Photovoltaic glass production process optimization system based on detection data feedback

Abstract

The invention discloses a photovoltaic glass production process optimization system based on detection data feedback, and relates to the technical field of photovoltaic production, and the system comprises a production process and data management end which is responsible for extracting predefined production process parameter groups from a database, and constructing an initial photovoltaic glass group set based on the parameter groups; afterwards, the light transmittance fluctuation characteristics of the photovoltaic glass are quantitatively evaluated through professional detection equipment in the photovoltaic glass intelligent detection end, and a comprehensive light transmittance fluctuation broken line is obtained through mathematical simulation; the collaborative optimization end combines the normalized light transmittance and the fluctuation characteristics to calculate the comprehensive index of each process group, so that the optimal process parameter group is accurately positioned; and finally, the iteration feedback end dynamically updates the optimal process parameters in a preset period to ensure that the production is always in an optimal state. The whole system remarkably improves the efficiency and quality of photovoltaic glass production through a closed-loop feedback mechanism, and meanwhile, the system adapts to the continuous improvement requirement of the performance of a photovoltaic material.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic production technology, specifically, it relates to a photovoltaic glass production process optimization system based on feedback from detection data. Background Technology

[0002] With the advancement of global energy transition, the photovoltaic industry, as an important pillar of clean energy, is developing rapidly. As the core component of photovoltaic modules, the optimization of photovoltaic glass production processes has become a key link in improving photovoltaic conversion efficiency and reducing costs.

[0003] In existing technologies, the optimization of photovoltaic glass production processes typically relies on operator experience or fixed parameter settings, lacking a systematic data acquisition and feedback mechanism. For example, using a single transmittance testing method or sampling measurements only at specific wavelengths instead of covering the entire visible light spectrum leads to incomplete transmittance assessments, failing to capture performance fluctuations of photovoltaic glass at different wavelengths, thus affecting its accurate reflection of solar energy conversion efficiency. Furthermore, existing technology's process parameter adjustments are mostly based on offline analysis after mass production, rather than real-time or near-real-time testing data, resulting in optimization lags and difficulty in quickly responding to variations during production. For instance, traditional methods may only test transmittance on a few samples and then adjust the process based on experience, but lack comprehensive analysis of the entire photovoltaic glass assembly, failing to quantify the correlation between process parameters and transmittance fluctuations, easily introducing subjective biases, and leading to poor production consistency. At the same time, existing technologies often lack iterative update mechanisms; once process parameters are set, they are used for a long time, unable to adapt to dynamic factors such as changes in raw materials or equipment wear, causing the production process to gradually deviate from the optimal state, affecting the long-term quality and efficiency of photovoltaic glass.

[0004] To address the aforementioned issues, this invention proposes a photovoltaic glass manufacturing process optimization system based on feedback from detection data. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a photovoltaic glass production process optimization system based on detection data feedback, which solves the problems of insufficient process optimization, lack of data-driven approaches, and iterative updates in existing photovoltaic glass production technologies.

[0006] The objective of this invention can be achieved through the following technical solutions: A photovoltaic glass manufacturing process optimization system based on feedback from detection data, the system comprising: The production process and data management terminal retrieves the set of production process groups predefined by the operators from the production process parameter database, and constructs an initial photovoltaic glass group set corresponding to the total number of production process groups in the production process group set. Each initial photovoltaic glass assembly is processed using different production process groups to determine the final set of photovoltaic glass assemblies. The photovoltaic glass intelligent inspection terminal places each photovoltaic glass in the photovoltaic glass group in the photovoltaic glass group set on a pre-constructed inspection platform, performs inspection operations, evaluates the transmittance fluctuation line associated with each photovoltaic glass, and fits the transmittance fluctuation line of the photovoltaic glass in the same photovoltaic glass group to determine the comprehensive transmittance fluctuation line associated with the production process group corresponding to the photovoltaic glass group. On the collaborative optimization side, the overall transmittance fluctuation lines associated with each production process group in the production process group set are analyzed collaboratively to determine the optimal overall transmittance fluctuation line and its corresponding optimal production process parameter set, and then the photovoltaic glass is processed using the optimal production process parameter set. The iterative feedback end performs update operations on the optimal production process parameter set based on a preset update cycle.

[0007] As a further aspect of the present invention, the specific method for constructing the initial photovoltaic glass group set corresponding to the total number of production process groups in the production process group set in the production process and data management terminal is as follows: Determine the target specifications for photovoltaic glass; Extract the predefined set of production process groups A1, A2, ..., Aj associated with the photovoltaic glass of the target specification from the production process parameter database, where j is the total number of production process groups. Any production process parameter group Ai corresponds to all production process parameters involved in the entire processing flow of the photovoltaic glass of the target specification, where i is the counting index, 1≤i≤j. For a production process parameter group Ai, an initial photovoltaic glass group Bi is constructed, wherein the total number of photovoltaic glasses in the initial photovoltaic glass group Bi is d, where d is a preset integer. Similarly, construct the initial photovoltaic glass sets corresponding to each production process group in the production process group set A1, A2, ..., Aj, and sort them according to the order of the production process group set A1, A2, ..., Aj to obtain the initial photovoltaic glass set set B1, B2, ..., Bj.

[0008] As a further aspect of the present invention, the specific method for determining the photovoltaic glass assembly obtained after processing in the production process and data management terminal is as follows: Extract any one initial photovoltaic glass group Bi from the initial photovoltaic glass group set B1, B2, ..., Bj, and process d photovoltaic glass units in the initial photovoltaic glass group Bi using the production process parameter group Ai corresponding to the initial photovoltaic glass group Bi. Extract d photovoltaic glass units processed by production process parameter group Ai, and denot them as photovoltaic glass group Ci: c1, c2, ..., cd according to the processing order, where c1 to cd represent the first to last photovoltaic glass units processed; Similarly, the photovoltaic glass groups associated with each initial photovoltaic glass group in the initial photovoltaic glass group set B1, B2, ..., Bj are determined and summarized as the photovoltaic glass group set C1, C2, ..., Cj.

[0009] As a further embodiment of the present invention, the detection station in the photovoltaic glass intelligent inspection terminal integrates a photometer and a tunable visible light source. The photometer is located below the detection stage, the tunable visible light source is located above the detection stage, and the detection stage includes an opening for embedding photovoltaic glass; The tunable visible light source has a wavelength range of 380nm to 750nm.

[0010] As a further aspect of the present invention, the specific method for evaluating the transmittance fluctuation curve associated with each photovoltaic glass in the photovoltaic glass intelligent inspection terminal is as follows: S51. Obtain any photovoltaic glass group Ci: c1, c2, ..., cd from the set of photovoltaic glass groups C1, C2, ..., Cj; S52. Extract any photovoltaic glass co from c1, c2, ..., cd, where o is the counting index and 1 ≤ o ≤ d; S53. Embed the photovoltaic glass co into the opening of the detection stage, turn on the tunable visible light source, and record the current time t_str; S54. Adjust the wavelength of the tunable visible light source to increase uniformly from 380nm to 750nm. The rate of increase is determined by the operator. Record the time t_end when the wavelength of the light source is adjusted to 750nm. Sum the total number of time intervals from time t_str to time t_end, and denot it as m. S55. Using a photometer, record the m transmittance values ​​of the photovoltaic glass co at m time points in chronological order to obtain the transmittance sequence T1, T2, ..., Tm associated with the photovoltaic glass co. S56. Construct a two-dimensional coordinate system with the timeline as the horizontal axis and the transmittance as the vertical axis. The time span on the horizontal axis is from the first time to the mth time. Determine the wavelength of the tunable visible light source at the first time and assign it to the first time on the horizontal axis. Similarly, by assigning the corresponding light source wavelength to each of the m time points, we obtain a two-dimensional coordinate system P1 with the light source wavelength and time point coexisting on the horizontal axis and the transmittance on the vertical axis. S57. Plot the transmittance sequence T1, T2, ..., Tm associated with photovoltaic glass co in the form of data points in the two-dimensional coordinate system P1 according to the time sequence to obtain m data points. Connect the adjacent data points with short lines to obtain a broken line, which is denoted as the transmittance fluctuation broken line Z_co. S58. Similarly, determine the transmittance fluctuation line associated with each of the d photovoltaic glass units in the photovoltaic glass group Ci: c1, c2, ..., cd, and represent it as Z_c1, Z_c2, ..., Z_cd; S59. Similarly, determine the transmittance fluctuation curve of all photovoltaic glasses in all photovoltaic glass groups in the photovoltaic glass group set C1, C2, ..., Cj.

[0011] As a further aspect of the present invention, the specific method for determining the comprehensive transmittance fluctuation line associated with the production process group corresponding to the photovoltaic glass group in the photovoltaic glass intelligent inspection terminal is as follows: From the set of photovoltaic glass groups C1, C2, ..., Cj, obtain any photovoltaic glass group Ci, and extract the transmittance fluctuation lines associated with each of the d photovoltaic glass groups in Ci: Z_c1, Z_c2, ..., Z_cd; Take the average of the transmittance corresponding to the first time point of the d transmittance fluctuation lines, and use it as the comprehensive transmittance corresponding to the first time point. Similarly, determine the comprehensive transmittance corresponding to each of the m time points, and sort the m comprehensive transmittances in time order to obtain the comprehensive transmittance sequence. Copy any two-dimensional coordinate system P1 and clear it; Using the comprehensive transmittance sequence, following the steps described in step S57, construct a comprehensive transmittance fluctuation line associated with the photovoltaic glass group Ci, denoted as Z_Ci; Similarly, the overall transmittance fluctuation lines associated with each photovoltaic glass group in the photovoltaic glass group set C1, C2, ..., Cj are determined and denoted as the overall transmittance fluctuation line set Z_C1, Z_C2, ..., Z_Cj.

[0012] As a further aspect of the present invention, the specific method for determining the optimal comprehensive transmittance fluctuation line and its corresponding optimal production process parameter set in the collaborative optimization end is as follows: Determine the average transmittance associated with each of the comprehensive transmittance fluctuation lines in the set Z_C1, Z_C2, ..., Z_Cj, and denote them as the average transmittance sequence T_avg_C1, T_avg_C2, ..., T_avg_Cj in the order of the comprehensive transmittance fluctuation line set Z_C1, Z_C2, ..., Z_Cj; The j average transmittance values ​​in the average transmittance sequence T_avg_C1, T_avg_C2, ..., T_avg_Cj are normalized to the zero-to-one interval according to the method that 0% transmittance corresponds to 0 and 100% transmittance corresponds to 1, so as to obtain the normalized transmittance sequence T_gy_C1, T_gy_C2, ..., T_gy_Cj; Take any one of the comprehensive transmittance fluctuation lines Z_Ci from the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj and its corresponding two-dimensional coordinate system P1; By passing through the scale of the average transmittance T_avg_Ci on the vertical axis of the two-dimensional coordinate system P1, a straight line L1 is constructed that is parallel to the horizontal axis and perpendicular to the vertical axis. A straight line is constructed by passing through the data points corresponding to the first time point and the data points corresponding to the m-th time point on the comprehensive transmittance fluctuation line Z_Ci, respectively. The line intersects the straight line L1 and is perpendicular to the straight line L1, perpendicular to the horizontal axis, and parallel to the vertical axis. These lines are denoted as L2 and L3, respectively. Extract the area values ​​of several closed regions formed by the comprehensive transmittance fluctuation line Z_Ci, line L1, line L2 and line L3, and denote them as fluctuation characteristics R_Ci; Similarly, determine the fluctuation characteristics associated with each of the comprehensive transmittance fluctuation lines in the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj, and summarize them as the fluctuation characteristic set R_C1, R_C2, ..., R_Cj; Normalize j fluctuation features in the fluctuation feature set R_C1,R_C2,...,R_Cj to the zero-one interval to obtain the normalized fluctuation feature set R_gy_C1,R_gy_C2,...,R_gy_Cj; Subtract each normalized fluctuation feature from 1 in turn to obtain the set of normalized complementary fluctuation features R_gb_C1, R_gb_C2, ..., R_gb_Cj; Take the set of broken lines Z_Ci associated with the overall transmittance fluctuation of any photovoltaic glass group Ci; Extract the normalized transmittance T_gy_Ci and the normalized complement fluctuation feature R_gb_Ci associated with the set of broken lines of comprehensive transmittance fluctuation Z_Ci. Calculate the comprehensive index K_Ci using T_gy_Ci×α+R_gb_Ci×β=K_Ci, where α and β are preset calculation weights, α>0, β>0, α+β=1; Similarly, the comprehensive index associated with each comprehensive transmittance fluctuation line in the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj is determined and denoted as K_C1, K_C2, ..., K_Cj, where K_Ci corresponds to Z_Ci; Extract the comprehensive transmittance fluctuation line Z_Ci corresponding to the maximum comprehensive index K_Ci, obtain the production process parameter group Ai corresponding to the comprehensive transmittance fluctuation line Z_Ci, and mark the comprehensive transmittance fluctuation line Z_Ci and the production process parameter group Ai as the optimal comprehensive transmittance fluctuation line and the optimal production process parameter group, respectively.

[0013] As a further aspect of the present invention, the collaborative optimization end also includes processing the subsequent photovoltaic glass with the determined optimal production process parameter set.

[0014] As a further aspect of the present invention, the specific method for performing the update operation on the optimal production process parameter set in the iterative feedback terminal is as follows: When the optimal production process parameter set is determined, record the current time and extract the update cycle preset by the operator. Use the current time as the start time of the update cycle and update the current time in real time. If the time interval between the current time and the start time reaches the update cycle, an update is triggered, and the optimal production process parameter set is re-determined.

[0015] The beneficial effects of this invention are: This invention utilizes data-driven decision-making, accurately assessing process performance through intelligent detection and fitting of transmittance fluctuation lines, thereby quickly identifying optimal production process parameters and reducing reliance on human experience. Simultaneously, an iterative feedback mechanism ensures dynamic optimization of process parameters, adapting to changes in the production environment, effectively reducing volatility and scrap rates, and enhancing product consistency and reliability. Overall, it achieves automation, precision, and adaptive optimization in the photovoltaic glass production process, significantly improving production efficiency and product competitiveness. This invention achieves high-precision, full-visible-spectrum detection of the transmittance of photovoltaic glass by integrating a photometer and a tunable visible light source, thereby comprehensively evaluating its optical performance. Its advantages include an automated measurement process that reduces human error and improves detection efficiency; simultaneously, by constructing a transmittance fluctuation line and aggregating multiple glass data into a comprehensive line, it can intuitively reveal the overall quality fluctuations of the production process group, providing reliable data support for optimizing the production process and effectively improving the consistency and quality control level of photovoltaic glass. This invention normalizes the average transmittance and extracts its fluctuation characteristics, then calculates a comprehensive index to quantitatively evaluate the performance of each production process parameter group. This accurately determines the optimal comprehensive transmittance fluctuation line and the corresponding optimal production process parameter group. It achieves data-driven objective decision-making, comprehensively considering the average value and fluctuation of transmittance to ensure that photovoltaic glass not only has high transmittance but also stable performance. Secondly, by applying the optimal parameter group for subsequent processing, the consistency and reliability of photovoltaic glass are improved, variations and defects in the production process are reduced, production quality is optimized, and scrap rate and costs are reduced, providing a solid guarantee for sustainable production. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 This is a schematic diagram of the system described in this invention; Figure 2 This is a flowchart illustrating the method described in Embodiment 2 of the present invention; Figure 3 This is a flowchart illustrating the method described in Embodiment 3 of the present invention; Figure 4 This is a flowchart illustrating the method described in Embodiment 4 of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1 A photovoltaic glass manufacturing process optimization system based on feedback from testing data, such as Figure 1 As shown, this system includes the following: This system is a photovoltaic glass production process optimization system based on detection data feedback. It mainly optimizes the production processes involved in the production of photovoltaic glass. The principle is as follows: In the process of converting solar energy into electrical energy, photovoltaic panels mainly collect visible light from sunlight and convert the radiation of visible light into electrical energy. Based on this characteristic, it can be known that photovoltaic glass needs to have a high transmittance in the visible light band (380nm to 750nm) as much as possible to improve the absorption efficiency of solar energy by photovoltaic panels. Therefore, the transmittance of photovoltaic glass in the visible light band can become a core indicator for measuring its quality. Based on this, this method was constructed to optimize the photovoltaic glass production process. It should be noted that all photovoltaic glasses involved in this solution are photovoltaic glasses that have undergone preliminary inspection and do not have defective glass, such as obvious defects such as cracks, scratches, and bubbles that would render them unusable.

[0020] This system mainly includes: a production process and data management terminal, a photovoltaic glass intelligent inspection terminal, a collaborative optimization terminal, and an iterative feedback terminal. In addition, it also includes a production process parameter database pre-built by operators.

[0021] The production process and data management terminal is mainly used to obtain the set of production process groups predefined by the operators from the production process parameter database, and at the same time construct an initial photovoltaic glass group set corresponding to the total number of production process groups in the production process group set. Then, each initial photovoltaic glass assembly is processed using different production process groups, and the processed initial photovoltaic glass assembly set is marked as the photovoltaic glass assembly set.

[0022] In the photovoltaic glass intelligent inspection terminal, the photovoltaic glass group set obtained after processing by the production process and data management terminal is first acquired. Then, each photovoltaic glass in all photovoltaic glass groups in the photovoltaic glass group set is placed on the inspection table pre-constructed by the operator. The inspection table is used to perform inspection operations on each photovoltaic glass. During the inspection process, the transmittance of each photovoltaic glass in different visible light bands is recorded in real time, and the transmittance fluctuation line associated with each photovoltaic glass is evaluated accordingly. Next, the transmittance fluctuation lines of all photovoltaic glass in the same group (photovoltaic glass group) are extracted and fitted. After fitting, a unique line is obtained, which is used as the comprehensive transmittance fluctuation line associated with the production process group corresponding to this photovoltaic glass group (photovoltaic glass group and production process group correspond).

[0023] In the collaborative optimization process, the comprehensive transmittance fluctuation line obtained after processing by the photovoltaic glass intelligent inspection end is acquired. Since each photovoltaic glass group (production process group) corresponds to a comprehensive transmittance fluctuation line, Concord ultimately obtains several comprehensive transmittance fluctuation lines. The several comprehensive transmittance fluctuation lines are collaboratively analyzed to finally determine the unique optimal comprehensive transmittance fluctuation line and the optimal production process parameter set corresponding to this optimal comprehensive transmittance fluctuation line. This optimal production process parameter set is then used as the production process parameter set for subsequent processing of photovoltaic glass.

[0024] The iterative feedback terminal mainly performs iterative updates on the determined optimal production process parameter set. The optimal production process parameter set is updated every certain period of time, which is preset by the operator. All the above steps are repeated to perform one update.

[0025] This system aims to optimize the photovoltaic glass manufacturing process to improve its transmittance in the visible light band (380nm to 750nm), thereby enhancing the photovoltaic panel's solar energy absorption efficiency. By acquiring the set of production process groups predefined by the operators, an initial photovoltaic glass group set is constructed, and the initial photovoltaic glass group is processed using different production process groups to form a photovoltaic glass group set. Subsequently, the transmittance of each photovoltaic glass in the photovoltaic glass array is tested, the transmittance fluctuation line is evaluated, and the comprehensive transmittance fluctuation line is obtained by fitting. Through collaborative analysis of these lines, the system determines the optimal comprehensive transmittance fluctuation line and its corresponding optimal production process parameter set for subsequent production. Meanwhile, this system has an iterative feedback function, which updates the optimal production process parameter set according to a preset cycle to ensure continuous optimization and stability of the production process. This enables closed-loop management from data acquisition and detection analysis to process optimization, reducing reliance on human experience and promoting the development of photovoltaic glass production towards refinement and intelligence. It provides the photovoltaic industry with an efficient and dynamic production process optimization solution.

[0026] Example 2 This embodiment further discloses the methods and steps involved in the production process and data management terminal based on embodiment 1, such as... Figure 2 As shown, it specifically includes the following: First, determine any target specification of photovoltaic glass. After determining the target specification of photovoltaic glass, obtain the set of production process groups associated with this target specification of photovoltaic glass (predefined by the operator) from the production process parameter database pre-built by the operator. All production process groups in the set of production process groups are sorted according to the order of acquisition. The sorted result is represented as: A1, A2, ..., Aj, where j represents the total number of production process groups in the set of production process groups, and A1 to Aj represent the first acquired production process group to the j-th acquired production process group. The production process parameter group Ai in the production process group set A1, A2, ..., Aj corresponds to the production process parameters involved in the entire processing of photovoltaic glass of the target specification (such as melting temperature, time, ratio, rolling speed, pressure, mold temperature, etc.), where i is a counting index and the value of i ranges from 1 to j.

[0027] Based on any one of the production process parameter groups Ai in the production process group set A1, A2, ..., Aj, construct the corresponding initial photovoltaic glass group and denote it as Bi, where the total number of photovoltaic glasses in the initial photovoltaic glass group Bi is d, and d is an integer preset by the operator.

[0028] The above operations determine the initial photovoltaic glass group Bi corresponding to the production process parameter group Ai. Similarly, the initial photovoltaic glass groups corresponding to all production process groups in the production process group set A1, A2, ..., Aj can be determined and recorded as the initial photovoltaic glass group set according to the sorting order of the production process group set A1, A2, ..., Aj, denoted as: B1, B2, ..., Bj.

[0029] Thus, the initial photovoltaic glass set B1, B2, ..., Bj corresponding to the production process set A1, A2, ..., Aj was determined; Then, extract any one initial photovoltaic glass group Bi (corresponding to the production process parameter group Ai) from the determined initial photovoltaic glass group set B1, B2, ..., Bj, and use the production process parameter group Ai to process all the photovoltaic glass in the initial photovoltaic glass group Bi through the entire process. Obtain all the photovoltaic glass in the initial photovoltaic glass group Bi after being processed by the production process parameter group Ai, for a total of d photovoltaic glass. According to the order of processing the d photovoltaic glass, it is denoted as photovoltaic glass group Ci, corresponding to the initial photovoltaic glass group Bi and the production process parameter group Ai. The photovoltaic glass group Ci is represented as: c1, c2, ..., cd, where c1 to cd represent the first photovoltaic glass to the last photovoltaic glass processed; It is important to explain here that in actual implementation, there may be more than one production line for producing photovoltaic glass. Different production lines must not mix with each other, and the uniqueness of the data source must be determined. By analogy, determine the photovoltaic glass groups associated with each initial photovoltaic glass group in the initial photovoltaic glass group set B1, B2, ..., Bj, and sum them up to obtain the photovoltaic glass group set corresponding to the initial photovoltaic glass group set B1, B2, ..., Bj, denoted as: C1, C2, ..., Cj, where Bi corresponds to Ci.

[0030] At this point, the methods and steps involved in the production process and data management have been completed, resulting in the initial photovoltaic glass group set B1, B2, ..., Bj, the photovoltaic glass group set C1, C2, ..., Cj, and the production process group set A1, A2, ..., Aj.

[0031] This embodiment generates a photovoltaic glass group and a complete product set by using a predefined set of production process parameters. Starting with the photovoltaic glass of the target specification, it obtains the set of production process groups related to the specification from the pre-built production process parameter database, and then sorts them in a predetermined order to form the set of production process groups A1, A2, ..., Aj.

[0032] Based on each production process parameter group Ai, a corresponding initial photovoltaic glass group Bi is constructed, and the final photovoltaic glass group Ci is generated during the processing.

[0033] Finally, all initial photovoltaic glass assemblies and their corresponding processing results are summarized into complete initial photovoltaic glass assembly sets B1, B2, ..., Bj and photovoltaic glass assembly sets C1, C2, ..., Cj. Through standardized processes and data management, traceability and data collection capabilities are provided for the photovoltaic glass production process, while ensuring the uniqueness of production data sources from different production lines, providing reliable support for subsequent production optimization and quality analysis.

[0034] Example 3 This embodiment further discloses the methods and steps involved in the intelligent inspection terminal of photovoltaic glass based on embodiment 2, such as... Figure 3 As shown, it specifically includes the following: First, the detection station described in Example 1 is improved. The detection station integrates a photometer and a tunable visible light source. The wavelength range of the tunable visible light source is 380nm to 750nm (corresponding to the visible light band). The photometer is located below the testing platform, and the tunable visible light source is located above the testing platform. The testing platform includes an opening for embedding and placing photovoltaic glass. The photovoltaic glass is illuminated from below by the tunable visible light source, and then the light transmittance of the corresponding photovoltaic glass is measured using the photometer.

[0035] Based on the content described in Example 2, a set of photovoltaic glass groups C1, C2, ..., Cj is obtained, and any photovoltaic glass group Ci is extracted from the set of photovoltaic glass groups C1, C2, ..., Cj, where the photovoltaic glass group Ci is represented as c1, c2, ..., cd.

[0036] Next, any photovoltaic glass 'co' is extracted from the d photovoltaic glass clocks in the photovoltaic glass group Ci: c1, c2, ..., cd, where 'o' is the counting index, with a value ranging from 1 to d.

[0037] Next, the photovoltaic glass CO is embedded into the opening of the detection stage, then the tunable visible light source is turned on, and the current time is recorded as t_str.

[0038] The wavelength of the tunable visible light source is increased uniformly from 380nm to 750nm. The rate of increase of the wavelength is determined by the operator based on the actual situation. The moment when the wavelength of the light source increases uniformly to 750nm is recorded and denoted as t_end. Next, the total number of moments between time t_str and time t_end is summarized and denoted as m.

[0039] The transmittance associated with the photovoltaic glass (co) is measured in real time using a photometer located above the photovoltaic glass. The transmittance is then obtained by arranging the m transmittance values ​​associated with the m times (t_str to t_end). These m transmittance values ​​are then sorted according to their time sequence to obtain a sequence, which is denoted as the transmittance sequence associated with the photovoltaic glass (co): T1, T2, ..., Tm.

[0040] Next, a two-dimensional coordinate system is constructed with the timeline as the horizontal axis and the transmittance as the vertical axis. It should be noted that the time span on the horizontal axis of the two-dimensional coordinate system corresponds to m time intervals between time t_str and time t_end. Then, the wavelength of the tunable visible light source, 380nm, is mapped to m times corresponding to 750nm. The wavelength of the light source associated with each time time is determined. The wavelength of the light source at each time time is then assigned to the horizontal axis of the constructed two-dimensional coordinate system. For example, if the wavelength of the light source at the first time time is 380nm, then the wavelength of the light source, 380nm, is assigned to the first time time on the horizontal axis of the two-dimensional coordinate system.

[0041] By analogy, the wavelength of the light source associated with each moment on the horizontal axis can be determined. Thus, a horizontal axis in which the wavelength of the light source and the time coexist can be obtained, which together with the vertical axis of transmittance forms a new two-dimensional coordinate system, which is denoted as the two-dimensional coordinate system P1.

[0042] Next, the transmittance sequence T1, T2, ..., Tm associated with any photovoltaic glass co in the photovoltaic glass group Ci: c1, c2, ..., cd is extracted. The determined transmittance sequence T1, T2, ..., Tm is plotted in the two-dimensional coordinate system P1 in the form of data points according to the time sequence. Finally, m data points can be obtained. By connecting the adjacent data points with short lines, a broken line can be obtained, which is recorded as the transmittance fluctuation broken line associated with photovoltaic glass co, denoted as: Z_co.

[0043] By repeating the above steps, the transmittance fluctuation lines associated with each photovoltaic glass in the photovoltaic glass group Ci: c1, c2, ..., cd can be determined. The transmittance fluctuation lines associated with each of the d photovoltaic glasses in c1, c2, ..., cd are then sorted according to the order of c1, c2, ..., cd. The sorted result is denoted as the transmittance fluctuation lines, represented as: Z_c1, Z_c2, ..., Z_cd.

[0044] By repeating the above steps, the transmittance fluctuation curves of all photovoltaic glasses in all photovoltaic glass groups within the photovoltaic glass group set C1, C2, ..., Cj can be determined.

[0045] Then, any photovoltaic glass group Ci is obtained from the photovoltaic glass group set C1, C2, ..., Cj. The subsequent steps are to process the photovoltaic glass group Ci as an example. The remaining photovoltaic glass groups in the photovoltaic glass group set C1, C2, ..., Cj are processed in the same way as follows. Extract the transmittance fluctuation lines associated with each of the photovoltaic glass groups Ci:c1,c2,...,cd, which are Z_c1,Z_c2,...,Z_cd; Then, extract the transmittance corresponding to the first time point for each transmittance fluctuation line, for a total of d transmittances corresponding to the first time point. Average the d transmittances and record the average value as the comprehensive transmittance corresponding to the first time point. Repeat this step to determine the comprehensive transmittance corresponding to each of the m time points, and summarize them in chronological order to obtain the comprehensive transmittance sequence.

[0046] Next, a two-dimensional coordinate system P1 is copied, and the original polyline in the two-dimensional coordinate system P1 is cleared, leaving only the horizontal and vertical axis structures of the two-dimensional coordinate system P1. The m comprehensive transmittance values ​​in the comprehensive transmittance sequence are plotted as data points in the two-dimensional coordinate system P1 determined in this step according to the time sequence, resulting in m data points. Short lines are used to connect the adjacent data points to obtain a polyline, which is denoted as the comprehensive transmittance fluctuation polyline associated with the photovoltaic glass group Ci and its corresponding production process group Ai, and is represented as: Z_Ci.

[0047] Repeat the above steps for each photovoltaic glass group in the set C1, C2, ..., Cj, to determine the overall transmittance fluctuation line associated with each photovoltaic glass group (and its corresponding production process group). Then, sort and combine the photovoltaic glass groups according to the sorting order of C1, C2, ..., Cj to obtain the set of overall transmittance fluctuation lines, denoted as: Z_C1, Z_C2, ..., Z_Cj.

[0048] In this embodiment, a detection station integrating a photometer and a tunable visible light source is used to detect the transmittance of each photovoltaic glass in the photovoltaic glass assembly. The wavelength of the tunable visible light source increases uniformly from 380nm to 750nm, and the photometer measures the transmittance in real time to obtain the transmittance value at each moment, forming a transmittance fluctuation line.

[0049] Next, the transmittance fluctuation curves of all photovoltaic glasses in the same photovoltaic glass group are averaged to obtain the comprehensive transmittance fluctuation curve.

[0050] Repeat this process to obtain the set of comprehensive transmittance fluctuation lines for all photovoltaic glass groups and their corresponding production process groups. This allows for a systematic evaluation of the transmittance performance of photovoltaic glass under different production process parameters, providing data support and technical basis for optimizing the production process.

[0051] Example 4 This embodiment further discloses the methods and steps involved in the collaborative optimization process based on embodiment 3, such as... Figure 4 As shown, it specifically includes the following: First, based on the content described in Example 3, the average transmittance associated with each transmittance fluctuation line in the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj is extracted (averaging all transmittance values ​​on the transmittance fluctuation lines), resulting in a total of j average transmittance values. Then, the j average transmittance values ​​are sorted according to the sorting order of the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj to obtain the average transmittance sequence, represented as: T_avg_C1, T_avg_C2, ..., T_avg_Cj, where the average transmittance T_avg_Ci corresponds to the comprehensive transmittance fluctuation line Z_Ci.

[0052] Next, following the rule that 0% transmittance corresponds to 0 and 100% transmittance corresponds to 1, the j average transmittance values ​​in the average transmittance sequence T_avg_C1, T_avg_C2, ..., T_avg_Cj are normalized to the zero-to-one interval, resulting in the normalized transmittance sequence T_gy_C1, T_gy_C2, ..., T_gy_Cj after normalization.

[0053] Next, obtain any one of the comprehensive transmittance fluctuation lines Z_Ci from the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj, and the two-dimensional coordinate system P1 in which the comprehensive transmittance fluctuation line Z_Ci is located. The following steps are an example of processing the comprehensive transmittance fluctuation line Z_Ci. All comprehensive transmittance fluctuation lines in the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj are processed in the same way.

[0054] By taking the scale of the transmittance value on the vertical axis of the two-dimensional coordinate system P1, where the transmittance fluctuation line Z_Ci is located, equal to the average transmittance T_avg_Ci, a straight line parallel to the horizontal axis and perpendicular to the vertical axis is constructed, and this straight line is denoted as L1. Then, a line L2 is constructed by taking the data point corresponding to the first moment on the comprehensive transmittance fluctuation broken line Z_Ci. This line intersects with line L1 and is perpendicular to line L1, perpendicular to the horizontal axis, and parallel to the vertical axis. Then, construct a straight line L3 by taking the data point corresponding to the last moment (mth moment) on the comprehensive transmittance fluctuation broken line Z_Ci. This line intersects the straight line L1 and is perpendicular to the straight line L1, perpendicular to the horizontal axis, and parallel to the vertical axis. At this point, the combined transmittance fluctuation line Z_Ci, line L1, line L2, and line L3 will form several closed regions. The area of ​​each closed region will be calculated, and the value of the area will be recorded as the fluctuation characteristic R_Ci associated with the combined transmittance fluctuation line Z_Ci.

[0055] Repeat the above steps to determine the fluctuation characteristics associated with each of the comprehensive transmittance fluctuation lines in the set Z_C1, Z_C2, ..., Z_Cj, for a total of j fluctuation characteristics. Then, sort the determined j fluctuation characteristics according to the sorting order of the comprehensive transmittance fluctuation line set Z_C1, Z_C2, ..., Z_Cj to obtain the fluctuation characteristic set, denoted as: R_C1, R_C2, ..., R_Cj.

[0056] Then, the j fluctuation features in the fluctuation feature set R_C1, R_C2, ..., R_Cj are normalized to the zero-one interval to obtain the normalized fluctuation feature set, denoted as: R_gy_C1, R_gy_C2, ..., R_gy_Cj.

[0057] Then, subtract each normalized fluctuation feature in the normalized fluctuation feature set from 1 in turn to obtain the normalized complement fluctuation feature set, represented as: R_gb_C1,R_gb_C2,...,R_gb_Cj, where R_gb_Ci=1-R_gy_Ci.

[0058] In this way, the set of integrated transmittance fluctuation lines Z_Ci associated with any photovoltaic glass group Ci and production process group Ai can be determined, and the normalized transmittance T_gy_Ci and normalized complement fluctuation feature R_gb_Ci associated with the set of integrated transmittance fluctuation lines Z_Ci can be further determined (it should be noted that the normalized transmittance T_gy_Ci and the normalized complement fluctuation feature R_gb_Ci are both normalized values, dimensionless, and therefore can be calculated directly).

[0059] Next, by adopting: T_gy_Ci×α+R_gb_Ci×β=K_Ci Calculate the comprehensive index K_Ci associated with the set of comprehensive transmittance fluctuation lines Z_Ci, where α and β are preset calculation weights, and both α and β are greater than 0, α+β=1. The larger the value of the comprehensive index K_Ci, the better the overall transmittance of the photovoltaic glass in the photovoltaic glass group Ci corresponding to the set of comprehensive transmittance fluctuation lines Z_Ci, which also means that the production process group Ai corresponding to the photovoltaic glass group Ci is better.

[0060] Repeat the above steps to determine the comprehensive index associated with each comprehensive transmittance fluctuation line in the set Z_C1, Z_C2, ..., Z_Cj, and represent it according to the sorting order of the comprehensive transmittance fluctuation line set Z_C1, Z_C2, ..., Z_Cj as: K_C1, K_C2, ..., K_Cj, where K_Ci corresponds to Z_Ci.

[0061] Then, the maximum comprehensive index K_Ci is determined from K_C1, K_C2, ..., K_Cj, and the comprehensive transmittance fluctuation line Z_Ci associated with the maximum comprehensive index K_Ci is further determined. The photovoltaic glass group Ci corresponding to the comprehensive transmittance fluctuation line Z_Ci and the production process group Ai corresponding to this photovoltaic glass group Ci are extracted.

[0062] The overall transmittance fluctuation line Z_Ci and the production process parameter group Ai are marked as the optimal overall transmittance fluctuation line and the optimal production process parameter group, respectively. The photovoltaic glass is then processed using the determined optimal production process parameter group.

[0063] All data in the formulas described above are numerical calculations performed after removing their dimensions. Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0064] The above description is merely an example and illustration of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

[0065] It should be stated that all user data collected in this application was collected with the user's consent and authorization. Furthermore, the uses of user data are legal and compliant, and the use and processing of user data comply with the relevant laws, regulations, and standards of the relevant regions.

Claims

1. A photovoltaic glass manufacturing process optimization system based on detection data feedback, characterized in that, The system includes: The production process and data management terminal retrieves the set of production process groups predefined by the operators from the production process parameter database, and constructs an initial photovoltaic glass group set corresponding to the total number of production process groups in the production process group set. Each initial photovoltaic glass assembly is processed using different production process groups to determine the final set of photovoltaic glass assemblies. The photovoltaic glass intelligent inspection terminal places each photovoltaic glass in the photovoltaic glass group in the photovoltaic glass group set on a pre-constructed inspection platform, performs inspection operations, evaluates the transmittance fluctuation line associated with each photovoltaic glass, and fits the transmittance fluctuation line of the photovoltaic glass in the same photovoltaic glass group to determine the comprehensive transmittance fluctuation line associated with the production process group corresponding to the photovoltaic glass group. On the collaborative optimization side, the overall transmittance fluctuation lines associated with each production process group in the production process group set are analyzed collaboratively to determine the optimal overall transmittance fluctuation line and its corresponding optimal production process parameter set, and then the photovoltaic glass is processed using the optimal production process parameter set. The iterative feedback end performs update operations on the optimal production process parameter set based on a preset update cycle.

2. The system according to claim 1, characterized in that, In the production process and data management terminal, the specific method for constructing the initial photovoltaic glass group set corresponding to the total number of production process groups in the production process group set is as follows: Determine the target specifications for photovoltaic glass; Extract the predefined set of production process groups A1, A2, ..., Aj associated with the photovoltaic glass of the target specification from the production process parameter database, where j is the total number of production process groups. Any production process parameter group Ai corresponds to all production process parameters involved in the entire processing flow of the photovoltaic glass of the target specification, where i is the counting index, 1≤i≤j. For a production process parameter group Ai, an initial photovoltaic glass group Bi is constructed, wherein the total number of photovoltaic glasses in the initial photovoltaic glass group Bi is d, where d is a preset integer. Similarly, construct the initial photovoltaic glass sets corresponding to each production process group in the production process group set A1, A2, ..., Aj, and sort them according to the order of the production process group set A1, A2, ..., Aj to obtain the initial photovoltaic glass set set B1, B2, ..., Bj.

3. The system according to claim 2, characterized in that, In the aforementioned production process and data management terminal, the specific method for determining the resulting photovoltaic glass assembly after processing is as follows: Extract any one initial photovoltaic glass group Bi from the initial photovoltaic glass group set B1, B2, ..., Bj, and process d photovoltaic glass units in the initial photovoltaic glass group Bi using the production process parameter group Ai corresponding to the initial photovoltaic glass group Bi. Extract d photovoltaic glass units processed by production process parameter group Ai, and denot them as photovoltaic glass group Ci: c1, c2, ..., cd according to the processing order, where c1 to cd represent the first to last photovoltaic glass units processed; Similarly, the photovoltaic glass groups associated with each initial photovoltaic glass group in the initial photovoltaic glass group set B1, B2, ..., Bj are determined and summarized as the photovoltaic glass group set C1, C2, ..., Cj.

4. The system according to claim 3, characterized in that, The testing station in the photovoltaic glass intelligent inspection terminal integrates a photometer and a tunable visible light source. The photometer is located below the detection stage, the tunable visible light source is located above the detection stage, and the detection stage includes an opening for embedding photovoltaic glass; The tunable visible light source has a wavelength range of 380nm to 750nm.

5. The system according to claim 4, characterized in that, In the aforementioned photovoltaic glass intelligent inspection terminal, the specific method for evaluating the transmittance fluctuation line associated with each photovoltaic glass is as follows: S51. Obtain any photovoltaic glass group Ci: c1, c2, ..., cd from the set of photovoltaic glass groups C1, C2, ..., Cj; S52. Extract any photovoltaic glass co from c1, c2, ..., cd, where o is the counting index and 1 ≤ o ≤ d; S53. Embed the photovoltaic glass co into the opening of the detection stage, turn on the tunable visible light source, and record the current time t_str; S54. Adjust the wavelength of the tunable visible light source to increase uniformly from 380nm to 750nm. The rate of increase is determined by the operator. Record the time t_end when the wavelength of the light source is adjusted to 750nm. Sum the total number of time intervals from time t_str to time t_end, and denot it as m. S55. Using a photometer, record the m transmittance values ​​of the photovoltaic glass co at m time points in chronological order to obtain the transmittance sequence T1, T2, ..., Tm associated with the photovoltaic glass co. S56. Construct a two-dimensional coordinate system with the timeline as the horizontal axis and the transmittance as the vertical axis. The time span on the horizontal axis is from the first time to the mth time. Determine the wavelength of the tunable visible light source at the first time and assign it to the first time on the horizontal axis. Similarly, by assigning the corresponding light source wavelength to each of the m time points, we obtain a two-dimensional coordinate system P1 with the light source wavelength and time point coexisting on the horizontal axis and the transmittance on the vertical axis. S57. Plot the transmittance sequence T1, T2, ..., Tm associated with photovoltaic glass co in the form of data points in the two-dimensional coordinate system P1 according to the time sequence to obtain m data points. Connect the adjacent data points with short lines to obtain a broken line, which is denoted as the transmittance fluctuation broken line Z_co. S58. Similarly, determine the transmittance fluctuation line associated with each of the d photovoltaic glass units in the photovoltaic glass group Ci: c1, c2, ..., cd, and represent it as Z_c1, Z_c2, ..., Z_cd; S59. Similarly, determine the transmittance fluctuation curve of all photovoltaic glasses in all photovoltaic glass groups in the photovoltaic glass group set C1, C2, ..., Cj.

6. The system according to claim 5, characterized in that, In the aforementioned photovoltaic glass intelligent inspection terminal, the specific method for determining the comprehensive transmittance fluctuation line associated with the production process group corresponding to the photovoltaic glass group is as follows: From the set of photovoltaic glass groups C1, C2, ..., Cj, obtain any photovoltaic glass group Ci, and extract the transmittance fluctuation lines associated with each of the d photovoltaic glass groups in Ci: Z_c1, Z_c2, ..., Z_cd; Take the average of the transmittance corresponding to the first time point of the d transmittance fluctuation lines, and use it as the comprehensive transmittance corresponding to the first time point. Similarly, determine the comprehensive transmittance corresponding to each of the m time points, and sort the m comprehensive transmittances in time order to obtain the comprehensive transmittance sequence. Copy any two-dimensional coordinate system P1 and clear it; Using the comprehensive transmittance sequence, following the steps described in step S57, construct a comprehensive transmittance fluctuation line associated with the photovoltaic glass group Ci, denoted as Z_Ci; Similarly, the overall transmittance fluctuation lines associated with each photovoltaic glass group in the photovoltaic glass group set C1, C2, ..., Cj are determined and denoted as the overall transmittance fluctuation line set Z_C1, Z_C2, ..., Z_Cj.

7. The system according to claim 6, characterized in that, In the collaborative optimization process, the specific method for determining the optimal overall transmittance fluctuation line and its corresponding optimal production process parameter set is as follows: Determine the average transmittance associated with each of the comprehensive transmittance fluctuation lines in the set Z_C1, Z_C2, ..., Z_Cj, and denote them as the average transmittance sequence T_avg_C1, T_avg_C2, ..., T_avg_Cj in the order of the comprehensive transmittance fluctuation line set Z_C1, Z_C2, ..., Z_Cj; The j average transmittance values ​​in the average transmittance sequence T_avg_C1, T_avg_C2, ..., T_avg_Cj are normalized to the zero-to-one interval according to the method that 0% transmittance corresponds to 0 and 100% transmittance corresponds to 1, so as to obtain the normalized transmittance sequence T_gy_C1, T_gy_C2, ..., T_gy_Cj; Take any one of the comprehensive transmittance fluctuation lines Z_Ci from the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj and its corresponding two-dimensional coordinate system P1; By passing through the scale of the average transmittance T_avg_Ci on the vertical axis of the two-dimensional coordinate system P1, a straight line L1 is constructed that is parallel to the horizontal axis and perpendicular to the vertical axis. A straight line is constructed by passing through the data points corresponding to the first time point and the data points corresponding to the m-th time point on the comprehensive transmittance fluctuation line Z_Ci, respectively. The line intersects the straight line L1 and is perpendicular to the straight line L1, perpendicular to the horizontal axis, and parallel to the vertical axis. These lines are denoted as L2 and L3, respectively. Extract the area values ​​of several closed regions formed by the comprehensive transmittance fluctuation line Z_Ci, line L1, line L2 and line L3, and denote them as fluctuation characteristics R_Ci; Similarly, determine the fluctuation characteristics associated with each of the comprehensive transmittance fluctuation lines in the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj, and summarize them as the fluctuation characteristic set R_C1, R_C2, ..., R_Cj; Normalize j fluctuation features in the fluctuation feature set R_C1,R_C2,...,R_Cj to the zero-one interval to obtain the normalized fluctuation feature set R_gy_C1,R_gy_C2,...,R_gy_Cj; Subtract each normalized fluctuation feature from 1 in turn to obtain the set of normalized complementary fluctuation features R_gb_C1, R_gb_C2, ..., R_gb_Cj; Take the set of broken lines Z_Ci associated with the overall transmittance fluctuation of any photovoltaic glass group Ci; Extract the normalized transmittance T_gy_Ci and the normalized complement fluctuation feature R_gb_Ci associated with the set of broken lines of comprehensive transmittance fluctuation Z_Ci. Calculate the comprehensive index K_Ci using T_gy_Ci×α+R_gb_Ci×β=K_Ci, where α and β are preset calculation weights, α>0, β>0, α+β=1; Similarly, the comprehensive index associated with each comprehensive transmittance fluctuation line in the set of comprehensive transmittance fluctuation lines Z_C1, Z_C2, ..., Z_Cj is determined and denoted as K_C1, K_C2, ..., K_Cj, where K_Ci corresponds to Z_Ci; Extract the comprehensive transmittance fluctuation line Z_Ci corresponding to the maximum comprehensive index K_Ci, obtain the production process parameter group Ai corresponding to the comprehensive transmittance fluctuation line Z_Ci, and mark the comprehensive transmittance fluctuation line Z_Ci and the production process parameter group Ai as the optimal comprehensive transmittance fluctuation line and the optimal production process parameter group, respectively.

8. The system according to claim 7, characterized in that, The collaborative optimization process also includes processing the photovoltaic glass using the determined optimal production process parameter set.

9. The system according to claim 8, characterized in that, In the iterative feedback terminal, the specific method for performing the update operation on the optimal production process parameter set is as follows: When the optimal production process parameter set is determined, record the current time and extract the update cycle preset by the operator. Use the current time as the start time of the update cycle and update the current time in real time. If the time interval between the current time and the start time reaches the update cycle, an update is triggered, and the optimal production process parameter set is re-determined.

Images