Enamel electrostatic powder coating real-time monitoring and feedback control system

Abstract

The invention discloses a real-time monitoring and feedback control system for enamel electrostatic powder coating, and relates to the technical field of industrial enamel coating. The system comprises a data acquisition and monitoring module, a feedback control decision module, an execution mechanism control module, a spraying path planning and adjusting module, a parameter optimization and adaptive adjustment module and a multi-target cooperative adjustment strategy module. The system builds a standardized state vector and recognizes abnormity by collecting spraying enamel electrostatic powder state parameters, deposition error signals and environment interference data, and closed-loop control and dynamic adjustment of spray gun electrode voltage, powder supply amount and spray gun path parameters are achieved. Intelligent algorithms such as particle swarm optimization are introduced to optimize a controller and process parameters, the coating uniformity and the deposition efficiency are automatically balanced in combination with a multi-target collaborative strategy, and stable spraying quality is achieved under complex working conditions. The method has the advantages of self-learning, self-adaption and expandability, and the intelligent level and the production performance of the enamel electrostatic powder spraying process are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial enamel coating, in particular to a real-time monitoring and feedback control system for enamel electrostatic powder coating. Background Art

[0002] As an important surface treatment process, electrostatic enamel spraying is inherently a complex dynamic control system involving multiple variables and disturbances. Widely used in the automotive, home appliance, and construction industries, this process places stringent demands on coating uniformity, spraying efficiency, and energy consumption. However, due to the diverse morphologies of sprayed objects and volatile operating conditions, coordinated control of various execution units, real-time parameter adjustment, and environmental disturbance suppression during the spraying process are key factors affecting product quality and production efficiency.

[0003] In existing spray control systems, mainstream solutions rely on fixed control programs or manual experience to set control parameters. These solutions lack a real-time closed-loop feedback mechanism, making it difficult to achieve dynamic adaptive control of the spray process. This is especially true when faced with complex curved surfaces, spatial obstructions, and fluctuating environmental conditions. The system's control accuracy is insufficient, making it prone to quality issues such as uneven coating, overspray, and underspray. Furthermore, while some studies have introduced sensor monitoring, image recognition, or coating thickness detection modules, these functional units lack efficient linkage and coordination with the control decision-making and execution mechanisms, preventing the formation of a complete automated closed-loop control system.

[0004] Current control strategies generally have the following shortcomings: first, the coupling degree between data acquisition, analysis and execution is low, and information transmission delays lead to delayed control response; second, control logic is mostly statically set and lacks adaptive optimization capabilities based on real-time data; third, path planning and spray parameter adjustment fail to coordinate operations under a unified control framework, which cannot fully reflect the advantages of process control systems in multi-variable coordinated optimization.

[0005] In response to the above problems, the present invention provides a real-time monitoring and feedback control system for enamel electrostatic powder coating. Through multi-source perception, dynamic modeling, real-time feedback and closed-loop control of the actuator, combined with the collaborative self-optimization of path and process parameters, it realizes intelligent control and adaptive operation of the entire spraying process, significantly improving the control accuracy and stability of the system. Summary of the Invention

[0006] In response to the above problems, the present invention provides a real-time monitoring and feedback control system for enamel electrostatic powder coating to solve the problems of the existing technology in the enamel electrostatic powder spraying process, such as lack of real-time feedback regulation, disconnection between path and deposition state, and insufficient parameter optimization capability.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a real-time monitoring and feedback control system for enamel electrostatic powder coating, comprising the following modules: The data acquisition and monitoring module includes a spray gun parameter sensing unit, a particle deposition monitoring unit, and an environmental state perception unit, which respectively collect spray state parameters, deposition error signals, and environmental interference parameters; the signal processing and anomaly detection unit generates a standardized state vector and a weighted error signal vector; The feedback control decision module consists of an error calculation and reference signal reconstruction unit, an amplitude and phase compensation unit, a control algorithm unit, and an output signal adjustment unit. It receives and processes the standardized state vector and the multi-dimensional error signal in sequence to obtain the final execution control signal. The actuator control module splits and maps the final execution control signal to different control objects, driving the high-voltage electrostatic output control unit to adjust the spray gun electrode voltage, the powder supply and air pressure control unit to adjust the powder flow and conveying gas pressure, and the motion path execution unit to control the spray gun motion posture and trajectory speed; The spray path planning and adjustment module generates the initial posture control path of the spray gun based on the three-dimensional morphology of the workpiece and the desired deposition specifications, and adjusts the path parameters in real time based on the coating deposition feedback signal during the spraying process; The parameter optimization and adaptive adjustment module is responsible for intelligent adjustment, dynamic coordination and knowledge accumulation of control parameters and process set points, maintaining stability and optimal coating performance under variable working conditions; The multi-objective collaborative adjustment strategy module is used to comprehensively balance various control objectives during operation and adjust control parameters and execution parameters in real time.

[0008] Compared with the prior art, the advantages of the present invention are: The present invention constructs a data acquisition and monitoring module that integrates spray gun parameters, particle deposition, and environmental status. It introduces a multi-channel synchronous acquisition and signal fusion mechanism for the first time in enamel electrostatic powder spraying. Through standardized state vector construction and abnormal interference identification, it not only improves the integrity and consistency of state data, but also effectively eliminates the influence of external disturbances on the spraying process, solves the problem that traditional single-point monitoring methods are difficult to capture systematic errors, and provides high-reliability basic data for subsequent feedback control.

[0009] The present invention designs a multi-layer feedback control structure including state reconstruction, gain adjustment, control algorithm and output correction. It not only corrects the spraying parameters in real time based on the instantaneous error signal, but also has the advantages of strong disturbance suppression ability, fast adjustment response and high control accuracy. In particular, it can still maintain good coating consistency and uniformity on complex workpiece surfaces and under extreme environmental conditions.

[0010] The present invention refines the feedback control signal into three-dimensional execution actions: high-voltage power supply regulation, powder supply / air pressure linkage regulation, and path motion control. Combined with high-precision sensors and fast-response actuators, it achieves millisecond-level implementation of control instructions, constructs a complete "instruction-execution-feedback" closed-loop chain, and the spray gun electrode voltage, powder flow rate and movement trajectory are highly coordinated, which greatly reduces overspray and missed spraying, improves material utilization and reduces the recycling burden.

[0011] The present invention effectively solves the problem of uneven coating caused by the "Faraday cage" effect in complex and special-shaped workpieces or edge areas through the fusion mechanism of offline path generation and online trajectory fine-tuning. The online trajectory adjustment unit can dynamically correct the spray gun speed, posture and dwell time according to the coating deviation, realizing "re-spraying" and "avoiding spraying" on demand.

[0012] The present invention introduces an improved particle swarm optimization algorithm and a multi-objective collaborative optimization strategy to perform global optimization search and dynamic adjustment of controller parameters and process parameters. The algorithm supports self-learning and online updates, and can adapt to different workpieces, powder types or environmental conditions in real time, avoiding manual trial and error adjustment, greatly shortening debugging time and improving system stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. It is understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 It is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described in conjunction with the drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but is merely a selected embodiment of the present invention.

[0016] Please refer to Figure 1 , Figure 1 This is an architecture diagram of a real-time monitoring and feedback control system for enamel electrostatic powder coating provided by an embodiment of the present invention, including the following modules: The data acquisition and monitoring module includes a spray gun parameter sensing unit, a particle deposition monitoring unit and an environmental status perception unit, which respectively collect spray state parameters, deposition error signals and environmental interference parameters; the collected data is denoised and filtered, standardized and analyzed by the signal processing and anomaly detection unit, and then organized to generate a standardized state vector and a weighted error signal vector.

[0017] The spray gun parameter sensing unit collects spray state parameters through high-voltage voltage sensors, corona discharge current sensors, and atomization and delivery air pressure sensors. The spray state parameters are specifically: high-voltage electrode voltage U(t), current I(t), and air pressure P(t) during the spraying process. Deviation analysis is performed on the spray state parameters and the set target parameters to determine whether the current spray gun is in a stable working state, and then determine whether the working point of the enamel electrostatic powder spraying equipment is normal. Among them, the working point refers to the operating state collection point of the electrostatic spray gun body and its high-voltage power supply and pneumatic interface, which is the judgment basis of the control logic. It should be noted that the high-voltage voltage sensor is connected to the output end of the high-voltage power module of the spray gun, and the corona discharge current sensor is embedded in the discharge electrode path of the spray gun; the atomization and delivery air pressure sensor is installed in the air supply pipe of the spray gun; U(t) is used to determine whether the spray gun has reached the expected electric field strength; I(t) is used to reflect the spray corona intensity and powder charging efficiency, and P(t) is used to confirm whether the pneumatic delivery is stable; in addition, the working point of the spray equipment is the real-time operating status point of the electrostatic spray gun and its high-voltage power supply and air supply interface, and the judgment is based on are all within the set threshold range; if the working point deviates, it is determined to be "abnormal operation of the spray gun". This state will trigger the feedback control module to adjust the spraying parameters and issue an alarm signal.

[0018] The particle deposition monitoring unit uses a laser thickness gauge placed near the enamel electrostatic powder spray gun to continuously measure coating thickness. It also uses an industrial camera to capture spray coverage images and extract uniformity features from the spray images. A deposition rate estimation model based on current fluctuations is introduced to indirectly infer the change in powder application per unit time (i.e., deposition rate per unit time), comprehensively reflecting coating quality and powder utilization efficiency. The deposition rate per unit time is as shown in the formula: in, is the deposition rate per unit time, is the current fluctuation value per unit time, k is the charging efficiency constant, As unit time.

[0019] The collected coating thickness, uniformity characteristics and deposition rate per unit time are compared with the target values ​​to generate a deposition error signal.

[0020] Environmental status sensing unit, used to detect interference factors in the spraying environment that may affect the stability of the coating, using a digital humidity sensor to obtain relative humidity RH(t) and a thermoelectric sensor to obtain ambient temperature , and the electric field disturbance sensor detects background interference voltage from non-gun sources , the three together constitute the environmental interference parameters; When any indicator of the environmental interference parameters exceeds the set tolerance threshold, this state is marked as "external interference activated", and the subsequent control module is triggered to dynamically adjust parameters such as spraying voltage or air pressure to maintain coating stability.

[0021] The signal processing and anomaly detection unit performs unified preprocessing, anomaly detection and weighted error signal generation on the collected spray state parameters, deposition error signals and environmental interference parameters. First, the three types of collected raw data are organized into spray state parameter vector S(t), deposition effect parameter vector D(t) and environmental interference parameter vector E(t), and the above three vectors are unified to form the original state data set. , and then perform fusion processing on the input data set based on the time window. The specific operations are as follows: First, the three types of state vectors are denoised separately by a sliding average filter to eliminate transient deviations caused by high-frequency acquisition fluctuations and device jitter; then the Z-score-based normalization method is used to normalize the state vector to a zero-mean unit variance. , eliminate the deviation between different dimensions; For the standardized state vector, three anomaly identification strategies are further designed: 1. Set dynamic control limits for each parameter in the standardized state vector, and construct an upper limit based on the historical average and process fluctuation threshold With lower limit ,when Any parameter item in exceeds the range When , it is regarded as an out-of-bounds exception and the deviation range is recorded; Second, calculate the perturbation growth rate of each state parameter and compare it with its sliding historical mean. If the rate of change increases or decreases abnormally, it is marked as a trend drift anomaly; 3. Establishing the state residual function ,in, is the residual value of the i-th state parameter at time t, reflecting the difference between the current value and the predicted value; is the actual observed parameter value of item i, is the estimated value based on the autoregressive sliding window forecast; when the residual Exceeding the seatbelt threshold , it is determined to be a soft fault risk.

[0022] All abnormal results identified by the above three abnormality identification strategies will be uniformly encoded into a multi-dimensional error signal vector ,Each item in the vector not only represents the abnormal type of the ,corresponding indicator, but also comes with a specific deviation value, ,triggering time and confidence score, which are used to express specific ,scenarios, including “instantaneous drop in the spray gun voltage”, “uneven powder deposition”, and “excessive ,temperature interference in the workshop”.

[0023] In order to further enhance the response accuracy of the feedback loop, this unit constructs a parameter priority matrix , where m is the dimension of the control variable and n is the dimension of the error signal. The multidimensional error signal vector is weighted, and each element in W is It represents the influence strength of the j-th error signal on the i-th control variable, which can be set based on historical control data, expert experience and fault impact statistics. After the matrix acts on the multidimensional error signal vector, the output is a weighted error signal vector. , while retaining multi-source anomaly information, it highlights the impact of key control variables, thereby driving the feedback adjustment mechanism of the subsequent control strategy generation module in a more targeted manner.

[0024] The feedback control decision module consists of an error calculation and reference signal reconstruction unit, an amplitude and phase compensation unit, a control algorithm unit, and an output signal adjustment unit. It receives and processes the standardized state vector and the multidimensional error signal in sequence to obtain the final execution control signal.

[0025] Error calculation and reference signal reconstruction unit, based on the standardized state vector output by the data acquisition and monitoring module and weighted error signal vector , further process the current state of the system and build a smoother and more dynamically adaptable control reference signal , and based on this, the instantaneous error signal vector that can be directly used by the controller is calculated .

[0026] First, introduce a sliding time window The historical state sequence on , using weighted quadratic spline interpolation function to generate the current reference state trajectory , where the interpolation weight vector According to the current weighted error signal Dynamic adjustment prioritizes strengthening the fitting accuracy of high-weight abnormal variables, and the expression for generating the reference trajectory is: Where, is the quadratic spline basis function with the i-th moment as the control node, is the weight factor, satisfying The above method ensures that the reference signal not only reflects the historical change trend, but also has the ability to adjust in real time to adapt to the control needs under different disturbance levels.

[0027] After obtaining the current reference state vector After that, the unit further compares the real-time state vector For comparison, calculate the instantaneous error signal vector of the system: in, is the instantaneous error signal vector, is the real-time state vector, is the current reference state vector; vector Each dimension of It indicates the instantaneous deviation degree of the corresponding state parameters, including the spray gun output voltage, current, powder deposition thickness, environmental disturbance intensity, etc. The error signal of this unit is generated based on the adaptively constructed reference trajectory, and has stronger time consistency and anti-interference ability.

[0028] In order to enhance the efficiency of the controller's response to abnormal changes, Perform weighted mapping processing to construct an adjustment signal that can be used to stimulate the control strategy model, as shown in the formula: in, To adjust the signal, is the parameter priority matrix is the instantaneous error signal vector, and the dimension m is the number of parameters of the state vector. This vector not only retains the amplitude characteristics of the error information; ultimately, this unit outputs two core variable adaptive reference state trajectories and weighted control error signal .

[0029] The amplitude-phase compensation unit, located in the signal preprocessing stage of the feedback control module, is used to correct the system amplitude-frequency and phase-frequency distortion caused by transmission delay, perception lag, and dynamic mismatch in the spray control chain before the control error signal enters the controller calculation. By designing a compensation filter in the frequency domain to reversely correct the gain attenuation and phase lag generated during the transmission process, the entire control channel is equivalent to an ideal linear system, thereby enhancing the controller's response speed and adjustment accuracy to the error signal; First, through the frequency domain experimental identification in the trial operation phase, the frequency response model of the spray control path is obtained, which is recorded as the transfer function , which represents the response relationship from the spray gun control parameter input to the deposition error signal output. The function can be decomposed into the amplitude-frequency response Phase frequency response , as shown in the formula: According to this model, a compensation filter is constructed , its goal is to achieve amplitude-frequency inverse gain and phase reverse feedforward, so that: The entire transmission chain is made to approach unity gain and zero phase lag within the control frequency band, thereby achieving instant amplification and phase alignment under the linear response condition of the error signal.

[0030] When the filter acts on the error signal and generates the compensated control reference input This ensures that the signal received by the controller is no longer distorted by the dynamic lag of the spray gun-deposition-sensing channel, thereby improving the dynamic controllability and stability margin of the closed-loop control system.

[0031] For multi-gun channel scenarios, this unit constructs the channel amplitude and phase compensation matrix , and adjust the filter coefficients of each channel online to keep the response bandwidth of each channel consistent and the cross-interference minimized in the multi-input multi-output system. The compensation formula is expanded to: in, and They represent Fourier transform and inverse Fourier transform operations, respectively, which are used to complete frequency domain filtering operations.

[0032] It should be noted that the accuracy and response stability of the compensation filter are limited by the accuracy of the transfer function model and the sampling frequency. Therefore, in the initial deployment stage, the system is tested and identified through actual measurement, model fitting, and parameter self-tuning. and Perform iterative optimization to ensure that the compensation effect in the main working frequency band meets the performance requirements of rapid convergence of deposition error, and finally output the compensated error signal vector .

[0033] The control algorithm unit is responsible for receiving the error signal processed by the amplitude and phase compensation unit , and combined with the current reference state vector , generating the master output signal This system is used to guide the spraying equipment to achieve precise adjustment of coating thickness. Considering the significant nonlinear disturbances, operating point drift, and environmental interference in the electrostatic powder spraying process, a linear active disturbance rejection control (LADRC) strategy is used to construct the main control line. Its core advantages lie in its weak model dependence and strong anti-disturbance robustness. The LADRC control structure consists of the following three parts: Tracking differentiator TD, used to process the reference state vector The mutation terms or discontinuous boundary values ​​in the θ are transformed smoothly by first-order or second-order dynamic structures to avoid the high-frequency components from entering the controller. The output smooth reference trajectory is recorded as , which preserves the desired behavior trend while eliminating signal jitter.

[0034] The extended state observer (ESO) uses feedback signals and control signals to jointly estimate the current state of the system (including deposition thickness and deposition rate) and the total disturbance term (including unmodeled dynamics, powder supply instability, and environmental wind disturbances). The ESO is usually designed in second-order or third-order form, and the state estimate is expressed as follows: Estimated sediment thickness, Thickness change rate, The total perturbation estimate, all the estimated results constitute the expanded state vector , providing an observational basis for the feedback law.

[0035] Linear state error feedback LSEF, by calculating the desired trajectory With estimated state The deviation between them is used to construct the feedback law to generate the control output: in, They are proportional gain and disturbance compensation gain respectively. The feedback adjustment process has the ability of structural decoupling, rapid response and active disturbance suppression. The overall control strategy is designed as a control gain adjustable structure, which is automatically adjusted during operation according to the system transient response and control accuracy requirements, so that the system can maintain a stable deposition thickness for different workpiece structures.

[0036] In addition, in order to improve the output controllability, the unit also introduces a limiting function And control rate constraints to limit the control quantity from changing suddenly or exceeding the physical boundary: Finally, the unit outputs the control main quantity The signal will be passed to the output signal conditioning unit for pre-processing and is the core instruction flow that drives the adjustment of the spray gun's operating status.

[0037] Output signal conditioning unit, responsible for receiving the main control output signal generated by the control algorithm unit , and complete a series of safety restrictions and response adaptation operations before the command is sent to the actuator module, and finally generate an execution command flow that can directly drive the spray gun execution component .

[0038] Control signal safety limiting mechanism: The controller generates Ideal control instructions may exceed the operating range allowed by the spray gun hardware in engineering practice, posing risks such as overvoltage, overcurrent, and overheating. Therefore, physical limit constraints are first applied to ensure that each control dimension is within the set safety range: in, These are the physical lower and upper limits of each execution variable (such as electrostatic voltage, powder supply valve opening, guide rail speed, etc.). This process ensures that all sent signals do not cause equipment overload, operational instability, or control failure.

[0039] Actuator dynamic adaptation mechanism: Different types of actuators have their own specific dynamic response characteristics. Figure 1 To further Perform the following compensation corrections: The slope limiter suppresses signal mutations to prevent equipment impact or coating scattering caused by a sudden increase in the command. The expression is as follows: The hysteresis compensation term pre-compensates for valve response delay, voltage rise inertia, etc., to improve the tracking accuracy of the control response; the offset correction term compensates for static errors and constructs output zero point alignment; the feedforward gain factor adds gain correction in combination with the trend direction of the control quantity to improve the response bandwidth.

[0040] The above mechanisms act in combination , and obtain the final execution control signal after dynamic matching: in, represents the combined effect of all compensation functions and can be regarded as a "device-friendly control mapping function". In addition, in order to achieve data closed-loop management, this unit will The time series value is synchronously transmitted back to the data acquisition and monitoring module as the control signal execution log; the final output signal adjustment unit will finally execute the control signal sent to the actuator module.

[0041] The actuator control module receives the final execution control signal output by the feedback control module, splits it and maps it to different control objects, and drives the high-voltage electrostatic output control unit to adjust the spray gun electrode voltage, the powder supply and air pressure control unit to adjust the powder flow and conveying gas pressure, and the motion path execution unit to control the spray gun movement posture and trajectory speed. They work together on the electrical, pneumatic and mechanical components of the spraying equipment to realize the spatial distribution control and thickness adjustment of the powder deposition process.

[0042] In the high voltage output control unit, it is used to analyze the control signal components for the parameter part in the feedback control module , and map it to the high voltage output setting value of the spray gun electrode in real time , in order to adjust the charge state and migration driving force of the powder particles, and thus accurately control the deposition behavior of the enamel electrostatic powder on the workpiece surface; this unit constructs a high-voltage electrode control interface layer, connects the high-voltage generation module, and realizes closed-loop adjustment of the electric field strength between the spray gun electrode needle and the workpiece.

[0043] Specifically, the control signal After signal decoding and safety verification, the upper and lower limits of the electrostatic output are first set through the dynamic limiter module, and the voltage slope limiter is introduced at the same time. , to ensure that the voltage changes smoothly and avoid sudden changes that may cause powder accumulation instability or gun penetration; then, the adjusted electrostatic instruction drives the high-voltage power supply through the closed-loop interface to make its output voltage Closely track the control set value; to ensure closed-loop performance, a local feedback control loop is built through the integrated voltage / current sensor to monitor the electrical status of the spray gun end in real time. If the set threshold is exceeded, the system will automatically reduce the voltage output and activate the overcurrent protection mechanism.

[0044] During the dynamic operation of the spraying process, if the control system detects that the powder deposition in a certain area is insufficient, the feedback decision module will increase the , thereby enhancing the charge level of the powder particles, increasing the electric field driving force for their migration to the workpiece, and achieving enhanced local deposition; conversely, if the deposition is too thick, reduce the gun electrode voltage to limit the adhesion of excess powder.

[0045] The high voltage control quantity output by this unit The data will be synchronously transmitted to the data acquisition and monitoring module for operation record of spraying electrical parameters and subsequent fault tracing. At the same time, the response delay and adjustment deviation of the electrostatic control interface will be periodically checked to ensure that the control closed loop has high bandwidth, high precision and low latency performance.

[0046] The powder supply and air pressure control unit is used to analyze the powder supply and air drive parts in the output signal of the feedback control decision module, that is, and The unit controls the powder discharge concentration and velocity distribution, directly determining the coating thickness uniformity and material utilization.

[0047] control signal After adjustment, it is used as the powder feeder execution command to control the driving frequency of the screw powder feed pump or vibration powder feeder The frequency determines the amount of powder delivered per unit time, and thus adjusts the actual powder flow rate at the spray gun nozzle. To prevent powder supply blockage or overshoot, a speed limiter and slope constraint module are introduced before signal transmission to ensure a smooth powder supply adjustment process. The system integrates a powder flow sensor and uses feedback to establish a closed-loop regulation to maintain the powder supply consistent with the set value.

[0048] control signal Analyzed as delivery pressure and atomizing air pressure The regulation instructions act on the electronically controlled proportional pressure regulating valve to control the gas flow rate and pressure, thereby affecting the propagation range and atomization quality of the powder jet. In addition, the air pressure configuration is dynamically adjusted according to process requirements. When insufficient powder is applied on complex contours or deep concave areas, the particle injection speed and dispersion are improved by increasing the atomization air pressure. On flat areas, the air pressure can be reduced to reduce powder rebound and waste.

[0049] By building a complete powder supply-air pressure coordinated control mechanism, the kinetic energy matching of powder particles and gas is adjusted, and their distribution state and deposition coverage during flight are controlled. In addition, the air pressure, flow rate, and powder concentration parameters are monitored in real time through a local feedback loop to ensure the spraying quality and actuator response performance; finally, the powder supply and air pressure control unit outputs the adjusted powder supply drive signal and air pressure setting signal.

[0050] Motion path execution unit, used to analyze the spatial path control instruction part in the output signal of the feedback control decision module , and convert it into real-time trajectory scheduling instructions for the mechanical actuator in the spraying equipment, driving the spray gun to run along the desired trajectory on the workpiece surface, completing the coating control in the spatial dimension.

[0051] The main objects controlled by this unit include industrial robot arms, linear slides, rotary platforms and other motion control devices. The input signals are in the form of expected spray gun position, attitude angle, and instantaneous velocity path parameters, forming a three-dimensional motion instruction set. The unit is equipped with a trajectory interpolator and a kinematic solver to convert discrete instructions into a continuous, smooth and physically reachable joint instruction sequence.

[0052] To ensure the coordination between the spray gun path and powder supply / voltage control, the unit constructs a spray synchronization mechanism to constrain the dynamic relationship between the spray gun position and the powder concentration and electric field strength. When the path passes through the edge of the workpiece or a deep concave area, if the data acquisition module detects insufficient deposition, the motion path execution unit responds to the deceleration command issued by the control module and reduces the local passing speed to increase the actual deposition thickness in the area. If overspray is found in a local area, the path execution immediately increases the spray gun speed to shorten the action time and suppress overspray.

[0053] It should be noted that this unit supports two path scheduling modes: offline path execution mode, which generates the spray trajectory in advance according to the workpiece CAD model and process requirements, stores it as a standard path file, and executes it according to a fixed timing during the spraying process; online path dynamic correction mode, which corrects the running path or local trajectory segment in real time based on the deposition error signal and position deviation feedback from the monitoring module to achieve adaptive spraying.

[0054] Finally, the motion path execution unit outputs real-time control commands to the robot controller or motion platform driver; at the same time, the actual position and speed information of the spray gun are transmitted back to the data acquisition and monitoring module in real time to assist in the subsequent analysis of the deviation relationship between deposition thickness and path execution.

[0055] The spray path planning and adjustment module is responsible for generating the initial posture control path of the spray gun according to the three-dimensional morphology of the workpiece and the desired deposition specifications, and adjusting the path parameters in real time based on the coating deposition feedback signal during the spraying process to achieve dynamic optimization of deposition uniformity and material utilization.

[0056] The offline path generation unit constructs the initial spray path point set based on the three-dimensional CAD model of the workpiece, the expected thickness mapping matrix, and the spray width characteristic parameters, and outputs the path control vector sequence.

[0057] Path planning utilizes a geometric algorithm based on voxel segmentation and surface scanning, integrating process constraints to ensure effective coverage of all critical surface areas while avoiding duplicate powder application or missed areas. The generated path vector sequence, including parameters such as pose points, transit speed, and preset attitude angles, is uploaded to the main control module before spraying, providing a basis for initial scheduling for the motion path execution unit.

[0058] The online trajectory adjustment unit receives the deposition thickness spatial distribution error signal output by the particle deposition monitoring unit during the spraying process, combines it with the offline path point set, identifies the deviation area in real time, executes the fine-tuning strategy, and generates the correction path instruction.

[0059] Specific adjustment strategies include: for areas with significant under-thickness, dynamically reducing the spray gun passing speed, increasing the number of repetitions of the path point, or appropriately shortening the distance from the spray gun to the workpiece; for areas detected to be over-thick, increasing the spray gun passing speed, reducing repeated path points, or temporarily disabling the path segment in this area; all trajectory fine-tuning operations are subject to a smoothing function to avoid actuator oscillation or position error diffusion caused by path jumps; the adjustment frequency is triggered according to the spray batch or cycle period to avoid unnecessary waiting or false triggering responses caused by frequent path reconstruction.

[0060] The multi-axis coordination and obstacle avoidance unit is used to coordinate and schedule the multi-axis joints of the spray actuator during the path planning and adjustment process, and to detect physical conflicts or constraint conflicts that may arise during the path adjustment process.

[0061] The input includes the corrected path control vector, the spray environment model, and the robot motion constraints; the inverse kinematics solver is used to generate joint space instructions, and an obstacle avoidance detector is run at each waypoint. Once a waypoint is found to approach or cross the obstacle volume boundary, path replanning or local avoidance is performed.

[0062] In extreme cases, it supports setting fault-tolerant rules or marking areas as manual re-spraying areas to avoid the risk of strange postures or collisions caused by forced path adjustments; the path control instructions output by the final module and the control instructions output by the feedback control decision module jointly drive the motion path execution unit to form a multi-channel fusion control chain of spatial path and process control, thereby improving the system's control ability over the consistency of surface coatings on complex workpieces.

[0063] The parameter optimization and adaptive adjustment module provides the system with continuous learning and self-improvement capabilities. It is responsible for intelligently adjusting control parameters and process setpoints, dynamically coordinating them, and accumulating knowledge, ensuring system stability and optimal coating performance under variable operating conditions. This module forms a loosely coupled closed loop with the control decision module and the actuator control module through parameter vector and state feedback, and features independent calculation, periodic updates, and event triggering mechanisms.

[0064] The controller parameter self-tuning unit takes the LADRC control structure in the feedback control decision module as the target, constructs a multi-performance index joint objective function around key parameters (i.e., observer gain and control gain), and uses an improved hybrid particle swarm optimization algorithm (M-PSO) for global search and local adaptive fine-tuning.

[0065] The input is performance feedback, namely the current control error signal, coating error convergence speed, system response waveform and disturbance recovery time; the output is the optimal control parameter set, which is written into the LADRC execution structure through the controller parameter update interface; a "simulation + short-time measurement" hybrid evaluation mode is used to verify particle performance to avoid pure model errors; chaotic perturbations and adaptive inertia weight adjustment mechanisms are introduced in the optimization iteration to improve the convergence rate and suppress premature traps; the self-tuning process can be executed by events (such as powder type switching), cycles (such as before the first spraying of the day) or trend triggers (such as a significant increase in control error); the results are synchronously written into the parameter update and knowledge accumulation units for long-term experience backtracking and rapid migration.

[0066] The process multi-parameter collaborative optimization unit is responsible for constructing a multi-objective optimal control strategy for the spray gun voltage setting value, powder supply rate, spray gun movement speed and air pressure control value, and outputting the process setting parameter vector, as follows: The inputs are the current particle deposition distribution error, powder consumption rate per unit area, and total spraying time; the optimization objective function is constructed with the square of the coating uniformity deviation, powder utilization rate, and spraying efficiency indicators; an improved multi-objective particle swarm optimization algorithm (MOPSO) is used, and it can be switched to alternative algorithms such as CPO to improve the global optimization capability; the output is a set of parameters in the optimal Pareto frontier solution that are close to the expected weight combination, which are pushed to the actuator control module after smoothing to adjust the system behavior in real time; an optional "online fine-tuning" mode is available: when the monitoring indicators deviate continuously, the optimal neighborhood is maintained through local updates within a limited calculation step; in scenarios with high humidity, low temperature, or complex special-shaped parts, it exhibits adaptive capabilities that are superior to static strategies, such as automatically increasing the voltage to hedge against decreased charging efficiency.

[0067] The parameter update and knowledge accumulation unit, as the central coordinator, is responsible for distributing the optimization results to each target module in an orderly manner, and recording and managing the parameter evolution trajectory in the background; the control parameter set and process parameters are smoothly injected into the controller and execution module through linear interpolation buffering to prevent jump disturbances; the results of each optimization iteration are archived together with the corresponding working condition labels (such as powder model, temperature and humidity, and workpiece number) to form a working condition-parameter mapping table; when the similarity between the current working condition and the historical optimization conditions is higher than the threshold, the corresponding historical optimal parameter solution can be called without re-iteration; long-term control and process indicator trends are recorded to provide a model correction basis for system evaluation; engineers are allowed to adjust the optimization target weight, set the iteration step size, and enable specific parameter versions to support the integration of human experience and algorithmic decision-making.

[0068] It should be noted that this module takes "controller parameter vector" and "process setting parameter vector" as the output core, and transmits them synchronously as input to the feedback control decision module and the actuator control module to build a cross-module, multi-objective, and multi-time domain dynamic optimization closed loop, providing adaptive enhancement capabilities and long-term evolution support for the entire spraying system.

[0069] The multi-objective collaborative adjustment strategy module is used to comprehensively balance various control objectives during system operation, adjust control and execution parameters in real time, and achieve dynamic coordination optimization between quality indicators and production efficiency.

[0070] The performance monitoring and evaluation unit receives the performance indicator data stream from the data acquisition and monitoring module, combines it with the current spraying process goals, and quantifies the degree of satisfaction of each control goal in real time: The input data includes deposition rate, coating thickness standard deviation, and material utilization rate; the data trend within the objective function window is used to filter out the impact of short-term fluctuations, and the current target satisfaction vector is output to indicate whether each performance target has been met; the indicator judgment logic is based on the set threshold interval and determines whether the trend is continuously deteriorating; the output result is used as the criterion for weight adjustment and parameter redistribution, and is passed to the target weight decision unit.

[0071] The target weight decision unit dynamically adjusts the weight configuration of multi-target control based on performance evaluation results and current process priorities, driving the system strategy to shift to the optimal preference direction: The input is the performance satisfaction vector and the historical optimal parameter candidate set; a fuzzy rule engine or experience-driven mapping function is used to output the current weight vector. When it is found that the uniformity index has degraded and is close to the quality threshold, its weight is appropriately increased and the efficiency weight is reduced to prioritize coating consistency; at the same time, the required parameter adjustment strategy set is decided to clarify the adjustment direction and amplitude of each parameter; the optimization adjustment direction can be matched with the Pareto parameter set output by the parameter optimization module to select the reference solution that best meets the current weight configuration; this unit forms a complete adjustment strategy package, submits it to the coordination execution unit for implementation, and enters the strategy execution closed loop.

[0072] The coordination execution unit is responsible for implementing parameter update tasks across modules and continuing to monitor system performance after adjustments, forming a real-time control closed loop: it receives parameter adjustment instruction sets and weight adjustment plans and sends them to the corresponding modules; For the feedback control decision module, dynamically modify the LADRC control target or feedback gain; For the actuator control module, write the setting values ​​of spray gun voltage, powder supply rate, moving speed, etc. For the parameter optimization module, update the multi-objective optimization weight vector to guide its new round of output; To avoid system disturbances, all instruction executions use a trapezoidal buffer injection method (with set transition time and ramp rate); After implementation, performance monitoring results should be continuously monitored. If the target indicators improve as expected, the strategy should be maintained. Otherwise, further fine-tuning should be done or a strategy rollback mechanism should be implemented.

[0073] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A real-time monitoring and feedback control system for enamel electrostatic powder coating, characterized in that: Includes the following modules: The data acquisition and monitoring module includes a spray gun parameter sensing unit, a particle deposition monitoring unit, and an environmental state perception unit, which respectively collect spray state parameters, deposition error signals, and environmental interference parameters; the signal processing and anomaly detection unit generates a standardized state vector and a weighted error signal vector; The feedback control decision module consists of an error calculation and reference signal reconstruction unit, an amplitude and phase compensation unit, a control algorithm unit, and an output signal adjustment unit. It receives and processes the standardized state vector and the multi-dimensional error signal in sequence to obtain the final execution control signal. The actuator control module splits and maps the final execution control signal to different control objects, driving the high-voltage electrostatic output control unit to adjust the spray gun electrode voltage, the powder supply and air pressure control unit to adjust the powder flow and conveying gas pressure, and the motion path execution unit to control the spray gun motion posture and trajectory speed; The spray path planning and adjustment module generates the initial posture control path of the spray gun based on the three-dimensional morphology of the workpiece and the desired deposition specifications, and adjusts the path parameters in real time based on the coating deposition feedback signal during the spraying process; The parameter optimization and adaptive adjustment module is responsible for intelligent adjustment, dynamic coordination and knowledge accumulation of control parameters and process set points, maintaining stability and optimal coating performance under variable working conditions; The multi-objective collaborative adjustment strategy module is used to comprehensively balance various control objectives during operation and adjust control parameters and execution parameters in real time.

2. The real-time monitoring and feedback control system for enamel electrostatic powder coating according to claim 1 is characterized in that: The spray gun parameter sensing unit collects spray state parameters through high-voltage voltage sensors, corona discharge current sensors, and atomization and delivery air pressure sensors. It analyzes the deviation between the spray state parameters and the set target parameters to determine whether the current spray gun is in a stable working state, and further determine whether the working point of the spray equipment is normal; The particle deposition monitoring unit continuously obtains coating thickness using a laser thickness gauge. It also uses an industrial camera to capture spray coverage images and extract uniformity features from the spray images. A deposition rate estimation model based on current fluctuations is introduced to indirectly infer the change in powder load per unit time, i.e., the deposition rate per unit time. The collected coating thickness, uniformity features, and deposition rate per unit time are compared with the target values ​​to generate a deposition error signal. The environmental state sensing unit is used to detect interference factors in the spraying environment. It uses a digital humidity sensor to obtain relative humidity, a thermoelectric sensor to obtain ambient temperature, and an electric field disturbance sensor to obtain background interference voltage from non-spray gun sources. The three together constitute the environmental interference parameters; The signal processing and anomaly detection unit organizes the spraying state parameters, deposition error signals and environmental interference parameters into a spraying state parameter vector, a deposition effect parameter vector and an environmental interference parameter vector, respectively, and uniformly forms an original state data set. The input data set is fused based on a time window, denoised by a sliding average filter, and normalized into a standardized state vector based on a Z-score standardization method. For the standardized state vector, three anomaly recognition strategies are further designed: all anomaly results identified by the three anomaly recognition strategies are uniformly encoded into a multidimensional error signal vector; a parameter priority matrix is ​​constructed, and the multidimensional error signal vector is weighted. After the matrix acts on the multidimensional error signal vector, a weighted error signal vector is output.

3. The real-time monitoring and feedback control system for enamel electrostatic powder coating according to claim 1, characterized in that: The error calculation and reference signal reconstruction unit constructs a smoother and more dynamically adaptable control reference signal based on the standardized state vector and the weighted error signal vector, and calculates the instantaneous error signal vector accordingly. It introduces the historical state sequence on the sliding time window and uses a weighted quadratic spline interpolation function to generate the current reference state trajectory, that is, the current reference state vector. After obtaining the current reference state vector, it compares it with the real-time state vector, calculates the instantaneous error signal vector, and performs weighted mapping processing to construct an adjustment signal that can be used to stimulate the control strategy model, and finally outputs an adaptive reference state trajectory and a weighted control error signal. The amplitude-phase compensation unit is used to correct the system amplitude-frequency and phase-frequency distortion before the weighted control error signal enters the controller for calculation. By designing a compensation filter in the frequency domain, the gain attenuation and phase lag generated during the transmission process are reversely corrected, making the entire control channel equivalent to an ideal linear system. Through frequency domain experimental identification in the trial operation phase, the frequency response model of the spray control path is obtained, which is recorded as the transfer function , the function can be decomposed into the amplitude-frequency response Phase frequency response ; and construct a compensation filter based on this Acting on the error signal; The control algorithm unit is responsible for receiving the error signal processed by the amplitude and phase compensation unit and generating the main control output signal based on the current reference state vector. It uses the linear active disturbance rejection control strategy to construct the control main line, introduces the limit function and control rate constraint to limit the control quantity from changing suddenly or exceeding the physical boundary, and finally outputs the main control quantity. The output signal conditioning unit is responsible for receiving the output signal of the control master and applying physical limit constraints to the ideal control instructions generated by the controller. Based on the specific dynamic response characteristics of different types of actuators, corresponding compensation corrections are performed on the physical limit constraints to obtain the final execution control signal after dynamic matching.

4. The real-time monitoring and feedback control system for enamel electrostatic powder coating according to claim 1 is characterized in that: High-voltage electrostatic output control unit, used to analyze the control signal components for the electrostatic parameters in the feedback control module , and map it to the high voltage output setting value of the spray gun electrode in real time , adjust the charge state and migration driving force of powder particles, and thus control the deposition behavior of powder on the workpiece surface; The powder supply and air pressure control unit is used to analyze the powder supply and air drive parts in the output signal of the feedback control decision module, that is, and , control signal After adjustment, it is used as the powder feeder execution command to control the driving frequency The frequency determines the amount of powder delivered per unit time, and thus adjusts the actual powder flow rate at the spray gun nozzle. ; Control signal Analyzed as delivery pressure and atomizing air pressure The regulating instructions act on the electronically controlled proportional pressure regulating valve to control the gas flow rate and pressure, thereby affecting the propagation range and atomization quality of the powder jet. By building a powder supply-air pressure coordinated control mechanism, the kinetic energy matching of the powder particles and the gas is adjusted, and the distribution state and deposition coverage of the powder particles during the flight are controlled. Finally, the powder supply and air pressure control unit outputs the adjusted powder supply drive signal and air pressure setting signal. Motion path execution unit, used to analyze the spatial path control instruction part in the output signal of the feedback control decision module , and convert it into real-time trajectory scheduling instructions for the mechanical actuator in the spraying equipment, driving the spray gun to run along the desired trajectory on the workpiece surface, completing the coating control in the spatial dimension.

5. The real-time monitoring and feedback control system for enamel electrostatic powder coating according to claim 1 is characterized in that: The offline path generation unit constructs the initial spray path point set based on the 3D CAD model of the workpiece, the desired thickness mapping matrix, and the spray width characteristic parameters, and outputs a path control vector sequence. The path planning uses a geometric algorithm based on voxel partitioning and surface scanning, and integrates process rule constraints to generate a path vector sequence. The online trajectory adjustment unit receives the deposition thickness spatial distribution error signal output by the particle deposition monitoring unit during the spraying process, combines it with the offline path point set, identifies the deviation area in real time, executes the fine-tuning strategy, and generates the correction path instruction; The multi-axis coordination and obstacle avoidance unit is used to coordinate the multi-axis joints of the spray actuator during the path planning and adjustment process, and detect physical conflicts or constraint conflicts that may arise during the path adjustment process; it uses the inverse kinematics solver to generate joint space instructions and runs an obstacle avoidance detector at each path point. Once it is found that the path point approaches or crosses the obstacle volume boundary, it executes path replanning or local avoidance; in extreme cases, it supports setting fault tolerance rules or marking as manual re-spraying area to avoid posture abnormalities or collision risks caused by forced path adjustment.

6. The real-time monitoring and feedback control system for enamel electrostatic powder coating according to claim 1, characterized in that: The controller parameter self-tuning unit takes the LADRC control structure in the feedback control decision module as the target, constructs a multi-performance index joint objective function around key parameters, and uses an improved hybrid particle swarm optimization algorithm for global search and local adaptive fine-tuning; The process multi-parameter collaborative optimization unit is responsible for building a multi-objective optimal control strategy for the spray gun voltage setting value, powder supply rate, spray gun movement speed and air pressure control value, and outputting the process setting parameter vector; The parameter update and knowledge accumulation unit is responsible for distributing the optimization results to each target module in an orderly manner and recording and managing the parameter evolution trajectory in the background. It smoothly injects the control parameter set and process parameters into the controller and execution module through linear interpolation buffering. The results of each optimization iteration are archived together with the corresponding working condition labels to form a working condition-parameter mapping table.

7. The electrostatic powder coating real-time monitoring and feedback control system according to claim 1, characterized in that: The performance monitoring and evaluation unit receives the performance indicator data stream from the data acquisition and monitoring module, combines it with the current spraying process goals, and quantifies the degree of satisfaction of each control goal in real time; Through the data trend within the objective function window, the influence of short-term fluctuations is filtered out and the current target satisfaction vector is output; The indicator judgment logic is based on setting the threshold interval and judging whether the trend is continuously deteriorating; The output results are used as the criteria for weight adjustment and parameter redistribution; The target weight decision unit dynamically adjusts the weight configuration of multi-target control based on performance evaluation results and current process priorities, driving the system strategy to shift to the optimal preference direction; The coordination execution unit is responsible for implementing parameter update tasks across modules and continuing to monitor system performance after adjustments, forming a real-time control closed loop: it receives parameter adjustment instruction sets and weight adjustment plans, and sends them to the corresponding modules.

Images