A control method and system for a vertical three-in-one production line used in PCB production

Abstract

The present invention discloses a control method and system for a vertical three-in-one production line for PCB production, which relates to the technical field of PCB production control. By acquiring vibration signals on the production line and using historical vibration data of the production line, a prediction model of vibration signals is established through a long short-term memory network. The vibration signals of the current production line in the future of producing PCB are predicted according to the vibration signals, and the production process parameters are adjusted and optimized according to the prediction results of the vibration signals and the real-time vibration signals; the optimization effect coefficient after the adjustment of the process parameters is calculated, and it is judged and evaluated whether the optimization of the process parameters is effective; if the optimization of the process parameters is effective, the vibration signals of the production line are acquired again, and it is judged whether it is necessary to continue to optimize the process parameters according to the prediction results of the vibration signals and the vibration signals of the production line, so as to realize the control of the vibration of the production line, reduce the possibility of causing potential impacts on the equipment of the production line, and improve the production quality of PCB boards.

Description

Technical Field

[0001] The present invention relates to the technical field of PCB production control, and particularly relates to a control method and system for a vertical three-in-one production line for PCB production. Background Art

[0002] In the production of PCBs, especially when it comes to PCB slag removal, drilling, and electroplating production, generally, the PCB boards are placed flat on the respective production lines for slag removal, drilling, and electroplating, and the processed PCB boards are transferred in sequence; this method requires manual or mechanical handling of the PCB boards between each process, which not only increases the production time but also may cause damage, contamination, or oxidation of the boards during handling, affecting the quality of the final product; in addition, since the boards are placed flat, the fluidity of the liquid medicine is limited in the copper plating and electroplating processes, which easily leads to uneven distribution of the plating layer, affecting the electrical conductivity and reliability, increasing the production cost, and reducing the overall production efficiency;

[0003] Moreover, during the operation of the PCB production line, normal vibration is an inherent physical phenomenon during the production process and will not have a substantial impact on production efficiency and equipment stability. However, if the vibration is high, that is, abnormal vibration may have an adverse impact on PCB processing accuracy, plating layer uniformity, and equipment operation status. Therefore, it is necessary to effectively monitor and analyze the vibration state; the existing method is to collect vibration signals of the production line in real time through high-precision sensors and analyze the spectral characteristics of the vibration signals using Fourier transform, focusing on the energy distribution changes in the high-frequency and low-frequency regions; if an abnormal vibration mode is detected, the system will automatically adjust relevant production process parameters, such as transmission speed, liquid flow pressure, and the pressure of the fixture fixing the PCB board, to optimize the production environment, reduce vibration interference, and ensure the stability of the PCB production process and product quality.

[0004] However, there is a problem with the above-mentioned solution. Fourier transform needs to collect sufficient vibration data through a certain time window and perform spectral conversion before evaluating the vibration state; when abnormal vibration occurs, the response lag of the system may lead to the inability to adjust process parameters in a timely manner, thus affecting the quality of PCB production; in addition, sometimes during the process of optimizing process parameters, only one optimization may not achieve the final expectation. If it is impossible to judge whether it is necessary to continue optimizing the process parameters during PCB slag removal, drilling, and electroplating production according to the actual situation, abnormal vibration may not be completely eliminated, which may have a potential impact on the equipment of the production line and result in low quality of PCB boards. Summary of the Invention

[0005] The object of the present invention is to solve the above-mentioned problems and provide a control method and system for a vertical three-in-one production line for PCB production.

[0006] In the first aspect of the implementation of the present invention, a control method for a vertical three-in-one production line for PCB production is first proposed. The method includes:

[0007] Deploy vibration sensors on the PCB vertical three-in-one production line to obtain vibration signals on the production line in real time. During the process of producing PCB on the PCB vertical three-in-one production line, the PCB board is clamped to be in a vertically suspended placement state, and slag removal, hole punching, and electroplating are continuously completed on the production line.

[0008] Utilize the historical vibration data during the production of PCB by the PCB vertical three-in-one production line to establish a prediction model of vibration signals through a long short-term memory network, predict the future vibration signals of the current production line during the production of PCB, and adjust and optimize the production process parameters according to the vibration signal prediction results and real-time vibration signals to control the production line to produce PCB boards.

[0009] Obtain the vibration signals of the production line and the working current of the production line within a period of time after the adjustment of the process parameters is completed, and calculate the optimization effect coefficient to evaluate whether the process parameter optimization is effective.

[0010] If the process parameter optimization is effective, obtain the vibration signals of the production line again, and judge whether it is necessary to continue to optimize the process parameters according to the vibration signal prediction results and the vibration signals of the production line, so as to realize the control of the vibration of the production line.

[0011] Optionally, the steps of adjusting and optimizing the production process parameters according to the vibration signal prediction results and real-time vibration signals to control the production line to produce PCB boards are as follows:

[0012] Compare the predicted vibration signal with the real-time obtained vibration signal, and calculate the deviation value between them. Once the deviation value is not less than the preset maximum allowable deviation value, immediately adjust and optimize the production process parameters to control the production line to produce PCB boards.

[0013] Optionally, the steps of adjusting and optimizing the production process parameters to control the production line to produce PCB boards are as follows:

[0014] The production process parameters include transmission speed, liquid flow pressure, and the pressure of the fixture for fixing the PCB board.

[0015] Take the deviation value between the predicted vibration signal and the real-time vibration signal as the input item of fuzzy logic, and divide them into fuzzy sets respectively.

[0016] Take the production process parameters as the output item of fuzzy logic, and divide them into different fuzzy sets.

[0017] Formulate fuzzy rules to describe the influence of the deviation value between the predicted vibration signal and the real-time vibration signal on the production process parameters.

[0018] Perform fuzzy inference according to fuzzy rules, optimize the production process parameters of the current production line, and control the production line to produce PCB boards.

[0019] Optionally, the steps for calculating the optimization effect coefficient based on the vibration signal and the working current of the production line within a period of time after the process parameters are adjusted are as follows:

[0020] Obtain the real-time vibration signal before the process parameters are adjusted, and calculate the average error before optimization ;

[0021] Obtain the real-time vibration signal within a period of time after the process parameters are adjusted, and calculate the average error after optimization ;

[0022] Calculate the root mean square error before optimization and the root mean square error after optimization ;

[0023] According to , , , Calculate the error attenuation coefficient , and the calculation formula is: ;

[0024] Calculate the vibration standard deviation before and after optimization and , and calculate the standard deviation ratio and according to , which is used to measure the degree of vibration dispersion: the calculation formula is: ;

[0025] Calculate the vibration optimization effective coefficient according to the error attenuation coefficient and the standard deviation ratio , and the calculation formula is: ;

[0026] Calculate the optimization effect coefficient based on the vibration optimization effective coefficient and the working current of the production line.

[0027] Optionally, the steps for calculating the optimization effect coefficient based on the vibration optimization effective coefficient and the working current of the production line are as follows:

[0028] Obtain the motor current of the production line at multiple moments before the process parameters are optimized , and subtract the motor current at the previous moment from the motor current at the current moment to obtain the current change rate at the corresponding moment before optimization , indicating the The current change rate at a moment;

[0029] Obtain the motor current of the production line at multiple moments after optimizing the process parameters , and subtract the motor current at the previous moment from the motor current at the current moment to obtain the current change rate at the corresponding moment after optimization , Indicates the current change rate at the th moment after optimization;

[0030] Divide and into intervals, calculate the probability distribution of each interval, and the calculation formula is: , , where in the formula, and respectively represent the probabilities that the current change rates before and after optimization fall within the th interval;

[0031] Calculate the information entropy before and after optimization respectively, and the calculation formula is: , , where in the formula, and are the information entropies before and after optimization respectively;

[0032] Calculate the ratio of the information entropy before and after optimization, and the calculation formula is: , if is less than 1, it means that the optimization reduces the disorder degree of the system and the optimization effect is better;

[0033] Calculate the optimization order degree improvement coefficient, and the calculation formula is: , where in the formula, is the optimization order degree improvement coefficient;

[0034] Calculate the optimization effect coefficient according to the vibration optimization effective coefficient and the optimization order degree improvement coefficient.

[0035] Optionally, the steps to calculate the optimization effect coefficient according to the vibration optimization effective coefficient and the optimization order degree improvement coefficient are:

[0036] ;

[0037] where in the formula, is the optimization effect coefficient, and are the vibration optimization effective coefficient and the optimization order degree improvement coefficient respectively, are respectively and preset ratio values, and All are greater than 0.

[0038] Optionally, evaluating whether the process parameter optimization is effective includes:

[0039] Comparing the optimization effect coefficient with the preset optimization effect coefficient threshold. If the optimization effect coefficient is not less than the preset optimization effect coefficient threshold, it indicates that the process parameter optimization to reduce the abnormal vibration of the production line is effective;

[0040] If the optimization effect coefficient is less than the preset optimization effect coefficient threshold, it indicates that the process parameter optimization to reduce the abnormal vibration of the production line is ineffective. At this time, control the production line to stop running and issue an alarm signal.

[0041] Optionally, the steps of judging whether to continue to optimize the process parameters according to the vibration signal prediction result and the vibration signal of the production line to realize the control of the production line vibration are:

[0042] Calculate the deviation value between the predicted vibration signal and the real-time vibration signal after the process parameter optimization. If the deviation value is still not less than the preset maximum allowable deviation value, continue to adjust and optimize the production process parameters until the deviation value is less than the preset maximum allowable deviation value, and the optimization effect coefficient after optimizing the process parameters again is not less than the preset optimization effect coefficient threshold. Then control the production line to produce PCB boards according to the corresponding production process parameters.

[0043] In the second aspect of the implementation of the present invention, a control system for a vertical three-in-one production line for PCB production is proposed. The system includes:

[0044] Data acquisition module: Deploy vibration sensors on the PCB vertical three-in-one production line to obtain the vibration signal on the production line in real time; during the process of producing PCB on the PCB vertical three-in-one production line, the PCB board is clamped and placed in a vertically suspended state, and deburring, drilling, and electroplating are continuously completed on the production line;

[0045] Preliminary optimization module: Use the historical vibration data of the PCB vertical three-in-one production line during the production of PCB to establish a prediction model of the vibration signal through a long short-term memory network, predict the future vibration signal of the current production line during the production of PCB, and adjust and optimize the production process parameters according to the vibration signal prediction result and the real-time vibration signal to control the production line to produce PCB boards;

[0046] Judgment and evaluation module: Obtain the vibration signal of the production line and the working current of the production line within a period of time after the process parameter adjustment is completed, and calculate the optimization effect coefficient to evaluate whether the process parameter optimization is effective;

[0047] Optimization control module: If the process parameter optimization is effective, obtain the vibration signal of the production line again, and determine whether it is necessary to continue optimizing the process parameters based on the vibration signal prediction result and the vibration signal of the production line, so as to achieve the control of the vibration of the production line.

[0048] Advantages of the present invention:

[0049] 1. The present invention proposes a control method and system for a vertical three-in-one production line for PCB production. By vertically suspending the PCB board and continuously completing slag removal, drilling, and electroplating on the same production line, not only the manual or mechanical handling link is eliminated, the production time is greatly reduced, but also the plating solution can evenly cover the entire board surface, improving the consistency of the plating quality and electrical performance; it can significantly improve production efficiency and reduce costs.

[0050] 2. When abnormal vibration occurs on the production line, the system can respond in a timely manner and adjust the process parameters in a timely manner, thereby reducing the impact on the quality of PCB production; in addition, it can dynamically optimize the process parameters during PCB slag removal, drilling, and electroplating according to the actual situation, achieving the ultimate goal of suppressing abnormal vibration, reducing the possibility of potential impact on the equipment of the production line, and improving the quality of PCB boards. Description of the drawings

[0051] The following further describes the present invention with reference to the accompanying drawings.

[0052] Figure 1 It is a flowchart of a control method for a vertical three-in-one production line for PCB production;

[0053] Figure 2 It is a framework diagram of a control system for a vertical three-in-one production line for PCB production. Specific embodiments

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0055] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0056] The embodiments of the present invention provide a control method for a vertical three-in-one production line for PCB production. Refer to Figure 1 , Figure 1Flowchart of a control method for a vertical three-in-one production line for PCB production provided by an embodiment of the present invention. The method includes the following steps:

[0057] Deploy vibration sensors on the PCB vertical three-in-one production line to obtain vibration signals on the production line in real time; during the process of producing PCBs on the PCB vertical three-in-one production line, the PCB board is clamped and placed in a vertically suspended state, and slag removal, drilling, and electroplating are continuously completed on the production line;

[0058] Utilize the historical vibration data of the PCB vertical three-in-one production line during PCB production to establish a prediction model of vibration signals through a long short-term memory network, predict the future vibration signals of the current production line during PCB production, and adjust and optimize the production process parameters according to the vibration signal prediction results and real-time vibration signals to control the production line to produce PCB boards;

[0059] Obtain the vibration signals of the production line and the working current of the production line within a period of time after the process parameter adjustment is completed to calculate the optimization effect coefficient, which is used to evaluate whether the process parameter optimization is effective;

[0060] If the process parameter optimization is effective, obtain the vibration signals of the production line again, and determine whether it is necessary to continue optimizing the process parameters according to the vibration signal prediction results and the vibration signals of the production line to achieve the control of the vibration of the production line.

[0061] Based on the control method for a vertical three-in-one production line for PCB production provided by an embodiment of the present invention, through the above method, when abnormal vibration occurs, the system can respond in a timely manner, and can adjust the process parameters in a timely manner, thereby reducing the impact on the quality of PCB production; in addition, sometimes during the process of optimizing process parameters, the process parameters during PCB slag removal, drilling, and electroplating production can be dynamically optimized according to the actual situation, achieving the ultimate goal of suppressing abnormal vibration, reducing the possibility of potential impact on the equipment of the production line, and improving the quality of PCB boards.

[0062] In one embodiment, the step of deploying vibration sensors on the PCB vertical three-in-one production line to obtain vibration signals on the production line in real time is as follows:

[0063] Determine the acquisition target of vibration signals: First, it is necessary to clarify which parts of the PCB vertical three-in-one production line to install vibration sensors to obtain vibration signals in real time; these signals should include key parts related to the operating state of the production line, such as the vibration characteristics of the transmission device, fixture fixing points, copper plating tank, electroplating tank, etc.; it is also necessary to determine the vibration frequency range and signal intensity to be monitored.

[0064] Choose the right vibration sensor: Choose the right accelerometer, displacement sensor or velocity sensor according to the frequency and amplitude of the vibration signal and the required sensitivity of the sensor; generally, the vibration frequency range of the PCB production line is between tens of Hz and hundreds of Hz, so choose a sensor suitable for this range; choose the sensor installation position reasonably according to the vibration characteristics of each production process equipment to ensure that the most important vibration signal can be captured; a three-axis acceleration sensor can be selected to capture vibration signals in three dimensions. For specific processes, single-axis or multi-axis sensors can be selected as needed. The output signal of the sensor is transmitted to the data acquisition system through suitable cables and connectors. When wiring, ensure to avoid mechanical interference and interference with other parts of the production line.

[0065] Deploy a data acquisition system to receive and store vibration signals collected by sensors in real time; these signals can be analog or digital, and are processed using an appropriate analog-to-digital conversion (ADC) device depending on the output type of the sensor. Ensure that the data acquisition frequency is high enough to capture key vibration changes; in addition, the real-time transmission of sensor signals can use wireless transmission technology (such as Wi-Fi, Bluetooth) or wired transmission technology (such as RS485, Ethernet) to transmit the signal to the data processing platform in real time; the data acquisition system should have certain signal processing functions (such as denoising, filtering) to ensure signal stability during transmission; monitor the vibration signals of the production line in real time through the data processing platform, and sample, store and preliminarily analyze the vibration data. This process should minimize signal delays to ensure real-time data transmission and feedback.

[0066] Vibration signal feature extraction: Extract key features from real-time vibration signals, such as root mean square (RMS) value, peak value, spectral characteristics, etc. These features will serve as input for subsequent vibration prediction models.

[0067] In one embodiment, the historical vibration data of the PCB vertical three-in-one production line during normal operation of the PCB production is used to establish a vibration signal prediction model through a long short-term memory network to predict the vibration signal of the current production line in the future when producing PCBs, and the production process parameters are adjusted and optimized according to the vibration signal prediction results and the real-time vibration signal to control the production line to produce PCB boards;

[0068] Among them, the historical vibration data of the PCB vertical three-in-one production line when producing qualified PCBs is used to establish a prediction model of vibration signals through a long short-term memory network. The specific steps for predicting the vibration signal of the current production line in the future when producing PCBs are as follows:

[0069] First, collect the historical vibration data of the production line; this data usually includes timestamps, raw data of vibration signals (such as acceleration, velocity, or displacement), the positions of sensors, etc.; ensure that the data covers a period of time to capture the periodic characteristics and trends of the vibration signals.

[0070] Data cleaning and denoising: Clean the collected vibration data, remove missing values and outliers, and perform normalization to eliminate the influence of different dimensions, making the contributions of each feature to the model more balanced.

[0071] Divide the collected historical data into training set, validation set, and test set. Usually, the training set is used to train the model, the validation set is used for hyperparameter tuning and mid-term evaluation, and the test set is used for verifying the performance of the final model. The ratio of data division is generally 70% (training set), 15% (validation set), and 15% (test set).

[0072] Define the input-output structure: According to the time series characteristics of the vibration signal, design the input-output structure of the LSTM model. Assume that the goal of the model is to predict the vibration signal in the future for a certain period of time. A certain time step (such as the vibration data in the past 10 minutes or 30 minutes) can be selected as the input, and the output is the vibration prediction value in the future for 1 minute, 10 minutes, or longer. For example, the input of the model may include the vibration data of the past few time steps, and the output is the vibration value at the future time step (such as the vibration acceleration or velocity in the future 1 minute).

[0073] Design the number of layers and neurons of the LSTM network. The number of layers of the LSTM model generally ranges from 1 to 3, and the number of neurons in each layer can be adjusted according to the complexity of the data. Usually, more complex data may require more LSTM layers and a larger number of neurons. For example, use 2 layers of LSTM, with 128 neurons in each layer.

[0074] Select activation functions and optimizers: The activation functions in the LSTM model usually use the tanh and sigmoid functions. The former is used for the hidden state, and the latter is used for the forget gate. The Adam optimizer is often selected as the optimizer because it can adaptively adjust the learning rate and performs stably during training. The loss function usually uses the mean squared error (MSE) for regression tasks.

[0075] Use the training set data to train the LSTM model. The goal is to adjust the weights and biases in the network through the backpropagation algorithm to minimize the error between the predicted vibration signal and the true vibration signal. The training process will process the data in batches and update the model parameters once for each training.

[0076] Iterative Optimization: Through multiple rounds of training, the parameters of the model are continuously optimized using the training set data. During training, the change of the loss function can be monitored to ensure that the model converges to a better solution. Usually, after each round of training, the performance of the model on the validation set is evaluated, and hyperparameters such as the learning rate and batch size are adjusted to avoid overfitting or underfitting.

[0077] By using the early stopping strategy to monitor the performance on the validation set, the training is stopped when the loss on the validation set no longer improves. At the same time, regularization methods such as dropout layers or L2 regularization can also be used to avoid the model overfitting the training data.

[0078] Model Evaluation: By evaluating the model on the validation set and the test set, the prediction effect of the model is judged. Evaluation metrics such as the mean squared error (MSE) and mean absolute error (MAE) can be calculated to measure the prediction error. The better the performance of the model, the stronger its ability to predict future vibration signals.

[0079] Vibration Signal Prediction: After training is completed, the trained LSTM model is used to predict the vibration signals in the future time period. Inputting the vibration data at the current moment, the model will generate the predicted values of the vibration signals at future time steps. These predicted values can be used to evaluate the vibration trend in the production process, predict future vibration changes, predict the vibration signals of the current production line during the production of PCBs in the future for a period of time, and adjust and optimize the production process parameters according to the vibration signal prediction results and real-time vibration signals to control the production line to produce PCBs;

[0080] Among them, the steps of adjusting and optimizing the production process parameters according to the vibration signal prediction results and real-time vibration signals to control the production line to produce PCBs are as follows:

[0081] Compare the predicted vibration signal with the real-time acquired vibration signal, and calculate the deviation value between them. Once the deviation value is not less than the preset maximum allowable deviation value, immediately adjust and optimize the production process parameters to control the production line to produce PCBs.

[0082] It should be noted that the preset maximum allowable deviation value is set by professionals according to the actual situation, and no specific limitation and elaboration are made.

[0083] In one implementation method, through the above method, it is possible to timely detect and respond to potential production anomalies based on the combination of real-time data and prediction information, rather than relying solely on historical data or static rules, thereby improving the stability and efficiency of the production process. When abnormal vibration occurs, the system can respond in a timely manner to ensure that process parameters can be adjusted in a timely manner, reducing the impact on the quality of PCB production. By dynamically adjusting process parameters, the production line can maintain optimal performance in a fluctuating working environment, not only reducing equipment failures caused by vibration problems, but also ensuring that the quality of the produced PCB boards is more consistent, improving the overall production efficiency and product qualification rate. At the same time, this closed-loop control method helps with real-time feedback and optimization, reducing risks in the production process and increasing the adaptive ability and intelligent level of the production line.

[0084] In one embodiment, the steps for adjusting and optimizing production process parameters to control the production line to produce PCB boards are as follows:

[0085] The production process parameters include transmission speed, liquid flow pressure, and the pressure of the fixture for fixing the PCB board.

[0086] Take the deviation value between the predicted vibration signal and the real-time vibration signal as the input item of fuzzy logic, and divide them into fuzzy sets respectively.

[0087] Take the production process parameters as the output item of fuzzy logic, and divide them into different fuzzy sets.

[0088] Formulate fuzzy rules to describe the influence of the deviation value between the predicted vibration signal and the real-time vibration signal on the production process parameters.

[0089] Perform fuzzy reasoning according to the fuzzy rules to optimize the production process parameters of the current production line and control the production line to produce PCB boards.

[0090] It should be noted that the steps for adjusting and optimizing production process parameters, combined with the specific implementation process of fuzzy logic control, can be specifically divided into the following key links:

[0091] 1. Calculation of the vibration signal deviation value and division of fuzzy sets;

[0092] First, it is necessary to compare the future vibration signal predicted by the long short-term memory network (LSTM) model with the vibration signal collected in real time, and calculate the deviation value between the two. This deviation value represents the difference between the actual vibration and the predicted vibration, reflecting the working state of the production line. If the deviation value is large, it means that the current production state deviates from the expected target, which may lead to abnormal vibration. Next, input this deviation value into the fuzzy logic system and divide it into different fuzzy sets according to the preset criteria, such as "small deviation", "medium deviation", and "large deviation".

[0093] 2. Fuzzy set partitioning of production process parameters;

[0094] In fuzzy logic control, the production process parameters (transmission speed, liquid flow pressure, and fixture fixing pressure) are used as output items and need to be partitioned into different fuzzy sets. For example:

[0095] Transmission speed: can be partitioned into "slow", "moderate", and "high speed";

[0096] Liquid flow pressure: can be partitioned into "low pressure", "moderate", and "high pressure";

[0097] Fixture fixing pressure: can be partitioned into "light pressure", "moderate", and "heavy pressure".

[0098] These fuzzy sets represent the adjustment ranges of different process parameters and are used to control the vibration impact during the production process.

[0099] 3. Formulation of fuzzy rules:

[0100] Based on actual production experience and experimental data, formulate fuzzy rules to describe the relationship between the deviation value and the process parameters. These rules will guide how to adjust the production process parameters according to the magnitude of the deviation value. For example:

[0101] If the deviation value is "large deviation", then "reduce the transmission speed", "increase the liquid flow pressure", and "increase the fixture fixing pressure" should be taken to reduce vibration;

[0102] If the deviation value is "medium deviation", then "appropriately reduce the transmission speed" and "adjust the liquid flow pressure" can be taken;

[0103] If the deviation value is "small deviation", then it may not be necessary to adjust the process parameters and keep the existing state.

[0104] The fuzzy rules are based on the experience accumulation in production and propose corresponding process adjustment suggestions for different vibration states.

[0105] 4. Fuzzy inference and process parameter optimization;

[0106] Fuzzy inference is to perform inference calculations on the input fuzzy values (i.e., deviation values) and the fuzzy rules to obtain the corresponding output fuzzy sets (i.e., optimized production process parameters); for example, inputting "large deviation" may result in "low speed", "high pressure", and "heavy pressure" as the output of the optimized process parameters. Through the results of fuzzy inference, the optimized production process parameters can adjust the operating state of the production line, control the vibration during the production process, and reduce the impact of abnormal vibration on production quality and efficiency.

[0107] 5. Process parameter control and vibration monitoring;

[0108] Finally, the optimized process parameters obtained through fuzzy inference are fed back into the production line to adjust parameters such as the transmission speed, liquid flow pressure, and fixture fixing pressure, so that the production line is in a more stable working state. Monitor the vibration signal in real time to ensure the effectiveness of the optimization measures. If the optimized process parameters still fail to significantly reduce the vibration, re-optimization is required;

[0109] Among them, fuzzy inference calculates based on the input fuzzy values (i.e., deviation values) and the set fuzzy rules to derive the optimized production process parameters. This process is mainly achieved through a fuzzy inference system (Fuzzy Inference System, FIS). Commonly used methods include Mamdani-type fuzzy inference and Sugeno-type fuzzy inference. In this process, the deviation value is used as the input, and it is inferred with relevant fuzzy rules, and the corresponding output is obtained through the intersection and union operations of fuzzy sets. For example, if the deviation value of the vibration signal is "large deviation", it may indicate that the production line is in an abnormal state. Then, according to the rule inference, the optimized production process parameters may be "low-speed transmission", "high-pressure liquid flow", and "heavy-pressure fixture". The specific inference steps are as follows:

[0110] Fuzzification: First, convert the input precise data (such as the deviation value of the vibration signal) into a fuzzy set. For example, the deviation value between 0 and 100 may be divided into "small deviation", "medium deviation", and "large deviation". These fuzzy values are calculated through a membership function, which maps the numerical value into the corresponding fuzzy set. Commonly used types of membership functions include triangular, trapezoidal, Gaussian, etc.

[0111] Rule base application: Based on the input fuzzy values and the formulated fuzzy rules, infer the output; the rules can be derived through a fuzzy inference algorithm (such as Mamdani-type inference) to determine the output fuzzy set.

[0112] Defuzzification: Through defuzzification operation, convert the output value of the fuzzy set into specific and operable process parameters. Commonly used defuzzification methods include the centroid method and the max membership method. For example, the process parameter such as "low speed" calculated by the centroid method may be quantified as a specific transmission speed, such as 2 meters per minute, and "high pressure" may be quantified as a specific liquid flow pressure value, such as 15 bar.

[0113] Through these steps, fuzzy inference can effectively optimize the production process parameters, enabling the production line to quickly take adjustment measures in case of abnormal vibration, avoiding further quality problems and efficiency reduction.

[0114] In one implementation manner, the optimization of the production process based on fuzzy logic control can not only respond to the changes in vibration signals in real time, but also be flexibly adjusted according to the continuously changing states in the actual production process, improving production efficiency, ensuring the quality of PCB boards, and reducing the potential risks brought by abnormal vibrations.

[0115] In one embodiment, within a period of time after the adjustment of process parameters is completed, the vibration signal of the production line and the working current of the production line are used to calculate an optimization effect coefficient for evaluating whether the process parameter optimization is effective.

[0116] Among them, the steps of calculating the optimization effect coefficient using the vibration signal of the production line and the working current of the production line within a period of time after the adjustment of process parameters is completed are as follows:

[0117] Obtain the real-time vibration signal before the adjustment of process parameters , and calculate the average error before optimization, that is, the deviation between the measured vibration data and the predicted vibration data. The calculation formula is: ; in the formula, is the average error before optimization, is the th real-time vibration signal before optimization, is the predicted th vibration signal before optimization, represents the number of sampling points before optimization;

[0118] Obtain the real-time vibration signal within a period of time after the adjustment of process parameters , and calculate the average error after optimization, that is, the deviation between the measured vibration data and the predicted vibration data. The calculation formula is: ; in the formula, is the average error after optimization, is the th real-time vibration signal after optimization, is the predicted th vibration signal after optimization, represents the number of sampling points after optimization;

[0119] Calculate the root mean square error before and after optimization to measure the overall deviation of vibration and can amplify larger errors. The calculation formula is: , in the formula, is the root mean square error before optimization; ; is the root mean square error after optimization;

[0120] RMSE calculates the square root of the mean of the squared errors, which can reduce the influence of small errors and highlight the contribution of large errors.

[0121] The optimized RMSE should be lower than the RMSE before optimization, indicating that the optimization reduces the vibration deviation;

[0122] Calculate the error attenuation coefficient , and the calculation formula is: ;

[0123] Calculate the vibration standard deviation before and after optimization and , according to and calculate the standard deviation ratio , which is used to measure the degree of vibration dispersion: the calculation formula is: ;

[0124] According to the error attenuation coefficient and the standard deviation ratio calculate the vibration optimization effectiveness coefficient , and the calculation formula is: ;

[0125] Calculate the optimization effect coefficient according to the vibration optimization effectiveness coefficient and the working current of the production line.

[0126] It should be noted that the smaller the error attenuation coefficient and the smaller the standard deviation ratio, the better the optimization effect of the process parameter vibration, which is used to measure the degree of error reduction, the smaller it is, the more effective the optimization; which is used to measure the stability of vibration, the smaller it is, the more effective the optimization;

[0127] It should be noted that the real-time vibration signals before and after the adjustment of the process parameters can be obtained through the installed sensors; the predicted vibration signals before and after optimization can be directly obtained through the output of the trained model;

[0128] It should be noted that the larger the vibration optimization effectiveness coefficient and the closer it is to 1, the smaller the impact of the vibration of the production line after adjusting process parameters on the production line and the production of PCB boards. The vibration optimization effectiveness coefficient is jointly determined by the error attenuation coefficient and the vibration stability. Among them, the error attenuation coefficient reflects the reduction of vibration error after optimization, while the vibration stability measures the dispersion degree of vibration signals. If the vibration optimization effectiveness coefficient is close to 1, it indicates that the vibration error after optimization is significantly reduced, the vibration signal tends to be stable, and the impact on production equipment and PCB board production is significantly reduced, and the optimization effect is ideal; on the contrary, if the optimization effectiveness coefficient is small, it indicates that the effect of the optimization measures is not good, the vibration is still relatively intense or the error attenuation is not obvious. Combining the root mean square error, standard deviation ratio, and error attenuation coefficient of vibration before and after optimization can comprehensively evaluate the optimization effect of process parameter adjustment on vibration signals, so as to guide further parameter optimization and improve the stability of the production process and product quality.

[0129] In one implementation method, by calculating the average error and root mean square error of vibration signals, the deviation of vibration signals before and after process adjustment can be measured, and then the change of the vibration state of the production line before and after optimization can be revealed. This calculation method can effectively highlight large errors and reduce the interference of small errors, so that the significant changes during the optimization process are more prominent and easy to identify and analyze. Secondly, through the calculation of the error attenuation coefficient, the actual effect of the optimization measures in reducing vibration deviation can be understood, while the standard deviation ratio helps to evaluate the vibration volatility and dispersion degree after optimization, which is crucial for judging whether the optimization measures bring a more stable production environment. Finally, by comprehensively calculating these indicators to obtain the vibration optimization effectiveness coefficient, the comprehensive effect of the optimization measures can be fully reflected, ensuring that the vibration of the production line is made more stable through process adjustment, so as to provide data support for subsequent optimization decisions.

[0130] In one embodiment, the steps for calculating the optimization effect coefficient according to the vibration optimization effectiveness coefficient and the working current of the production line are as follows:

[0131] Obtain the motor currents of the production line at multiple moments before optimizing process parameters , and subtract the motor current at the previous moment corresponding to the current moment from the motor current at the current moment to obtain the current change rate at the corresponding moment before optimization , denotes the current change rate at the th moment before optimization;

[0132] Obtain the motor currents of the production line at multiple moments after optimizing process parameters , and subtract the motor current at the previous moment corresponding to the current moment from the motor current at the current moment to obtain the current change rate at the corresponding moment after optimization , denotes the Rate of change of current at a moment

[0133] Divide and into intervals, calculate the probability distribution of each interval, and the calculation formula is: , , where and respectively represent the probabilities that the rate of change of current before and after optimization falls into the th interval; represents the number of divided intervals (which can be selected according to the data distribution);

[0134] Calculate the information entropy before and after optimization respectively, and the calculation formula is: , , where and are the information entropies before and after optimization respectively;

[0135] Calculate the ratio of the information entropy before and after optimization, and the calculation formula is: , if is less than 1, it means that the optimization reduces the disorder degree of the system and the optimization effect is better;

[0136] Calculate the optimization order degree improvement coefficient, and the calculation formula is: , where is the optimization order degree improvement coefficient;

[0137] Calculate the optimization effect coefficient according to the vibration optimization effective coefficient and the optimization order degree improvement coefficient.

[0138] It should be noted that the current sensors or power monitoring devices on the production line will collect the current values of the motor at different time points and store them in the database or industrial control system (such as SCADA or PLC); subsequently, data cleaning is performed through data processing software or programming languages (such as Python, MATLAB) to remove outliers, and the rate of change of current at adjacent moments is calculated in chronological order. Then, statistical analysis is performed on the data, and the rate of change of current is divided into different intervals by using the histogram or kernel density estimation method to calculate the probability distribution. Finally, the change in the disorder degree of the system before and after optimization is calculated through the information entropy formula, so as to obtain the optimization order degree improvement coefficient.

[0139] It should be noted that the larger the optimization order degree improvement coefficient is, the smaller the impact of the vibration of the production line after adjusting the process parameters on the production line and the PCB board production is. Because the decrease in information entropy means that the uncertainty or disorder degree inside the system has been effectively controlled, which further makes the current change rate more stable and consistent. This stable current change usually indicates that the operation state of the production line after optimization is more stable, which can reduce mechanical shocks, vibrations and failures caused by irregular current fluctuations, thereby improving production efficiency and the production quality of PCB boards. In this way, the optimization not only improves the electrical performance of the production line, but also greatly reduces the potential negative impacts on equipment, processes and the final product, thus realizing a more precise and efficient production process.

[0140] In one implementation method, by calculating the current change rate and dividing the interval to calculate the probability distribution, the patterns of current fluctuations before and after optimization can be intuitively understood, so as to better evaluate whether the optimization measures have produced substantial effects in reducing system volatility and improving system stability. Next, through the calculation of information entropy, the "disorder degree" of the system can be effectively quantified, and the change in the disorder degree of the system before and after optimization can be calculated through the ratio, which is very crucial for judging whether the process adjustment has brought more orderly and stable current fluctuations. And the situation where the information entropy ratio is less than 1 indicates that the optimization has indeed played a role in reducing system fluctuations and improving system stability. Finally, calculating the optimization order degree improvement coefficient can synthesize these data and provide a comprehensive and clear quantitative result for the effect of production line optimization, enabling further optimization decisions to be based on reliable data support.

[0141] In one embodiment, the steps for calculating the optimization effect coefficient according to the vibration optimization effective coefficient and the optimization order degree improvement coefficient are as follows:

[0142] ;

[0143] In the formula, is the optimization effect coefficient, and are the vibration optimization effective coefficient and the optimization order degree improvement coefficient respectively, are respectively and 's preset proportional values, and are both greater than 0;

[0144] It should be noted that is set by professionals according to the actual situation. Generally, 's sum is 1. For example, can be 0.5, 0.5 respectively, or other numbers, and no specific limitation is made.

[0145] In one embodiment, the vibration signal of the production line and the working current of the production line within a period of time after the process parameter adjustment is completed are obtained to calculate the optimization effect coefficient, which is used to evaluate whether the process parameter optimization is effective;

[0146] Among them, the evaluation of whether the process parameter optimization is effective includes:

[0147] The optimization effect coefficient is compared with the preset optimization effect coefficient threshold. If the optimization effect coefficient is not less than the preset optimization effect coefficient threshold, it means that the process parameter optimization is effective in reducing the abnormal vibration of the production line.

[0148] If the optimization effect coefficient is less than the preset optimization effect coefficient threshold, it means that the process parameter optimization to reduce the abnormal vibration of the production line is ineffective. At this time, the production line is controlled to stop running and an alarm signal is issued.

[0149] It should be noted that the preset optimization effect coefficient threshold is set by professionals based on actual conditions and is not specifically limited or elaborated.

[0150] In one implementation, it is a key link in stable operation and efficient production in the vertical three-in-one production line of PCB production.

[0151] By comparing the optimization effect coefficient with the preset optimization effect coefficient threshold, it is possible to scientifically and quantitatively determine whether the process parameter adjustment has achieved the expected effect. If the optimization effect coefficient is not less than the preset threshold, it means that by adjusting the process parameters, the abnormal vibration of the production line has been effectively suppressed, and the stability and operating efficiency of the production line have been improved; at this time, the production line can continue to operate efficiently and maintain its normal production rhythm.

[0152] On the contrary, if the optimization effect coefficient is less than the preset optimization effect coefficient threshold, it means that the process parameter adjustment has not effectively reduced the vibration, and the production line still has large fluctuations, which may affect the production quality and even damage the equipment. In this case, in order to prevent greater losses, the control system will automatically stop the production line and send out an alarm signal to remind the staff to carry out maintenance; this measure can avoid continuing production in an unstable vibration environment to the greatest extent, ensure the long-term safe operation of the equipment, and provide time and space for subsequent adjustments and optimizations to further ensure the stability of the production line and production quality.

[0153] This evaluation method based on the optimization effect coefficient can not only detect problems in a timely manner and take measures, but also provide data support for the continuous optimization of process parameters, ultimately ensuring the efficient and stable operation of the vertical three-in-one production line in the PCB production process.

[0154] In one embodiment, if the optimization of process parameters is effective, the vibration signal of the production line is acquired again, and it is determined whether to continue optimizing the process parameters based on the vibration signal prediction result and the vibration signal of the production line, so as to control the vibration of the production line;

[0155] Among them, the steps of determining whether to continue optimizing the process parameters based on the vibration signal prediction result and the vibration signal of the production line to control the vibration of the production line are as follows:

[0156] Calculate the deviation value between the predicted vibration signal and the real-time vibration signal after the optimization of process parameters. If the deviation value is still not less than the preset maximum allowable deviation value, continue to adjust and optimize the production process parameters until the deviation value is less than the preset maximum allowable deviation value, and the optimization effect coefficient after optimizing the process parameters again is not less than the preset optimization effect coefficient threshold, then control the production line to produce PCB boards according to the corresponding production process parameters.

[0157] It should be noted that by calculating the deviation value between the predicted vibration signal after optimization and the real-time vibration signal, the vibration state in the current production process is judged. If the deviation value is still greater than the preset maximum allowable deviation value, it indicates that the current process parameters still cannot completely eliminate the vibration problem of the production line. Therefore, it is necessary to further adjust and optimize the production process parameters. Through this process, the production line can be continuously adjusted until the vibration deviation value drops within the preset range, and at the same time, the optimization effect coefficient must also reach the preset threshold to ensure that the optimization effect of the process parameters meets the expectations. When the vibration reaches the acceptable range and the optimization effect coefficient meets the requirements, the system will continue to control the operation of the production line according to the final production process parameters to complete the production of PCB boards.

[0158] In one implementation manner, the vibration is reduced by continuously optimizing the process parameters to avoid the decrease in production efficiency or equipment damage caused by vibration problems. Secondly, through refined control, the accuracy of PCB production can be greatly improved, ensuring the consistency and reliability of product quality. If not adjusted in time, excessive vibration may affect the quality of PCB boards and even cause the shutdown of the production line. Through this method, each step in the production process can be carried out under the optimized process parameters, thereby improving the working efficiency of the production line, reducing the fluctuations in production, reducing the occurrence of rework and defective products, and ultimately enhancing the overall production efficiency.

[0159] Based on the same inventive concept, the embodiment of the present invention also provides a control system for a vertical three-in-one production line for PCB production. Refer to Figure 2 , Figure 2 which is a framework diagram of a control system for a vertical three-in-one production line for PCB production provided by the embodiment of the present invention. The system includes:

[0160] Data acquisition module: Vibration sensors are deployed on the PCB vertical three-in-one production line to obtain vibration signals on the production line in real time. During the production of PCBs, the PCB vertical three-in-one production line clamps the PCB boards to place them in a vertical hanging state, and continuously completes slag removal, drilling and electroplating on the production line.

[0161] Preliminary optimization module: Using the historical vibration data of the PCB vertical three-in-one production line when producing PCBs, a vibration signal prediction model is established through a long short-term memory network to predict the future vibration signals of the current production line when producing PCBs. The production process parameters are adjusted and optimized based on the vibration signal prediction results and real-time vibration signals to control the production line to produce PCB boards;

[0162] Judgment and evaluation module: Obtain the vibration signal of the production line and the working current of the production line within a period of time after the process parameter adjustment is completed to calculate the optimization effect coefficient, which is used to evaluate whether the process parameter optimization is effective;

[0163] Optimization control module: If the process parameter optimization is effective, the vibration signal of the production line is obtained again, and based on the vibration signal prediction result and the vibration signal of the production line, it is determined whether the process parameters need to be further optimized to achieve control of the production line vibration.

[0164] Based on a control system of a vertical three-in-one production line for PCB production provided by an embodiment of the present invention, through the above-mentioned method, when abnormal vibration occurs, the system can respond in time and adjust the process parameters in time, thereby reducing the impact on the PCB production quality; in addition, sometimes in the process of optimizing the process parameters, the process parameters of PCB slag removal, drilling and electroplating production can be dynamically optimized according to the actual situation, so as to achieve the ultimate expectation of suppressing abnormal vibration, reduce the possibility of potential impact on the equipment of the production line, and improve the production quality of PCB boards.

[0165] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be used to artificially limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A control method for a vertical three-in-one production line used in PCB production, characterized in that, The method includes the following steps: Deploy vibration sensors on the PCB vertical triple-in-one production line to obtain vibration signals on the production line in real time; during the process of producing PCB on the PCB vertical triple-in-one production line, the PCB board is clamped and placed in a vertically suspended state, and slag removal, drilling, and electroplating are continuously completed on the production line; Utilize the historical vibration data of the PCB vertical triple-in-one production line during the production of PCB, establish a prediction model of vibration signals through a long short-term memory network, predict the future vibration signals of the current production line during the production of PCB, and adjust and optimize the production process parameters according to the vibration signal prediction results and real-time vibration signals to control the production line to produce PCB boards; Obtain the vibration signals of the production line and the working current of the production line within a period of time after the adjustment of the process parameters is completed, and calculate the optimization effect coefficient to evaluate whether the process parameter optimization is effective; If the process parameter optimization is effective, obtain the vibration signals of the production line again, and judge whether it is necessary to continue optimizing the process parameters according to the vibration signal prediction results and the vibration signals of the production line to achieve the control of the vibration of the production line; The steps of obtaining the vibration signals of the production line and the working current of the production line within a period of time after the adjustment of the process parameters is completed to calculate the optimization effect coefficient are as follows: Obtain the real-time vibration signal before adjusting the process parameters and calculate the average error before optimization , that is, the deviation between the measured vibration data and the predicted vibration data; Obtain the real-time vibration signal within a period of time after adjusting the process parameters, and calculate the optimized average error , that is, the deviation between the measured vibration data and the predicted vibration data; Root mean square error before calculation optimization and root mean square error after optimization ; According to , , , calculate the error attenuation coefficient , and the calculation formula is: ; Calculate the standard deviation of vibration before and after optimization and , according to and calculate the standard deviation ratio , used to measure the degree of dispersion of vibration: the calculation formula is: ; According to the error attenuation coefficient and the standard deviation ratio calculate the vibration optimization effective coefficient , and the calculation formula is: ; Calculate the optimization effect coefficient according to the vibration optimization effective coefficient and the working current of the production line.

2. The control method of a vertical three-in-one production line for PCB production according to claim 1, characterized in that, The steps of adjusting and optimizing the production process parameters according to the vibration signal prediction results and real-time vibration signals to control the production line to produce PCB boards are as follows: Compare the predicted vibration signal with the real-time obtained vibration signal, and calculate the deviation value between them. Once the deviation value is not less than the preset maximum allowable deviation value, immediately adjust and optimize the production process parameters to control the production line to produce PCB boards.

3. The control method of a vertical three-in-one production line for PCB production according to claim 2, characterized in that The steps of adjusting and optimizing the production process parameters to control the production line to produce PCB boards are as follows: The production process parameters include transmission speed, liquid flow pressure, and the pressure of the fixture for fixing the PCB board; Take the deviation value between the predicted vibration signal and the real-time vibration signal as the input item of fuzzy logic, and divide them into fuzzy sets respectively; Take the production process parameters as the output item of fuzzy logic, and divide them into different fuzzy sets; Formulate fuzzy rules to describe the influence of the deviation value between the predicted vibration signal and the real-time vibration signal on the production process parameters; Conduct fuzzy reasoning according to the fuzzy rules to optimize the production process parameters of the current production line and control the production line to produce PCB boards.

4. The control method of a vertical three-in-one production line for PCB production according to claim 1, characterized in that, The steps of calculating the optimization effect coefficient according to the vibration optimization effective coefficient and the working current of the production line are as follows: Obtain the motor currents of the production line at multiple moments before process parameter optimization , and subtract the motor current at the previous moment from the motor current at the current moment to obtain the current change rate corresponding to the moment before optimization , denote the current change rate at the th moment before optimization; Obtain the motor currents of the production line at multiple moments after the process parameters are optimized , and subtract the motor current at the previous moment from the motor current at the current moment to obtain the current change rate at the corresponding moment after optimization , denote the current change rate at the th moment after optimization; Divide and into intervals, calculate the probability distribution of each interval, and the calculation formula is: , , where and respectively represent the probabilities that the current change rates before and after optimization fall within the th interval; Calculate the information entropy before and after optimization respectively. The calculation formula is as follows: , , where and are the information entropy before and after optimization respectively; Calculate the ratio of the information entropy before and after optimization , and the calculation formula is as follows: ; Calculate the optimized order degree improvement coefficient, and the calculation formula is: , where is the optimized order degree improvement coefficient; Calculate the optimization effect coefficient according to the vibration optimization effective coefficient and the optimization order improvement coefficient.

5. The control method of a vertical three-in-one production line for PCB production according to claim 4, characterized in that, The steps of calculating the optimization effect coefficient according to the vibration optimization effective coefficient and the optimization order improvement coefficient are as follows: ； In the formula, is the optimization effect coefficient, and are respectively the effective coefficient of vibration optimization and the coefficient of improvement of optimization orderliness, are respectively and preset proportional values, and are both greater than 0.

6. The control method of a vertical three-in-one production line for PCB production according to claim 1, characterized in that, Evaluating whether the process parameter optimization is effective includes: Compare the optimization effect coefficient with the preset optimization effect coefficient threshold. If the optimization effect coefficient is not less than the preset optimization effect coefficient threshold, it means that the process parameter optimization to reduce the abnormal vibration of the production line is effective; If the optimization effect coefficient is less than the preset optimization effect coefficient threshold, it means that the optimization of process parameters to reduce the abnormal vibration of the production line is ineffective. At this time, the production line is controlled to stop running and an alarm signal is issued.

7. The control method of a vertical three-in-one production line for PCB production according to claim 6, characterized in that, The steps to determine whether to continue optimizing process parameters based on the vibration signal prediction result and the vibration signal of the production line to control the vibration of the production line are as follows: Calculate the deviation value between the predicted vibration signal after optimizing the process parameters and the real-time vibration signal. If the deviation value is still not less than the preset maximum allowable deviation value, continue to adjust and optimize the production process parameters until the deviation value is less than the preset maximum allowable deviation value, and the optimization effect coefficient after re-optimizing the process parameters is not less than the preset optimization effect coefficient threshold. Then, control the production line to produce PCB boards according to the corresponding production process parameters.

8. A control system for a vertical three-in-one production line for PCB manufacturing, which is used to implement the control method for a vertical three-in-one production line for PCB manufacturing according to any one of the above claims 1-7, characterized in that, The system includes: Data acquisition module: Deploy vibration sensors on the PCB vertical three-in-one production line to obtain the vibration signal on the production line in real time. During the process of producing PCB on the PCB vertical three-in-one production line, the PCB board is clamped and placed in a vertically suspended state, and slag removal, hole punching, and electroplating are continuously completed on the production line. Preliminary optimization module: Use the historical vibration data of the PCB vertical three-in-one production line during PCB production to establish a prediction model of the vibration signal through a long short-term memory network, predict the future vibration signal of the current production line during PCB production, and adjust and optimize the production process parameters according to the vibration signal prediction result and the real-time vibration signal to control the production line to produce PCB boards. Judgment and evaluation module: Obtain the vibration signal of the production line and the working current of the production line within a period of time after the adjustment of the process parameters is completed, and calculate the optimization effect coefficient to evaluate whether the optimization of the process parameters is effective. Optimization control module: If the optimization of the process parameters is effective, obtain the vibration signal of the production line again, and determine whether to continue optimizing the process parameters according to the vibration signal prediction result and the vibration signal of the production line to control the vibration of the production line.

Images