A method and system for controlling the quality of laser cutting of photovoltaic cells

Abstract

The present invention relates to a laser cutting processing quality control method and system for photovoltaic cells, the method comprising: P1. In the process of laser cutting photovoltaic cells, real-time acquisition of processing parameter data information of laser cutting photovoltaic cells; P2. Collecting historical processing parameter data information of laser cutting photovoltaic cells, and based on the historical processing parameter data information of laser cutting photovoltaic cells, using a BP neural network regression prediction algorithm based on improved aurora optimization to predict the processing quality of laser cutting photovoltaic cells. The present invention can not only accurately control and adjust the laser cutting of photovoltaic cells, thereby improving the production stability and production efficiency of photovoltaic cells, but also the entire cutting process of photovoltaic cells does not require manual participation, reducing labor costs, thereby further improving the production efficiency of photovoltaic cells.

Description

Technical Field

[0001] The present invention relates to the field of laser cutting technology, and in particular to a laser cutting quality control method and system for photovoltaic cells. Background Art

[0002] During the use of photovoltaic cells, in order to reduce the power loss inside the photovoltaic module, larger photovoltaic cells are usually cut into suitable sizes, and then the cut cells are welded to achieve the effect of reducing their internal losses.

[0003] Usually, photovoltaic cells need to be cut with the help of a dicing machine. The cell material is cut by laser. According to production requirements, larger cells are cut into sizes that meet actual needs in order to ensure the accuracy of the dicing position and reduce losses, as well as the corresponding beat requirements. The processing quality of laser cutting needs to be controlled and adjusted. The laser cutting processing quality control method refers to the process parameters of the laser cutting of photovoltaic cells can be adaptively adjusted according to the actual processing quality of the laser cutting of photovoltaic cells according to the changes in the processing conditions of the photovoltaic cells on site, so as to achieve the optimal control of the laser cutting processing quality of photovoltaic cells. Thereby improving the problems of poor accuracy and poor adaptability of traditional laser cutting. Therefore, how to accurately control the cutting quality of laser cutting has become a problem that we urgently need to solve. Summary of the invention

[0004] In view of the above problems, the present invention provides a method and system for controlling the quality of laser cutting of photovoltaic cells, which can not only accurately control and adjust the laser cutting of photovoltaic cells, thereby improving the production stability and production efficiency of photovoltaic cells, but also the entire cutting process of photovoltaic cells does not require human participation, reducing labor costs, thereby further improving the production efficiency of photovoltaic cells.

[0005] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows:

[0006] A method for controlling the quality of laser cutting of photovoltaic cells, the method comprising:

[0007] P1. During the laser cutting process of photovoltaic cells, real-time acquisition of processing parameter data information of laser cutting photovoltaic cells;

[0008] P2. Collect historical processing parameter data information of laser cutting photovoltaic cells, and based on the historical processing parameter data information of laser cutting photovoltaic cells, use the improved Aurora-optimized BP neural network regression prediction algorithm to predict the processing quality of laser cutting photovoltaic cells, and obtain the predicted processing quality data information of laser cutting photovoltaic cells;

[0009] P3. Based on the predicted processing quality data information of the laser-cut photovoltaic cell and the processing parameter data information of the laser-cut photovoltaic cell, the processing parameters of the laser-cut photovoltaic cell are optimized by using the Newton-Raphson optimization algorithm based on the Gaussian kernel function to obtain the optimized processing parameter data information of the laser-cut photovoltaic cell;

[0010] P4. Based on the optimized processing parameter data information of laser cutting photovoltaic cells, a processing quality assessment model of laser cutting photovoltaic cells is constructed, the assessment value of the processing quality of laser cutting photovoltaic cells is characterized, and data correction information of the processing quality of laser cutting photovoltaic cells is obtained. The correction information is used to assist in adjusting the laser cutting processing technology of photovoltaic cells.

[0011] Furthermore, the method further comprises:

[0012] P5. Based on the data correction information of the processing quality of laser-cut photovoltaic cells, a preset threshold is set. If the data correction information of the processing quality of laser-cut photovoltaic cells is greater than the preset threshold, the optimized processing parameter data information of laser-cut photovoltaic cells is output. If the data correction information of the processing quality of laser-cut photovoltaic cells is less than the preset threshold, the requirement is not met and the process returns to step P3.

[0013] Furthermore, in step P2, the use of the improved Aurora-optimized BP neural network regression prediction algorithm to predict the processing quality of laser-cut photovoltaic cells includes:

[0014] P21. Input the historical processing parameter data information of the laser cutting photovoltaic cell into the BP neural network regression prediction model for training and learning, initialize the weight and bias of the BP neural network, and construct the weight matrix and bias matrix of the BP neural network to obtain the data information of the weight matrix and bias matrix of the BP neural network;

[0015] P22. Based on the data information of the weight matrix and bias matrix of the BP neural network, the aurora population is initialized, the boundary and population parameters of the solution space are determined, and the initialized aurora population data information is obtained;

[0016] P23. Based on the initialized aurora population data information, establish the target optimization function Q of the aurora population,

[0017] ,

[0018] Among them, x is the data information of the aurora population after initialization, α 1 , α 2 and α 3 For the adaptive weight of the aurora population, the weight and bias of the BP neural network are optimized to obtain the data information of the weight and bias of the optimized BP neural network;

[0019] P24. Based on the weight and bias data information of the optimized BP neural network, a prediction function W of the BP neural network regression prediction model is established.

[0020] ,

[0021] Among them, f i is the activation function of the i-th neuron in the BP neural network regression prediction model, β i is the weight of the BP neural network after optimization of the i-th neuron function, δ i is the bias of the optimized BP neural network of the i-th neuron function, n is the sample capacity, and the processing quality of the laser-cut photovoltaic cells is predicted to obtain the predicted processing quality data information of the laser-cut photovoltaic cells.

[0022] Furthermore, the activation function f of the i-th neuron of the BP neural network regression prediction is i for,

[0023] ,

[0024] ,

[0025] Among them, y is the historical processing parameter data information of laser cutting photovoltaic cells, γ i is the activation factor of the activation function of the i-th neuron, and n is the sample size.

[0026] Furthermore, the weight β of the optimized BP neural network of the i-th neuron function is i and the bias δ ​​of the BP neural network after optimization of the i-th neuron function i The constraints are:

[0027] ,

[0028] Where n is the sample size.

[0029] Furthermore, in step P3, the optimization of the processing parameters of laser cutting photovoltaic cells by using the Newton-Raphson optimization algorithm based on Gaussian kernel function includes:

[0030] P31. Based on the predicted processing quality data information of the laser-cut photovoltaic cell and the processing parameter data information of the laser-cut photovoltaic cell, a Gaussian kernel function R of the laser cutting process is established.

[0031] ,

[0032] Wherein, a is the predicted processing quality data information of laser cutting photovoltaic cells, b is the processing parameter data information of laser cutting photovoltaic cells, η j is the Gaussian constant parameter of laser cutting photovoltaic cells, J is a positive integer, and the multi-dimensional space data of the processing quality and processing parameters of laser cutting photovoltaic cells are characterized to obtain the data information of the multi-dimensional space matrix of the processing quality and processing parameters of laser cutting photovoltaic cells;

[0033] P32. Based on the data information of the multidimensional space matrix of the processing quality and processing parameters of the laser-cut photovoltaic cell, the Newton-Raphson population is initialized, the upper and lower bounds of the population variables are determined, and the data information of the initialized Newton-Raphson population is obtained;

[0034] P33. Based on the data information of the initialized Newton-Raphson population, an optimization function G of the Newton-Raphson population is established.

[0035] ,

[0036] Among them, c is the data information of the initialized Newton-Raphson population, λ 1 , 2 and λ 3 The optimized constant parameters of the population are used to optimize the processing parameters of the laser cutting photovoltaic cells, and the optimized processing parameter data information of the laser cutting photovoltaic cells is obtained.

[0037] Furthermore, the optimization constant parameter λ of the population 1 , 2 and λ 3 The constraint function g is,

[0038] ,

[0039] ,

[0040] Among them, the value range of the constraint function g is (2,3).

[0041] Furthermore, the Gaussian constant parameter η of the laser cutting photovoltaic cell is j The constraints are:

[0042] ,

[0043] Wherein, J is a positive integer.

[0044] Furthermore, in step P4, the construction of the processing quality assessment model for laser cutting photovoltaic cells and characterizing the assessment value of the processing quality of laser cutting photovoltaic cells include:

[0045] P41. Based on the optimized laser cutting photovoltaic cell processing parameter data information, construct a laser cutting photovoltaic cell processing parameter data set;

[0046] P42. Inputting the processing parameter data set of the laser cutting photovoltaic cell into the processing quality assessment model of the laser cutting photovoltaic cell for training and learning, and determining the processing quality assessment function H of the laser cutting photovoltaic cell,

[0047] ,

[0048] Among them, r is the processing parameter data set of laser cutting photovoltaic cells, µ 1 、µ 2 and µ 3 is an evaluation factor of the processing quality of laser cutting photovoltaic cells, and a trained processing quality evaluation model of laser cutting photovoltaic cells is obtained;

[0049] P43. Based on the trained processing quality assessment model for laser cutting of photovoltaic cells, the optimized processing parameter data information for laser cutting of photovoltaic cells is input, the evaluation value of the processing quality of laser cutting of photovoltaic cells is characterized, and data correction information on the processing quality of laser cutting of photovoltaic cells is obtained. The correction information is used to assist in adjusting the laser cutting processing technology of photovoltaic cells.

[0050] Furthermore, the evaluation factor µ of the processing quality of the laser cutting photovoltaic cell is 1 、µ 2 and µ 3 for,

[0051] ,

[0052] ,

[0053] ,

[0054] Among them, r is the processing parameter data set of laser cutting photovoltaic cells, r 0 is the mean value of the processing parameter data for laser cutting of photovoltaic cells.

[0055] In order to achieve the above-mentioned and other related purposes, the present invention also provides a laser cutting processing quality control system for photovoltaic cells, including a computer device, which is programmed or configured to execute any one of the steps of the laser cutting processing quality control method for photovoltaic cells.

[0056] The present invention has the following positive effects:

[0057] 1. The present invention predicts the processing quality of laser-cut photovoltaic cells by adopting a BP neural network regression prediction algorithm based on improved Aurora optimization, and optimizes the processing parameters of laser-cut photovoltaic cells by combining a Newton-Raphson optimization algorithm based on a Gaussian kernel function. It can not only adaptively adjust the processing parameters of laser cutting of photovoltaic cells according to the changes in the processing conditions of on-site photovoltaic cells according to the actual processing quality of laser cutting of photovoltaic cells, so as to achieve optimal control of the laser cutting processing quality of photovoltaic cells, but also accurately control and adjust the laser cutting of photovoltaic cells, thereby improving the production stability and production efficiency of photovoltaic cells.

[0058] 2. The present invention constructs a processing quality evaluation model for laser cutting of photovoltaic cells, characterizes the evaluation value of the processing quality of laser cutting of photovoltaic cells, and judges the processing quality of laser cutting of photovoltaic cells in combination with a preset threshold, thereby ensuring the processing stability of laser cutting of photovoltaic cells. Not only does the entire cutting process of photovoltaic cells require no human intervention, reducing labor costs and further improving the production efficiency of photovoltaic cells, but it also has a high degree of automation and is easy to operate. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0060] Figure 2 It is a flow chart of the BP neural network regression prediction algorithm based on improved aurora optimization of the present invention;

[0061] Figure 3 It is a schematic diagram of the process of the Newton-Raphson optimization algorithm based on Gaussian kernel function of the present invention;

[0062] Figure 4 A schematic diagram of the process of constructing a processing quality assessment model for laser cutting photovoltaic cells according to the present invention;

[0063] Figure 5Schematic diagram of the system architecture of the present invention. DETAILED DESCRIPTION

[0064] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0065] Example 1: Figure 1 As shown, a laser cutting quality control method for photovoltaic cells, the method comprising:

[0066] P1. During the laser cutting process of photovoltaic cells, historical processing parameter data information of laser cutting photovoltaic cells is collected, and the processing parameter data information of laser cutting photovoltaic cells is obtained in real time;

[0067] P2. Based on the historical processing parameter data information of the laser cutting photovoltaic cell, the processing quality of the laser cutting photovoltaic cell is predicted by using the BP neural network regression prediction algorithm based on the improved Aurora optimization to obtain the predicted processing quality data information of the laser cutting photovoltaic cell;

[0068] P3. Based on the predicted processing quality data information of the laser-cut photovoltaic cell and the processing parameter data information of the laser-cut photovoltaic cell, the processing parameters of the laser-cut photovoltaic cell are optimized by using the Newton-Raphson optimization algorithm based on the Gaussian kernel function to obtain the optimized processing parameter data information of the laser-cut photovoltaic cell;

[0069] P4. Based on the optimized processing parameter data information of laser cutting photovoltaic cells, a processing quality assessment model of laser cutting photovoltaic cells is constructed, the assessment value of the processing quality of laser cutting photovoltaic cells is characterized, and data correction information of the processing quality of laser cutting photovoltaic cells is obtained. The correction information is used to assist in adjusting the laser cutting processing technology of photovoltaic cells.

[0070] In this embodiment, the method further includes:

[0071] P5. Based on the data correction information of the processing quality of the laser-cut photovoltaic cell, a preset threshold is set. If the data correction information of the processing quality of the laser-cut photovoltaic cell is greater than the preset threshold, the optimized processing parameter data information of the laser-cut photovoltaic cell is output. If the data correction information of the processing quality of the laser-cut photovoltaic cell is less than the preset threshold, the requirement is not met and the process returns to step P3.

[0072] In this embodiment, if Figure 2 As shown, in step P2, the prediction of the processing quality of laser-cut photovoltaic cells by using the BP neural network regression prediction algorithm based on improved Aurora optimization includes:

[0073] P21. Input the historical processing parameter data information of the laser cutting photovoltaic cell into the BP neural network regression prediction model for training and learning, initialize the weight and bias of the BP neural network, and construct the weight matrix and bias matrix of the BP neural network to obtain the data information of the weight matrix and bias matrix of the BP neural network;

[0074] P22. Based on the data information of the weight matrix and bias matrix of the BP neural network, the aurora population is initialized, the boundary and population parameters of the solution space are determined, and the initialized aurora population data information is obtained;

[0075] P23. Based on the initialized aurora population data information, establish the target optimization function Q of the aurora population,

[0076] ,

[0077] Among them, x is the data information of the aurora population after initialization, α 1 , α 2 and α 3 For the adaptive weight of the aurora population, the weight and bias of the BP neural network are optimized to obtain the data information of the weight and bias of the optimized BP neural network;

[0078] P24. Based on the weight and bias data information of the optimized BP neural network, a prediction function W of the BP neural network regression prediction model is established.

[0079] ,

[0080] Among them, f i is the activation function of the i-th neuron in the BP neural network regression prediction model, β i is the weight of the BP neural network after optimization of the i-th neuron function, δ iis the bias of the optimized BP neural network of the i-th neuron function, n is the sample capacity, and the processing quality of the laser-cut photovoltaic cells is predicted to obtain the predicted processing quality data information of the laser-cut photovoltaic cells.

[0081] In this embodiment, the activation function f of the i-th neuron of the BP neural network regression prediction is i for,

[0082] ,

[0083] ,

[0084] Among them, y is the historical processing parameter data information of laser cutting photovoltaic cells, γ i is the activation factor of the activation function of the i-th neuron, and n is the sample size.

[0085] In this embodiment, the weight β of the optimized BP neural network of the i-th neuron function is i and the bias δ ​​of the BP neural network after optimization of the i-th neuron function i The constraints are:

[0086] ,

[0087] Where n is the sample size.

[0088] Laser cutting process optimization and quality control is a complex multi-objective optimization problem, which requires balancing many processing quality evaluation indicators such as section roughness, kerf width and slag amount. There are often contradictory relationships between these indicators. For example, although increasing the auxiliary gas blowing pressure can reduce the residue, it may also lead to an increase in the kerf width, which in turn affects the processing accuracy. The method based on Pareto multi-objective optimization theory is difficult to deal with this contradiction because it is difficult to find a solution that can simultaneously meet the optimal solution of all indicators. On the other hand, although the weighted comprehensive scoring method is simple and intuitive, it is difficult to give a unified weight due to the differences in the dimensions and importance of different indicators, especially when the processing requirements are changeable and the optimization objectives are not fixed. The weighted comprehensive scoring method often leads to large fluctuations in the numerical range of the evaluation results. Therefore, the use of the BP neural network regression prediction algorithm based on the improved Aurora optimization to predict the processing quality of laser-cut photovoltaic cells brings higher compatibility and stability requirements to the subsequent optimization algorithm.

[0089] Example 2: Based on the quality control method for laser cutting of photovoltaic cells in Example 1, the present invention is further illustrated and described below.

[0090] like Figure 1As shown, a laser cutting quality control method for photovoltaic cells, the method comprising:

[0091] P1. During the laser cutting process of photovoltaic cells, real-time acquisition of processing parameter data information of laser cutting photovoltaic cells;

[0092] P2. Collect historical processing parameter data information of laser cutting photovoltaic cells, and based on the historical processing parameter data information of laser cutting photovoltaic cells, use the improved Aurora-optimized BP neural network regression prediction algorithm to predict the processing quality of laser cutting photovoltaic cells, and obtain the predicted processing quality data information of laser cutting photovoltaic cells;

[0093] P3. Based on the predicted processing quality data information of the laser-cut photovoltaic cell and the processing parameter data information of the laser-cut photovoltaic cell, the processing parameters of the laser-cut photovoltaic cell are optimized by using the Newton-Raphson optimization algorithm based on the Gaussian kernel function to obtain the optimized processing parameter data information of the laser-cut photovoltaic cell;

[0094] P4. Based on the optimized processing parameter data information of laser cutting photovoltaic cells, a processing quality assessment model of laser cutting photovoltaic cells is constructed, the assessment value of the processing quality of laser cutting photovoltaic cells is characterized, and data correction information of the processing quality of laser cutting photovoltaic cells is obtained. The correction information is used to assist in adjusting the laser cutting processing technology of photovoltaic cells.

[0095] In this embodiment, if Figure 3 As shown, in step P3, the optimization of the processing parameters of laser cutting photovoltaic cells by using the Newton-Raphson optimization algorithm based on Gaussian kernel function includes:

[0096] P31. Based on the predicted processing quality data information of the laser-cut photovoltaic cell and the processing parameter data information of the laser-cut photovoltaic cell, a Gaussian kernel function R of the laser cutting process is established.

[0097] ,

[0098] Wherein, a is the predicted processing quality data information of laser cutting photovoltaic cells, b is the processing parameter data information of laser cutting photovoltaic cells, η j is the Gaussian constant parameter of laser cutting photovoltaic cells, J is a positive integer, and the multi-dimensional space data of the processing quality and processing parameters of laser cutting photovoltaic cells are characterized to obtain the data information of the multi-dimensional space matrix of the processing quality and processing parameters of laser cutting photovoltaic cells;

[0099] P32. Based on the data information of the multidimensional space matrix of the processing quality and processing parameters of the laser-cut photovoltaic cell, the Newton-Raphson population is initialized, the upper and lower bounds of the population variables are determined, and the data information of the initialized Newton-Raphson population is obtained;

[0100] P33. Based on the data information of the initialized Newton-Raphson population, an optimization function G of the Newton-Raphson population is established.

[0101] ,

[0102] Among them, c is the data information of the initialized Newton-Raphson population, λ 1 , 2 and λ 3 The optimized constant parameters of the population are used to optimize the processing parameters of the laser cutting photovoltaic cells, and the optimized processing parameter data information of the laser cutting photovoltaic cells is obtained.

[0103] In this embodiment, the optimization constant parameter λ of the population 1 , 2 and λ 3 The constraint function g is,

[0104] ,

[0105] ,

[0106] Among them, the value range of the constraint function g is (2,3).

[0107] In this embodiment, the Gaussian constant parameter η of the laser cutting photovoltaic cell sheet is j The constraints are:

[0108] ,

[0109] Wherein, J is a positive integer.

[0110] In this embodiment, if Figure 4 As shown, in step P4, the construction of the processing quality assessment model for laser cutting photovoltaic cells and characterization of the assessment value of the processing quality for laser cutting photovoltaic cells include:

[0111] P41. Based on the optimized laser cutting photovoltaic cell processing parameter data information, construct a laser cutting photovoltaic cell processing parameter data set;

[0112] P42. Inputting the processing parameter data set of the laser cutting photovoltaic cell into the processing quality assessment model of the laser cutting photovoltaic cell for training and learning, and determining the processing quality assessment function H of the laser cutting photovoltaic cell,

[0113] ,

[0114] Among them, r is the processing parameter data set of laser cutting photovoltaic cells, µ 1 、µ 2 and µ 3 is an evaluation factor of the processing quality of laser cutting photovoltaic cells, and a trained processing quality evaluation model of laser cutting photovoltaic cells is obtained;

[0115] P43. Based on the trained processing quality assessment model for laser cutting of photovoltaic cells, the optimized processing parameter data information for laser cutting of photovoltaic cells is input, the evaluation value of the processing quality of laser cutting of photovoltaic cells is characterized, and data correction information on the processing quality of laser cutting of photovoltaic cells is obtained. The correction information is used to assist in adjusting the laser cutting processing technology of photovoltaic cells.

[0116] In this embodiment, the evaluation factor µ of the processing quality of the laser cutting photovoltaic cell is 1 、µ 2 and µ 3 for,

[0117] ,

[0118] ,

[0119] ,

[0120] Among them, r is the processing parameter data set of laser cutting photovoltaic cells, r 0 is the mean value of the processing parameter data for laser cutting of photovoltaic cells.

[0121] In this embodiment, the present invention provides a laser cutting quality control system for photovoltaic cells, including a computer device programmed or configured to execute any one of the steps of the laser cutting quality control method for photovoltaic cells.

[0122] In this embodiment, if Figure 5 As shown, a laser cutting quality control system for photovoltaic cells, the system comprising:

[0123] The laser cutting data acquisition module is used to collect historical processing parameter data information of laser cutting photovoltaic cells, and obtain the processing parameter data information of laser cutting photovoltaic cells in real time;

[0124] A laser cutting photovoltaic cell processing quality prediction module is connected to the laser cutting data acquisition module and is used to predict the processing quality of the laser cutting photovoltaic cell by using a BP neural network regression prediction algorithm based on improved Aurora optimization to obtain the predicted processing quality data information of the laser cutting photovoltaic cell;

[0125] A processing parameter optimization module for laser cutting photovoltaic cells, connected to the processing quality prediction module for laser cutting photovoltaic cells, is used to optimize the processing parameters for laser cutting photovoltaic cells by using a Newton-Raphson optimization algorithm based on a Gaussian kernel function, and obtain optimized processing parameter data information for laser cutting photovoltaic cells;

[0126] A processing quality assessment module for laser cutting photovoltaic cells, connected to the processing parameter optimization module for laser cutting photovoltaic cells, is used to construct a processing quality assessment model for laser cutting photovoltaic cells, characterize the assessment value of the processing quality of laser cutting photovoltaic cells, and obtain data information of the assessment value of the processing quality of laser cutting photovoltaic cells;

[0127] The threshold judgment module for laser cutting of photovoltaic cells is connected to the processing quality assessment module for laser cutting of photovoltaic cells and is used to set a preset threshold. If the assessment value of the processing quality of the laser cutting of photovoltaic cells is greater than the preset threshold, the optimized processing parameter data information of the laser cutting of photovoltaic cells is output; if the assessment value of the processing quality of the laser cutting of photovoltaic cells is less than the preset threshold, the requirement is not met.

[0128] In this embodiment, the present invention provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the methods for controlling the quality of laser cutting of photovoltaic cells.

[0129] Any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0130] In summary, the present invention can not only accurately control and adjust the laser cutting of photovoltaic cells, thereby improving the production stability and production efficiency of photovoltaic cells, but also the entire cutting process of photovoltaic cells does not require human participation, reducing labor costs, thereby further improving the production efficiency of photovoltaic cells.

[0131] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for controlling the quality of laser cutting of photovoltaic cells, characterized in that: The method comprises: P1. During the laser cutting process of photovoltaic cells, real-time acquisition of processing parameter data information of laser cutting photovoltaic cells; P2. Collect historical processing parameter data information of laser cutting photovoltaic cells, and based on the historical processing parameter data information of laser cutting photovoltaic cells, use the improved Aurora-optimized BP neural network regression prediction algorithm to predict the processing quality of laser cutting photovoltaic cells, and obtain the predicted processing quality data information of laser cutting photovoltaic cells; P3. Based on the predicted processing quality data information of the laser-cut photovoltaic cell and the processing parameter data information of the laser-cut photovoltaic cell, the processing parameters of the laser-cut photovoltaic cell are optimized by using the Newton-Raphson optimization algorithm based on the Gaussian kernel function to obtain the optimized processing parameter data information of the laser-cut photovoltaic cell; P4. Based on the optimized processing parameter data information of the laser cutting photovoltaic cell, a processing quality assessment model of the laser cutting photovoltaic cell is constructed, the assessment value of the processing quality of the laser cutting photovoltaic cell is characterized, and data correction information of the processing quality of the laser cutting photovoltaic cell is obtained, and the correction information is used to assist in adjusting the laser cutting processing technology of the photovoltaic cell; In step P2, the prediction of the processing quality of laser-cut photovoltaic cells by using the BP neural network regression prediction algorithm based on improved Aurora optimization includes: P21. Input the historical processing parameter data information of the laser cutting photovoltaic cell into the BP neural network regression prediction model for training and learning, initialize the weight and bias of the BP neural network, and construct the weight matrix and bias matrix of the BP neural network to obtain the data information of the weight matrix and bias matrix of the BP neural network; P22. Based on the data information of the weight matrix and bias matrix of the BP neural network, the aurora population is initialized, the boundary and population parameters of the solution space are determined, and the initialized aurora population data information is obtained; P23. Based on the initialized aurora population data information, establish the target optimization function Q of the aurora population, ， Among them, x is the data information of the aurora population after initialization, α1, α2 and α3 are the adaptive weights of the aurora population, and the weights and biases of the BP neural network are optimized to obtain the data information of the weights and biases of the optimized BP neural network; P24. Based on the weight and bias data information of the optimized BP neural network, a prediction function W of the BP neural network regression prediction model is established. ， Among them, f i is the activation function of the i-th neuron in the BP neural network regression prediction model, β i is the weight of the BP neural network after optimization of the i-th neuron function, δ i is the bias of the optimized BP neural network of the i-th neuron function, n is the sample capacity, and the processing quality of the laser-cut photovoltaic cells is predicted to obtain the predicted processing quality data information of the laser-cut photovoltaic cells.

2. The laser cutting quality control method for photovoltaic cells according to claim 1, characterized in that: The method further comprises: P5. Based on the data correction information of the processing quality of the laser-cut photovoltaic cell, a preset threshold is set. If the data correction information of the processing quality of the laser-cut photovoltaic cell is greater than the preset threshold, the optimized processing parameter data information of the laser-cut photovoltaic cell is output. If the data correction information of the processing quality of the laser-cut photovoltaic cell is less than the preset threshold, the requirement is not met and the process returns to step P3.

3. The laser cutting quality control method for photovoltaic cells according to claim 1, characterized in that: The activation function f of the i-th neuron of the BP neural network regression prediction is i for, ， ， The weight β of the BP neural network after optimization of the i-th neuron function i and the bias δ ​​of the BP neural network after optimization of the i-th neuron function i The constraints are: ， Among them, y is the historical processing parameter data information of laser cutting photovoltaic cells, γ i is the activation factor of the activation function of the i-th neuron, and n is the sample size.

4. The laser cutting quality control method for photovoltaic cells according to claim 1, characterized in that: In step P3, the optimization of the processing parameters of laser cutting photovoltaic cells by using the Newton-Raphson optimization algorithm based on Gaussian kernel function includes: P31. Based on the predicted processing quality data information of the laser-cut photovoltaic cell and the processing parameter data information of the laser-cut photovoltaic cell, a Gaussian kernel function R of the laser cutting process is established. ， Wherein, a is the predicted processing quality data information of laser cutting photovoltaic cells, b is the processing parameter data information of laser cutting photovoltaic cells, η j is the Gaussian constant parameter of laser cutting photovoltaic cells, J is a positive integer, and the multi-dimensional space data of the processing quality and processing parameters of laser cutting photovoltaic cells are characterized to obtain the data information of the multi-dimensional space matrix of the processing quality and processing parameters of laser cutting photovoltaic cells; P32. Based on the data information of the multidimensional space matrix of the processing quality and processing parameters of the laser-cut photovoltaic cell, the Newton-Raphson population is initialized, the upper and lower bounds of the population variables are determined, and the data information of the initialized Newton-Raphson population is obtained; P33. Based on the data information of the initialized Newton-Raphson population, an optimization function G of the Newton-Raphson population is established. ， Among them, c is the data information of the initialized Newton-Raphson population, λ1, λ2 and λ3 are the optimization constant parameters of the population, and the processing parameters of the laser cutting photovoltaic cell are optimized to obtain the optimized processing parameter data information of the laser cutting photovoltaic cell.

5. The laser cutting quality control method for photovoltaic cells according to claim 4, characterized in that: The constraint function g of the optimization constant parameters λ1, λ2 and λ3 of the population is, ， ， Among them, the value range of the constraint function g is (2,3).

6. The laser cutting quality control method for photovoltaic cells according to claim 4, characterized in that: The Gaussian constant parameter η of the laser cutting photovoltaic cell j The constraints are: ， Wherein, J is a positive integer.

7. The laser cutting quality control method for photovoltaic cells according to claim 1, characterized in that: In step P4, the construction of the processing quality assessment model for laser cutting photovoltaic cells and characterizing the assessment value of the processing quality of laser cutting photovoltaic cells include: P41. Based on the optimized laser cutting photovoltaic cell processing parameter data information, construct a laser cutting photovoltaic cell processing parameter data set; P42. Inputting the processing parameter data set of the laser cutting photovoltaic cell into the processing quality assessment model of the laser cutting photovoltaic cell for training and learning, and determining the processing quality assessment function H of the laser cutting photovoltaic cell, ， Among them, r is the processing parameter data set of laser cutting photovoltaic cells, µ1, µ2 and µ3 are the evaluation factors of the processing quality of laser cutting photovoltaic cells, and the trained laser cutting photovoltaic cell processing quality evaluation model is obtained; P43. Based on the trained processing quality assessment model for laser cutting of photovoltaic cells, the optimized processing parameter data information for laser cutting of photovoltaic cells is input, the evaluation value of the processing quality of laser cutting of photovoltaic cells is characterized, and data correction information on the processing quality of laser cutting of photovoltaic cells is obtained. The correction information is used to assist in adjusting the laser cutting processing technology of photovoltaic cells.

8. The laser cutting quality control method for photovoltaic cells according to claim 7, characterized in that: The evaluation factors µ1, µ2 and µ3 of the processing quality of the laser cutting photovoltaic cell are: ， ， ， Among them, r is the processing parameter data set of laser cutting photovoltaic cells, and r0 is the mean value of the processing parameter data of laser cutting photovoltaic cells.

9. A laser cutting quality control system for photovoltaic cells, comprising a computer device, characterized in that: The computer device is programmed or configured to execute the steps of the laser cutting quality control method for photovoltaic cells as claimed in any one of claims 1 to 8.

Images