MCU-based coating equipment monitoring and control system and method

Abstract

The invention discloses a coating equipment monitoring and control system and method based on an MCU, and belongs to the technical field of vacuum coating equipment. The system comprises a real-time monitoring module, an intelligent early warning module and a control module; the real-time monitoring module is used for collecting working state information of coating equipment in real time and transmitting the collected information to the intelligent early warning module for analysis and judgment; the intelligent early warning module receives the data transmitted by the real-time monitoring module, predicts and evaluates the running state of the coating equipment through data preprocessing, feature extraction and SVM model training, and sends out an early warning signal in time when the equipment is abnormal; and the control module receives an early warning signal of the intelligent early warning module, generates a corresponding control signal through the MCU, generates a direct-current high-voltage signal after inversion amplification, boosting and voltage doubling rectification, and automatically adjusts working parameters of the coating equipment, so that the coating equipment recovers to a normal working state.

Description

Technical Field

[0001] The present invention relates to the technical field of vacuum coating equipment, and particularly to a monitoring and control system and method for coating equipment based on MCU. Background Technique

[0002] In the field of modern industrial manufacturing, coating equipment, as a key link in precision processing, the stability of its performance and the coating quality directly determine the quality of the final product. Currently, vacuum coating technologies, especially magnetron sputtering and evaporation coating technologies, have become the mainstream coating methods. Magnetron sputtering coating equipment is widely used in the surface coating treatment of various products due to its high coating efficiency and good film layer strength; evaporation coating equipment decomposes coating materials with high energy through an electron gun to complete the coating process. Although its application range is relatively local, it still occupies an important position in specific fields.

[0003] However, as the usage time of coating equipment increases, a series of problems gradually emerge: for magnetron sputtering coating equipment and evaporation coating equipment, it becomes increasingly difficult to maintain the vacuum condition, resulting in a decline in coating quality. The MASK uniformity of evaporation coating equipment gradually deteriorates, the surface roughness increases, and even arcing occurs, seriously affecting the coating effect and the stable operation of the equipment; the connection parts of electrical components are prone to burnout due to long-term operation, and the resistance value rises, which not only increases the maintenance cost but also may cause safety accidents; in addition, the distortion of each component requires a large amount of time and manpower for calibration, further increasing the operating cost. Summary of the Invention

[0004] The purpose of the present invention is to provide a monitoring and control system and method for coating equipment based on MCU to solve the problems raised in the above background technique.

[0005] To solve the above technical problems, the present invention provides the following technical solution: A monitoring and control system for coating equipment based on MCU, the system includes a real-time monitoring module, an intelligent warning module, and a control module;

[0006] The real-time monitoring module is used to collect the working state information of the coating equipment in real time and transmit the collected information to the intelligent warning module for analysis and judgment;

[0007] The intelligent warning module receives the data transmitted by the real-time monitoring module, predicts and evaluates the operating state of the coating equipment through data preprocessing, feature extraction, and SVM model training. When the equipment shows an abnormality, it timely issues a warning signal and transmits it to the control module;

[0008] The control module receives the warning signal from the intelligent warning module. When it detects that the coating equipment is in an abnormal state, it generates corresponding control signals through the MCU. After inversion amplification, boosting, and voltage-doubling rectification, a DC high-voltage signal is generated to automatically adjust the working parameters of the coating equipment, so that the coating equipment returns to the normal working state.

[0009] Furthermore, the real-time monitoring module includes a data acquisition unit and a data processing unit;

[0010] The data acquisition unit installs a timer, a temperature sensor, a vacuum gauge, a power sensor, and a film thickness tester on the coating equipment; the timer starts timing from the first startup of the equipment and records the cumulative operation time; the temperature sensor is used to measure the temperature T of each component of the equipment in real time; the vacuum gauge is used to measure the vacuum degree V; the power sensor is used to measure the real-time power P of the equipment; the film thickness tester measures the film thickness M by optical interference method; the timer is synchronized with the temperature sensor, the vacuum gauge, the power sensor, and the film thickness tester, and records the time stamps at the time of each data acquisition.

[0011] The data processing unit amplifies and filters the temperature analog signal collected by the temperature sensor and the vacuum analog signal collected by the vacuum gauge, and then converts them into digital signals through an analog-to-digital converter and stores them in the time series database; the relevant data collected by the timer, the power sensor, and the film thickness tester are directly stored in the time series database in the form of digital signals.

[0012] In the above technical solution, by continuously capturing the working state information of the coating equipment, it provides real-time and detailed data support for the entire system, provides necessary input for the intelligent warning module, ensures the accuracy of the warning model and the timeliness of the warning signal, thereby improving the reliability and efficiency of the coating operation as a whole.

[0013] Furthermore, the intelligent warning module includes a data annotation unit, a model construction unit, and a warning trigger unit;

[0014] The data annotation unit first takes the cumulative operation of the equipment for x hours as a slice period, divides the collected data into data subsets for y time periods, and performs preprocessing operations of data cleaning and standardization for each subset respectively; sets the normal threshold range for each parameter, marks the data within the normal range as normal data, represented by 0; marks the data outside the range as abnormal data, represented by 1; combines the data subsets for the y time periods to determine the time period where the abnormal data is located.

[0015] Then, for each data point, calculate the ratio of the equipment operation time to each parameter: calculate the change in vacuum degree ΔV = V i+1 -V i, where i represents the i-th data point in the time series, and V i represents the vacuum degree of the i-th data point, and V i+1 represents the vacuum degree of the (i + 1)-th data point;

[0016] Let t be the time interval between adjacent time points, and calculate the ratio R of the running time to the change in vacuum degree vt = t / ΔV, where ΔV≠0; when ΔV = 0, R vt is assigned -1;

[0017] Similarly, calculate the ratio R of time to the change in temperature Tt = t / ΔT, where ΔT represents the change in temperature, ΔT≠0, and when ΔT = 0, R Tt is assigned -1;

[0018] Calculate the ratio R of time to the change in power Pt = t / ΔP, where ΔP represents the change in power, ΔP≠0; when ΔP = 0, R Pt is assigned -1;

[0019] Calculate the ratio R of time to the change in film thickness Mt = t / ΔM, where ΔM represents the change in film thickness, ΔM≠0; when ΔM = 0, R Mt is assigned -1;

[0020] Taking the past n hours as the time window, for each data point, collect all the temperature data points {T i , T i-1 ,..., T i-k} within the previous n hours, where T i represents the temperature value of the i-th data point, and T i-k represents the temperature value of the data point with a time difference of k hours from the i-th data point; within n hours, according to the standard deviation formula σ Tn = [1 / mΣ m K＝1 (T i-k - μT n ) 2 1 / 2 , where σ Tn represents the standard deviation of the temperature within the past n hours, m is the number of data points within the time window, and μT n is the mean value of the temperature within the time window; similarly, calculate the standard deviation σ Vn of the vacuum degree, the standard deviation σ Pn of the power, and the standard deviation σ Mn of the film thickness within the past n hours; integrate the constructed time-related features with the original parameters to form a new feature vector X = [T, V, P, M, R vt ​,R Tt ,R Pt ,R Mt ,σ Tn ,σ Vn ,σ Pn ,σ Mn ; At the same time, the corresponding normal or abnormal data status of the marked data is added to the feature vector;

[0021] The model construction unit takes the feature vector X integrated by the data annotation unit as the input feature, divides the data set into a training set and a test set according to a ratio of 7:3; uses the data in the training set to preliminarily train the SVM model, tunes the parameters through the penalty factor and kernel function parameters, defines a parameter grid including different values of the penalty factor and kernel function, uses the method of five-fold cross-validation to evaluate each parameter combination on the training set, and selects the parameter combination with the highest evaluation index as the final model parameter; uses the model to predict the data in the test set, and evaluates the model status by calculating the accuracy, precision, recall rate, and F1 value until all evaluation indicators reach the preset standards;

[0022] The early warning trigger unit inputs the real-time monitored data into the trained SVM model. The model judges the running status of the current device and outputs a normal or abnormal result. If an abnormal result is output, it is converted into an early warning signal and transmitted to the control module.

[0023] In the above technical solution, based on the data provided by the real-time monitoring module, an SVM model is constructed to deeply analyze and predict the running status of the coating equipment. It can not only discover potential fault risks in advance, so as to send out early warning signals before the occurrence of faults, reducing the unplanned downtime, but also improve the accuracy and reliability of early warning through continuous model optimization and learning;

[0024] Further, the control module includes a host power supply unit, a 24V power supply unit, a PLC control unit, an MCU unit, an inverter amplification circuit unit, a voltage multiplier rectification circuit unit, a discharge signal unit, a vacuum control display instrument, a step-up transformer, a timer, and a discharge needle;

[0025] The host power supply unit is used to provide power for the PLC control unit and the 24V power supply unit;

[0026] The 24V power supply unit is respectively connected to the MCU unit and the inverter amplification circuit unit to supply power to them;

[0027] The PLC control unit is used to preset the vacuum parameters at startup, obtain the vacuum degree information in the coating equipment through a vacuum gauge, and feedback the information to the MCU unit through the vacuum control display instrument;

[0028] The vacuum control and display instrument is used to display the vacuum degree information transmitted by the PLC control unit, and after information processing, it feeds back to the MCU unit;

[0029] The MCU unit selects a single-chip microcomputer. When the original signal output is a high voltage, the MCU unit starts the timer. When the original signal output is a negative voltage, the MCU unit turns off the timer and the signal output; the MCU unit changes the output by setting the interval time through the timer, so as to generate the double-pulse square wave signal;

[0030] The inverter amplification circuit unit receives the double-pulse square wave signal from the MCU unit, amplifies it, and converts the amplified double-pulse square wave signal into an AC high-voltage signal;

[0031] The voltage multiplier rectification circuit unit converts the AC high-voltage signal into a DC high-voltage signal through rectifier diodes and capacitors, and amplifies the voltage multiple;

[0032] The discharge signal unit receives the DC high-voltage signal output from the voltage multiplier rectification circuit unit, and forms a DC high voltage across the discharge needles through the received DC high-voltage signal.

[0033] Further, after the MCU unit in the control module receives the warning signal from the intelligent warning module, it starts the host power supply unit to supply power to the entire control module. The 24V power supply unit delivers electrical energy to the MCU unit and the inverter amplification circuit unit, enabling them to enter the standby working state; the PLC control unit sets the original starting vacuum parameters according to the preset program, and the vacuum gauge starts to work, senses the vacuum degree inside the film coating equipment through filament heating, and transmits the measurement information to the vacuum control and display instrument. After processing, it feeds back to the MCU unit, and the MCU unit obtains the initial vacuum degree data;

[0034] The interval time is set through the timer to change the output of the MCU unit, thereby generating a double-pulse square wave signal. The signal enters the inverter amplification circuit unit, which amplifies and converts the double-pulse square wave signal into an AC high-voltage signal; the AC high-voltage signal is further boosted by a step-up transformer and then enters the voltage multiplier rectification circuit unit; the voltage multiplier rectification circuit unit uses rectifier diodes and capacitors to convert the AC high-voltage signal into a DC high-voltage signal; the DC high-voltage signal is transmitted to the discharge signal unit, forming a DC high voltage across the discharge needles, generating a large number of positive ions, triggering the bombardment effect, and prompting the film coating equipment to enter the normal film coating working state.

[0035] In the above technical solution, according to the warning signal issued by the intelligent warning module, the working parameters of the coating equipment can be adjusted quickly and accurately to restore the normal working state of the equipment. This automatic control ability not only reduces manual intervention and improves the production automation level, but also can respond quickly after receiving the warning signal to effectively avoid the occurrence and expansion of faults. In addition, by precisely controlling the operation state of the equipment, the service life of the equipment can be extended, and equipment damage caused by excessive wear or improper operation can be reduced, thus improving the efficiency and sustainability of the coating operation as a whole.

[0036] A monitoring and control method for a coating equipment based on MCU, the method comprising:

[0037] Step S100: Real-time collect the working state information of the coating equipment, and store and transmit the collected information.

[0038] Step S200: Preprocess, extract features and train the SVM model for the collected data, predict and evaluate the operation state of the coating equipment, and issue a warning signal in a timely manner when the equipment shows abnormalities.

[0039] Step S300: Receive the warning signal, generate a corresponding control signal through the MCU, and after inversion amplification, boosting and voltage doubling rectification, generate a DC high voltage signal to automatically adjust the working parameters of the coating equipment to restore the coating equipment to a normal working state.

[0040] The step S100 includes:

[0041] Step S101: Install a timer, a temperature sensor, a vacuum gauge, a power sensor and a film thickness tester on the coating equipment to real-time collect the cumulative running time of the equipment, the temperature of each component, the vacuum degree, the real-time power and the film thickness, and synchronously record the time stamps of each data.

[0042] Step S102: Amplify and filter the temperature and vacuum analog signals, and then convert them into digital signals through analog-to-digital conversion; store the digital signals collected by the timer, the power sensor and the film thickness tester together in the time series database.

[0043] The step S200 includes:

[0044] Step S201: Take the cumulative running x hours of the equipment as a cycle, divide the collected data into y time period subsets, perform data cleaning and standardization, set the normal threshold range of the parameters, mark the normal and abnormal data, and determine the abnormal data time period.

[0045] Step S202: Calculate the ratio of the equipment running time to the change amount of each parameter, calculate the standard deviation of each parameter in the past n hours, integrate and construct a new feature vector, and add a data status mark.

[0046] Step S203: Divide the integrated feature vectors into a training set and a test set at a ratio of 7:3, initially train the SVM model, optimize the penalty factor and kernel function parameters, and select the best parameter combination using five-fold cross-validation; evaluate the model using the test set data, calculate the accuracy, precision, recall, and F1 value until the preset standard is reached; input the real-time monitoring data into the model to judge the device status, and issue a warning signal if it is abnormal.

[0047] The step S300 includes: The MCU receives the warning signal, starts the control process, powers on the entire control-related components, and at the same time, sets the vacuum parameters for the original start of the coating equipment, measures the vacuum degree inside the coating equipment, and feeds back the measurement information to the MCU so that the MCU obtains the initial vacuum degree data; the MCU sets the output to change at intervals through a timer to generate a double-pulse square wave signal, the signal is first amplified and converted into an AC high-voltage signal, then further boosted, and then rectified and converted into a DC high-voltage signal; the DC high-voltage signal is sent to the discharge device of the coating equipment to form a DC high-voltage electricity at both ends of the discharge needle, generating a large number of positive ions, triggering the bombardment effect, and prompting the coating equipment to return to the normal coating working state.

[0048] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0049] Through the collaborative work of the real-time monitoring module and the intelligent warning module, the present invention realizes the comprehensive and real-time monitoring and warning of the operation status of the coating equipment. Once the equipment has an abnormality, the system can quickly issue a warning signal, and automatically adjust the working parameters through the control module to restore the equipment to the normal working state, thereby effectively improving the production efficiency and the fault response speed, and reducing the risk of production interruption; at the same time, the present invention provides a stable DC high-voltage start signal for the coating equipment by increasing the inverter amplification circuit and the voltage multiplier rectification circuit, effectively shortening the waiting time before coating, solving problems such as difficult start-up, improving the coating efficiency of the coating equipment, and extending the service life of the equipment. Description of the Drawings

[0050] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0051] Figure 1 is a system module diagram of a coating equipment monitoring and control system based on MCU;

[0052] Figure 2 is a control module diagram of a coating equipment monitoring and control system based on MCU. Detailed Embodiments

[0053] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with 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. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0054] Please refer to Figure 1 - Figure 2 , the present invention provides a technical solution: a monitoring and control system for a coating device based on an MCU. The system includes a real-time monitoring module, an intelligent warning module, and a control module;

[0055] The real-time monitoring module is used to collect the working state information of the coating device in real time and transmit the collected information to the intelligent warning module for analysis and judgment;

[0056] The intelligent warning module receives the data transmitted by the real-time monitoring module, predicts and evaluates the running state of the coating device through data preprocessing, feature extraction, and SVM model training. When the device is abnormal, it issues a warning signal in time and transmits it to the control module;

[0057] The control module receives the warning signal from the intelligent warning module. When it detects that the coating device is in an abnormal state, it generates a corresponding control signal through the MCU, and after inversion amplification, boosting, and voltage doubling rectification, it generates a DC high-voltage signal to automatically adjust the working parameters of the coating device to make the coating device return to the normal working state.

[0058] Further, the real-time monitoring module includes a data acquisition unit and a data processing unit;

[0059] The data acquisition unit installs a timer, a temperature sensor, a vacuum gauge, a power sensor, and a film thickness tester on the coating device; the timer starts timing from the first startup of the device and records the cumulative running time; the temperature sensor is used to measure the temperature T of each component of the device in real time; the vacuum gauge is used to measure the vacuum degree V; the power sensor is used to measure the real-time power P of the device; the film thickness tester measures the film thickness M by optical interference method; the timer is synchronized with the temperature sensor, the vacuum gauge, the power sensor, and the film thickness tester to record the time stamps at the time of each data acquisition;

[0060] The data processing unit amplifies and filters the temperature analog signal collected by the temperature sensor and the vacuum analog signal collected by the vacuum gauge, and then converts them into digital signals through an analog-to-digital converter and stores them in the time series database; the relevant data collected by the timer, the power sensor, and the film thickness tester are directly stored in the time series database in the form of digital signals.

[0061] In the above technical solution, by continuously capturing the working state information of the coating equipment, real-time and detailed data support is provided for the entire system, necessary input is provided for the intelligent warning module, the accuracy of the warning model and the timeliness of the warning signal are ensured, thereby improving the reliability and efficiency of the coating operation as a whole.

[0062] Further, the intelligent warning module includes a data annotation unit, a model construction unit, and a warning trigger unit;

[0063] The data annotation unit first takes the cumulative operation of the equipment for x hours as a slice period, divides the collected data into data subsets for y time periods, and performs preprocessing operations of data cleaning and standardization for each subset respectively; sets a normal threshold range for each parameter, marks the data within the normal range as normal data, represented by 0; marks the data outside the range as abnormal data, represented by 1; combines the data subsets for the y time periods to determine the time period where the abnormal data is located;

[0064] Next, for each data point, calculate the ratio of the equipment operation time to each parameter: calculate the change in vacuum degree ΔV = V i+1 -V i , where i represents the i-th data point in the time series, V i represents the vacuum degree of the i-th data point, and V i+1 represents the vacuum degree of the (i + 1)-th data point;

[0065] Let t be the time interval between adjacent time points, and calculate the ratio R of the operation time to the change in vacuum degree vt = t / ΔV, where ΔV ≠ 0; when ΔV = 0, R vt is assigned -1;

[0066] Similarly, calculate the ratio R of time to the change in temperature Tt = t / ΔT, where ΔT represents the change in temperature, ΔT ≠ 0, and when ΔT = 0, R Tt is assigned -1;

[0067] Calculate the ratio R of time to the change in power Pt = t / ΔP, where ΔP represents the change in power, ΔP ≠ 0; when ΔP = 0, R Pt is assigned -1;

[0068] Calculate the ratio R of time to the change in film thickness Mt = t / ΔM, where ΔM represents the change in film thickness, ΔM ≠ 0; when ΔM = 0, R Mt is assigned -1;

[0069] Taking the past n hours as the time window, for each data point, collect all the temperature data points {T i , T i-1 ,..., T i-k} within the previous n hours, where T i represents the temperature value of the i-th data point, and T i-k represents the temperature value of the data point with a time difference of k hours from the i-th data point; within n hours, according to the standard deviation formula σ Tn = [1 / m Σ m K＝1 (T i-k - μT n ) 2 1 / 2 , where σ Tn represents the standard deviation of the temperature within the past n hours, m is the number of data points within the time window, and μT n is the mean value of the temperature within the time window; similarly, calculate the standard deviation σ Vn of the vacuum degree, the standard deviation σ Pn of the power, and the standard deviation σ Mn of the film thickness within the past n hours; integrate the constructed time-related features with the original parameters to form a new feature vector X = [T, V, P, M, R vt , R Tt , R Pt , R Mt , σ Tn , σ Vn , σ Pn , σ Mn ; at the same time, add the corresponding marked normal or abnormal data status to the feature vector;

[0070] The model construction unit takes the feature vector X integrated by the data annotation unit as the input feature, divides the data set into a training set and a test set according to a ratio of 7:3; uses the data in the training set to preliminarily train the SVM model, tunes the parameters through the penalty factor and the kernel function parameters, defines a parameter grid containing different values of the penalty factor and the kernel function, uses the five-fold cross-validation method to evaluate each parameter combination on the training set, and selects the parameter combination with the highest evaluation index as the final model parameter; uses the model to predict the data in the test set, and evaluates the model status by calculating the accuracy rate, precision rate, recall rate, and F1 value until all evaluation indicators reach the preset standards;

[0071] The warning trigger unit inputs the real-time monitored data into the trained SVM model. The model judges the running state of the current device and outputs a normal or abnormal result. If an abnormal result is output, it is converted into a warning signal and transmitted to the control module.

[0072] ​In the above technical solution, based on the data provided by the real-time monitoring module, the operation status of the coating equipment is deeply analyzed and predicted through the SVM model. It can not only detect potential failure risks in advance, thus sending out warning signals before the occurrence of failures, reducing the unplanned downtime, but also improve the accuracy and reliability of the warning through continuous model optimization and learning;

[0073] Further, the control module includes a host power supply unit, a 24V power supply unit, a PLC control unit, an MCU unit, an inverter amplification circuit unit, a voltage multiplier rectification circuit unit, a discharge signal unit, a vacuum control display instrument, a step-up transformer, a timer, and a discharge needle;

[0074] The host power supply unit is used to provide power for the PLC control unit and the 24V power supply unit;

[0075] The 24V power supply unit is respectively connected to the MCU unit and the inverter amplification circuit unit to supply power to them;

[0076] The PLC control unit is used to preset the vacuum parameters at startup, obtain the vacuum degree information in the coating equipment through a vacuum gauge, and feedback the information to the MCU unit through the vacuum control display instrument;

[0077] The vacuum control display instrument is used to display the vacuum degree information transmitted by the PLC control unit, and after information processing, it feeds back to the MCU unit;

[0078] The MCU unit selects a single-chip microcomputer. When the original signal output is a high voltage, the timer is started. When the original signal output is a negative voltage, the timer and the signal output are turned off; the MCU unit changes the output by setting the interval time through the timer, thereby generating the double-pulse square wave signal;

[0079] The inverter amplification circuit unit receives the double-pulse square wave signal from the MCU unit, amplifies it, and converts the amplified double-pulse square wave signal into an AC high-voltage signal;

[0080] The voltage multiplier rectification circuit unit converts the AC high-voltage signal into a DC high-voltage signal through rectifier diodes and capacitors, and amplifies the voltage multiple;

[0081] The discharge signal unit receives the DC high-voltage signal output from the voltage multiplier rectification circuit unit, and forms a DC high voltage across the discharge needle through the received DC high-voltage signal.

[0082] Further, after the MCU unit in the control module receives the warning signal from the intelligent warning module unit, it starts the host power supply unit to supply power to the entire control module. The 24V power supply unit delivers electrical energy to the MCU unit and the inverter amplification circuit unit, enabling them to enter the standby working state; the PLC control unit sets the original starting vacuum parameters according to the preset program, the vacuum gauge starts to work, senses the vacuum degree inside the filament heating induction coating equipment, and transmits the measurement information to the vacuum control display instrument, which is then processed and fed back to the MCU unit. The MCU unit obtains the initial vacuum degree data;

[0083] The output of the MCU unit is changed by setting the interval time through the timer, thereby generating a double-pulse square wave signal. The signal enters the inverter amplification circuit unit, where the double-pulse square wave signal is amplified and converted into an AC high-voltage signal; the AC high-voltage signal is further boosted by the step-up transformer and then enters the voltage multiplier rectification circuit unit; the voltage multiplier rectification circuit unit uses rectifier diodes and capacitors to convert the AC high-voltage signal into a DC high-voltage signal; the DC high-voltage signal is transmitted to the discharge signal unit, forming a DC high voltage at both ends of the discharge needle, generating a large number of positive ions, triggering the bombardment effect, and prompting the coating equipment to enter the normal coating working state.

[0084] In the above technical solution, according to the warning signal issued by the intelligent warning module, the working parameters of the coating equipment can be adjusted quickly and accurately, and the normal working state of the equipment can be restored. This automatic control ability not only reduces manual intervention and improves the level of production automation, but also can quickly respond after receiving the warning signal, effectively avoiding the occurrence and expansion of faults; in addition, by precisely controlling the operation state of the equipment, the service life of the equipment can be extended, and equipment damage caused by excessive wear or improper operation can be reduced, thereby improving the efficiency and sustainability of the coating operation as a whole.

[0085] A method for monitoring and controlling a coating equipment based on MCU, the method includes:

[0086] Step S100: Collect the working state information of the coating equipment in real time, and store and transmit the collected information;

[0087] Step S200: Preprocess, extract features and train the SVM model for the collected data, predict and evaluate the operation state of the coating equipment, and issue a warning signal in a timely manner when the equipment shows abnormalities;

[0088] Step S300: Receive the warning signal, generate corresponding control signals through the MCU, and after inverter amplification, boosting and voltage multiplier rectification, generate a DC high-voltage signal to automatically adjust the working parameters of the coating equipment to restore the coating equipment to the normal working state.

[0089] The step S100 includes:

[0090] Step S101: Install a timer, a temperature sensor, a vacuum gauge, a power sensor, and a film thickness tester on the coating equipment to collect the cumulative operation time of the equipment, the temperatures of various components, the vacuum degree, the real-time power, and the film thickness in real time, and synchronously record the timestamps of each data.

[0091] Step S102: Amplify and filter the temperature and vacuum analog signals, and then convert them into digital signals through analog-to-digital conversion; store the digital signals collected by the timer, the power sensor, and the film thickness tester together in a time series database.

[0092] The said step S200 includes:

[0093] Step S201: Take the cumulative operation of the equipment for x hours as a cycle, divide the collected data into y subsets of time periods, perform data cleaning and standardization, set the normal threshold range of parameters, mark normal and abnormal data, and determine the time periods of abnormal data.

[0094] Step S202: Calculate the ratio of the equipment operation time to the change amount of each parameter, calculate the standard deviation of each parameter within the past n hours, integrate and construct a new feature vector, and add a data status mark.

[0095] Step S203: Divide the integrated feature vector into a training set and a test set according to 7:3, preliminarily train the SVM model, optimize the penalty factor and kernel function parameters, select the best parameter combination by five-fold cross-validation; evaluate the model with the test set data, calculate the accuracy rate, precision rate, recall rate, and F1 value until reaching the preset standard; input the real-time monitoring data into the model to judge the equipment status, and send out a warning signal if it is abnormal.

[0096] The said step S300 includes: The MCU receives the warning signal, starts the control process, turns on the power supply to supply power to all the components related to the control. At the same time, set the vacuum parameters for the original start of the coating equipment, measure the vacuum degree inside the coating equipment, and feedback the measurement information to the MCU so that the MCU can obtain the initial vacuum degree data; the MCU sets the output at intervals through a timer to generate a double-pulse square wave signal, the signal is first amplified and converted into an AC high-voltage signal, then further boosted, and then rectified and converted into a DC high-voltage signal; the DC high-voltage signal is sent to the discharge device of the coating equipment to form a DC high voltage at both ends of the discharge needle, generating a large number of positive ions, triggering a bombardment effect, and prompting the coating equipment to return to the normal coating working state.

[0097] Embodiment of the present invention: Take a certain industrial coating equipment as the application object, which is used to deposit a functional film on the surface of electronic components;

[0098] Install on the coating equipment:

[0099] Timer, which starts timing from the first startup of the device and records the cumulative running time;

[0100] Temperature sensor, a high-precision thermocouple temperature sensor is selected to measure the temperature of the key components of the device in real time, including components such as the evaporation source and the substrate heating table;

[0101] Vacuum gauge, a hot cathode ionization vacuum gauge is used to measure the vacuum degree inside the coating equipment;

[0102] Power sensor, a Hall effect power sensor is selected to measure the real-time power of the device;

[0103] Film thickness tester, an optical interference film thickness tester is used to measure the film thickness by optical interference method; all sensors are synchronized with the timer to record the timestamps at the time of each data acquisition;

[0104] The temperature analog signal collected by the temperature sensor and the vacuum analog signal collected by the vacuum gauge are first amplified by an operational amplifier with an amplification factor of 10 times, and then filtered by a low-pass filter to remove high-frequency noise interference; the analog signal after amplification and filtering is converted into a digital signal by a 16-bit analog-to-digital converter and stored in the time series database; the relevant data collected by the timer, power sensor and film thickness tester are digital signals themselves and are directly stored in the time series database in digital signal form;

[0105] Taking 10 hours of cumulative operation of the device as a slice period, assuming the device has run for 50 hours, the data will be divided into 5 data subsets of time periods; for each subset, preprocessing operations of data cleaning and standardization are performed respectively, obvious outliers are removed, and the data is standardized to a range with a mean of 0 and a standard deviation of 1 using the Z-score standardization method;

[0106] Set the normal threshold range for each parameter:

[0107] Temperature T: The normal range is 200°C - 300°C;

[0108] Vacuum degree V: The normal range is 10 -4 Pa - 10 -6 Pa;

[0109] Power P: The normal range is 500W - 800W;

[0110] Film thickness M: According to the coating process requirements, the normal range is 50nm - 100nm;

[0111] Mark the data within the normal range as normal data, represented by 0; the data outside the range is marked as abnormal data, represented by 1, and combined with the divided time period data subsets, determine the time period where the abnormal data is located;

[0112] Calculate the ratio of the operating time of the computing device to each parameter: Calculate the change in vacuum degree ΔV = V i+1 -V i between adjacent time points. The vacuum degree V i at the i-th data point is 10 -3 Pa, and the vacuum degree V i+1 at the (i + 1)-th data point is 1.01×10 -3 Pa. ΔV = 0.01×10 -3 Pa.

[0113] Assume that the time interval t between adjacent time points is 1 s, and calculate the ratio R of the operating time to the change in vacuum degree vt = t / ΔV = 10 5 s / Pa;

[0114] Similarly, calculate the ratio R of time to the change in temperature Tt = t / ΔT; Calculate the ratio R of time to the change in power Pt = t / ΔP;

[0115] Taking the past 5 hours as a time window, for each data point, collect all temperature data points {T i-k} within the previous 5 hours; within 5 hours, calculate σ according to the standard deviation formula Tn = [1 / mΣ m K＝1 (T i-k - μT n ) 2 1 / 2 , where m is the number of data points within the time window, m = 18000; Similarly, calculate the standard deviation σ of the vacuum degree, the standard deviation σ of the power, and the standard deviation σ of the film thickness within the past n hours Vn ; Integrate the constructed time-related features with the original parameters to form a new feature vector X = [T, V, P, M, R Pn , R Mn , R vt , R Tt , R Pt , R Mt , σ Tn , σ Vn , σ Pn , σ Mn ; At the same time, add the corresponding marked normal or abnormal data status to the feature vector;

[0116] ​Take the integrated feature vector X as the input feature. There are a total of 10,000 samples. Divide the dataset into a training set and a test set according to a ratio of 7:3. Use the data in the training set to initially train the SVM model, and tune the parameters through the penalty factor C and the kernel function parameter γ. Define a parameter grid, where the penalty factor C = [0.1, 1, 10] and the kernel function γ = [0.01, 0.1, 1]. Use the five-fold cross-validation method to evaluate each parameter combination on the training set. Divide the training set into 5 subsets, take 4 subsets for training each time, and 1 subset for validation. Select the parameter combination with the highest evaluation index as the final model parameter. Use the model to predict the data in the test set, and evaluate the model status by calculating the accuracy, precision, recall, and F1 value until all evaluation indicators reach the preset standards. Input the real-time monitored data into the trained SVM model. The model judges the running status of the current device and outputs a normal or abnormal result. If an abnormal result is output, it is converted into a warning signal.

[0117] The control module includes a host power supply unit, a 24V power supply unit, a PLC control unit, an MCU unit, an inverter amplification circuit unit, a voltage multiplier rectification circuit unit, a vacuum control display instrument, a discharge signal unit, a step-up transformer, a timer, and a discharge needle.

[0118] The MCU unit selects a single-chip microcomputer, and the inverter amplification circuit unit is a 24V to 1500V DC / AC inverter unit.

[0119] When the original signal output of the MCU unit is a high voltage, start the 16μs timer. When the original signal output is a negative voltage, turn off the 16μs timer and the signal output. Change the output every 16 microseconds through the timer to generate the double-pulse square wave signal.

[0120] The MCU unit generates a double-pulse square wave signal with a frequency of 30kHz and a duty cycle of 50%. After the signal passes through the inverter amplification circuit unit, it is boosted to a 1500V AC high-voltage signal by the step-up transformer, and then converted into a 3000V DC high-voltage signal by the voltage multiplier rectification circuit unit. The DC high-voltage signal is connected to the discharge signal unit in the coating equipment to provide a DC high-voltage start signal for the coating equipment.

[0121] The voltage amplification multiple of the voltage multiplier rectification circuit is 2. Through rectifier diodes and capacitors, the 1500V AC high-voltage signal is converted into a 3000V DC high-voltage signal.

[0122] After the MCU unit in the control module receives the warning signal from the intelligent warning module, it starts the host power supply unit to supply power to the entire control module. The 24V power supply unit delivers electrical energy to the MCU unit and the inverter amplification circuit unit, enabling them to enter the standby working state; the PLC control unit sets the original starting vacuum parameters according to the preset program, and the vacuum gauge starts to work, senses the vacuum degree inside the filament heating induction coating equipment, and transmits the measurement information to the vacuum control display instrument. After being processed, it is fed back to the MCU unit, and the MCU unit obtains the initial vacuum degree data;

[0123] The output of the MCU unit is changed every 16 microseconds through a 16μs timer, thereby generating a double-pulse square wave signal with a frequency of 30kHz and a duty cycle of 50%. The signal enters the inverter amplification circuit unit, which amplifies and converts the double-pulse square wave signal into an AC high-voltage signal of 1500V; the AC high-voltage signal enters the voltage multiplier rectifier circuit unit after being further boosted by the step-up transformer; the voltage multiplier rectifier circuit unit uses rectifier diodes and capacitors to convert the 1500V AC high-voltage signal into a DC high-voltage signal of 3000V; the 3000V DC power is introduced into the coating equipment through a high-voltage wire and connected to the discharge needle, where the negative pole is grounded and the positive pole is connected to the discharge needle, with a spacing of about 10mm to prevent contact. When the coating equipment is working, a DC high voltage of 3000V is formed at both ends of the discharge needle, generating a large number of positive ions, thereby generating a corresponding bombardment effect, enabling the coating machine to quickly reach the normal working state.

[0124] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A coating equipment monitoring and control system based on MCU, characterized in that: The system includes a real-time monitoring module, an intelligent early warning module and a control module; The real-time monitoring module is used to collect the working status information of the coating equipment in real time, and transmit the collected information to the intelligent early warning module for analysis and judgment; The intelligent early warning module receives the data transmitted by the real-time monitoring module, and predicts and evaluates the operating status of the coating equipment through data preprocessing, feature extraction and SVM model training. When the equipment is abnormal, it promptly issues an early warning signal and transmits it to the control module; The control module receives the warning signal from the intelligent warning module. When it is detected that the coating equipment is in an abnormal state, the MCU generates a corresponding control signal. After inverter amplification, boosting and voltage doubling rectification, a DC high-voltage signal is generated to automatically adjust the working parameters of the coating equipment to restore the coating equipment to a normal working state.

2. According to the MCU-based coating equipment monitoring and control system of claim 1, it is characterized in that: The real-time monitoring module includes a data acquisition unit and a data processing unit; The data acquisition unit is equipped with a timer, a temperature sensor, a vacuum gauge, a power sensor, and a film thickness tester on the coating equipment; the timer starts timing from the first start of the equipment and records the cumulative running time; the temperature sensor is used to measure the temperature T of each component of the equipment in real time; the vacuum gauge is used to measure the vacuum degree V; the power sensor is used to measure the real-time power P of the equipment; the film thickness tester measures the film thickness M by optical interferometry; the timer is synchronized with the temperature sensor, vacuum gauge, power sensor, and film thickness tester to record the timestamp when each data is collected; The data processing unit amplifies and filters the temperature analog signal collected by the temperature sensor and the vacuum analog signal collected by the vacuum gauge, and then converts them into digital signals through an analog-to-digital converter and stores them in a time series database; the relevant data collected by the timer, power sensor and film thickness tester are directly stored in the time series database in the form of digital signals.

3. According to the MCU-based coating equipment monitoring and control system of claim 1, it is characterized in that: The intelligent early warning module includes a data annotation unit, a model building unit, and an early warning triggering unit; The data annotation unit first divides the collected data into data subsets of y time periods, taking the cumulative operation of the equipment for x hours as a slice cycle, and performs data cleaning and standardization preprocessing operations on each subset; Set a normal threshold range for each parameter, mark the data within the normal range as normal data, represented by 0; mark the data outside the range as abnormal data, represented by 1; combine the data subsets divided into y time periods to determine the time period where the abnormal data is located; Then, for each data point, calculate the ratio of the equipment operation time to each parameter: Calculate the vacuum degree change between adjacent time points ΔV = V i+1 -V i , where i represents the i-th data point in the time series, V i Represents the vacuum degree of the ith data point, V i+1 Indicates the vacuum degree of the i+1th data point; Let t be the time interval between adjacent time points, and calculate the ratio of running time to vacuum degree change R vt =t / ΔV, where ΔV≠0; when ΔV=0, R vt Assign a value of -1; Similarly, calculate the ratio of time to temperature change R Tt =t / ΔT, where ΔT represents the temperature change, ΔT≠0, when ΔT=0, R Tt Assign a value of -1; Calculate the ratio of time to power change R Pt =t / ΔP, where ΔP represents the power change, ΔP≠0; when ΔP=0, R Pt Assign a value of -1; Calculate the ratio of time to film thickness change R Mt =t / ΔM, where ΔM represents the change in film thickness, ΔM≠0; when ΔM=0, R Mt Assign a value of -1; Taking the past n hours as the time window, for each data point, collect all temperature data points in the previous n hours {T i ,T i-1 ,...,T i-k }, where T i Represents the temperature value of the i-th data point, T i-k Indicates the temperature value of the data point with a time difference of k hours from the i-th data point; within n hours, according to the standard deviation formula σ Tn =[1 / mΣ m K＝1 (T i-k -μT n ) 2 ] 1 / 2 , where σ Tn represents the standard deviation of the temperature over the past n hours, m is the number of data points in the time window, μT n is the mean temperature in the time window; similarly, calculate the standard deviation of vacuum degree σ in the past n hours Vn , standard deviation of power σ Pn , standard deviation of film thickness σ Mn ; Integrate the constructed time-related features with the original parameters to form a new feature vector X = [T, V, P, M, R vt ,R Tt ,R Pt ,R Mt ,σ Tn ,σ Vn ,σ Pn ,σ Mn ]; at the same time, the marked normal or abnormal data state is added to the feature vector; The model building unit uses the feature vector X integrated by the data annotation unit as an input feature, and divides the data set into a training set and a test set in a ratio of 7:3; uses the data of the training set to perform preliminary training on the SVM model, performs parameter tuning through the penalty factor and the kernel function parameter, defines a parameter grid, including penalty factors and kernel functions with different values, uses a five-fold cross-validation method to evaluate each parameter combination on the training set, and selects the parameter combination with the highest evaluation index as the final model parameter; uses the model to predict the data of the test set, and evaluates the model status by calculating the accuracy, precision, recall rate and F1 value until all evaluation indicators reach the preset standards; The early warning trigger unit inputs the real-time monitored data into the trained SVM model. The model determines the current operating status of the equipment and outputs normal or abnormal results. If the output is an abnormal result, it is converted into an early warning signal and transmitted to the control module.

4. The MCU-based coating equipment monitoring and control system according to claim 1, characterized in that: The control module includes a host power supply unit, a 24V power supply unit, a PLC control unit, an MCU unit, an inverter amplifier circuit unit, a voltage doubler rectifier circuit unit, a discharge signal unit, a vacuum control display instrument, a step-up transformer, a timer, and a discharge needle; The host power supply unit is used to provide power to the PLC control unit and the 24V power supply unit; The 24V power supply unit is respectively connected to the MCU unit and the inverter amplifier circuit unit to supply power to them; The PLC control unit is used to preset vacuum parameters at startup, obtain vacuum degree information in the coating equipment through a vacuum gauge, and feed the information back to the MCU unit through a vacuum control display instrument; The vacuum control display instrument is used to display the vacuum degree information transmitted by the PLC control unit, and feed back the information to the MCU unit after processing; The MCU unit uses a single chip microcomputer. When the original signal output is a high voltage, the MCU unit starts the timer, and when the original signal output is a negative voltage, the timer and signal output are turned off. The MCU unit changes the output by setting the interval time through the timer, thereby generating the double pulse square wave signal. The inverter amplifier circuit unit receives the double-pulse square wave signal from the MCU unit, amplifies it, and converts the amplified double-pulse square wave signal into an AC high-voltage signal; The voltage doubler rectifier circuit unit converts the AC high voltage signal into a DC high voltage signal through a rectifier diode and a capacitor, and amplifies the voltage multiple; The discharge signal unit receives the DC high voltage signal output from the voltage doubler rectifying circuit unit, and forms a DC high voltage voltage at both ends of the discharge needle through the received DC high voltage signal.

5. The MCU-based coating equipment monitoring and control system according to claim 1, characterized in that: After receiving the warning signal from the intelligent warning module, the MCU unit in the control module starts the host power supply unit to power the entire control module, and the 24V power supply unit transmits electric energy to the MCU unit and the inverter amplifier circuit unit, so that they enter the ready-to-work state; the PLC control unit sets the original starting vacuum parameters according to the preset program, and the vacuum gauge starts to work, senses the vacuum degree in the coating equipment through filament heating, and transmits the measurement information to the vacuum control display instrument, which is fed back to the MCU unit after processing, and the MCU unit obtains the initial vacuum degree data; The output of the MCU unit is changed by setting the interval time through the timer, thereby generating a double-pulse square wave signal. The signal enters the inverter amplifier circuit unit, amplifies the double-pulse square wave signal and converts it into an AC high-voltage signal; the AC high-voltage signal is further boosted by the step-up transformer and enters the voltage-doubling rectifier circuit unit; the voltage-doubling rectifier circuit unit converts the AC high-voltage signal into a DC high-voltage signal using a rectifier diode and a capacitor; the DC high-voltage signal is transmitted to the discharge signal unit, forming a DC high-voltage electricity at both ends of the discharge needle, generating a large number of positive ions, triggering a bombardment effect, and prompting the coating equipment to enter a normal coating working state.

6. A coating equipment monitoring and control method based on MCU, characterized in that: The method comprises: Step S100: collecting working status information of the coating equipment in real time, and storing and transmitting the collected information; Step S200: preprocessing, feature extraction and SVM model training of the collected data, predicting and evaluating the operating status of the coating equipment, and issuing a warning signal in time when the equipment is abnormal; Step S300: Receive the warning signal, generate a corresponding control signal through the MCU, generate a DC high-voltage signal after inverter amplification, boost and voltage doubling rectification, automatically adjust the working parameters of the coating equipment, and restore the coating equipment to a normal working state.

7. The MCU-based coating equipment monitoring and control method according to claim 6, characterized in that: The step S100 includes: Step S101: Install a timer, a temperature sensor, a vacuum gauge, a power sensor and a film thickness tester on the coating equipment to collect the cumulative running time of the equipment, the temperature of each component, the vacuum degree, the real-time power and the film thickness in real time, and synchronously record the timestamp of each data; Step S102: amplify and filter the temperature and vacuum analog signals, and then convert them into digital signals through analog-to-digital conversion; store the digital signals collected by the timer, power sensor and film thickness tester into a time series database.

8. The method for monitoring and controlling coating equipment based on MCU according to claim 6, characterized in that: The step S200 includes: Step S201: Taking the cumulative operation time of the equipment as a period of x hours, the collected data is divided into y time period subsets, data cleaning and standardization are performed, the normal threshold range of the parameters is set, normal and abnormal data are marked, and the abnormal data time period is determined; Step S202: Calculate the ratio of the equipment running time to the change of each parameter, calculate the standard deviation of each parameter in the past n hours, integrate and construct a new feature vector, and add a data status mark; Step S203: Divide the integrated feature vector into a training set and a test set at a ratio of 7:3, perform preliminary training on the SVM model, optimize the penalty factor and kernel function parameters, and select the best parameter combination using 5-fold cross validation; evaluate the model using the test set data, calculate the accuracy, precision, recall rate, and F1 value until the preset standard is reached; input the real-time monitoring data into the model to determine the equipment status, and issue a warning signal if an abnormality occurs.

9. The method for monitoring and controlling coating equipment based on MCU according to claim 6, characterized in that: The step S300 includes: the MCU receives the warning signal, starts the control process, starts the power supply to power the entire control-related components, and at the same time, sets the vacuum parameters of the original startup of the coating equipment, measures the vacuum degree in the coating equipment, and feeds back the measurement information to the MCU, so that the MCU obtains the initial vacuum degree data; the MCU changes the output by setting the interval time through the timer to generate a double-pulse square wave signal, the signal is first amplified and converted into an AC high-voltage signal, and then further boosted, and then converted into a DC high-voltage signal through rectification; the DC high-voltage signal is transmitted to the discharge device, and DC high-voltage electricity is formed at both ends of the discharge needle, generating a large number of positive ions, triggering a bombardment effect, and prompting the coating equipment to return to a normal coating working state.

Images