiments.

[0032] The processor 310 can be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, cells, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The communication unit can be a communication interface, transceiver, transceiver circuitry, etc., and the storage unit can be a memory.

[0033] The memory 320 can be volatile memory or non-volatile memory, or it can include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0034] It is understood that the controller 120 may include more or fewer structural elements than those shown in the block diagram above, such as a power module, physical buttons, a Wi-Fi module, a speaker, a Bluetooth module, sensors, a display module, etc., without limitation. It is understood that the controller 120 may be equipped with... Figure 1 The architecture of the photovoltaic cell repair system is described.

[0035] After understanding the software and hardware architecture of this application, the following will be combined with... Figure 4 This application describes a laser-enhanced contact repair method for photovoltaic cells. Figure 4 This is a flowchart illustrating a laser-enhanced contact repair method for photovoltaic cells provided in an embodiment of this application. The method is applied to a controller in a photovoltaic cell repair system and specifically includes the following steps: Step S401: Obtain first resistance detection data of the electrode region of the photovoltaic cell; and manufacturing process data of the photovoltaic cell electrode; the resistance detection data includes: contact resistance data and contact resistance distribution data.

[0036] The first resistance detection data refers to the detection data characterizing the contact performance state of the electrode area, collected by the Online Contact Resistance Monitoring (OCRM) module after the photovoltaic cell completes the passivation process. This includes contact resistance data and contact resistance distribution data. The contact resistance data represents the magnitude of the contact resistance between the metal electrode and the semiconductor in the electrode area, while the contact resistance distribution data represents the spatial distribution of contact resistance at different locations in the electrode area, characterizing the uniformity of the contact performance in the electrode area. The manufacturing process data refers to the process parameter data generated during the manufacturing process of the photovoltaic cell. This may include at least one of the following: slurry type, slurry batch, sintering peak temperature, sintering curve parameters, passivation process number, passivation film thickness, annealing temperature, and annealing time. These parameters are not limited here and are used to characterize the manufacturing process information corresponding to the current photovoltaic cell.

[0037] Specifically, firstly, after the photovoltaic cell completes the passivation process, the controller controls the online contact resistance monitoring module to perform online detection on the electrode area of ​​the photovoltaic cell. It collects contact resistance data corresponding to multiple detection locations according to a preset sampling method, and simultaneously scans the contact resistance changes between each detection location in the electrode area, generating corresponding contact resistance distribution data. The preset sampling method can be a 10Hz frequency sampling. Next, the contact resistance data and contact resistance distribution data are correlated to generate first resistance detection data corresponding to the electrode area, fully characterizing the contact performance state of the current electrode area. Then, the controller reads the manufacturing process data corresponding to the current photovoltaic cell from the Manufacturing Execution System (MES) or the production database, and completes the correlation between the manufacturing process data and the first resistance detection data based on the cell's unique identification information. Finally, the acquired first resistance detection data and manufacturing process data are organized and stored as input data for the subsequent laser-enhanced contact optimization model, for use in the generation of subsequent target repair data.

[0038] It should be noted that, due to the passivation process, the degree of contact degradation in photovoltaic cells varies between different manufacturing batches or under different manufacturing processes. Contact resistance data alone cannot fully reflect the process factors leading to changes in contact performance. Therefore, by simultaneously acquiring both the initial resistance detection data and manufacturing process data, the initial resistance detection data accurately reflects the actual contact performance state of the current electrode area. Furthermore, by combining the manufacturing process data, the subsequent laser-enhanced contact optimization model can comprehensively consider the electrode contact state and manufacturing process factors to generate targeted repair data. This avoids the problems of insufficient or over-repair caused by using fixed repair parameters, thereby improving the targeting, stability, and consistency of laser-enhanced contact repair.

[0039] Step S402: Input the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode area.

[0040] The laser-enhanced contact optimization model is a pre-built intelligent decision-making model used to generate laser-enhanced contact repair parameters for photovoltaic cells. It establishes a mapping relationship between resistance detection data, manufacturing process data, and laser-enhanced contact repair parameters based on historical repair samples, manufacturing process data, and corresponding repair effects. Target repair data refers to parameter data generated based on the current contact performance state of the electrode area and the manufacturing process background, used to control the laser-enhanced contact optimization repair equipment to perform repair processing. This data can include laser power, laser scanning speed, bias voltage magnitude, bias voltage waveform, spot size, spot shape, and scanning trajectory. It guides the laser-enhanced contact optimization repair equipment to perform targeted repair processing on the photovoltaic cell electrode area.

[0041] Specifically, firstly, the controller preprocesses the first resistance detection data to extract detection features characterizing the current contact performance state of the electrode area. Simultaneously, it reads the manufacturing process data of the corresponding photovoltaic cell and extracts process features characterizing the manufacturing process background. Next, the detection features and process features are used as model input data and fed into a preset laser-enhanced contact optimization model. The laser-enhanced contact optimization model comprehensively analyzes the contact performance state of the current electrode area and the manufacturing process background, and determines a repair strategy matching the current electrode area based on the data mapping relationship established within the model. This laser-enhanced contact optimization model can be a machine learning model or an advanced deep learning model, such as a model based on a convolutional neural network. Then, the laser-enhanced contact optimization model calculates the corresponding laser-enhanced contact repair parameters according to the determined repair strategy and generates target repair data for the electrode area. Finally, the target repair data is output and sent to the laser-enhanced contact optimization repair device to repair the electrode area.

[0042] It can be seen that by inputting the first resistance detection data and manufacturing process data into the laser-enhanced contact optimization model, the model can dynamically generate target repair data that matches the current electrode area by comprehensively considering the contact performance status of the current electrode area and the manufacturing process background of the photovoltaic cell. Instead of using fixed repair parameters for uniform repair, it can implement differentiated repair for different photovoltaic cells or different electrode areas, effectively improving the targeting and accuracy of laser-enhanced contact repair, avoiding the decline in electrode contact performance due to insufficient or over-repair, and further improving the repair quality and product consistency of photovoltaic cells.

[0043] In an optional embodiment, the step of inputting the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model to obtain target repair data corresponding to the electrode region includes: generating contact state information corresponding to the electrode region based on the first resistance detection data; generating process influence information corresponding to the electrode region based on the manufacturing process data; determining repair constraints for the electrode region based on the contact state information and the process influence information; determining repair data corresponding to the electrode region based on the repair constraints, the contact state information, and the process influence information to obtain first repair data; and inputting the first repair data into the laser-enhanced contact optimization model to obtain the target repair data.

[0044] The contact state information is feature information extracted from the first resistance detection data to characterize the current contact performance state of the electrode area, reflecting contact characteristics such as contact resistance level, contact uniformity, and contact degradation degree. The process influence information is process feature information extracted from the manufacturing process data to characterize the degree of influence of the manufacturing process on the electrode contact performance, which may include at least one of the following: sintering process influence features, slurry process influence features, passivation process influence features, and annealing process influence features. The repair constraint conditions are repair control conditions determined comprehensively based on the contact state information and the process influence information, used to limit the value range of repair parameters and the execution requirements of the repair strategy during the laser-enhanced contact repair process. The first repair data is the initial repair parameter data generated based on the repair constraint conditions, used as input to the laser-enhanced contact optimization model, and further used to generate target repair data applicable to the current electrode area.

[0045] Specifically, firstly, feature extraction is performed on the first resistance detection data to analyze the contact performance state reflected by the contact resistance data and contact resistance distribution data, generating corresponding contact state information. Simultaneously, manufacturing process data is parsed to extract manufacturing process features related to electrode contact performance, generating corresponding process influence information. Next, the contact state information and process influence information are comprehensively analyzed to identify limiting factors affecting the repair effect of the current electrode area. Based on these limiting factors, repair constraints corresponding to the current electrode area are determined, ensuring that subsequent repair parameters meet the process adaptation requirements of the current electrode area. Then, based on the repair constraints and combined with the contact state information and process influence information, first repair data matching the current electrode area is determined, ensuring that the first repair data meets the current contact performance recovery requirements. Finally, the first repair data is input into a preset laser-enhanced contact optimization model. The laser-enhanced contact optimization model, combined with parameter mapping relationships established from historical repair samples, further optimizes the first repair data, generating target repair data corresponding to the electrode area, which serves as the control basis for subsequent repair processing by the laser-enhanced contact repair equipment.

[0046] As can be seen, by adding contact state analysis, process influence analysis, and repair constraint control process, this embodiment enables the input data of the laser-enhanced contact optimization model to more comprehensively reflect the current contact performance state and manufacturing process background of the photovoltaic cell. This avoids the problem of insufficient targeting of the repair strategy due to directly using the original detection data to generate repair parameters, improves the accuracy of target repair data generation and the adaptability of laser-enhanced contact repair, and thus further improves the stability and repair consistency of photovoltaic cell electrode area contact repair.

[0047] In an optional embodiment, generating contact state information corresponding to the electrode region based on the first resistance detection data includes: determining multiple resistance distribution relationships corresponding to multiple detection areas within the electrode region based on the contact resistance distribution data; determining contact resistance values ​​corresponding to the multiple detection positions based on the contact resistance data to obtain multiple contact resistance values; constructing a contact state distribution spectrum corresponding to the electrode region based on the multiple resistance distribution relationships and the multiple contact resistance values; and generating the contact state information corresponding to the electrode region based on the contact state distribution spectrum.

[0048] Multiple detection areas refer to multiple local detection units within an electrode area divided according to preset detection rules. Each detection area corresponds to at least one detection position. The predicted detection rules can be determined based on one or more of the following: the geometric dimensions of the electrode area, the preset grid division method, the detection accuracy requirements, the distribution of process features, and the distribution of historical defects. Specifically, the electrode area can be divided into multiple regular grids according to preset dimensions as multiple detection areas. Alternatively, it can be adaptively divided based on the electrode edge area, center area, welding area, or other critical areas prone to defects, so that different detection areas correspond to different detection densities or the number of detection positions. The resistance distribution relationship represents the spatial distribution relationship of contact resistance between each detection area and the resistance change between adjacent detection areas. The contact resistance value is the actual contact resistance value measured at each detection position to reflect the contact performance of the corresponding detection position. The contact state distribution spectrum is a comprehensive state feature constructed based on the resistance distribution relationship and multiple contact resistance values ​​corresponding to multiple detection areas, used to describe the overall contact performance distribution of the electrode area.

[0049] Specifically, firstly, based on the contact resistance distribution data in the first resistance detection data, the electrode area is divided into regions, multiple detection areas are determined within the electrode area, and the distribution pattern and trend of contact resistance between each detection area are analyzed to obtain multiple resistance distribution relationships corresponding to multiple detection areas; nextly, the contact resistance data in the first resistance detection data is read, the contact resistance value corresponding to each detection position is extracted, and the correspondence between each detection position and its corresponding detection area is established to obtain multiple contact resistance values; then, the multiple resistance distribution relationships and multiple contact resistance values ​​are fused to establish the correlation between the spatial distribution characteristics of each detection area and the corresponding contact resistance value, and the contact state distribution spectrum corresponding to the electrode area is constructed based on the correlation to reflect the overall contact performance distribution state of the electrode area; finally, the state features that can characterize the contact performance of the electrode area are extracted based on the contact state distribution spectrum, and the contact state information corresponding to the electrode area is generated.

[0050] It can be seen that this embodiment does not directly use the contact resistance value as the basis for characterizing the contact state of the electrode area. Instead, it comprehensively considers the resistance distribution relationship between each detection area and the contact resistance value at the corresponding detection position to construct a contact state distribution spectrum that can reflect the overall contact performance distribution characteristics of the electrode area. Furthermore, it generates contact state information, which not only reflects the magnitude of the contact resistance but also reflects the spatial distribution characteristics and regional differences of the contact resistance. This improves the completeness and accuracy of characterizing the contact performance state of the electrode area, thereby enhancing the pertinence and repair effect of the laser-enhanced contact repair strategy.

[0051] In an optional embodiment, the step of determining the repair data corresponding to the electrode region based on the contact state information and the process influence information according to the repair constraints to obtain the first repair data includes: inputting the contact state information and the process influence information into a preset repair data and performance mapping model to obtain a first repair mapping relationship between the repair data corresponding to the electrode region and the repair performance; filtering the mapping relationship corresponding to the electrode region from the first repair mapping relationship according to the repair constraints to obtain a target repair mapping relationship; and determining the first repair data corresponding to the electrode region according to the target repair mapping relationship.

[0052] The repair data and performance mapping model is a predictive model established based on historical repair samples to describe the correspondence between repair data and repair performance. It predicts the repair effect by learning the correlation between repair data and final repair performance under different contact states and manufacturing processes. The first repair mapping relationship consists of multiple candidate mapping relationships output by the repair data and performance mapping model, representing the expected repair performance corresponding to different repair data. The target repair mapping relationship is the mapping relationship selected from multiple candidate mapping relationships based on repair constraints, which meets the repair requirements of the current electrode area. The first repair data is the initial repair parameter data determined based on the target repair mapping relationship, used as input data for further optimization of the laser-enhanced contact optimization model.

[0053] Specifically, firstly, contact state information and process influence information are input into a pre-defined repair data and performance mapping model as model input features. Next, based on the correlation between repair data and repair performance established in historical repair samples, the repair data and performance mapping model performs predictive analysis on the current electrode region, outputting multiple first repair mapping relationships between repair data and corresponding repair performance. Each repair mapping relationship includes a set of candidate repair data and a corresponding predicted repair performance. Then, based on the repair constraints corresponding to the current electrode region, the multiple first repair mapping relationships are filtered, eliminating candidate mapping relationships that do not meet the repair constraints, and retaining mapping relationships that meet the repair constraints and have better predicted repair performance, thus obtaining the target repair mapping relationship. Finally, the corresponding repair data is extracted based on the target repair mapping relationship, and the first repair data corresponding to the electrode region is determined.

[0054] It can be seen that by introducing a mapping mechanism between repair data and repair performance, the first repair data can not only meet the contact state requirements of the current electrode area, but also take into account manufacturing process factors and repair performance targets. This avoids the problem of unstable repair effect caused by directly using empirical parameters, improves the scientificity and accuracy of the first repair data determination process, and provides a more reliable input basis for the subsequent generation of target repair data by the laser-enhanced contact optimization model. This further improves the stability, targeting and repair quality of laser-enhanced contact repair of photovoltaic cells.

[0055] Step S403: Control the laser-enhanced contact optimization repair device to repair the electrode area according to the target repair data, and obtain the second resistance detection data of the repaired electrode area.

[0056] The laser-enhanced contact optimization and repair equipment is used to perform LECO repair treatment on the electrode area of ​​photovoltaic cells. It adjusts the corresponding repair parameters based on the target repair data sent by the controller and utilizes the synergistic effect of laser and bias electric field to improve the contact performance between the metal electrode and the semiconductor. The second resistance detection data refers to the contact performance detection data re-collected after the electrode area has undergone laser-enhanced contact optimization and repair, including repaired contact resistance data and contact resistance distribution data, characterizing the contact performance state of the repaired electrode area.

[0057] Specifically, firstly, the controller converts the generated target repair data into control commands that the laser-enhanced contact optimization repair device can recognize, and sends them to the device. Next, the device automatically adjusts the corresponding repair parameters based on the control commands, including laser power, laser scanning speed, bias voltage, bias waveform, spot size, and scanning path. It then performs laser-enhanced contact repair on the electrode area according to the repair strategy corresponding to the target repair data, resulting in an optimized metal-semiconductor contact structure. After completing the laser-enhanced contact repair, the controller again controls the OCRM to detect the repaired electrode area, acquiring the repaired contact resistance data and contact resistance distribution data, and generating second resistance detection data. Finally, the controller establishes a correspondence between the second resistance detection data and the first resistance detection data, storing the second resistance detection data in the controller for subsequent analysis of the electrode area repair effect.

[0058] Step S404: Perform repair analysis on the electrode region based on the first resistance detection data and the second resistance detection data to obtain the repair analysis result.

[0059] The repair analysis refers to the process of analyzing the improvement in contact performance of the photovoltaic cell electrode area after laser-enhanced contact repair based on resistance test data before and after repair. This analysis is used to evaluate the effectiveness of the current laser-enhanced contact repair strategy. The repair analysis results characterize the recovery of contact performance in the electrode area after laser-enhanced contact repair and may include at least one of the following: repair effect evaluation information, contact performance improvement information, and repair status evaluation information. The preset repair analysis rules refer to the repair quality evaluation rules established in advance based on the electrode repair process. These rules are used to comprehensively evaluate the repair effect corresponding to the target repair data based on the repair analysis information.

[0060] Specifically, firstly, the controller reads the first resistance detection data and the second resistance detection data, and establishes a correspondence between them to ensure that the detection data before and after repair originate from the same electrode area. Next, the first and second resistance detection data are compared and analyzed to determine the changes in the contact performance of the electrode area before and after repair, and corresponding repair analysis information is generated based on the changes in contact resistance and its distribution. Then, according to preset repair analysis rules, the repair analysis information is comprehensively evaluated to determine the repair effect corresponding to the current target repair data. The repair analysis rules comprehensively analyze various repair quality indicators based on the defect residue in the repair area, the size recovery of the repair area, electrode continuity, edge integrity, conductivity recovery, and changes in the detection data before and after repair, and combine the weights or quality thresholds corresponding to each indicator to determine the repair effect corresponding to the target repair data. Finally, the repair analysis result corresponding to the electrode area is generated based on the repair effect and stored in the controller for use in subsequent repair feedback data generation and laser-enhanced contact optimization model updates.

[0061] For easier understanding, please refer to Figure 5 , Figure 5 This is a schematic flowchart illustrating the electrode region repair analysis of a photovoltaic cell according to an embodiment of this application. As can be seen, firstly, after laser-enhanced contact repair is completed, the corresponding electrode region of the photovoltaic cell is subjected to pre-repair and post-repair detections to obtain first resistance detection data and second resistance detection data. A correspondence is then established between the first and second resistance detection data, ensuring that the detection data before and after repair correspond to the same electrode region. Next, the first and second resistance detection data are compared and analyzed (e.g., ...). Figure 5As shown (solid dots represent before repair, hollow dots represent after repair), by analyzing the contact resistance magnitude, contact resistance distribution, and local resistance changes before and after repair, the degree of improvement in electrode area contact performance is determined, and corresponding repair analysis information is generated. This repair analysis information can include multiple evaluation indicators corresponding to defect residue, dimensional recovery, continuity, edge integrity, conductivity, and distribution improvement. Then, the above repair analysis information is comprehensively evaluated according to preset repair analysis rules. The comprehensive evaluation is determined through (rules / weights); the repair analysis rules set corresponding evaluation weights or quality thresholds based on the importance of each repair quality indicator, and comprehensively calculate each evaluation indicator to determine the repair effect corresponding to the current laser-enhanced contact repair strategy. Further, a corresponding repair effect evaluation is generated based on the comprehensive evaluation results, forming the final repair analysis result. Finally, the repair analysis results are stored in a storage device so that subsequent repair feedback data, repair experience samples, and repair strategy sample libraries can be constructed based on the repair analysis results. These will serve as important input data for incremental learning of the laser-enhanced contact optimization model, enabling continuous optimization of the repair strategy.

[0062] It can be seen that by comparing and analyzing the first and second resistance detection data before and after repair, an objective evaluation of the laser-enhanced contact repair effect in the photovoltaic cell electrode area is achieved. This allows for quantitative analysis of the repair effect based on actual detection data, rather than relying solely on empirical judgments based on preset repair parameters, thereby improving the accuracy and reliability of the repair effect evaluation. Simultaneously, the obtained repair analysis results can serve as an important basis for subsequent repair feedback and continuous optimization of the laser-enhanced contact optimization model, forming a data closed loop of "detection, repair, and evaluation" in the entire laser-enhanced contact repair process. This provides reliable data support for subsequent model self-learning and repair strategy optimization, improving the intelligence level and repair consistency of photovoltaic cell laser-enhanced contact repair.

[0063] In an optional embodiment, the step of performing repair analysis on the electrode region based on the first resistance detection data and the second resistance detection data to obtain repair analysis results includes: determining the corresponding detection change data of the electrode region based on the first resistance detection data and the second resistance detection data; extracting detection change parameters corresponding to multiple detection areas in the electrode region from the detection change data to obtain multiple detection change parameters; determining the region repair response parameters corresponding to the multiple detection areas based on the multiple detection change parameters to obtain multiple repair response parameters; the multiple repair response parameters are used to characterize the contact performance recovery parameters of the electrode region after laser-enhanced repair optimization, and the multiple repair response parameters include at least one of: contact resistance improvement magnitude, contact resistance distribution balance parameter, and contact recovery trend parameter; determining the region repair state parameters corresponding to the multiple detection areas based on the multiple repair response parameters to obtain multiple repair state parameters; and generating the corresponding repair analysis results for the electrode region based on the multiple repair state parameters.

[0064] The detection change data is obtained by comparing the first resistance detection data and the second resistance detection data, and is used to characterize the changes in contact performance of the electrode area before and after repair. The detection change parameters are change feature parameters extracted from the detection change data to reflect the degree of change in contact resistance and its distribution characteristics in each detection area before and after repair. The repair response parameters are repair effect evaluation parameters calculated based on the detection change parameters, characterizing the degree of recovery of contact performance of the electrode area after laser-enhanced contact repair. These parameters may include the magnitude of contact resistance improvement, contact resistance distribution balancing parameters, and contact recovery trend parameters. The repair state parameters are state evaluation parameters comprehensively determined based on the repair response parameters, used to reflect the repair state and repair quality of each detection area, and serve as the basis for generating repair analysis results. In some embodiments, the repair response parameters may also be determined in conjunction with the electrical performance test results or defect detection results of the photovoltaic cell after repair. The electrical performance test results may include at least one of conversion efficiency, fill factor, open-circuit voltage, and short-circuit current, and the defect detection results may include electroluminescence detection results.

[0065] Specifically, firstly, the controller reads the first resistance detection data acquired before repair and the second resistance detection data acquired after repair, and establishes a correspondence between the detection data before and after repair according to the detection area. By calculating the changes in contact resistance and contact resistance distribution in each detection area, it generates detection change data corresponding to the electrode area. Secondly, the controller extracts features from the detection change data, extracting change features such as the amount of contact resistance change, the degree of change in resistance distribution, and the trend of regional change corresponding to multiple detection areas, generating multiple detection change parameters. Then, the controller calculates the repair response parameters corresponding to each detection area based on the multiple detection change parameters. Among them, the contact resistance improvement magnitude reflects the degree of contact resistance reduction, the contact resistance distribution equalization parameter reflects the uniformity of the contact resistance distribution in each detection area after repair, and the contact recovery trend parameter reflects the development trend of contact performance recovery in the electrode area. Subsequently, the controller comprehensively evaluates each detection area based on the multiple repair response parameters, determines the repair state parameters corresponding to each detection area, and characterizes whether the repair effect of each detection area meets the expected requirements. Finally, the controller integrates the repair state parameters corresponding to multiple detection areas to generate the repair analysis results corresponding to the electrode area, and outputs the repair analysis results to the subsequent repair feedback module for constructing repair experience samples and updating the laser-enhanced contact optimization model.

[0066] As can be seen, this embodiment analyzes the recovery of contact performance in the electrode region after repair by constructing detection change data, extracting detection change parameters, calculating repair response parameters, and determining repair state parameters. This allows the repair analysis results to comprehensively reflect the degree of contact performance improvement, the balance of contact resistance distribution, and the overall repair quality in each detection area. By establishing a multi-level evaluation mechanism from raw detection data to repair analysis results, not only is the objectivity and accuracy of the repair effect evaluation improved, but it also provides richer and more reliable evaluation data for the subsequent construction of repair experience samples and the continuous learning of the laser-enhanced contact optimization model. This further improves the adaptive optimization capability and long-term repair stability of the photovoltaic cell laser-enhanced contact repair strategy.

[0067] In one optional embodiment, generating the repair analysis result corresponding to the electrode region based on the plurality of repair status parameters includes: determining the repair status type corresponding to the electrode region based on the plurality of repair status parameters, wherein the repair status type includes at least one of insufficient repair, repair fit, and over-repair; determining the repair evaluation information corresponding to the electrode region based on the repair status type; and generating the repair analysis result based on the repair evaluation information.

[0068] The repair status type is a classification result of the repair effect determined by comprehensively considering the repair status parameters corresponding to multiple detection areas, characterizing the overall repair status of the current electrode area after laser-enhanced contact repair. Insufficient repair indicates that the improvement in contact performance after repair does not meet the preset repair target; repair fit indicates that the contact performance after repair meets the preset repair target and the repair effect meets the expected requirements; over-repair indicates that the repair intensity during laser-enhanced contact repair exceeds the fit range of the current electrode area, leading to abnormal changes in local contact performance or the risk of over-repair. Repair evaluation information is an evaluation result generated based on the repair status type, used to characterize the fit and repair quality of the current repair strategy, and serves as an important basis for generating repair analysis results and optimizing subsequent repair strategies. In some embodiments, the repair evaluation information may also include repair parameter adjustment suggestions, which indicate the direction of adjustment for repair parameters during subsequent laser-enhanced contact repair, including increasing laser power, decreasing laser power, adjusting scanning speed, adjusting bias parameters, or maintaining at least one of the current repair strategies.

[0069] Specifically, firstly, the controller establishes regional repair status distribution information based on multiple repair status parameters and the corresponding positional relationships of multiple detection areas, and performs comprehensive statistical analysis on the repair status of multiple detection areas. Next, according to the repair status determination rules, the controller comprehensively evaluates the multiple repair status parameters. When the repair status parameters of multiple detection areas all meet the preset repair requirements, the repair status type corresponding to the electrode area is determined to be repair fit. When multiple detection areas have insufficient contact performance recovery or an overall recovery level lower than the preset requirements, the repair status type is determined to be repair insufficiency. When multiple detection areas have local over-repair, abnormal changes in contact performance, or repair parameters exceeding the fit range, the repair status type is determined to be over-repair. Then, based on the determined repair status type, corresponding repair evaluation information is generated. This repair evaluation information includes at least one of the following: repair status description, repair quality level, repair strategy fit degree, and subsequent repair suggestions, to reflect the execution effect of the current laser-enhanced contact repair strategy. Finally, based on the repair evaluation information, repair analysis results corresponding to the electrode areas are generated, and the repair analysis results are output to the repair feedback module, providing an evaluation basis for subsequent repair experience sample construction and incremental learning of the laser-enhanced contact optimization model.

[0070] As can be seen, by comprehensively analyzing the repair status parameters corresponding to multiple detection areas, the overall repair status type of the electrode area is determined, and repair evaluation information and repair analysis results are further generated. This allows the repair analysis results to not only reflect the local repair effect of each detection area, but also to comprehensively evaluate the overall repair quality of the entire electrode area after laser-enhanced contact repair. Compared to judging the repair effect based on only a single detection parameter, this embodiment achieves the classification evaluation and quantitative analysis of the repair effect, improving the accuracy and reliability of the repair result determination.

[0071] With the above Figure 4 The embodiments shown are consistent; please refer to [link / reference]. Figure 6 , Figure 6 This is a flowchart illustrating an incremental learning method for a photovoltaic cell laser-enhanced contact repair model provided in an embodiment of this application. After performing repair analysis on the electrode region based on the first resistance detection data and the second resistance detection data to obtain the repair analysis result, the method further includes the following steps: S601. Generate repair feedback data corresponding to the electrode region based on the target repair data, the repair analysis results, and the second resistance detection data; S602. Construct a repair experience sample corresponding to the electrode region based on the repair feedback data; the repair experience sample includes the manufacturing process data, the first resistance detection data, the second resistance detection data, the target repair data, and the repair analysis results; S603. Write the repair experience sample into the repair strategy sample library to update the sample data in the repair strategy sample library and obtain the target repair strategy sample library. S604. Perform incremental learning on the laser-enhanced contact optimization model according to the target repair strategy sample library to obtain the target laser-enhanced contact optimization model; the target laser-enhanced contact optimization model is used to generate target repair data for photovoltaic cells.

[0072] The repair feedback data is generated by comprehensively considering the target repair data, repair analysis results, and second resistance detection data. It reflects the actual execution effect and performance improvement of the laser-enhanced contact repair strategy. Repair experience samples are training samples built from the repair feedback data for model learning. These samples include manufacturing process data, first resistance detection data, second resistance detection data, target repair data, and repair analysis results, comprehensively recording the data relationships in a single laser-enhanced contact repair process. The repair strategy sample library is a dataset storing multiple repair experience samples, continuously accumulating repair experience under different manufacturing process conditions, contact states, and repair strategies. Incremental learning refers to a learning method that continuously updates model parameters using newly added repair experience samples while maintaining the existing knowledge of the laser-enhanced contact optimization model, thereby improving the model's predictive and generalization abilities for different photovoltaic cell repair scenarios. The laser-enhanced contact optimization model is an adaptive repair algorithm model. Internally, it establishes a parameter-performance mapping model between repair parameters and repair performance, used to learn the correlation between different manufacturing process backgrounds, different electrode degradation states, and repair performance. In some embodiments, before writing the repair experience samples into the repair strategy sample library, the controller may also perform a quality assessment on the repair experience samples, and screen the repair experience samples based on at least one of the following: data integrity, repair effect reliability, and sample representativeness, and only write the repair experience samples that meet the preset quality requirements into the repair strategy sample library.

[0073] Specifically, firstly, the controller establishes the correlation between repair parameters, repair effects, and post-repair contact performance based on the target repair data, repair analysis results, and second resistance detection data. It then generates corresponding repair feedback data based on these correlations, ensuring the feedback comprehensively reflects the execution and effectiveness of the laser-enhanced contact repair process. Next, based on the repair feedback data, combined with the corresponding photovoltaic cell manufacturing process data, first resistance detection data, and second resistance detection data, the controller fully records the repair process and constructs a repair experience sample that includes manufacturing process characteristics, contact state characteristics, repair strategy characteristics, and repair effect characteristics. Then, the constructed repair experience sample is standardized and validated. Repair experience samples that meet preset conditions are written into the repair strategy sample library to update the sample data, resulting in the target repair strategy sample library. Finally, the controller performs incremental learning on the laser-enhanced contact optimization model based on the updated target repair strategy sample library. By adding new repair experience samples, the controller continuously adjusts the model parameters, enabling the laser-enhanced contact optimization model to learn the repair patterns corresponding to different manufacturing process backgrounds, different contact degradation states, and different repair strategies, ultimately obtaining the target laser-enhanced contact optimization model. Manufacturing process data from the target repair strategy sample library is input into the laser-enhanced contact optimization model as background information. During model inference, the manufacturing process background and contact degradation state are used as joint inputs. Based on preset optimization objectives, different combinations of repair parameters are analyzed and calculated to dynamically determine the target repair data applicable to the current photovoltaic cell or subsequent photovoltaic cells in the same batch. The preset optimization objectives may include at least one of the following: maximizing post-repair conversion efficiency, minimizing conversion efficiency dispersion, achieving optimal repair performance while meeting preset EL yield requirements, or maximizing repair success rate. The target repair data includes at least one of the following: laser power, scanning speed, bias voltage magnitude, bias waveform, and spot shape modulation parameters, to achieve adaptive optimization of laser-enhanced contact repair parameters under different manufacturing process conditions and contact degradation states.

[0074] For easier understanding, please refer to Figure 7 , Figure 7This is a schematic diagram of a closed-loop learning and optimization process for photovoltaic cell repair based on laser-enhanced contact optimization, provided in an embodiment of this application. As can be seen, in stage ①, the contact resistance detection device performs online detection on the photovoltaic cell electrode area to obtain pre-repair detection data, thereby quantifying the current contact degradation state of the electrode area. The detected data may include contact resistance magnitude, contact resistance distribution, local contact performance, and other information that reflects the degree of electrode contact degradation. Subsequently, in stage ②, the repair decision-making process, the controller calls the laser-enhanced contact optimization model, taking the pre-repair detection data and the manufacturing process data corresponding to the current photovoltaic cell as input. Based on the parameter-performance mapping relationship established within the model and the preset optimization target, different combinations of repair parameters are analyzed and calculated to dynamically generate the target repair parameters corresponding to the current photovoltaic cell. The target repair parameters may include at least one of laser power, scanning speed, bias voltage magnitude, bias voltage waveform, and spot shape modulation parameters, enabling the repair parameters to adaptively match the degradation degree of the current electrode area and the manufacturing process background. Next, in stage ③, laser repair, the restorative LECO device performs laser-enhanced contact repair according to the target repair parameters. By applying a laser to the electrode contact area, local contact defects are optimized and repaired, improving the conductivity and contact consistency of the electrode area. After repair, stage ④, re-inspection, is performed again on the repaired electrode area to obtain post-repair inspection data, ensuring that the pre-repair and post-repair inspection data correspond to the same inspection area. Then, stage ⑤, repair analysis, compares and analyzes the pre-repair and post-repair inspection data to determine the changes in contact performance before and after repair. Combining indicators such as defect residue, dimensional recovery, conductivity continuity, edge integrity, conductivity recovery, and changes in contact resistance distribution in the repaired area, a comprehensive evaluation is completed according to preset repair analysis rules, yielding the corresponding repair analysis results. The repair analysis results not only reflect the effectiveness of the current repair strategy but also characterize the degree of improvement in the contact performance of the repaired electrode area, providing a reliable basis for subsequent model optimization. Further, in stage ⑥, feedback and sample construction, the correlation between repair parameters, repair effects, and post-repair contact performance is established based on the repair analysis results. Corresponding repair experience samples are then constructed by combining manufacturing process data, pre-repair inspection data, post-repair inspection data, repair parameters, and repair analysis results. Each repair experience sample fully records the data correlations of one laser-enhanced contact repair process. After completing sample construction, a quality assessment is performed on the repair experience samples, judging sample quality based on indicators such as data completeness, reliability of repair effects, sample representativeness, and data consistency.When the quality assessment result is qualified, the repair experience sample is written into the repair strategy sample library, allowing the library to continuously accumulate repair experience corresponding to different manufacturing process backgrounds, different degradation states, and different repair parameters. When the quality assessment result is unqualified, the current sample is discarded or the repair strategy is readjusted to prevent low-quality samples from entering the repair strategy sample library, thereby ensuring the data quality of the sample library and the model learning effect. Finally, in the incremental learning and optimization stage, the controller calls the newly added repair experience samples in the repair strategy sample library to perform incremental learning on the laser-enhanced contact optimization model. This allows the model to continuously learn new repair experience while maintaining its existing knowledge base, continuously optimizing the parameter-performance mapping model. Simultaneously, the final test sorting results (efficiency, EL, etc.) of the photovoltaic cells are received as feedback on the actual repair performance. The test sorting results are compared and analyzed with the model prediction results to further correct the model parameters and improve the accuracy of the model's repair performance prediction for different manufacturing process backgrounds and different degradation states. After the model is updated, the new parameter-performance mapping relationship re-enters the repair decision stage, enabling subsequent photovoltaic cells to directly adopt the optimized repair strategy for laser-enhanced contact repair, thereby realizing the continuous evolution of the repair strategy from experience-driven to data-driven and adaptive optimization.

[0075] As can be seen, this embodiment constructs repair feedback data from target repair data, repair analysis results, and second resistance detection data. It further generates repair experience samples, updates the repair strategy sample library, and performs incremental model learning. This allows the laser-enhanced contact optimization model to continuously absorb new repair experience and optimize the internal repair strategy mapping relationship. Compared to using a fixed training model for repair parameter prediction, this embodiment establishes a closed-loop learning mechanism of "repair feedback, experience samples, sample library updates, and incremental model learning." This enables the model to continuously improve the generation accuracy and adaptability of target repair data as repair cases accumulate, thereby enhancing the stability, accuracy, and intelligence level of laser-enhanced contact repair of photovoltaic cells under different manufacturing process conditions and contact degradation states.

[0076] In one optional embodiment, constructing the repair experience sample corresponding to the electrode region based on the repair feedback data includes: establishing a correlation between the manufacturing process information, the target repair data information, and the repair result information; generating repair experience data corresponding to the electrode region based on the correlation; determining repair strategy data corresponding to the electrode region based on the repair experience data; and generating the repair experience sample based on the repair strategy data.

[0077] The manufacturing process information is characterized by the technological features of the photovoltaic cell manufacturing process, which may include at least one of the following: slurry type, sintering temperature, sintering time, passivation process parameters, annealing process parameters, etc. The target repair data information is the repair control parameter information output by the laser-enhanced contact optimization model, which may include at least one of the following: laser power, scanning speed, bias voltage parameters, scanning path, and spot parameters. The repair result information is the repair effect information determined based on the repair analysis results and the second resistance detection data, used to characterize the actual repair performance corresponding to the target repair data. Repair experience data is a dataset extracted based on the correlation between the manufacturing process information, target repair data information, and repair result information, used to characterize the correspondence between repair strategies and repair effects under different manufacturing process conditions. Repair strategy data is strategy feature data summarized from the repair experience data, used to describe the repair strategy rules applicable to different manufacturing process backgrounds and contact states. Repair experience samples are model training samples generated based on the repair strategy data, used for subsequent updates to the repair strategy sample library and incremental learning of the laser-enhanced contact optimization model. Repair strategy data can include manufacturing process characteristics, contact state characteristics, repair parameter characteristics, and mapping rules or strategy labels between corresponding repair effects, which are used to characterize the preferred repair strategies under different manufacturing conditions and contact states.

[0078] Specifically, firstly, the controller extracts the manufacturing process information, target repair data information, and repair result information of the corresponding photovoltaic cell based on the repair feedback data, and establishes a correspondence between the manufacturing process information, target repair data information, and repair result information, so that the process conditions, repair parameters, and repair effects in a complete repair process form a unified data association. Next, the controller fuses different information according to the established association, generating corresponding repair experience data, so that the repair experience data can fully represent the correspondence between repair parameters and repair effects under specific manufacturing process conditions. Then, based on the repair experience data, the controller analyzes the differences in repair effects corresponding to different combinations of repair parameters, extracts strategy features that can represent the repair rules, and determines the repair strategy data corresponding to the electrode area accordingly, so that the repair strategy data can reflect the configuration of the optimal repair strategy under different process conditions. Finally, the controller generates standardized repair experience samples based on the repair strategy data, and stores the repair experience samples as training samples for subsequent updates to the repair strategy sample library and for incremental learning of the laser-enhanced contact optimization model.

[0079] As can be seen, this embodiment, by associating the process features, repair parameters, and repair effects in a single repair process through modeling, enables the repair experience samples to not only record the repair results but also reflect the inherent correspondence between repair strategies and repair performance under different manufacturing process conditions. This improves the knowledge representation ability and training value of the repair experience samples, providing a more complete and reliable data foundation for subsequent updates to the repair strategy sample library and incremental learning of the laser-enhanced contact optimization model. Consequently, it further enhances the adaptive learning ability of the laser-enhanced contact optimization model for different photovoltaic cell repair scenarios and the prediction accuracy of target repair data.

[0080] For easier understanding, please refer to Figure 8 , Figure 8 This is a schematic diagram of a repair scenario for laser-enhanced contact optimization of photovoltaic cells provided in an embodiment of this application. As can be seen, the photovoltaic cell is used as the repair object. The OCRM module performs online contact state detection on the photovoltaic cell, obtains degradation state information corresponding to the electrode areas of the photovoltaic cell, and sends the detected degradation evaluation results to the adaptive algorithm engine. The adaptive algorithm engine combines the model library and empirical data to comprehensively analyze the current manufacturing process background, contact degradation state, and historical repair experience of the photovoltaic cell, generates corresponding repair parameter instructions, and controls the laser to perform laser-enhanced contact repair on the photovoltaic cell. After repair, the repaired photovoltaic cell enters a testing and sorting machine for performance testing, and the performance feedback obtained from the test is input again into the adaptive algorithm engine to update the model library and empirical data, realizing continuous learning and adaptive optimization of the photovoltaic cell laser-enhanced contact repair strategy.

[0081] Specifically, firstly, the photovoltaic cell to be repaired enters the online inspection station. The OCRM module performs non-contact inspection on the electrode area of ​​the photovoltaic cell, acquiring contact resistance data for the current electrode area, and establishing a corresponding degradation assessment result based on the inspection data. The degradation assessment result characterizes the degree of contact degradation in the current photovoltaic cell electrode area, and can include contact resistance magnitude, contact resistance distribution, changes in local contact performance, and other characteristic information reflecting the contact degradation state. Subsequently, the OCRM module sends the degradation assessment result to the adaptive algorithm engine as an important input for repair parameter optimization. Next, the adaptive algorithm engine receives the degradation assessment result and calls the parameter-performance mapping model stored in the model library. Simultaneously, it combines historical repair experience accumulated from empirical data to comprehensively analyze the manufacturing process background and contact degradation state of the current photovoltaic cell. The model library stores trained laser-enhanced contact optimization models and parameter-performance mapping relationships, while the empirical data stores historical repair samples corresponding to different manufacturing process conditions, different contact degradation states, and different repair strategies. The adaptive algorithm engine predicts and analyzes different repair parameter combinations based on the current degradation state, manufacturing process background, and historical experience. It dynamically determines the optimal repair parameters for the current photovoltaic cell according to preset optimization goals, generating corresponding repair parameter instructions. Then, the adaptive algorithm engine sends these instructions to the laser. The laser modulates the photovoltaic cell's electrode area according to the corresponding laser power, scanning speed, bias voltage, bias waveform, and spot shape modulation parameters, creating a more stable ohmic contact between the electrode and the silicon substrate, reducing contact resistance, and improving the conductivity and contact uniformity of the electrode area. During the repair process, the laser can perform differentiated repairs on different electrode areas based on the repair parameter instructions, allowing the repair strategy to adapt to photovoltaic cells with different degradation levels and different manufacturing processes. After repair, the repaired photovoltaic cells enter a testing and sorting machine for performance testing. The testing and sorting machine acquires the repaired performance data and generates corresponding performance feedback. This feedback may include photovoltaic cell conversion efficiency, open-circuit voltage, fill factor, EL test results, and other test data reflecting the repair effect. Subsequently, the test sorting machine sends performance feedback to the adaptive algorithm engine. Based on this feedback, the engine compares and analyzes the model's predicted results with the actual repair results, corrects the model parameters, and simultaneously updates the model library and experience data. This ensures that the newly added repair experience can continuously participate in subsequent model training and parameter optimization. Simultaneously, the adaptive algorithm engine can also regenerate repair parameter instructions more suitable for subsequent batches of photovoltaic cells based on the updated model, and continue to control the laser to complete the laser-enhanced contact repair of the next batch of photovoltaic cells.In this way, the repair strategy can be dynamically adjusted according to the actual degradation state of the photovoltaic cells, and the model prediction capability can be continuously optimized by using the real performance data after repair. This effectively improves the conversion efficiency, repair consistency and mass production stability of the photovoltaic cells after repair, while enhancing the system's adaptability to different manufacturing process conditions and different degradation scenarios.

[0082] As can be seen, this embodiment obtains first resistance detection data and manufacturing process data of the photovoltaic cell electrode region through the laser-enhanced contact repair method for photovoltaic cells provided above. Then, the first resistance detection data and manufacturing process data are input into a preset laser-enhanced contact optimization model to obtain the target repair parameters corresponding to the electrode region. Next, the laser-enhanced contact repair equipment is controlled to repair the electrode region according to the target repair parameters, and the repaired second resistance detection data is obtained. Finally, the electrode region is repaired based on the first and second resistance detection data to obtain the repair analysis results. In this way, the problem of insufficient repair and over-repair of some cells after the passivation process in photovoltaic cell manufacturing, resulting in inconsistent cell conversion efficiency and low product yield, can be solved.

[0083] The above primarily describes the solutions of the embodiments of this application from the perspective of the method execution process. It is understood that, in order to achieve the above functions, the controller includes the corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments provided herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0084] This application embodiment can divide the controller into functional units according to the above method example. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0085] When dividing each function into modules according to its corresponding function. Figure 9This is a functional module block diagram of a laser-enhanced contact repair device for photovoltaic cells provided in the application embodiment. The laser-enhanced contact repair device 900 for photovoltaic cells is used as a controller in a photovoltaic cell repair system and includes: The detection unit 910 is used to acquire first resistance detection data of the electrode area of ​​the photovoltaic cell; and manufacturing process data of the photovoltaic cell electrode; and to acquire second resistance detection data of the electrode area after repair; the resistance detection data includes: contact resistance data and contact resistance distribution data; The calculation unit 920 is used to input the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode area; Control unit 930 is used to control the laser-enhanced contact optimization repair equipment to repair the electrode area according to the target repair data; The repair unit 940 is used to perform repair analysis on the electrode area based on the first resistance detection data and the second resistance detection data, and obtain the repair analysis result.

[0086] In an optional embodiment, the computing unit 920, in inputting the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode region, is specifically used for: The contact state information corresponding to the electrode region is generated based on the first resistance detection data; and the process influence information corresponding to the electrode region is generated based on the manufacturing process data. Based on the contact state information and the process influence information, determine the repair constraints for the electrode region; Based on the repair constraints, the contact state information, and the process influence information, the repair data corresponding to the electrode region is determined to obtain the first repair data; The first repair data is input into the laser-enhanced contact optimization model to obtain the target repair data.

[0087] In an optional embodiment, the computing unit 920, in generating contact state information corresponding to the electrode region based on the first resistance detection data, is specifically configured to: Based on the contact resistance distribution data, determine the multiple resistance distribution relationships corresponding to multiple detection areas within the electrode area; and based on the contact resistance data, determine the contact resistance values ​​corresponding to the multiple detection positions to obtain multiple contact resistance values; Construct a contact state distribution spectrum corresponding to the electrode region based on the multiple resistance distribution relationships and the multiple contact resistance values; The contact state information corresponding to the electrode region is generated based on the contact state distribution spectrum.

[0088] In an optional embodiment, the calculation unit 920, in determining the repair data corresponding to the electrode region based on the contact state information and the process influence information according to the repair constraints, and obtaining the first repair data, is specifically used for: The contact state information and the process influence information are input into a preset repair data and performance mapping model to obtain the first repair mapping relationship between the repair data and repair performance corresponding to the electrode area; Based on the repair constraints, the mapping relationships corresponding to the electrode regions are filtered from the first repair mapping relationships to obtain the target repair mapping relationship; The first repair data corresponding to the electrode region is determined based on the target repair mapping relationship.

[0089] In an optional embodiment, the repair unit 940, in performing repair analysis on the electrode region using the first resistance detection data and the second resistance detection data to obtain repair analysis results, is specifically used for: The corresponding detection change data of the electrode region is determined based on the first resistance detection data and the second resistance detection data; From the detected change data, multiple detection change parameters corresponding to multiple detection areas in the electrode region are extracted to obtain multiple detection change parameters; Based on the multiple detection change parameters, the corresponding regional repair response parameters of the multiple detection areas are determined to obtain multiple repair response parameters; the multiple repair response parameters are used to characterize the contact performance recovery parameters of the electrode area after laser-enhanced repair optimization, and the multiple repair response parameters include at least one of the following: contact resistance improvement magnitude, contact resistance distribution equalization parameter, and contact recovery trend parameter. Based on the multiple repair response parameters, the region repair status parameters corresponding to the multiple detection regions are determined to obtain multiple repair status parameters; The repair analysis results corresponding to the electrode region are generated based on the multiple repair status parameters.

[0090] In an optional embodiment, the repair unit 940, in generating the repair analysis results corresponding to the electrode region based on the plurality of repair state parameters, is specifically used for: The repair state type corresponding to the electrode region is determined based on multiple repair state parameters, and the repair state type includes at least one of insufficient repair, repair fit, and over-repair. Determine the repair evaluation information corresponding to the electrode region based on the repair status type; The repair analysis results are generated based on the repair evaluation information.

[0091] In an optional embodiment, after the calculation unit performs repair analysis on the electrode region based on the first resistance detection data and the second resistance detection data to obtain the repair analysis result, it is further configured to: Based on the target repair data, the repair analysis results, and the second resistance detection data, repair feedback data corresponding to the electrode region is generated; Based on the repair feedback data, a repair experience sample corresponding to the electrode region is constructed; the repair experience sample includes the manufacturing process data, the first resistance detection data, the second resistance detection data, the target repair data, and the repair analysis results; The repair experience samples are written into the repair strategy sample library to update the sample data in the repair strategy sample library, thereby obtaining the target repair strategy sample library; Incremental learning is performed on the laser-enhanced contact optimization model based on the target repair strategy sample library to obtain the target laser-enhanced contact optimization model; the target laser-enhanced contact optimization model is used to generate target repair data for photovoltaic cells.

[0092] In an optional embodiment, the computing unit 920, in constructing the repair experience sample corresponding to the electrode region based on the repair feedback data, is specifically used for: Establish the association between the manufacturing process information, the target repair data information, and the repair result information; Based on the aforementioned correlation, repair experience data corresponding to the electrode region is generated; Based on the repair experience data, determine the repair strategy data corresponding to the electrode region; The repair experience sample is generated based on the repair strategy data.

[0093] As can be seen, this embodiment provides a laser-enhanced contact repair device for photovoltaic cells. It acquires first resistance detection data of the photovoltaic cell electrode region and manufacturing process data of the photovoltaic cell electrode. Then, it inputs the first resistance detection data and manufacturing process data into a preset laser-enhanced contact optimization model to obtain target repair parameters corresponding to the electrode region. Next, it controls the laser-enhanced contact repair equipment to repair the electrode region according to the target repair parameters and acquires second resistance detection data after repair. Finally, it performs repair analysis on the electrode region based on the first and second resistance detection data to obtain the repair analysis results. In this way, it can match the actual contact degradation degree formed under different manufacturing process conditions, reducing under-repair or over-repair phenomena caused by fixed repair parameters, thereby improving the consistency of photovoltaic cell contact repair and enhancing the consistency of photovoltaic cell conversion efficiency and product yield.

[0094] This application also provides a computer-readable storage medium storing a computer program for electronic data interchange, the computer program causing a computer to perform some or all of the steps of any of the methods described in the above method embodiments, the computer including a controller.

[0095] This application also provides a laser-enhanced contact repair system for photovoltaic cells, which can perform some or all of the steps of any of the methods described in the above method embodiments.

[0096] It should be noted that, for the sake of simplicity, the above embodiments are all described as a series of actions. Those skilled in the art should understand that this application is not limited to the described order of actions, as some steps in the embodiments of this application can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions, steps, modules, or units involved are not necessarily essential to the embodiments of this application. In the above embodiments, the descriptions of each embodiment in this application have different focuses; parts not described in detail in a certain embodiment can be referred to the relevant descriptions of other embodiments.

[0097] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0098] The steps of the methods or algorithms described in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disk, portable hard disk, read-only optical disk (CD-ROM), or any other form of storage medium well known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Furthermore, the ASIC can reside in a terminal device or management device. Alternatively, the processor and storage medium can exist as discrete components in the terminal device or management device.

[0099] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in the embodiments of this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0100] The modules / units included in the various devices and products described in the above embodiments can be software modules / units, hardware modules / units, or a combination of both. For example, for devices and products applied to or integrated into a chip, all modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits. For devices and products applied to or integrated into a chip module, all modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip module, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0101] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the embodiments of this application. It should be understood that the above descriptions are merely specific embodiments of the embodiments of this application and are not intended to limit the protection scope of the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solutions of the embodiments of this application should be included within the protection scope of the embodiments of this application.

Claims

1. A laser-enhanced contact repair method for photovoltaic cells, characterized in that, A controller applied to a photovoltaic cell repair system, the method comprising: Acquire first resistance detection data of the electrode region of the photovoltaic cell; and manufacturing process data of the photovoltaic cell electrode; the resistance detection data includes: contact resistance data and contact resistance distribution data; The first resistance detection data and the manufacturing process data are input into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode area; The laser-enhanced contact optimization repair equipment is controlled to repair the electrode area according to the target repair data, and the second resistance detection data of the repaired electrode area is obtained. The electrode region is repaired and analyzed based on the first resistance detection data and the second resistance detection data to obtain the repair analysis results.

2. The method as described in claim 1, characterized in that, The step of inputting the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode region includes: The contact state information corresponding to the electrode region is generated based on the first resistance detection data; and the process influence information corresponding to the electrode region is generated based on the manufacturing process data. Based on the contact state information and the process influence information, determine the repair constraints for the electrode region; Based on the repair constraints, the contact state information, and the process influence information, the repair data corresponding to the electrode region is determined to obtain the first repair data; The first repair data is input into the laser-enhanced contact optimization model to obtain the target repair data.

3. The method as described in claim 2, characterized in that, The step of generating contact state information corresponding to the electrode region based on the first resistance detection data includes: Based on the contact resistance distribution data, determine the multiple resistance distribution relationships corresponding to multiple detection areas within the electrode area; and based on the contact resistance data, determine the contact resistance values ​​corresponding to the multiple detection positions to obtain multiple contact resistance values; Construct a contact state distribution spectrum corresponding to the electrode region based on the multiple resistance distribution relationships and the multiple contact resistance values; The contact state information corresponding to the electrode region is generated based on the contact state distribution spectrum.

4. The method as described in claim 2, characterized in that, The step of determining the repair data corresponding to the electrode region based on the contact state information and the process influence information according to the repair constraints, to obtain the first repair data, includes: The contact state information and the process influence information are input into a preset repair data and performance mapping model to obtain the first repair mapping relationship between the repair data and repair performance corresponding to the electrode area; Based on the repair constraints, the mapping relationships corresponding to the electrode regions are filtered from the first repair mapping relationships to obtain the target repair mapping relationship; The first repair data corresponding to the electrode region is determined based on the target repair mapping relationship.

5. The method according to any one of claims 1-4, characterized in that, The step of performing repair analysis on the electrode region based on the first resistance detection data and the second resistance detection data to obtain repair analysis results includes: The corresponding detection change data of the electrode region is determined based on the first resistance detection data and the second resistance detection data; From the detected change data, multiple detection change parameters corresponding to multiple detection areas in the electrode region are extracted to obtain multiple detection change parameters; Based on the multiple detection change parameters, the corresponding regional repair response parameters of the multiple detection areas are determined to obtain multiple repair response parameters; the multiple repair response parameters are used to characterize the contact performance recovery parameters of the electrode area after laser-enhanced repair optimization, and the multiple repair response parameters include at least one of the following: contact resistance improvement magnitude, contact resistance distribution equalization parameter, and contact recovery trend parameter. Based on the multiple repair response parameters, the region repair status parameters corresponding to the multiple detection regions are determined to obtain multiple repair status parameters; The repair analysis results corresponding to the electrode region are generated based on the multiple repair status parameters.

6. The method as described in claim 5, characterized in that, The step of generating the corresponding repair analysis results for the electrode region based on the multiple repair status parameters includes: The repair state type corresponding to the electrode region is determined based on multiple repair state parameters, and the repair state type includes at least one of insufficient repair, repair fit, and over-repair. Determine the repair evaluation information corresponding to the electrode region based on the repair status type; The repair analysis results are generated based on the repair evaluation information.

7. The method according to any one of claims 1-4, characterized in that, After performing repair analysis on the electrode region based on the first resistance detection data and the second resistance detection data to obtain the repair analysis result, the method further includes: Based on the target repair data, the repair analysis results, and the second resistance detection data, repair feedback data corresponding to the electrode region is generated; Based on the repair feedback data, a repair experience sample corresponding to the electrode region is constructed; the repair experience sample includes the manufacturing process data, the first resistance detection data, the second resistance detection data, the target repair data, and the repair analysis results; The repair experience samples are written into the repair strategy sample library to update the sample data in the repair strategy sample library, thereby obtaining the target repair strategy sample library; Incremental learning is performed on the laser-enhanced contact optimization model based on the target repair strategy sample library to obtain the target laser-enhanced contact optimization model; the target laser-enhanced contact optimization model is used to generate target repair data for photovoltaic cells.

8. The method as described in claim 7, characterized in that, The step of constructing a repair experience sample corresponding to the electrode region based on the repair feedback data includes: Establish the association between the manufacturing process information, the target repair data information, and the repair result information; Based on the aforementioned correlation, repair experience data corresponding to the electrode region is generated; Based on the repair experience data, determine the repair strategy data corresponding to the electrode region; The repair experience sample is generated based on the repair strategy data.

9. A laser-enhanced contact repair device for photovoltaic cells, characterized in that, A controller used in a photovoltaic cell repair system, the device comprising: The detection unit is used to acquire first resistance detection data of the electrode region of the photovoltaic cell; and manufacturing process data of the photovoltaic cell electrode; the resistance detection data includes: contact resistance data and contact resistance distribution data; The calculation unit is used to input the first resistance detection data and the manufacturing process data into a preset laser-enhanced contact optimization model to obtain the target repair data corresponding to the electrode area; The control unit is used to control the laser-enhanced contact optimization repair equipment to repair the electrode area according to the target repair data; The detection unit is also used to acquire second resistance detection data of the electrode area after repair; The repair unit is used to perform repair analysis on the electrode area based on the first resistance detection data and the second resistance detection data, and obtain the repair analysis results.

10. A controller, characterized in that, include: Processor, memory, communication interface, and one or more programs; The one or more programs are stored in the memory and configured to be executed by the processor, the programs including instructions for performing the steps of the method as described in any one of claims 1-8.

Images