A multi-source data fusion-based intelligent monitoring system for a whole process of paint production
By using multi-source data fusion technology, high-precision control and dynamic adjustment of the powder coating bonding process are achieved, solving the problems of low control accuracy and slow response of traditional bonding machines, and improving production efficiency and product quality.
Patent Information
- Application Number
- CN202610932515.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-26
- Publication Date
- 2026-08-25
AI Technical Summary
Traditional bonding machines suffer from low control precision, slow response, and poor coordination in powder coating production, resulting in unstable product quality and low production efficiency. This is mainly due to low data processing precision, lag in glass transition estimation, and disconnect between temperature control and torque control.
By employing multi-source data fusion technology, data preprocessing is performed using the Kalman filter algorithm and recursive least squares method. Combined with adaptive impedance control and a piecewise PID-fuzzy composite temperature control algorithm, the glass transition degree of powder coatings can be estimated and dynamically adjusted in real time, thereby optimizing control parameters.
It achieves high-precision control of the powder coating bonding process (±0.3℃, ±0.2 N·m), improves response speed to the millisecond level, enhances process stability, shortens single cycle by 28%, reduces unit energy consumption by 22%, and increases product qualification rate to 98%.
Smart Images
Figure CN122632784A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of coating production technology, specifically to an intelligent monitoring system for the entire coating production process based on multi-source data fusion. Background Technology
[0002] Powder coatings, especially metallic powder coatings, are widely used in home appliances, automotive wheels, and electronic products due to their excellent environmental performance and unique metallic sheen. Currently, the main industrial preparation methods for metallic powder coatings include dry mixing, melt extrusion, and bonding. Among these, bonding effectively avoids the inherent defects of dry mixing (easily causing powder shedding and separation) and melt extrusion (easily producing solvent residue and high energy consumption). The resulting powder coatings exhibit excellent color stability and a vivid, realistic metallic sheen during spraying, making it a key technology for preparing high-end metallic powder coatings.
[0003] The core of the bonding process involves precisely controlling the bonding temperature (which must be close to but slightly below the glass transition temperature of the powder coating) and applying appropriate mechanical shear force (manifested as stirring torque) in a mixture of powder coating (base powder) and metallic pigment (metal powder) to soften the surface of the base powder, thereby "bonding" or "coating" the metallic pigment onto the surface. The precision of this process directly determines the final performance of the product, such as gloss, adhesion, and the amount of pigment added.
[0004] However, traditional bonding machines face the following key technical bottlenecks in actual production, resulting in low control precision, slow response, and poor coordination, which limits further improvements in product quality and production efficiency:
[0005] Low data processing accuracy: During actual operation, the bonding machine experiences strong mechanical vibrations and complex temperature fluctuations, leading to severe multi-band composite noise interference in the key parameters (especially torque and temperature) collected by the sensors. Traditional signal processing methods, such as simple mean filtering or moving average filtering, cannot effectively separate the signal from the noise, resulting in significant errors in parameter measurements (e.g., temperature errors exceeding ±2℃). These distorted data severely affect the accuracy of subsequent judgment and control of the powder's glass transition state.
[0006] Glass transition temperature estimation lag: Current technologies mainly rely on offline detection methods (such as differential scanning calorimetry, DSC) to pre-determine the glass transition temperature range of materials and use it as a fixed reference value during production. This method cannot reflect in real time the dynamic changes in the glass transition temperature of powder coatings caused by component changes, uneven heat transfer, and mechanical effects during the bonding process. This estimation lag leads to untimely adjustments to control system parameters, easily causing local agglomeration or insufficient bonding. To mitigate this risk, manufacturers typically need to conservatively control the amount of metallic pigment added (often below 12%), which limits the performance of high-end products.
[0007] The cumbersome switching of control modes is prone to overshoot and oscillation: To achieve effective control of flexible and rigid powder mixtures, existing technologies often employ a force / position hybrid control mode. Position control is used when the powder is in a free-flowing state (non-vitrified), and force control is switched when it enters the contact extrusion state (near vitrification). However, during rapid changes in powder state, frequent mode switching procedures are complex and can easily cause sluggish system response or command conflicts, manifesting as large overshoot in blade torque, system oscillation, and batch-to-batch quality instability.
[0008] Disconnect between temperature and torque control: Temperature is the most important variable in the bonding process. Traditional temperature control methods often use a single PID algorithm with a fixed target value. This method does not fully consider the changes in material viscosity and shear resistance (i.e., glass transition state changes) reflected in the real-time blade torque and contact force, leading to a disconnect between the temperature control strategy and the actual state of the material. The consequence is a large temperature difference within the bonding tank (especially between hot and cold sections). In some areas, the powder "burns" or clumps due to excessively high temperatures, while in other areas, bonding is poor due to insufficient temperatures. The overall product qualification rate cannot be effectively guaranteed. Therefore, we propose an intelligent monitoring system for the entire coating production process based on multi-source data fusion. Summary of the Invention
[0009] The purpose of this invention is to provide an intelligent monitoring system for the entire coating production process based on multi-source data fusion, so as to solve the problems mentioned in the background art.
[0010] To achieve the above objectives, the present invention provides the following technical solution: an intelligent monitoring system for the entire coating production process based on multi-source data fusion, comprising the following steps:
[0011] S1. Real-time acquisition of multi-source data: Data is acquired in real time by sensors installed on the powder coating bonding machine, and the communication link between the sensors and the bonding machine is configured using an industrial control platform to establish variable mapping to store the acquired data;
[0012] S2. Multi-source data preprocessing: Based on the Kalman filter algorithm, a high-precision filtering model suitable for the bonding machine is constructed. The data collected in step S1 is input into the filtering model for denoising to obtain accurate data after denoising.
[0013] S3. Online estimation of glass transition degree of powder coating: A glass transition degree estimation algorithm is constructed based on the recursive least squares method. The denoised accurate data obtained in step S2 is used as input, and the glass transition degree of powder coating is calculated in real time through recursive update.
[0014] S4. Adaptive Impedance Control Parameter Adjustment: Based on the estimation results of the glass transition degree in step S3, an adaptive impedance control model is constructed. The adaptive impedance control model is based on the "mass-damping-spring" physical model, and the impedance control parameters of the adaptive impedance control model are dynamically adjusted according to the glass transition degree.
[0015] S5. Bonding process closed-loop control: Based on the impedance control parameters adjusted in step S4, combined with the segmented PID-fuzzy composite temperature control algorithm, the blade rotation speed of the bonding machine, the temperature supply of the hot and cold blocks of the bonding barrel, and the single-cycle duration of the bonding process are adjusted in real time.
[0016] Preferably, step S1 specifically includes the following working steps:
[0017] Data collection:
[0018] By deploying a high-precision six-dimensional force / torque sensor at the connection end between the mixing shaft and the drive motor of the bonding machine, dynamic torque signals, axial / radial contact force signals and blade displacement signals are simultaneously collected during the mixing process;
[0019] Real-time temperatures are collected at various locations by temperature sensors deployed in the hot and cold partitioning area, the powder material core area, and the inlet and outlet of the blade cooling holes of the bonding barrel.
[0020] The real-time rotational speed of the blade is collected by an integrated speed sensor deployed on the blade drive shaft;
[0021] Communication links and data storage:
[0022] Based on the CodeSys platform, the communication link between the sensor and the programmable logic controller of the bonding machine is configured. The industrial Ethernet protocol is used to perform dynamic variable mapping on the collected data to build the coupling relationship between the data, and the data is stored through a dynamic data caching mechanism.
[0023] Preferably, step S2 specifically includes the following working steps:
[0024] S2.1. Constructing a high-precision filtering model: The high-precision filtering model is constructed based on the Kalman filtering algorithm. The torque, force, displacement, and temperature data in the collected data are used as the input of the filtering model. The noise caused by equipment vibration and environmental interference is eliminated by constructing state equations and observation equations.
[0025] S2.2. Data verification and correction: The temperature sensor is calibrated. If the data deviation between two temperature sensors in the same hot and cold segment exceeds a preset threshold, faulty data is removed and replaced with a moving average, based on the temperature change trend of adjacent areas. At the same time, the filtered torque data is linearly corrected based on the coupling relationship between the blade speed and the torque.
[0026] Preferably, step S3 specifically includes the following working steps:
[0027] S3.1. Constructing a correlation model: Based on the denoised and accurate data, the degree of glass transition is initially calculated using a correlation model that includes a torque normalization function, a temperature normalization function, a contact force correction function, and a displacement correction function;
[0028] S3.2. Online update of weight coefficients: The weight coefficients in the association model are updated in real time using the recursive least squares method, and the updated and more accurate glassiness degree is calculated recursively.
[0029] Preferably, the steps in step S3.2 using the recursive least squares method are as follows:
[0030] First, data initialization is performed, using the torque normalization function, temperature normalization function, contact force correction function, and displacement correction function as the initial output vector, and setting the initial weight coefficient vector;
[0031] Then, the augmented matrix for the current time step is calculated based on the input vector from the previous time step.
[0032] Finally, based on the augmented matrix at the current moment and the weight coefficient vector at the previous moment, the weight coefficient vector at the current moment is calculated and updated, and then the updated glassiness degree is calculated.
[0033] Preferably, step S4 specifically includes the following working steps:
[0034] Based on the dynamic characteristics of the bonding machine stirring system, a "mass-damping-spring" impedance control physical model is constructed with equivalent mass, damping coefficient and spring stiffness as parameters, wherein the damping coefficient and spring stiffness are positively correlated with the updated glass transition degree.
[0035] The updated glass transition value is compared with a preset threshold range, and the equivalent mass, damping coefficient, and spring stiffness are dynamically adjusted based on the comparison result.
[0036] Based on the adjusted physical model and the actual contact force, angular displacement, angular velocity, and angular acceleration, the desired contact force is determined.
[0037] Preferably, step S5, "combining the segmented PID-fuzzy composite temperature control algorithm," specifically includes the following steps:
[0038] First-stage control: When the detected temperature deviates significantly from the target glass transition temperature, fuzzy control is adopted, taking the temperature deviation and the rate of change of deviation as inputs, and rapidly heating or cooling according to the preset fuzzy rule table.
[0039] Second-stage control: When the temperature deviation is reduced to a smaller range, switch to PID control mode to continue precise temperature fine-tuning.
[0040] Preferably, the specific steps for "adjusting the blade rotation speed of the bonding machine" in step S5 are as follows:
[0041] The updated degree of vitrification is compared with a preset first threshold and a second threshold.
[0042] If the degree of glass transition is lower than the first threshold, then increase the blade rotation speed;
[0043] If the degree of glass transition is between the first threshold and the second threshold, then maintain a moderate blade rotation speed;
[0044] If the degree of glass transition is higher than the second threshold, the blade rotation speed is reduced.
[0045] Preferably, the specific steps for "adjusting the single-bundle production duration" in step S5 are as follows:
[0046] Based on the time required for the glass transition to reach the target value during the bonding process of the current batch, calculate and adjust the bonding duration for the next batch: specifically, add the time required for the current batch to reach the target value to the reserved stabilization time as the reference duration for the next batch.
[0047] Preferably, the step of "adjusting the single-cycle duration of bonding production" further includes:
[0048] If the bonding length correction value of several consecutive batches tends to be stable and meets the expected conditions, then the correction value is fixed as the benchmark bonding length for subsequent batches; if it does not meet the conditions, then the calculation and dynamic adjustment continue.
[0049] Compared with the prior art, the beneficial effects of the present invention are:
[0050] This invention constructs an integrated intelligent bonding control system encompassing "perception-estimation-decision-execution." Through high-precision fusion of multi-source data and real-time online estimation of glass transition state, it achieves dynamic collaborative optimization of impedance control parameters and temperature control processes. This overcomes the bottlenecks of lag and discreteness in traditional control modes, enabling a systematic leap in five dimensions of powder coating bonding processes: control accuracy (±0.3℃, ±0.2 N·m), response speed (millisecond-level feedback), process stability (adaptive correction of single-cycle duration), material limits (metal powder addition up to 17.2%), and system robustness (fault self-healing, zero-delay synchronization). Ultimately, it achieves the intelligent manufacturing goals of a 28% reduction in single-cycle time, a 22% reduction in unit energy consumption, and a product qualification rate exceeding 98%. Attached Figure Description
[0051] Figure 1 This is a corrected flowchart of the intelligent monitoring system for the entire coating production process of the present invention. Detailed Implementation
[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0053] Example 1
[0054] Please see Figure 1 The diagram illustrates an intelligent monitoring system for the entire coating production process based on multi-source data fusion, comprising the following steps:
[0055] S1. Real-time acquisition of multi-source data: Data is acquired in real time by sensors installed on the powder coating bonding machine, and the communication link between the sensors and the bonding machine is configured using an industrial control platform to establish variable mapping to store the acquired data;
[0056] S2. Multi-source data preprocessing: Based on the Kalman filter algorithm, a high-precision filtering model suitable for the bonding machine is constructed. The data collected in step S1 is input into the filtering model for denoising to obtain accurate data after denoising.
[0057] S3. Online estimation of glass transition degree of powder coating: A glass transition degree estimation algorithm is constructed based on the recursive least squares method. The denoised accurate data obtained in step S2 is used as input, and the glass transition degree of powder coating is calculated in real time through recursive update.
[0058] S4. Adaptive Impedance Control Parameter Adjustment: Based on the estimation results of the glass transition degree in step S3, an adaptive impedance control model is constructed. The adaptive impedance control model is based on the "mass-damping-spring" physical model, and the impedance control parameters of the adaptive impedance control model are dynamically adjusted according to the glass transition degree.
[0059] S5. Bonding process closed-loop control: Based on the impedance control parameters adjusted in step S4, combined with the segmented PID-fuzzy composite temperature control algorithm, the blade rotation speed of the bonding machine, the temperature supply of the hot and cold blocks of the bonding barrel, and the single-cycle duration of the bonding process are adjusted in real time.
[0060] Preferably, step S1 specifically includes the following working steps:
[0061] Data collection:
[0062] By deploying a high-precision six-dimensional force / torque sensor at the connection end between the mixing shaft and the drive motor of the bonding machine, dynamic torque signals, axial / radial contact force signals and blade displacement signals are simultaneously collected during the mixing process;
[0063] Real-time temperatures are collected at various locations by temperature sensors deployed in the hot and cold partitioning area, the powder material core area, and the inlet and outlet of the blade cooling holes of the bonding barrel.
[0064] The real-time rotational speed of the blade is collected by an integrated speed sensor deployed on the blade drive shaft;
[0065] Communication links and data storage:
[0066] Based on the CodeSys platform, the communication link between the sensor and the programmable logic controller of the bonding machine is configured. The industrial Ethernet protocol is used to perform dynamic variable mapping on the collected data to build the coupling relationship between the data, and the data is stored through a dynamic data caching mechanism.
[0067] Preferably, step S2 specifically includes the following working steps:
[0068] S2.1. Constructing a high-precision filtering model: The high-precision filtering model is constructed based on the Kalman filtering algorithm. The torque, force, displacement, and temperature data in the collected data are used as the input of the filtering model. The noise caused by equipment vibration and environmental interference is eliminated by constructing state equations and observation equations.
[0069] S2.2. Data verification and correction: The temperature sensor is calibrated. If the data deviation between two temperature sensors in the same hot and cold segment exceeds a preset threshold, faulty data is removed and replaced with a moving average, based on the temperature change trend of adjacent areas. At the same time, the filtered torque data is linearly corrected based on the coupling relationship between the blade speed and the torque.
[0070] Preferably, step S3 specifically includes the following working steps:
[0071] S3.1. Constructing a correlation model: Based on the denoised and accurate data, the degree of glass transition is initially calculated using a correlation model that includes a torque normalization function, a temperature normalization function, a contact force correction function, and a displacement correction function;
[0072] S3.2. Online update of weight coefficients: The weight coefficients in the association model are updated in real time using the recursive least squares method, and the updated and more accurate glassiness degree is calculated recursively.
[0073] Preferably, the steps in step S3.2 using the recursive least squares method are as follows:
[0074] First, data initialization is performed, using the torque normalization function, temperature normalization function, contact force correction function, and displacement correction function as the initial output vector, and setting the initial weight coefficient vector;
[0075] Then, the augmented matrix for the current time step is calculated based on the input vector from the previous time step.
[0076] Finally, based on the augmented matrix at the current moment and the weight coefficient vector at the previous moment, the weight coefficient vector at the current moment is calculated and updated, and then the updated glassiness degree is calculated.
[0077] Preferably, step S4 specifically includes the following working steps:
[0078] Based on the dynamic characteristics of the bonding machine stirring system, a "mass-damping-spring" impedance control physical model is constructed with equivalent mass, damping coefficient and spring stiffness as parameters, wherein the damping coefficient and spring stiffness are positively correlated with the updated glass transition degree.
[0079] The updated glass transition value is compared with a preset threshold range, and the equivalent mass, damping coefficient, and spring stiffness are dynamically adjusted based on the comparison result.
[0080] Based on the adjusted physical model and the actual contact force, angular displacement, angular velocity, and angular acceleration, the desired contact force is determined.
[0081] Preferably, step S5, "combining the segmented PID-fuzzy composite temperature control algorithm," specifically includes the following steps:
[0082] First-stage control: When the detected temperature deviates significantly from the target glass transition temperature, fuzzy control is adopted, taking the temperature deviation and the rate of change of deviation as inputs, and rapidly heating or cooling according to the preset fuzzy rule table.
[0083] Second-stage control: When the temperature deviation is reduced to a smaller range, switch to PID control mode to continue precise temperature fine-tuning.
[0084] Preferably, the specific steps for "adjusting the blade rotation speed of the bonding machine" in step S5 are as follows:
[0085] The updated degree of vitrification is compared with a preset first threshold and a second threshold.
[0086] If the degree of glass transition is lower than the first threshold, then increase the blade rotation speed;
[0087] If the degree of glass transition is between the first threshold and the second threshold, then maintain a moderate blade rotation speed;
[0088] If the degree of glass transition is higher than the second threshold, the blade rotation speed is reduced.
[0089] Preferably, the specific steps for "adjusting the single-bundle production duration" in step S5 are as follows:
[0090] Based on the time required for the glass transition to reach the target value during the bonding process of the current batch, calculate and adjust the bonding duration for the next batch: specifically, add the time required for the current batch to reach the target value to the reserved stabilization time as the reference duration for the next batch.
[0091] Preferably, the step of "adjusting the single-cycle duration of bonding production" further includes:
[0092] If the bonding length correction value of several consecutive batches tends to be stable and meets the expected conditions, then the correction value is fixed as the benchmark bonding length for subsequent batches; if it does not meet the conditions, then the calculation and dynamic adjustment continue.
[0093] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0094] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. An intelligent monitoring system for the entire coating production process based on multi-source data fusion, comprising the following steps: S1. Real-time acquisition of multi-source data: Data is acquired in real time by sensors installed on the powder coating bonding machine, and the communication link between the sensors and the bonding machine is configured using an industrial control platform to establish variable mapping to store the acquired data; S2. Multi-source data preprocessing: Based on the Kalman filter algorithm, a high-precision filtering model suitable for the bonding machine is constructed. The data collected in step S1 is input into the filtering model for denoising to obtain accurate data after denoising. S3. Online estimation of glass transition degree of powder coating: A glass transition degree estimation algorithm is constructed based on the recursive least squares method. The denoised accurate data obtained in step S2 is used as input, and the glass transition degree of powder coating is calculated in real time through recursive update. S4. Adaptive Impedance Control Parameter Adjustment: Based on the estimation results of the glass transition degree in step S3, an adaptive impedance control model is constructed. The adaptive impedance control model is based on the "mass-damping-spring" physical model, and the impedance control parameters of the adaptive impedance control model are dynamically adjusted according to the glass transition degree. S5. Bonding process closed-loop control: Based on the impedance control parameters adjusted in step S4, combined with the segmented PID-fuzzy composite temperature control algorithm, the blade rotation speed of the bonding machine, the temperature supply of the hot and cold blocks of the bonding barrel, and the single-cycle duration of the bonding process are adjusted in real time.
2. The intelligent monitoring system for the entire coating production process based on multi-source data fusion as described in claim 1, characterized in that: Step S1 specifically includes the following working steps: Data collection: By deploying a high-precision six-dimensional force / torque sensor at the connection end between the mixing shaft and the drive motor of the bonding machine, dynamic torque signals, axial / radial contact force signals and blade displacement signals are simultaneously collected during the mixing process; Real-time temperatures are collected at various locations by temperature sensors deployed in the hot and cold partitioning area, the powder material core area, and the inlet and outlet of the blade cooling holes of the bonding barrel. The real-time rotational speed of the blade is collected by an integrated speed sensor deployed on the blade drive shaft; Communication links and data storage: Based on the CodeSys platform, the communication link between the sensor and the programmable logic controller of the bonding machine is configured. The industrial Ethernet protocol is used to perform dynamic variable mapping on the collected data to build the coupling relationship between the data, and the data is stored through a dynamic data caching mechanism.
3. The intelligent monitoring system for the entire coating production process based on multi-source data fusion as described in claim 2, characterized in that: Step S2 specifically includes the following working steps: S2.
1. Constructing a high-precision filtering model: The high-precision filtering model is constructed based on the Kalman filtering algorithm. The torque, force, displacement, and temperature data in the collected data are used as the input of the filtering model. The noise caused by equipment vibration and environmental interference is eliminated by constructing state equations and observation equations. S2.
2. Data verification and correction: The temperature sensor is calibrated. If the data deviation between two temperature sensors in the same hot and cold segment exceeds a preset threshold, faulty data is removed and replaced with a moving average, based on the temperature change trend of adjacent areas. At the same time, the filtered torque data is linearly corrected based on the coupling relationship between the blade speed and the torque.
4. The intelligent monitoring system for the entire coating production process based on multi-source data fusion as described in claim 3, characterized in that: Step S3 specifically includes the following working steps: S3.
1. Constructing a correlation model: Based on the denoised and accurate data, the degree of glass transition is initially calculated using a correlation model that includes a torque normalization function, a temperature normalization function, a contact force correction function, and a displacement correction function; S3.
2. Online update of weight coefficients: The weight coefficients in the association model are updated in real time using the recursive least squares method, and the updated and more accurate glassiness degree is calculated recursively.
5. The intelligent monitoring system for the entire coating production process based on multi-source data fusion as described in claim 4, characterized in that: The steps in step S3.2 using the recursive least squares method are as follows: First, data initialization is performed, using the torque normalization function, temperature normalization function, contact force correction function, and displacement correction function as the initial output vector, and setting the initial weight coefficient vector; Then, the augmented matrix for the current time step is calculated based on the input vector from the previous time step. Finally, based on the augmented matrix at the current moment and the weight coefficient vector at the previous moment, the weight coefficient vector at the current moment is calculated and updated, and then the updated glassiness degree is calculated.
6. The intelligent monitoring system for the entire coating production process based on multi-source data fusion as described in claim 5, characterized in that: Step S4 specifically includes the following working steps: Based on the dynamic characteristics of the bonding machine stirring system, a "mass-damping-spring" impedance control physical model is constructed with equivalent mass, damping coefficient and spring stiffness as parameters, wherein the damping coefficient and spring stiffness are positively correlated with the updated glass transition degree. The updated glass transition value is compared with a preset threshold range, and the equivalent mass, damping coefficient, and spring stiffness are dynamically adjusted based on the comparison result. Based on the adjusted physical model and the actual contact force, angular displacement, angular velocity, and angular acceleration, the desired contact force is determined.
7. The intelligent monitoring system for the entire coating production process based on multi-source data fusion as described in claim 5, characterized in that: The "combination of segmented PID-fuzzy composite temperature control algorithm" in step S5 specifically includes the following steps: First-stage control: When the detected temperature deviates significantly from the target glass transition temperature, fuzzy control is adopted, taking the temperature deviation and the rate of change of deviation as inputs, and rapidly heating or cooling according to the preset fuzzy rule table. Second-stage control: When the temperature deviation is reduced to a smaller range, switch to PID control mode to continue precise temperature fine-tuning.
8. The intelligent monitoring system for the entire coating production process based on multi-source data fusion according to claim 7, characterized in that: The specific steps for "adjusting the blade rotation speed of the bonding machine" in step S5 are as follows: The updated degree of vitrification is compared with a preset first threshold and a second threshold. If the degree of glass transition is lower than the first threshold, then increase the blade rotation speed; If the degree of glass transition is between the first threshold and the second threshold, then maintain a moderate blade rotation speed; If the degree of glass transition is higher than the second threshold, the blade rotation speed is reduced.
9. The intelligent monitoring system for the entire coating production process based on multi-source data fusion as described in claim 8, characterized in that: The specific steps for "adjusting the single-bundle production duration" in step S5 are as follows: Based on the time required for the glass transition to reach the target value during the bonding process of the current batch, calculate and adjust the bonding duration for the next batch: specifically, add the time required for the current batch to reach the target value to the reserved stabilization time as the reference duration for the next batch.
10. The intelligent monitoring system for the entire coating production process based on multi-source data fusion according to claim 1, characterized in that: The steps for "adjusting the single-bundle production duration" also include: If the bonding length correction value of several consecutive batches tends to be stable and meets the expected conditions, then the correction value is fixed as the benchmark bonding length for subsequent batches; if it does not meet the conditions, then the calculation and dynamic adjustment continue.