BC battery assembly preparation method and device based on optimal sorting and detection

Abstract

The invention provides a BC battery assembly preparation method and device based on optimal sorting and detection, and belongs to the technical field of photovoltaic cells. Cut BC battery pieces are preprocessed, multi-dimensional parameter acquisition is performed on the preprocessed qualified battery pieces, and parameter weights are determined by combining an analytic hierarchy process with an entropy weight method; dividing the battery pieces into a plurality of grades according to a comprehensive score obtained by weighting calculation; performing infrared thermal imaging and electroluminescence (EL) detection on the battery pieces of the selected grade, screening out qualified repairable products, and sequentially performing series welding, forming detection and packaging reinspection to obtain qualified products; and importing the whole-process data of the qualified product into a quality prediction model constructed by adopting a random forest algorithm for feedback optimization. Therefore, through multi-dimensional parameter sorting, full-process closed-loop detection and dynamic optimization, the sorting precision is improved, the defect omission ratio is reduced, the process is optimized, and efficient and reliable production is achieved.

Description

Technical Field

[0001] This application belongs to the field of photovoltaic cell technology, and specifically relates to a method and apparatus for preparing BC cell modules based on optimized sorting and detection. Background Technology

[0002] In the global transition to a clean and low-carbon energy structure, photovoltaic energy, as a core component of renewable energy, is experiencing rapid technological iteration. Back-Contact (BC) cells, with their unique structure where electrodes are embedded on the back of the cell, eliminate the shading of incident light by the front electrodes and possess bi-sided light-receiving capability. Their photoelectric conversion efficiency is 3-5 percentage points higher than traditional PERC (passivated emitter and back contact) cells, making them one of the mainstream development directions for high-efficiency photovoltaic cell technology. The manufacturing quality of BC cell modules directly determines their photoelectric performance and reliability. Cell sorting and inspection are crucial in the manufacturing process; the sorting stage determines the performance matching degree of cells within the same module, while the inspection stage directly screens out defective cells and monitors process stability. However, the embedded electrode structure and bi-sided light-receiving characteristics of BC cells make traditional cell sorting and inspection technologies difficult to adapt, becoming a core bottleneck restricting the large-scale mass production and performance improvement of BC cell modules.

[0003] The existing process is as follows: cell cutting → surface cleaning → sorting based on binary parameters of open-circuit voltage (Voc) and short-circuit current (Isc) → string bonding → EL (electroluminescence) single-point inspection → module packaging. In the sorting stage, only two basic electrical performance parameters—open-circuit voltage and short-circuit current—are collected using a pulse IV tester, classifying cells with parameter deviations within ±2% into the same grade. The inspection stage only uses EL inspection equipment to perform a single defect scan of the cells after string bonding, identifying obvious microcracks or broken grid defects. This process suffers from low sorting accuracy and a high rate of missed defect detection. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a BC battery module preparation method and apparatus based on optimized sorting and detection, which can improve sorting accuracy and reduce defect missed detection rate.

[0005] In a first aspect, this application provides a method for preparing BC battery modules based on optimized sorting and detection, the method comprising the following steps: A closed-loop process of plasma cleaning, cleanliness testing, passivation layer testing, and secondary treatment is performed on the cut BC solar cells to obtain pre-treated qualified solar cells. Multi-dimensional parameters were collected from pre-processed qualified solar cells, and the analytic hierarchy process (AHP) combined with the entropy weight method was used to determine the parameter weights. Based on the comprehensive score obtained by weighted calculation, the solar cells were divided into multiple grades. The multi-dimensional parameters collected included bifacial spectral response parameters, open-circuit voltage Voc, short-circuit current Isc, fill factor FF, and buried electrode contact resistance parameters. Infrared thermal imaging and electroluminescence (EL) testing were performed on selected grade solar cells, and qualified repairable products were screened out and then sequentially subjected to stringing, forming testing, and packaging re-inspection to obtain qualified products. The entire process data of qualified products is imported into a quality prediction model built using the random forest algorithm for feedback optimization.

[0006] In some embodiments, the closed-loop process of performing plasma cleaning, cleanliness testing, passivation layer testing, and secondary processing on the cut BC solar cells to obtain pre-treated qualified solar cells includes the following steps: The cut BC solar cells were cleaned using a plasma cleaner, and the amount of residual oil on the surface was tested by gravimetric method. BC solar cells that did not meet the requirements for the amount of residual oil were re-cleaned by plasma until the requirements were met. For BC solar cells with sufficient residual oil on the surface, five sampling points were selected at the center and four corners using an atomic force microscope to test the surface roughness. Among them, BC solar cells with satisfactory surface roughness were judged to have a qualified passivation layer and were considered as qualified pre-treatment solar cells. BC solar cells with unsatisfactory surface roughness were subjected to passivation treatment again.

[0007] In some embodiments, the process of collecting multi-dimensional parameters from pre-processed qualified solar cells, determining parameter weights using the analytic hierarchy process (AHP) combined with the entropy weight method, and calculating a comprehensive score based on the weighted average includes the following steps: For pre-treated qualified solar cells, the bifacial spectral response values ​​are collected using a xenon lamp monochromator and the bifacial spectral response difference coefficient is calculated; the open-circuit voltage Voc and short-circuit current Isc are collected using a pulsed light source IV test system and the fill factor FF is calculated; the contact resistance values ​​are collected using a four-probe tester and the standard deviation of the contact resistance values ​​is calculated. A single-cell parameter dataset is constructed by integrating the bifacial spectral response difference coefficient, fill factor FF, and standard deviation of contact resistance. A hierarchical model was established using the analytic hierarchy process combined with the entropy weight method to determine the weights of the spectral response difference coefficient, the fill factor FF, and the standard deviation of the contact resistance value. The single cell parameter dataset was then input into the hierarchical module to calculate the comprehensive score.

[0008] In some embodiments, the process of performing infrared thermal imaging and electroluminescence (EL) detection on selected grades of solar cells, and then sequentially performing stringing, forming inspection, and packaging re-inspection on qualified repairable products to obtain qualified products includes the following steps: Infrared thermal imaging and electroluminescence (EL) detection are performed on selected grade solar cells, and the temperature distribution and electroluminescence images are fused and analyzed to select qualified repairable products based on the defect area. The qualified and repairable products selected are then serially welded to obtain the finished components. During the serial welding process, a laser contour sensor and a high-definition vision inspection module are simultaneously activated to adjust the serial welding parameters based on real-time monitoring of the solder strip alignment deviation and welding temperature fluctuations. The obtained molded components were tested for sealing performance by ultraviolet light irradiation and IV curves were tested by a steady-state solar simulator. Photovoltaic performance parameters were collected to determine whether the molded components were qualified. After conducting thermal cycling tests on qualified molded components, the photoelectric performance is retested. Based on the calculated power attenuation rate, qualified products are determined and packaged for shipment.

[0009] In some embodiments, the solar cells are divided into five grades based on the defect area. Grades 4-5 are determined to be defective products and are rejected; grades 1-3 are determined to be qualified and repairable products and proceed to the stringing process.

[0010] In some embodiments, if the weld strip alignment deviation is detected to be 0.05-0.1mm, the position of the welding torch is finely adjusted; if the welding temperature fluctuates by ±3-5℃, the power of the welding torch is linearly adjusted.

[0011] In some embodiments, importing the full-process data of qualified products into a quality prediction model constructed using a random forest algorithm for feedback optimization includes the following steps: The entire process data of qualified products is integrated to build a sample database, and a quality prediction model is built using the random forest algorithm, with cell parameters as input and component efficiency as output for training. Feedback optimization is performed based on the analysis results of the quality prediction model; the parameters for feedback optimization include multi-dimensional parameter sorting thresholds and string welding parameters.

[0012] Secondly, this application also provides a BC battery module fabrication apparatus based on optimized sorting and detection, the apparatus comprising: The pretreatment module is used to perform a closed-loop process of plasma cleaning, cleanliness detection, passivation layer detection and secondary treatment on the cut BC solar cells to obtain qualified pretreatment solar cells. The multi-dimensional parameter sorting module is used to collect multi-dimensional parameters of pre-processed qualified solar cells, and uses the analytic hierarchy process combined with the entropy weight method to determine the parameter weights. Based on the comprehensive score obtained by weighted calculation, the solar cells are divided into multiple grades. The multi-dimensional parameters collected include bifacial spectral response parameters, open circuit voltage Voc, short circuit current Isc, fill factor FF, and buried electrode contact resistance parameters. The full-process non-destructive testing module is used to perform infrared thermal imaging and electroluminescence (EL) testing on selected grades of solar cells, and to screen out qualified and repairable products for sequential stringing, forming inspection, and packaging re-inspection to obtain qualified products. The feedback optimization module is used to import the full-process data of qualified products into a quality prediction model built using the random forest algorithm for feedback optimization.

[0013] Thirdly, this application also provides an electronic device, including: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the steps of the BC battery module preparation method based on optimized sorting and detection described in any one of the first aspects are executed.

[0014] Fourthly, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the BC battery module preparation method based on optimized sorting and detection as described in any one of the first aspects.

[0015] This application describes a BC (Branch-Based Cell) module manufacturing method based on optimized sorting and inspection. The method involves a closed-loop process of plasma cleaning, cleanliness inspection, passivation layer inspection, and secondary processing on the cut BC cells to obtain pre-processed, qualified cells. Multi-dimensional parameters are collected from the pre-processed cells, and the analytic hierarchy process (AHP) combined with entropy weighting is used to determine parameter weights. Based on the weighted comprehensive score, the cells are classified into multiple grades. The collected multi-dimensional parameters include bifacial spectral response parameters, open-circuit voltage (Voc), short-circuit current (Isc), fill factor (FF), and buried electrode contact resistance. Infrared thermal imaging and electroluminescence (EL) inspection are performed on the selected grades of cells, and qualified, repairable products are selected for sequential stringing, forming inspection, and encapsulation re-inspection to obtain qualified products. The entire process data of the qualified products is imported into a quality prediction model constructed using a random forest algorithm for feedback optimization. Thus, through multi-dimensional parameter sorting, closed-loop inspection throughout the process, and dynamic optimization, sorting accuracy is improved, defect detection rate is reduced, and the process is optimized, achieving efficient and reliable production. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of the BC battery module fabrication method based on optimized sorting and detection described in the embodiments of this application is shown; Figure 2 A flowchart illustrating the pre-processed qualified battery cells obtained according to an embodiment of this application is shown; Figure 3 A flowchart illustrating the comprehensive score obtained by weighted calculation according to an embodiment of this application is shown; Figure 4 A flowchart illustrating the process of obtaining a qualified product according to an embodiment of this application is shown; Figure 5 This document illustrates a flowchart of an embodiment of the present application, showing how the entire process data of qualified products is imported into a quality prediction model constructed using a random forest algorithm for feedback optimization. Figure 6 This paper shows a schematic diagram of the BC battery module fabrication apparatus based on optimized sorting and detection according to an embodiment of this application; Figure 7 A schematic diagram of the structure of the electronic device described in an embodiment of this application is shown. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0019] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0021] Addressing the shortcomings of existing technologies. (1) Low sorting accuracy: large performance dispersion of modules: sorting is based solely on binary parameters of open-circuit voltage and short-circuit current, without considering the core characteristics of BC cells. Difference in bifacial spectral response and contact resistance of buried electrodes. In actual mass production, for BC cells with the same Voc and Isc, the difference coefficient of bifacial spectral response in the 300-1200nm wavelength range can be as high as 15%, and the standard deviation of buried electrode contact resistance can exceed 0.2Ω, resulting in a deviation of more than 1.5% in the photoelectric conversion efficiency of modules composed of cells of the same grade, which seriously affects the consistency of module performance. (2) High rate of missed detection of defects and insufficient reliability: only a single EL test is performed after string welding, and no other detection methods are combined. EL detection has low sensitivity to defects such as microcracks (width < 5μm) and buried electrode connection, and cannot detect sealing defects of the encapsulation layer after encapsulation. In practical applications, the defect omission rate of this technology exceeds 8%, among which hidden defects such as electrode connection and encapsulation layer leakage will cause the power decay of the module to exceed 10% after 1-2 years of use. (3) No closed-loop feedback mechanism and poor process coordination: the sorting data and the detection results are independent of each other and are not related to the subsequent stringing process. When the module performance is found to be substandard, it is impossible to locate whether it is due to improper matching of sorting level or abnormal stringing parameters. The process can only be adjusted by experience, resulting in large fluctuations in module yield (mass production yield is only 88-90%), and dynamic optimization of process parameters cannot be achieved. (4) Rough pretreatment process and insufficient basic quality control: only conventional water washing is used for surface cleaning, which cannot completely remove the silicon slag and organic oil stains remaining after cutting; and the integrity of the passivation layer is not detected. Defects in the passivation layer will increase the surface recombination rate of the cell, causing a loss of 0.5-0.8% in the photoelectric conversion efficiency of the module. This application provides a method and apparatus for preparing BC battery modules based on optimized sorting and detection, which can improve sorting accuracy, reduce defect missed detection rate, optimize process, and achieve efficient and reliable production.

[0022] See the instruction manual appendix Figure 1This application provides a method for preparing BC battery modules based on optimized sorting and detection, the method comprising the following steps: S1. Perform a closed-loop process of plasma cleaning, cleanliness detection, passivation layer detection, and secondary treatment on the cut BC solar cells to obtain pre-treated qualified solar cells. S2. Multi-dimensional parameters are collected for pre-processed qualified solar cells, and the analytic hierarchy process combined with the entropy weight method is used to determine the parameter weights. Based on the comprehensive score obtained by weighted calculation, the solar cells are divided into multiple levels. Among them, the collected multi-dimensional parameters include bifacial spectral response parameters, open circuit voltage Voc, short circuit current Isc, fill factor FF, and buried electrode contact resistance parameters. S3. Perform infrared thermal imaging and electroluminescence (EL) detection on the selected grade of solar cells, and screen out qualified repairable products for sequential stringing, forming detection, and packaging re-inspection to obtain qualified products. S4. Import the full-process data of qualified products into the quality prediction model built using the random forest algorithm for feedback optimization.

[0023] Step S1 mainly addresses the shortcomings of existing technologies, such as the inability of conventional water washing to remove silicon slag and oil stains and the failure to detect the passivation layer. By pre-processing, the surface cleanliness of the BC solar cell is ensured and the passivation layer is intact, so that the spectral response, contact resistance and other data collected in subsequent multi-dimensional parameter sorting are true and effective, and the interference of impurities or passivation layer defects with the accuracy of parameter detection is avoided.

[0024] See the instruction manual appendix Figure 2 The closed-loop process of performing plasma cleaning, cleanliness testing, passivation layer testing, and secondary treatment on the cut BC solar cells to obtain pre-treated qualified solar cells includes the following steps: S101. Use a plasma cleaner to perform plasma cleaning on the cut BC battery cells, and use the gravimetric method to test the amount of residual oil on the surface; wherein, BC battery cells whose residual oil on the surface does not meet the requirements are subjected to plasma cleaning again until the requirements are met. S102. For BC solar cells with sufficient residual oil on the surface, five sampling points are selected at the center and four corners using an atomic force microscope to test the surface roughness. Among them, BC solar cells with sufficient surface roughness are judged to have a qualified passivation layer and are considered as qualified pre-treatment solar cells. BC solar cells with insufficient surface roughness are subjected to passivation treatment again.

[0025] Specifically, in step S101, the cut BC solar cells are received and inspected for any obvious damage or cracks. They are then placed into the reaction chamber of a medium-frequency plasma cleaning device. A mixture of argon and oxygen (3:1 volume ratio) is introduced into the chamber. The cleaning power is set to 280-320W and the cleaning time to 50-70s. Through plasma bombardment, the silicon slag and organic oil residue remaining on the surface of the solar cells are thoroughly removed, laying a clean foundation for subsequent testing and sorting. After cleaning, the solar cells are removed, and the amount of residual oil on the surface is tested using a gravimetric method. The test data is recorded. If the residual oil content is <0.01mg / cm², the passivation layer is tested; if it is ≥0.01mg / cm², a second plasma cleaning at 320W for 80s is performed until the residual oil content meets the standard.

[0026] In step S102, for the battery cells that passed the cleaning in step S1, an atomic force microscope is used to select 5 sampling points at the center and four corners of the battery cell to test the surface roughness at each point and calculate the average roughness of the 5 sampling points to determine the passivation layer quality: if the average roughness is ≤0.5nm, the pretreatment is qualified; if it is >0.5nm, the battery cell is sent to a high-temperature processing device and kept at 580-620℃ for 25-35 minutes for secondary passivation treatment. Battery cells that still fail the test are transferred to the rejection process; qualified battery cells enter the multi-dimensional parameter sorting stage in step S2.

[0027] Step S2 mainly addresses the shortcomings of existing technologies that rely solely on Voc / Isc binary parameters for sorting and ignore the differences in bifacial spectral response and contact resistance of BC batteries. By using multi-dimensional parameter sorting, performance consistency can be improved.

[0028] See the instruction manual appendix Figure 3 The process of collecting multi-dimensional parameters from pre-processed qualified battery cells, determining parameter weights using the analytic hierarchy process (AHP) combined with the entropy weight method, and calculating a comprehensive score based on the weighted average includes the following steps: S201. For pre-treated qualified solar cells, use a xenon lamp monochromator to collect bifacial spectral response values ​​and calculate the bifacial spectral response difference coefficient; use a pulsed light source IV test system to collect open-circuit voltage Voc and short-circuit current Isc and calculate the fill factor FF; use a four-probe tester to collect contact resistance values ​​and calculate the standard deviation of contact resistance values. S202. Integrate the bifacial spectral response difference coefficient, fill factor FF, and standard deviation of contact resistance to construct a single cell parameter dataset; S203. A hierarchical model is established using the analytic hierarchy process combined with the entropy weight method to determine the weights of the spectral response difference coefficient, the fill factor FF, and the standard deviation of the contact resistance value; and the single cell parameter dataset is input into the hierarchical module to calculate the comprehensive score.

[0029] In step S201, the pre-treated qualified solar cells are fixed on the test stage, ensuring good contact between the cells and the test electrodes. Targeting the core characteristic of BC cells being illuminated from both sides, a xenon lamp monochromator is used to provide segmented monochromatic light within a wavelength range of 300-1200nm (50nm intervals) to illuminate the front and back sides of the cells respectively. A photodetector synchronously collects the bifacial spectral response values ​​at each wavelength, calculates the bifacial spectral response difference coefficient, and records it. Furthermore, a pulsed light source IV test system is used, with a pulse width of 10ms and a light intensity of 1000W / m², to collect the open-circuit voltage Voc, short-circuit current Isc, and fill factor FF electrical performance parameters of the cells, controlling the test error within ±0.5%. Finally, a four-probe tester is used, setting 20 test points at the grid line intersections of the cells to collect the contact resistance parameters of the buried electrodes, calculate the average value and standard deviation, and record them, comprehensively covering key performance dimensions.

[0030] In step S202, all collected parameters, including spectral response difference coefficient, Voc, Isc, FF, average contact resistance, and standard deviation, are integrated to construct a single-cell parameter dataset. A grading model is established using the analytic hierarchy process (AHP) combined with the entropy weighting method, with weights of 0.42 for the spectral response difference coefficient, 0.29 for the fill factor, and 0.29 for the contact resistance standard deviation. The single-cell parameter data is input into the grading model, and a comprehensive score is calculated. Based on the comprehensive score, the cells are divided into six levels, ensuring that the coefficient of variation of parameters for cells within the same level is ≤3%. The graded cells are then labeled and classified. Levels 1-3 are considered qualified and repairable products and proceed to the full-process non-destructive testing stage in step S3. Levels 4-5 are considered severely defective products and are transferred to a secondary passivation or rejection process.

[0031] Step S3 mainly addresses the problem of high false negative rate in single EL testing and inability to identify microcracks and packaging defects in existing technologies. It ensures component reliability through full-process non-destructive testing.

[0032] See the instruction manual appendix Figure 4 The process of performing infrared thermal imaging and electroluminescence (EL) detection on selected grade battery cells, and then screening out qualified and repairable products for sequential stringing, forming inspection, and packaging re-inspection to obtain qualified products includes the following steps: S301. Perform infrared thermal imaging and electroluminescence (EL) detection on selected grade solar cells, and fuse and analyze the temperature distribution and electroluminescence images to screen out qualified repairable products based on the defect area. S302. The selected qualified repairable products are serially welded to obtain the formed components. During the serial welding process, a laser contour sensor and a high-definition vision inspection module are activated simultaneously to adjust the serial welding parameters based on the real-time monitoring of the solder strip alignment deviation and welding temperature fluctuation. S303. The obtained molded components are tested for sealing performance by ultraviolet light irradiation and the IV curve is tested by steady-state solar simulator. Photoelectric performance parameters are collected to determine whether the molded components are qualified. S304. After conducting cold and hot cycle tests on qualified molded components, retest the photoelectric performance, determine qualified products based on the calculated power attenuation rate, and package and ship them out of the warehouse.

[0033] In step S301, the sorted grade 1-3 solar cells are sent to the testing equipment. An infrared thermal imaging module with a temperature resolution ≥0.02℃ is activated to scan and acquire images of the cell temperature distribution. Simultaneously, an EL detection module with a pixel resolution ≥1024×1024 is activated to acquire electroluminescence images of the cells. The infrared thermal imaging and EL detection data are fused, and the solar cells are divided into 5 grades based on defect area. Grade 1-3 products proceed to step S302 for string bonding, while grade 4-5 products undergo secondary passivation or rejection.

[0034] In step S302, the same-level solar cells are arranged and positioned according to the grading results. After the welding ribbon is laid, the stringing equipment is started, and the initial welding temperature, pressure, and transmission speed parameters are set. During the stringing process, the laser contour detection module and the high-definition vision inspection module are started simultaneously to monitor the welding ribbon alignment deviation and welding status in real time. If the detected alignment deviation is 0.05-0.1mm, the welding gun position is automatically fine-tuned; if the deviation is >0.1mm, the machine is stopped, manually adjusted, and then restarted; if the welding temperature fluctuates by ±3-5℃, the welding gun power is linearly adjusted; if the fluctuation is >±5℃, a process review is triggered. After the stringing is completed to form the solar cell string, a visual inspection is performed. After confirming that there are no obvious defects such as welding ribbon detachment or incomplete welding, the module is moved to the module forming stage. This achieves a real-time closed loop of inspection-adjustment-forming.

[0035] In step S303, after the component molding is completed, an ultraviolet fluorescence detection device is used to scan the encapsulation layer to check whether the sealing performance meets the standard. Simultaneously, the IV curve testing system is activated to test the component's open-circuit voltage, short-circuit current, fill factor, and conversion efficiency, and the test results are recorded. If the encapsulation layer sealing performance is unqualified or the photoelectric performance parameters do not meet the preset standards, the process is transferred to the rework stage; qualified components enter the encapsulation stage. This forms a seamless logic of molding-testing-optimization.

[0036] In step S304, the qualified components are placed in a high and low temperature test chamber, and the cyclic conditions are set from -40℃ to 85℃. Five cycles of thermal cycling are performed, each lasting 120 minutes. After the cycle test, the photoelectric performance of the components is tested again, and the power attenuation rate is calculated. If the attenuation rate is ≤2%, it is considered qualified; if it is >2%, the process proceeds to analysis and rework.

[0037] The process involves preparing EVA / POE composite encapsulation material with a POE content of 30%. After inspecting the material for damage and impurities, it is fed into a laminator. Components that have passed thermal cycling tests are placed into the laminator, and the lamination temperature is set to 145-155℃, the pressure to 0.08-0.12MPa, and the lamination time to 110-130s. Lamination and encapsulation are then initiated. After lamination, the components are removed, and the edges are trimmed to remove excess encapsulation material. The encapsulation surface is inspected for defects such as bubbles and wrinkles. An aluminum alloy frame and junction box are installed, and cables are welded out, ensuring secure wiring and adequate insulation performance. A unique QR code containing the component sorting grade, full-process testing data, and key process parameters is generated and laser-marked on a designated location on the component frame or backplate. Components with completed encapsulation and marking undergo final appearance and performance sampling inspections. Components that pass the inspections are then stored for shipment.

[0038] Step S4 primarily addresses the shortcomings of existing technologies, such as lack of closed-loop feedback, inability to pinpoint the causes of substandard performance, and large yield fluctuations. Through data integration and model optimization, it accurately identifies process problems and dynamically optimizes process parameters.

[0039] See the instruction manual appendix Figure 5 The process of importing the full-process data of qualified products into a quality prediction model constructed using the random forest algorithm for feedback optimization includes the following steps: S401. Integrate the full-process data of qualified products, construct a sample database, and use the random forest algorithm to construct a quality prediction model, using cell parameters as input and component efficiency as output for training. S402. Feedback optimization is performed based on the analysis results of the quality prediction model; among which, the parameters for feedback optimization include multi-dimensional parameter sorting thresholds and string welding parameters.

[0040] In step S401, cleanliness test data and passivation layer roughness data from the pretreatment stage are integrated, along with parameters and grading results from the multi-dimensional sorting stage, and supplementary data from each stage of the non-destructive testing process, stringing process adjustment parameters, and photoelectric performance data after module molding and thermal cycling are collected. A standardized database containing over 100,000 samples is constructed. Based on the database data, a quality prediction model is built using the random forest algorithm, with cell parameters as input and module efficiency as output. Model training is completed through processes such as bagging, random feature selection, and majority voting, keeping the prediction error within 0.5%.

[0041] In step S402, based on the model analysis results, precise optimization is performed and fed back to the corresponding nodes: if the proportion of electrode connection defects after stringing of a certain grade of solar cells is ≥2.5%, the contact resistance sorting threshold for that grade is automatically increased by 5-8%; if the module fill factor is 78-80%, the stringing temperature is finely adjusted by ±3℃ and the pressure by ±0.02MPa; if the fill factor is ≤78%, the sorting grade and process parameters are rematched. The optimized sorting threshold and process parameters are fed back to the multi-dimensional parameter sorting stage and the stringing process stage to achieve dynamic iteration of the process. This achieves full collaboration between data-driven, process optimization, and process improvement.

[0042] This application provides a BC cell module manufacturing method based on optimized sorting and inspection. On the one hand, through multi-dimensional parameter sorting, the performance dispersion coefficient of the cells is reduced from 12% to below 5%, and the efficiency deviation of modules of the same grade is ≤0.3%, which is more than 4 times better than the prior art. On the other hand, through closed-loop inspection, the defect missed rate is reduced from 8% to below 1.2%, and the accuracy of composite defect identification reaches 99%; cold and hot cycle re-inspection extends the service life of the module to more than 25 years. Furthermore, through a closed-loop feedback mechanism, the sorting and stringing processes are dynamically matched, resulting in an average improvement of 2.1% in module efficiency, an increase in mass production yield from 89% to 95.3%, and a reduction of 8-10% in cost per watt. Additionally, through a pretreatment closed-loop process, the surface recombination rate of the cells is reduced by 30%, and the performance loss caused by passivation layer defects is reduced from 0.7% to below 0.2%.

[0043] Based on the same inventive concept, this application also provides a BC battery module preparation apparatus based on optimized sorting and detection. Since the principle of the apparatus in this application is similar to the BC battery module preparation method based on optimized sorting and detection described above, the implementation of the apparatus can refer to the implementation of the method, and the repeated parts will not be described again.

[0044] As per the instruction manual Figure 6 As shown in the illustration, this application also provides a BC battery module fabrication apparatus based on optimized sorting and detection, the apparatus comprising: The pretreatment module 601 is used to perform a closed-loop process of plasma cleaning, cleanliness detection, passivation layer detection and secondary treatment on the cut BC solar cells to obtain qualified pretreatment solar cells. The multi-dimensional parameter sorting module 602 is used to collect multi-dimensional parameters of pre-processed qualified solar cells, and uses the analytic hierarchy process combined with the entropy weight method to determine the parameter weights. Based on the comprehensive score obtained by weighted calculation, the solar cells are divided into multiple grades. The collected multi-dimensional parameters include bifacial spectral response parameters, open circuit voltage Voc, short circuit current Isc, fill factor FF, and buried electrode contact resistance parameters. The full-process non-destructive testing module 603 is used to perform infrared thermal imaging and electroluminescence (EL) testing on selected grades of solar cells, and to screen out qualified and repairable products for sequential stringing, forming inspection, and packaging re-inspection to obtain qualified products. The feedback optimization module 604 is used to import the full-process data of qualified products into the quality prediction model built using the random forest algorithm for feedback optimization.

[0045] In some embodiments, the pretreatment module 601 performs a closed-loop process of plasma cleaning, cleanliness detection, passivation layer detection, and secondary treatment on the cut BC solar cells to obtain pretreated qualified solar cells. This includes: using a plasma cleaner to perform plasma cleaning on the cut BC solar cells and testing the surface oil residue using a gravimetric method; wherein, BC solar cells whose surface oil residue does not meet the requirements are re-plasma cleaned until the requirements are met; for BC solar cells whose surface oil residue meets the requirements, five sampling points are selected at the center and four corners using an atomic force microscope to test the surface roughness; wherein, BC solar cells whose surface roughness meets the requirements are determined to have a qualified passivation layer and are considered as pretreated qualified solar cells; BC solar cells whose surface roughness does not meet the requirements are re-passivated.

[0046] In some embodiments, the multi-dimensional parameter sorting module 602 collects multi-dimensional parameters of pre-processed qualified solar cells and uses the analytic hierarchy process (AHP) combined with the entropy weight method to determine the parameter weights. Based on the comprehensive score obtained by weighted calculation, the process includes: for pre-processed qualified solar cells, collecting bifacial spectral response values ​​using a xenon lamp monochromator and calculating the bifacial spectral response difference coefficient; collecting open-circuit voltage Voc and short-circuit current Isc using a pulsed light source IV test system and calculating the fill factor FF; collecting contact resistance values ​​using a four-probe tester and calculating the standard deviation of the contact resistance value; integrating the bifacial spectral response difference coefficient, fill factor FF, and standard deviation of the contact resistance value to construct a single-cell parameter dataset; establishing a grading model using the AHP combined with the entropy weight method to determine the weights of the spectral response difference coefficient, fill factor FF, and standard deviation of the contact resistance value; and inputting the single-cell parameter dataset into the grading module to calculate the comprehensive score.

[0047] In some embodiments, the full-process non-destructive testing module 603 performs infrared thermal imaging and electroluminescence (EL) testing on selected grades of solar cells, and selects qualified repairable products for sequential stringing, forming inspection, and packaging re-inspection to obtain qualified products. This includes: performing infrared thermal imaging and EL testing on selected grades of solar cells, fusing and analyzing temperature distribution and EL images, and selecting qualified repairable products based on defect area; stringing the selected qualified repairable products to obtain formed components; wherein, during the stringing process, a laser contour sensor and a high-definition vision inspection module are simultaneously activated, and stringing parameters are adjusted based on real-time monitoring of solder strip alignment deviation and welding temperature fluctuations; the obtained formed components are subjected to ultraviolet light irradiation for sealing inspection, and IV curves are tested using a steady-state solar simulator to collect photoelectric performance parameters and determine whether the formed components are qualified; after the qualified formed components undergo cold and hot cycle testing, the photoelectric performance is retested, and qualified products are determined based on the calculated power attenuation rate, and then packaged and shipped.

[0048] In some embodiments, the feedback optimization module 604 imports the full-process data of qualified products into a quality prediction model constructed using a random forest algorithm for feedback optimization, including: integrating the full-process data of qualified products to construct a sample database, and constructing a quality prediction model using a random forest algorithm, training it with cell parameters as input and component efficiency as output; and performing feedback optimization based on the analysis results of the quality prediction model; wherein the parameters for feedback optimization include multi-dimensional parameter sorting thresholds and stringing parameters.

[0049] The BC battery module manufacturing apparatus based on optimized sorting and detection described in this application improves sorting accuracy, reduces defect omission rate, and optimizes process through multi-dimensional parameter sorting, closed-loop detection throughout the process, and dynamic optimization, thereby achieving efficient and reliable production.

[0050] Based on the same concept of the present invention, as shown in the appendix to the specification. Figure 7 As shown in the figure, an embodiment of this application provides the structure of an electronic device 700, which includes: at least one processor 701, at least one network interface 704 or other user interface 703, a memory 705, and at least one communication bus 702. The communication bus 702 is used to realize the connection and communication between these components. The electronic device 700 may optionally include a user interface 703, including a display (e.g., touch screen, LCD, CRT, holographic imaging, or projector, etc.), a keyboard, or a clicking device (e.g., mouse, trackball, touchpad, or touch screen, etc.).

[0051] Memory 705 may include read-only memory and random access memory, and provides instructions and data to processor 701. A portion of memory 705 may also include non-volatile random access memory (NVRAM).

[0052] In some implementations, memory 705 stores executable modules or data structures, or subsets thereof, or extended sets thereof: The 7051 operating system contains various system programs used to implement various basic business functions and handle hardware-based tasks. Application module 7052 contains various applications, such as desktop (launcher), media player (MediaPlayer), browser (Browser), etc., to implement various application services.

[0053] In this embodiment of the application, the processor 701 is used to execute steps such as a BC battery assembly preparation method based on optimized sorting and detection by calling the program or instructions stored in the memory 705.

[0054] This application also provides a computer-readable storage medium storing a computer program that is executed by a processor, such as the steps in a method for preparing a BC battery module based on optimized sorting and detection.

[0055] Specifically, the storage medium can be a general-purpose storage medium, such as a portable disk or hard disk. When the computer program on the storage medium is run, it can improve sorting accuracy and reduce the defect detection rate.

[0056] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the mutual coupling or direct coupling or communication connection shown or discussed may be through some communication interface, and the indirect coupling or communication connection of the apparatus or units may be electrical, mechanical, or other forms.

[0057] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0058] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0059] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0060] Finally, it should be noted that the above embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for preparing BC battery modules based on optimized sorting and detection, characterized in that, The method includes the following steps: A closed-loop process of plasma cleaning, cleanliness testing, passivation layer testing, and secondary treatment is performed on the cut BC solar cells to obtain pre-treated qualified solar cells. Multi-dimensional parameters were collected from pre-processed qualified solar cells, and the analytic hierarchy process (AHP) combined with the entropy weight method was used to determine the parameter weights. Based on the comprehensive score obtained by weighted calculation, the solar cells were divided into multiple grades. The multi-dimensional parameters collected included bifacial spectral response parameters, open-circuit voltage Voc, short-circuit current Isc, fill factor FF, and buried electrode contact resistance parameters. Infrared thermal imaging and electroluminescence (EL) testing were performed on selected grade solar cells, and qualified repairable products were screened out and then sequentially subjected to stringing, forming testing, and packaging re-inspection to obtain qualified products. The entire process data of qualified products is imported into a quality prediction model built using the random forest algorithm for feedback optimization.

2. The method for preparing BC battery modules based on optimized sorting and detection according to claim 1, characterized in that, The closed-loop process of performing plasma cleaning, cleanliness testing, passivation layer testing, and secondary treatment on the cut BC solar cells to obtain pre-treated qualified solar cells includes the following steps: The cut BC solar cells were cleaned using a plasma cleaner, and the amount of residual oil on the surface was tested by gravimetric method. BC solar cells that did not meet the requirements for the amount of residual oil were re-cleaned by plasma until the requirements were met. For BC solar cells with sufficient residual oil on the surface, five sampling points were selected at the center and four corners using an atomic force microscope to test the surface roughness. Among them, BC solar cells with satisfactory surface roughness were judged to have a qualified passivation layer and were considered as qualified pre-treatment solar cells. BC solar cells with unsatisfactory surface roughness were subjected to passivation treatment again.

3. The method for preparing a BC battery module based on optimized sorting and detection according to claim 1, characterized in that, The process of collecting multi-dimensional parameters from pre-processed qualified solar cells, determining parameter weights using the analytic hierarchy process (AHP) combined with the entropy weight method, and calculating a comprehensive score based on the weighted average includes the following steps: For pre-treated qualified solar cells, the bifacial spectral response values ​​are collected using a xenon lamp monochromator and the bifacial spectral response difference coefficient is calculated; the open-circuit voltage Voc and short-circuit current Isc are collected using a pulsed light source IV test system and the fill factor FF is calculated; the contact resistance values ​​are collected using a four-probe tester and the standard deviation of the contact resistance values ​​is calculated. A single-cell parameter dataset is constructed by integrating the bifacial spectral response difference coefficient, fill factor FF, and standard deviation of contact resistance. A hierarchical model was established using the analytic hierarchy process combined with the entropy weight method to determine the weights of the spectral response difference coefficient, the fill factor FF, and the standard deviation of the contact resistance value. The single cell parameter dataset was then input into the hierarchical module to calculate the comprehensive score.

4. The method for preparing a BC battery module based on optimized sorting and detection according to claim 1, characterized in that, The process of performing infrared thermal imaging and electroluminescence (EL) testing on selected grade solar cells, and then screening out qualified and repairable products for sequential stringing, forming inspection, and packaging re-inspection to obtain qualified products includes the following steps: Infrared thermal imaging and electroluminescence (EL) detection are performed on selected grade solar cells, and the temperature distribution and electroluminescence images are fused and analyzed to select qualified repairable products based on the defect area. The qualified and repairable products selected are then serially welded to obtain the finished components. During the serial welding process, a laser contour sensor and a high-definition vision inspection module are simultaneously activated to adjust the serial welding parameters based on real-time monitoring of the solder strip alignment deviation and welding temperature fluctuations. The obtained molded components were tested for sealing performance by ultraviolet light irradiation and IV curves were tested by a steady-state solar simulator. Photovoltaic performance parameters were collected to determine whether the molded components were qualified. After conducting thermal cycling tests on qualified molded components, the photoelectric performance is retested. Based on the calculated power attenuation rate, qualified products are determined and packaged for shipment.

5. The method for preparing a BC battery module based on optimized sorting and detection according to claim 4, characterized in that, in, Based on the defect area, the solar cells are divided into five levels. Levels 4 and 5 are considered defective products and are rejected; levels 1 to 3 are considered qualified and repairable products and proceed to the stringing process.

6. The method for preparing a BC battery module based on optimized sorting and detection according to claim 4, characterized in that, in, If the weld strip alignment deviation is detected to be 0.05-0.1mm, the welding gun position should be finely adjusted. If the welding temperature fluctuates by ±3-5℃, adjust the welding torch power linearly.

7. The method for preparing a BC battery module based on optimized sorting and detection according to claim 1, characterized in that, The process of importing the full-process data of qualified products into a quality prediction model constructed using the random forest algorithm for feedback optimization includes the following steps: The entire process data of qualified products is integrated to build a sample database, and a quality prediction model is built using the random forest algorithm, with cell parameters as input and component efficiency as output for training. Feedback optimization is performed based on the analysis results of the quality prediction model; the parameters for feedback optimization include multi-dimensional parameter sorting thresholds and string welding parameters.

8. A BC battery module fabrication apparatus based on optimized sorting and detection, characterized in that, The device includes: The pretreatment module is used to perform a closed-loop process of plasma cleaning, cleanliness detection, passivation layer detection and secondary treatment on the cut BC solar cells to obtain qualified pretreatment solar cells. The multi-dimensional parameter sorting module is used to collect multi-dimensional parameters of pre-processed qualified solar cells, and uses the analytic hierarchy process combined with the entropy weight method to determine the parameter weights. Based on the comprehensive score obtained by weighted calculation, the solar cells are divided into multiple grades. The multi-dimensional parameters collected include bifacial spectral response parameters, open circuit voltage Voc, short circuit current Isc, fill factor FF, and buried electrode contact resistance parameters. The full-process non-destructive testing module is used to perform infrared thermal imaging and electroluminescence (EL) testing on selected grades of solar cells, and to screen out qualified and repairable products for sequential stringing, forming inspection, and packaging re-inspection to obtain qualified products. The feedback optimization module is used to import the full-process data of qualified products into a quality prediction model built using the random forest algorithm for feedback optimization.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of a BC battery assembly fabrication method based on optimized sorting and detection as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of a BC battery module preparation method based on optimized sorting and detection as described in any one of claims 1 to 7.

Images