Supercritical CO2 energy-saving spraying process control system for ship parts

Abstract

The invention relates to the technical field of spraying process control, and particularly discloses a supercritical CO2 energy-saving spraying process control system for ship parts, which is provided with an equipment executable judgment module, a CO2 chain control stabilization module, a paint chain control stabilization module and a spray gun spraying task execution module. The method comprises the following steps: firstly, enabling supercritical CO2 spraying equipment to enter a task connection period, and judging an executable state in the period; then a CO2 storage tank, a purifier, a supercharger and a buffer tank link are started, an admission contract is output, and meanwhile, an allowable setting interval is limited for the downstream in a contract form; then, feedforward is executed based on the admission contract, and a coating side steady state is formed; and finally, the supercritical CO2 is preferentially released by the coordinator to enter the jet mixer to form a main jet flow, then the coating is released by a small-flow step to be shaped and mixed and then is sprayed, and the coupling kinematics realizes the minimization of energy consumption and solvent consumption and the constraint of tail end concentration on the premise of meeting the appearance / adhesive force.

Description

Technical Field

[0001] This invention relates to the field of spraying process control technology, specifically to a supercritical CO2 energy-saving spraying process control system for ship components. Background Technology

[0002] Supercritical CO2 energy-saving spraying technology refers to using carbon dioxide at a temperature above its critical point (approximately 31.1°C and 7.38 MPa) as an adjustable solvent / propellant / atomizing agent. This carbon dioxide is mixed with the coating under high pressure and then sprayed out under reduced pressure. Utilizing the characteristics of supercritical fluids—high diffusion, low viscosity, and certain solubility—the apparent viscosity of the system is reduced, droplets are refined, and deposition efficiency is improved, thus achieving film formation at lower pre-baking temperatures and shorter drying times. This process significantly reduces the amount of traditional thinners used and VOC emissions. Combined with CO2 recovery and circulation and dew point temperature-linked control, it can reduce energy consumption per unit area while ensuring film thickness and appearance quality. It is suitable for spraying applications with high energy consumption and environmental protection requirements, such as metal components, ship / marine engineering, and 3C product casings.

[0003] Existing spraying processes typically include: degreasing, rust removal, grinding, and dust removal of the workpiece, and surface pretreatment such as phosphating / priming as needed; adjusting the proportion, viscosity, and solids content of solvent-based or water-based coatings according to the formula, filtering, and then pumping them to the spray gun by a metering pump; atomizing using methods such as air spraying, airless / airless / windless spraying, or electrostatic spraying, and applying the coating in stages with appropriate spray distance, gun speed, and overlap (multi-layer systems of primer, intermediate, and topcoat are required); then entering a flash evaporation zone for initial evaporation and leveling of solvents or water, followed by curing under conditions such as hot air / infrared / UV according to the resin system; after exiting the oven, inspecting the film thickness, appearance, adhesion, and hardness, and reworking any areas that fail to meet the standards; and finally, supporting solvent recovery and VOCs treatment (activated carbon / condensation / catalytic combustion, etc.), as well as overspray recovery and line cleaning, forming a closed-loop process from pretreatment to curing and environmental protection.

[0004] For example, Chinese invention patent application CN115373341A discloses a method, device and system for adjusting surface spraying process parameters. It can receive product characteristics, ambient temperature, interruption time and defect information in real time, respond to the above information in sequence according to the set response priority, select spraying process parameters, and automatically adjust different process parameters for different products and different working conditions, so as to realize the simultaneous control of multiple spraying equipment.

[0005] For example, Chinese invention patent CN118466417B discloses an adaptive adjustment method and system for spraying process based on temperature measurement, belonging to the field of spraying control. The method includes: reading product data of the part to be sprayed and generating a preset spraying path and a preset spraying response scheme; acquiring spraying speed data, configuring the monitoring anchor frame of the part to be sprayed, establishing a set of positional temperatures within the monitoring anchor frame, performing extreme value-free temperature evaluation, and generating evaluation results; if the result is not satisfactory, performing scheme optimization search and establishing an adaptive adjustment scheme; and completing the spraying within the monitoring anchor frame of the part to be sprayed through the adaptive adjustment scheme, thus completing the spraying operation of the part to be sprayed.

[0006] Based on the above technical solutions, it was found that existing spraying process control schemes generally adopt parameter regulation driven by narrowband feedback sets: even if multiple types of measurement values ​​are collected, they are mostly operated in a single closed loop or sequential cascade manner with several main indicators, lacking cross-level mutual verification, such as real-time consistency verification between multiple physical quantities such as temperature / wet film thickness / pressure / flow rate / charge, thus exposing the structural deficiencies of the device control layer under scene migration and operating condition fluctuations.

[0007] Specifically, the spraying site presents strong disturbances and complex geometry: kinematics exist, such as the spray gun's speed, acceleration, attitude, and trajectory curvature, which are not linked to the medium supply, such as paint volumetric flow rate, high-pressure CO2 mass flow rate, back pressure, and mixing section pressure difference. This causes a mismatch between wet film thickness (undried film thickness) and droplet spectrum due to velocity / attitude drift, resulting in decreased film thickness uniformity and edge coverage. This negative impact is particularly pronounced in CO2 spraying. Furthermore, the implementation of closed-loop optimization of energy efficiency KPIs at the equipment level is insufficient, failing to minimize energy and solvent consumption and emission concentration while meeting appearance and adhesion requirements, leading to increased energy consumption and VOCs treatment costs. Ultimately, this amplifies quality fluctuations and inherent safety risks. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a supercritical CO2 energy-saving spraying process control system for ship components, which can effectively solve the problems mentioned in the background technology.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a supercritical CO2 energy-saving spraying process control system for ship components, comprising: an executable determination module for the supercritical CO2 spraying equipment group and its associated observers entering a task transition period, during which the supercritical CO2 spraying equipment group and its associated observers complete the executable status determination; and a CO2 chain control stabilization module for, upon receiving an executable status signal, the CO2 chain controller formally starts the CO2 storage tank to transport internal CO2 through a purifier into a CO2 booster compressor, where it is pressurized into supercritical CO2. Critical CO2 enters the CO2 buffer tank, where it dissolves to form a stable CO2 chain. The CO2 chain controller sends an access contract. The paint chain control stabilization module is used by the paint chain controller to receive the access contract and execute control feedforward. The paint in the paint tank is metered by the paint pressurization metering pump and sent into the paint buffer tank, where it dissolves to form a stable paint chain. The spray gun spraying task execution module is used by the coordinator to prioritize the release of supercritical CO2 into the jet mixer based on the stable CO2 chain. Subsequently, the paint is released into the jet mixer for shaping and mixing, and then enters the spray gun to execute the current round of spraying task.

[0010] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) This invention provides a supercritical CO2 energy-saving spraying process control system for ship components. First, the equipment's executable determination module enables the supercritical CO2 spraying equipment group and its associated observers to enter the task connection cycle. During this cycle, executable status determinations such as sensor health, time synchronization, zero point, and dynamic fingerprint are completed. False signals / bad channels are eliminated before entering the process closed loop to avoid false releases caused by single-point drift. The entire link clock and unit caliber are unified to reduce the difficulty of subsequent control coupling. After receiving the executable signal, the CO2 chain control stabilization module starts the CO2 storage tank, purifier, booster, and buffer tank link to establish a steady state of back pressure, mass flow rate, and temperature and dampen vibrations. It outputs an access contract to first pin the boundary between atomization and phase behavior, reducing... The system minimizes whitening / pinhole triggering windows and simultaneously limits the downstream allowable setting range through a contract. Subsequently, the coating chain control stabilization module executes feedforward based on the access contract, driving the coating pressurization metering pump to deliver the metered coating into the coating buffer tank and eliminate pulsations, forming a coating-side steady state. This stabilizes the coating volume flow rate and outlet pressure in the return flow path, ensuring film thickness consistency at the moment of entry and preventing end-point gelling. Finally, the spray gun spraying task execution module prioritizes the release of supercritical CO2 into the jet mixer to form the main jet, followed by the release of coating in small flow steps for shaping and mixing before spraying. The CO2-first / coating-later threshold release reduces the synthesis disturbance at the mixing chamber inlet. Coupled kinematics minimizes energy and solvent consumption and constrains end-point concentration while satisfying appearance / adhesion requirements.

[0011] (2) This invention makes an admission-based judgment on controllable state indicators such as back pressure stability, mass flow fluctuation, temperature boundary, time alignment error, local pressure gradient baseline of mixing chamber, and nozzle channel speed balance during the task connection cycle. The spraying system quantifies whether it is controllable into an executable / unexecutable conclusion and confidence level. Compared with the traditional method of adjusting immediately upon power-on, it can detect hidden dangers such as slow drift, frosting, cavitation and cycle overlap before spraying, significantly reducing the probability of mis-spraying and rework, and solidifying the debugging time into a predictable threshold time window.

[0012] (3) In this invention, the back pressure main value, main frequency and phase, and temperature boundary of the CO2 chain output are reused in the coating chain as feedforward references for the anti-phase micro-throttling and recirculation proportional valves; the transient flow deviation formed by the coating chain and the stability of the buffer tank outlet pressure are recirculated back to the target alignment at the mixing chamber inlet; the dynamic fingerprint / baseline previously generated in the equipment judgment stage runs through the health weighting and anomaly backoff in the spraying and cruising stages. This kind of cross-module parameter reuse reduces the number of independent calibrations and redundant measurement points, shortens the transition time from cold start to stable spraying, and achieves seamless degradation in case of failure.

[0013] (4) Traditional solutions mainly rely on single-chain stabilization to direct spraying or single-point measurement closed loop, which is difficult to handle the pulsation superposition and inlet coupling under phase change conditions. This solution solidifies the physical sequence of CO2 stabilization followed by mixing into the control process through access contract and threshold release. It also replaces one-time switching with dual-chain parallel stabilization, common inlet target, and small flow step entry, which significantly suppresses the synthetic disturbance at the mixing chamber inlet. At the same time, it introduces multi-source fusion and health weighting to replace single-channel signals, making the closed loop insensitive to single-point drift. On the energy efficiency / environmental protection side, the coupling control driven by unified state quantity incorporates unit area energy consumption, carbon dioxide recovery rate and end concentration constraints into the device-level closed loop. Compared with the existing technology that only has quality closed loop and energy efficiency offline process, this solution achieves simultaneous convergence of the three objectives of quality, energy efficiency and emissions. In summary, this solution has quantifiable and reproducible system-level improvements over the existing technology in terms of controllability, stability, defect suppression and comprehensive energy efficiency / environmental protection indicators. Attached Figure Description

[0014] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the system module connections of the present invention.

[0016] Figure 2 This is a curve showing the volumetric flow rate of the coating entering the cavity.

[0017] Figure 3This is an industrial-scale equipment for instant mixing and spraying of supercritical CO2.

[0018] Figure 4 A schematic flowchart illustrating the control method for supercritical CO2 energy-saving spraying process.

[0019] Attached reference numerals: 1. CO2 storage tank; 2. Purifier; 3. CO2 booster; 4. High-pressure CO2 buffer tank; 5. Jet mixer; 6. Air compressor; 7. Paint tank; 8. Paint pressurizing and metering pump; 9. Paint buffer tank; 10. Hardener storage tank; 11. Hardener pressurizing and metering pump; 12. Spray gun; 13. Control cabinet. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] In this embodiment of the invention, the CO2 spraying equipment group belonging to the supercritical CO2 energy-saving spraying process is specifically as follows: Figure 3 As shown, Figure 3 This is an industrial-scale equipment for instant mixing and spraying supercritical CO2 spraying, in which... Figure 3 The equipment can use the surface response method to control the supercritical CO2 spraying process for epoxy primers and polyurethane topcoats used in the coating of ships and marine engineering equipment under different compositions, pressures, temperatures, spray gun sizes and spraying distances. By characterizing the performance of the coating under different spraying conditions, the optimal process parameters for supercritical CO2 spraying of various ship and marine engineering equipment coatings can be optimized.

[0023] See Figure 1 As shown, this embodiment of the invention provides a technical solution: a supercritical CO2 energy-saving spraying process control system for ship components, comprising an equipment executable determination module, a CO2 chain control stabilization module, a paint chain control stabilization module, a spray gun spraying task execution module, and a spraying control information database. The spraying control information database stores and extracts preset values ​​for delay definition duration, reference dwell time, and various parameters.

[0024] The process control process in this embodiment is based on Figure 4The schematic flow of the supercritical CO2 energy-saving spraying process control method starts from task connection: First, the observer is executed. If it fails, it is downgraded / repaired and re-inspected. If it passes, the execution status of the spraying equipment is determined. If there is an anomaly, an alarm is triggered and a conservative configuration is loaded. After the equipment passes, it enters the CO2 chain steady-state, with back pressure limiting the speed of pressurization and suppressing vibration. Then, an access contract containing the entry target and entry time slot is generated. According to the contract, the coating chain is steady-state and an entry ramp is given. Then, the entry coordination with the CO2 chain is completed (scheduled time point + predicted sequence). After confirmation, spraying starts with CO2 first, then coating, and small steps. During the spraying process, kinematic and media coupling and quality permission monitoring are performed. If the indicators meet the standards, the system continues to cruise. If the standards do not meet the standards, the system adaptively corrects itself according to the rules (mixing ratio, pre-baking temperature, spray gun spacing, etc.) and returns to the monitoring loop. At the end, the spray stop cycle is executed (paint cutting, CO2 pressurized purging, and slow back pressure reduction), and the data is archived and the next round is preset (termination zero point value, dynamic fingerprint, and next version of the contract), forming a closed-loop and reproducible control process.

[0025] It should be explained that the preset relationships stored in the above-mentioned spraying control information database include, but are not limited to, preset, matching, mapping, and association relationships. In this embodiment, the specific method of obtaining preset relationships is to take the termination zero point value and the observer data weighting factor as an example. First, based on the original and fused observations (including working condition labels, set values, and quality results) from the previous round and multiple rounds of historical operation, data cleaning, time alignment, and unit / caliber unification are completed. Robust statistics and health assessment (drift, noise, similarity variance, harmonic characteristics) are performed on key channels. Then, candidate relationships are formed by organizing them in three dimensions: object, working condition, and stage. Subsequently, a mapping is established by combining mechanistic constraints with statistical learning (such as piecewise linear / table lookup, kernel regression, or rule tree). The usability area and confidence are verified through cross-validation and playback simulation. Finally, the data is solidified into the database in a versioned manner, and the applicable boundaries and rollback strategies are recorded. The termination zero-point value is obtained by using the robust median of the zero-load, zero-flow, and isobaric / reference temperature windows after shutdown (supplemented by Hampshire / quantile de-extremes), and is associated with the equipment / environment fingerprint to generate a zero-point comparison with the operating conditions; the observer weighting factor is derived from the deviation between the current round and the historical baseline in terms of similarity variance, dominant frequency / harmonic consistency, drift, etc., and is mapped to a 0–1 confidence weight by the health model, and matched / associated with the channel redundancy consistency rule for automatic weighting in subsequent fusion and admission determination.

[0026] The equipment executable determination module is connected to the CO2 chain control and stabilization module, which in turn is connected to the paint chain control and stabilization module. The paint chain control and stabilization module is also connected to the spray gun spraying task execution module. All of these modules are connected to the spraying control information database.

[0027] The executable determination module is used when the supercritical CO2 spraying equipment group and its associated observers enter the mission transition period. During the mission transition period, the supercritical CO2 spraying equipment group and its associated observers complete the executable status determination.

[0028] Specifically, during the mission transition period, the supercritical CO2 spraying equipment group and its associated observers completed the executable status determination. The specific analysis process is as follows: The executable status determination includes the executable status determination of the observer and the executable status determination of the CO2 spraying equipment group.

[0029] In this embodiment of the invention, the executable status of the observer is first determined, and then the executable status of the CO2 spraying equipment group is determined based on the observer data: at the beginning of the task connection cycle, the health and consistency of the observer are checked first. Only when the observer output meets the executable requirements and publishes a stable and confident unified process status quantity is the equipment-side determination performed. Subsequently, based on the reliable observation data, the CO2 storage tank, purifier, booster, buffer tank and related valve group / interlock are subjected to equipment-level executable verification (back pressure and mass flow stability, temperature boundary, leakage and interlock integrity, actuator reachability and response fingerprint matching, etc.). Finally, the conclusion of whether the equipment is executable or not is given and serves as the sole basis for subsequent formal start-up and access contract.

[0030] The observer refers to the measurement and state sensing unit installed on the supercritical CO2 spraying equipment assembly and its pipelines / spray guns / environment, used to generate unified process state quantities that can be used for control and access determination. These are described as physical sensors (e.g., pressure, mass flow rate, volumetric flow rate, temperature, nozzle channel micro-pressure / flow rate, vibration / acoustics, wet film thickness, droplet size, visual camera / spectrum, VOC / CO2 concentration, dew point, and ambient wind speed).

[0031] The executable state determination of the observer is carried out through the following process: Extract the historical observation data sequence corresponding to each observer of each CO2 spraying equipment in the CO2 spraying equipment group to obtain the historical observation data sequence of each observer of the CO2 spraying equipment. The historical observation data sequence can be extracted from the historical observation record of the observer.

[0032] During the task transition period, the system extracts historical observation data sequences from the multi-channel observers corresponding to each piece of equipment in the CO2 spraying equipment group (historical refers to the data stream archived under the same operating window in the previous spraying operation cycle), such as: back pressure sensors A / B of the booster station, dual pressure gauges at the outlet of the buffer tank, mass flow meter and soft estimation channel, temperature probe and micro-pressure orifice array, etc. Each piece of equipment is equipped with more than one observer, thus forming multiple sets of independent data for mutual verification: on the one hand, consistency / drift checks are performed (such as dual gauge comparison, hardware and soft estimation mutual verification), and on the other hand, similarity variance is determined (comparing different historical sequences from the previous round in the same operating segment) to confirm that the measurement noise level and dynamic characteristics have not deviated abnormally. Only when the consistency and similarity variance between these multi-observers and multi-sequences all pass the threshold are the observation data of the current round marked as reliable and used for subsequent equipment executable state determination and access decision.

[0033] Based on historical observation data sequences, the termination zero-point values ​​of each observer in the CO2 spraying equipment are extracted and their zero-point values ​​are checked. If zero, data similarity variance is determined; otherwise, a data weighting factor is matched based on the termination zero-point values ​​of each observer in the CO2 spraying equipment. This is represented as a confidence weight (a coefficient between 0 and 1, with smaller values ​​indicating lower confidence) assigned to each observer. It is used in data fusion, admission determination, and closed-loop control to reduce the impact of abnormal channels and make the spraying control system more stable.

[0034] The termination zero-point value refers to the equipment baseline reading recorded by the observer at the end of the previous cycle, after the device has completed unloading and recirculation according to the shutdown procedure, under zero load, zero flow, isobaric, or specified reference conditions. It is used as the zero-point reference for the next cycle of task transition. It reflects the actual zero bias of the sensor chain at the end of the last cycle and can be used for zero-point compensation and health assessment in the current cycle: if the current cold start static reading is significantly deviated from zero after subtracting this value, it indicates anomalies such as drift, residual pressure, or cavitation; if the termination zero-point value itself does not meet the reference conditions (e.g., incomplete venting or temperature not returning to the baseline), it is marked as low confidence, does not participate in automatic zero-point application, and is only retained as a diagnostic record.

[0035] The data similarity variance determination process is as follows: The historical observation data sequences of each observer of the CO2 spraying equipment are sequentially processed to calculate data sequence similarity, specifically based on cosine similarity calculation. The variance of each observer's historical sequence similarity is then processed to obtain the data similarity variance of the CO2 spraying equipment observers. This variance is compared with a predefined data similarity variance threshold. If the data similarity variance of the CO2 spraying equipment observer is less than the threshold, the observer's executable status is determined to be passed. This process is repeated for the entire CO2 spraying equipment group; otherwise, it is determined to be failed, and an early warning is issued for this round of spraying tasks. The alarm status is displayed in a pop-up window on the spraying system management interface.

[0036] When the data similarity variance is less than the data similarity variance threshold, it indicates that the historical observation noise level is consistent with the dynamic response characteristics, and the measurement chain has not experienced drift, inaccuracy, or amplified abnormal pulsations. This ensures the reliability of the unified process state quantity from the source and reduces the risk of single-point anomalies causing closed loops. Compared with existing judgment methods that only look at instantaneous values ​​or simple limit exceedances, introducing a similarity variance threshold can significantly reduce false alarms / false alarms, shorten the adjustment and verification time in the task connection cycle, and focus alarm triggering on statistically significant real anomalies (such as spectral width expansion caused by cavitation and frosting), thereby improving the consistency and reproducibility of spraying startup and indirectly reducing quality defects and energy waste in the early stages of spraying.

[0037] Furthermore, the specific analysis process for determining the executable status of the CO2 spraying equipment group is as follows: CO2 spraying equipment, including a CO2 booster, a CO2 buffer tank, a paint pressurizing and metering pump, a paint buffer tank, a jet mixer, and a spray gun.

[0038] When the executable status of the observer is determined to be passed, the historical observation data sequence of each observer of the CO2 spraying equipment is fused. If a certain observer has an observer data weighting factor, the data fusion process is configured based on the observer data weighting factor to obtain and record the historical execution dataset of the CO2 spraying equipment.

[0039] Based on the historical execution dataset of the CO2 spraying equipment, the controllable status indicators of the CO2 spraying equipment are evaluated and compared with the predefined controllable status threshold indicators. If all the controllable status indicators of the CO2 spraying equipment are greater than or equal to the controllable status threshold indicators, the executable status of the CO2 spraying equipment group is judged as passed; otherwise, it is judged as failed, and an early warning is issued for this round of spraying tasks.

[0040] Specifically, the controllable state indicators of CO2 spraying equipment are evaluated, and the specific analysis process is as follows: The historical execution dataset of the CO2 spraying equipment includes the historical delay duration of the CO2 spraying equipment, the historical time-series CO2 back pressure of the CO2 spraying equipment, the historical time-series pipeline pressure difference of the CO2 spraying equipment, and the historical residence time of the CO2 spraying equipment, which is expressed as the historical storage time of CO2 in the equipment; the historical execution dataset can be extracted from the historical execution records of the CO2 spraying equipment.

[0041] The historical time-series CO2 back pressure and historical time-series pipeline pressure difference of the CO2 spraying equipment were processed by standard deviation to obtain the historical CO2 back pressure fluctuation coefficient and the historical pipeline pressure difference fluctuation coefficient of the CO2 spraying equipment.

[0042] Extract the delay definition duration and baseline dwell time from the spraying control information database.

[0043] The historical CO2 back pressure fluctuation coefficient and the historical pipeline pressure difference fluctuation coefficient of the CO2 spraying equipment were preprocessed separately. The data preprocessing included normalization and de-normalization to obtain the data preprocessing results.

[0044] The historical delay time of the CO2 spraying equipment is compared with the delay definition time to obtain the ratio processing result.

[0045] The historical dwell time of the CO2 spraying equipment was compared with the baseline dwell time to obtain the deviation processing result.

[0046] By introducing an influencing factor, the data preprocessing results, ratio processing results, and deviation processing results are correlated and merged to obtain the controllable state index of the CO2 spraying equipment.

[0047] The specific analysis process is as follows: In the formula, Ct is the controllable state index of the CO2 spraying equipment, Dd is the historical delay time of the CO2 spraying equipment, Dd' is the delay definition time, Bp is the historical CO2 back pressure fluctuation coefficient of the CO2 spraying equipment, Pp is the historical pipeline pressure difference fluctuation coefficient of the CO2 spraying equipment, Ds is the historical residence time of the CO2 spraying equipment, Ds' is the baseline residence time, y1 is the influence factor corresponding to the predefined historical delay time in the spraying control information database, y2 is the influence factor corresponding to the predefined historical CO2 back pressure fluctuation coefficient in the spraying control information database, y3 is the influence factor corresponding to the predefined historical pipeline pressure difference fluctuation coefficient in the spraying control information database, and y4 is the influence factor corresponding to the predefined historical residence time in the spraying control information database.

[0048] In this embodiment, multivariate analysis specifically considers the correlation between these parameters. An increase in historical delay time will weaken feedforward / feedback synchronization and amplify execution overshoot and undershoot. An increase in historical CO2 back pressure fluctuation coefficient means that the phase window is unstable and the atomization critical condition drifts. An increase in historical pipeline pressure difference fluctuation coefficient indicates that the flow resistance and transient flow velocity fluctuations are aggravated. An increase in the deviation between historical residence time and reference residence time will directly lead to the drift of mixing ratio and dissolution / flash window.

[0049] Increased time delay jitter reduces the controller's phase margin, making it difficult to suppress back pressure and differential pressure pulsations in a timely manner, thus increasing back pressure fluctuations and pipeline differential pressure fluctuations. Back pressure / differential pressure pulsations, in turn, alter intra-segment volume exchange through instantaneous flow fluctuations, further amplifying residence time deviations and significantly increasing the controllability difficulty of CO2 spraying equipment. The accumulation of residence time deviations also causes control calculations to be based on outdated states, further exacerbating the sensitivity to time delay. Overall, these four increases all lower the controllable state score, and their interactions often cause individual problems to amplify at the system level.

[0050] The CO2 chain control stabilization module is used to formally start the CO2 storage tank after the CO2 chain controller receives the executable status signal. The internal CO2 is transported through the purifier into the CO2 booster, where it is pressurized into supercritical CO2 and enters the CO2 buffer tank. The CO2 chain is then dissolved to form a stable CO2 chain. The CO2 chain controller then sends an access contract.

[0051] The aforementioned formal startup refers to the process where, during the task transition period, after each unit completes self-checks such as health and clock consistency, the CO2 chain controller switches from the pre-start (standby) state to the process operation state. It then issues flow authorization and setpoint activation commands to the CO2 storage tank, purifier, and booster compressor, removing the limitation of only small circulation / inertial pressure holding. The target back pressure and mass flow rate are then pushed into the equipment's process zone in a smooth, fixed-speed ramp-up manner, triggering the necessary valve positions and temperature controls to ensure a sustainable supercritical CO2 process flow. The prerequisites for this pre-start state are: the CO2 storage tank maintains effective safety valves and interlocks; the pipelines and purifier complete leak and valve position self-checks in low-pressure small circulation or inertial pressure holding; the booster compressor is in low-speed standby and lubrication / sealing preheating; and online sensors and purification units reach a stable baseline but are not allowed to send large flow rates out. Only after the executable status signal is issued does the system switch from pre-start to formal startup, entering the controlled pressure boosting and steady-state vibration damping process.

[0052] Furthermore, to maintain the stable operation of the CO2 chain, the specific analysis process is as follows: The CO2 booster pressurizes the initial CO2 into supercritical CO2 and introduces it into the CO2 buffer tank. The observer of the CO2 booster monitors the outlet pressure of the CO2 booster in real time and uploads it to the edge analysis terminal. The edge analysis terminal analyzes the mechanical cycle frequency of the CO2 booster based on the real-time outlet pressure.

[0053] The edge analysis end first samples the high-frequency pressure signal at the CO2 booster compressor outlet at millisecond levels to establish a sliding time window, performs baseline de-drift and bandpass noise reduction (preserving the known operating bandwidth of the equipment and suppressing low-frequency operating condition changes and electromagnetic interference), and then performs amplitude normalization and transient glitch removal. Subsequently, fast spectrum analysis (such as short-time fast Fourier transform) and autocorrelation / envelope spectrum verification are performed in parallel within the sliding window to identify the fundamental peak with the most concentrated energy and remove known stray lines such as those from the power grid and wind turbine. Peak tracking and harmonic consistency checks are performed on the main peaks of adjacent windows (the frequency ratio and relative energy stability of the fundamental and second and third harmonics). If necessary, a phase-locked loop frequency tracker is used to smooth the main frequency drift. The observed main frequency is mapped to the mechanical beat frequency by combining the nominal transmission ratio of the equipment and the number of cylinders / impeller blades. Finally, the mechanical beat frequency is output.

[0054] The mechanical cycle frequency of the CO2 booster is verified in real time against the predefined mechanical cycle reference frequency. Based on the difference between the mechanical cycle frequency of the CO2 booster and the mechanical cycle reference frequency, the compensation allocation ratio is mapped to obtain the compensation allocation ratio.

[0055] When the mechanical beat frequency is greater than the reference frequency at a certain moment (indicating that the beat energy has shifted upward and can excite stronger narrowband pulsations), the difference between the two, Δf = current frequency − reference frequency, is used to increase the compensation allocation ratio β (the proportional representation of the proportional vent valve) by converting Δf into compensation allocation ratio β (the proportional representation of the proportional vent valve). This means that more dynamic compensation is allocated to the proportional vent valve for anti-phase vibration suppression. When the mechanical beat frequency is less than or equal to the reference frequency (the pulsation risk is normal or reduced), β decreases and converges towards β_min (the minimum set percentage of the proportional vent valve). At the same time, more of the slow-varying correction weight α (the proportional representation of the active back pressure valve) = 1 − β is given to the active back pressure valve to stabilize the mean (back pressure / average mass flow rate). The two valves each perform their respective functions.

[0056] The compensation allocation ratio refers to the weighting of dynamic compensation between the two actuators: β is allocated to the proportional vent valve to handle the rapid phase reversal and peak reduction of the main frequency / harmonics, and α is allocated to the active back pressure valve to handle the mean and low-frequency steady-state correction (α+β=1, both with amplitude and rate limiting, saturation and health reduction weights); its function is to reduce pressure / flow fluctuations below the threshold by dividing the work into fast and slow parts without changing the target setting, while maintaining the stability and attainability of back pressure and ratio.

[0057] Based on the compensation allocation ratio, the correction and compensation process of the active back pressure valve and proportional vent valve of the CO2 booster is configured.

[0058] The edge analysis end first calculates the back pressure deviation using the target back pressure and the real-time back pressure, and decomposes it into a slow-varying component (mean drift obtained from the low-pass filter) and a fast-varying component (pulsation of the turbocharger's mechanical cycle and its first harmonic locked in the band-pass filter). Then, it generates two compensation paths according to online adaptive allocation coefficients (e.g., the active back pressure valve handles 70–90% of the slow-varying correction and a small portion of the low-frequency pulsation with α, and the proportional vent valve handles 10–30% of the fast anti-phase vibration suppression with β, α+β=1). ① Apply a slope-limiting, anti-saturation PI correction to the active back pressure valve, which only acts on the slow variable component and is superimposed with a small amount of low-frequency feedforward, to ensure that the mean returns to the target and there is no overshoot.

[0059] ② The proportional vent valve applies a small-amplitude anti-phase opening (amplitude adaptive, rate limited, and dead zone set to prevent noise drive) based on the main frequency and phase of the fast-changing component, for rapid peak shaving and valley filling.

[0060] Before execution, the two-way compensation performs a seamless switching of impact and a boundary check of the opening synthesis (not exceeding the upper and lower limits, meeting the minimum opening and minimum closing time), and dynamically adjusts the weight of the two valves based on their health status (automatically reducing the weight of either valve when it becomes saturated or its health status decreases, and the other valve takes over). When the operating condition changes or the main frequency drifts, the allocation coefficient and phase advance are re-estimated in real time to ensure that the average value is stabilized while suppressing pulsation. Finally, the synthesized valve position command is sent to the active back pressure valve and the proportional vent valve to achieve steady-state and dynamic coordinated correction and compensation of the back pressure.

[0061] The flow rate of CO2 in the pipeline and the inlet pressure of CO2 buffer tank are monitored in real time. The deviation between the current flow rate of CO2 in the pipeline and the previous flow rate of CO2 in the pipeline is processed and recorded as the CO2 flow rate fluctuation amplitude. The inlet pressure of CO2 buffer tank is compared with the outlet pressure of CO2 booster to calculate the CO2 pipeline pressure difference.

[0062] The CO2 flow rate fluctuation amplitude is compared with the predefined flow rate fluctuation benchmark amplitude in real time to obtain the flow rate fluctuation comparison result. The CO2 pipeline pressure difference is compared with the predefined pipeline reference pressure difference to obtain the pressure difference comparison result. Based on the flow rate fluctuation comparison result and the pressure difference comparison result, the CO2 chain is kept stable. The CO2 chain controller sends an access contract to the coating chain control stabilization module.

[0063] The specific digestion process is as follows: The above comparison results of flow velocity fluctuations include a first comparison result and a second comparison result. The first comparison result indicates that the CO2 flow velocity fluctuation amplitude is greater than the baseline amplitude, and the second comparison result indicates that the CO2 flow velocity fluctuation amplitude is less than or equal to the baseline amplitude.

[0064] When the flow velocity fluctuation comparison result shows the second flow velocity fluctuation comparison result, the residence time of CO2 in the pipeline is continuously monitored, and the residence time is waited for to enter the target window to maintain the steady state of the CO2 chain.

[0065] When the flow velocity fluctuation comparison result shows the first flow velocity fluctuation comparison result, the allocation ratio adjustment factor is obtained based on the difference between the CO2 flow velocity fluctuation amplitude and the flow velocity fluctuation reference amplitude. Based on the allocation ratio adjustment factor, specifically, the allocation ratio adjustment factor is multiplied by the compensation allocation ratio to obtain the compensation allocation correction ratio. Based on the compensation allocation correction ratio, the correction and compensation process of the active back pressure valve and proportional vent valve of the CO2 booster is reconfigured.

[0066] The above pressure difference comparison results include a first pressure difference comparison result and a second pressure difference comparison result. The first pressure difference comparison result indicates that the CO2 pipeline pressure difference is greater than the pipeline reference pressure difference, and the second pressure difference comparison result indicates that the CO2 pipeline pressure difference is less than or equal to the pipeline reference pressure difference.

[0067] When the differential pressure comparison result shows the second differential pressure comparison result, the residence time of CO2 in the pipeline is continuously monitored and the residence time is waited for to enter the target window for digestion to form a stable CO2 chain.

[0068] When the differential pressure comparison result is displayed as the first differential pressure comparison result, the throttle valve opening correction factor is obtained based on the difference between the CO2 pipeline differential pressure and the pipeline reference differential pressure. This factor is then multiplied by the current opening of the inlet electronic throttle valve of the CO2 buffer tank to obtain the appropriate opening of the inlet electronic throttle valve of the CO2 buffer tank.

[0069] Specifically, the CO2 chain controller sends the admission contract, and the detailed analysis process is as follows: The time-series CO2 back pressure of the CO2 booster compressor during the CO2 booster cycle is obtained, expressed as the controlled static pressure established and maintained at the CO2 booster compressor outlet. This back pressure is then compared with the CO2 booster cycle to obtain the back pressure change rate. The back pressure change rate is represented by the absolute value of the difference between the CO2 back pressure at the start and end of the CO2 booster cycle, calculated as the ratio of this difference to the CO2 booster cycle duration. Based on a mapping table between the back pressure change rate and the upper limit of the back pressure slope, the upper limit of the back pressure slope is obtained and denoted as the upper limit of the coating back pressure slope. The upper limit of the back pressure slope represents the maximum permissible rate of change of pressure within the pressurized metering pump chamber with respect to time when the coating enters the chamber.

[0070] The timing CO2 back pressure of the CO2 booster is compared with a predefined target back pressure. Specifically, the difference between the timing CO2 back pressure and the target back pressure is processed to obtain the timing CO2 back pressure deviation of the CO2 booster. The average error of the CO2 back pressure is obtained by averaging the deviation. Based on the mapping relationship between the average CO2 back pressure error and the upper limit of the initial inlet mass flow rate, the upper limit of the initial inlet mass flow rate is obtained, denoted as the upper limit of the initial inlet mass flow rate of the coating. It is expressed as the maximum allowable change in mass flow rate when the coating initially enters the pressurized metering pump chamber.

[0071] The mass flow rate fluctuation coefficient of the CO2 pipeline is obtained. This involves extracting the time-series mass flow rate of the CO2 pipeline from the observer associated with the CO2 pipeline, and then comparing the current mass flow rate with the previous mass flow rate to obtain the mass flow rate fluctuation coefficient. Based on a mapping table between the mass flow rate fluctuation coefficient and the upper limit of the growth rate, the upper limit of the growth rate is determined and denoted as the maximum growth rate of the coating entering the cavity. This represents the maximum permissible rate of change of the mass flow rate within the pressurized metering pump cavity over time during the process of the coating entering the cavity.

[0072] Based on the upper limit of the coating back pressure slope, the upper limit of the initial mass flow rate of the coating entering the cavity, and the maximum growth rate of the coating entering the cavity, an access contract is formed to enter the coating chain control stabilization module.

[0073] The coating chain control stabilization module is used by the coating chain controller to receive the admission contract execution control feedforward, and then the coating in the coating tank is metered and sent into the coating buffer tank by the coating pressurization metering pump, which dissolves and forms the coating chain steady state.

[0074] Furthermore, to maintain the stable operation of the coating chain, the specific analysis process is as follows: The coating chain controller generates a coating volumetric flow rate inlet ramp curve based on the admission contract, and obtains the coating volumetric flow rate prediction sequence by parsing the inlet ramp curve.

[0075] After receiving the access contract, the coating chain controller first reads the target volumetric flow rate, maximum acceleration and deceleration rates, allowable transient fluctuations, and mixing chamber inlet boundary constraints to generate an inlet ramp setting: smoothly rising from zero return flow to the target flow rate at a limited slope (stepping and allowing dwell time if necessary). Subsequently, within millisecond-level control cycles, the controller uses a discrete first-order inertial device model combined with transport delay to perform forward analysis on this setting, obtaining a coating volumetric flow rate prediction sequence (considering valve / pump response, maximum single-cycle change, dead zone, and small delay). In each cycle, the prediction is compared and corrected with real-time observations to determine whether the step can be amplified, whether reverse micro-throttling or return compensation is needed, until the prediction sequence meets the inlet target and quality allowance within a stable window before spraying begins.

[0076] The specific details of the inlet ramp curve are as follows: Figure 2 As shown, Figure 2 The graph shows the inlet flow rate ramp curve for the coating volumetric flow rate, comparing the set ramp rate with the predicted inlet flow rate. The horizontal axis represents time (s), and the vertical axis represents the coating volumetric flow rate (mL / s). The solid line represents the inlet ramp setting generated by the controller: smoothly ramping from 0 at a limited slope, reaching the target flow rate and remaining constant at approximately 5 seconds. The dashed line represents the predicted actual inlet flow rate obtained based on device dynamics (transport delay and first-order inertia): slightly lagging behind the setting (with an initial delay of about ten milliseconds), with a gentler ramp rate (due to the time constant of the valve / pump and pipeline), gradually merging with the setting and stabilizing near the target value after about 6 seconds.

[0077] The slope of the ramp represents the permissible upflow velocity; the steeper the ramp, the faster the inflow build-up, but the greater the impact on the paint pressurization metering pump chamber. The area / gap between the solid and dashed lines reflects the system's dynamic response difference (delay and time constant); an excessively large gap indicates the need to slow the ramp, increase the feedforward, or optimize valve / pump parameters. No steady-state deviation at the end of the two curves indicates that the setting is achievable and the steady-state response is followable; if long-term deviations occur, the reflux / throttling settings or the pump's achievable flow rate should be checked.

[0078] During the permitted entry period, the paint chain controller sends the scheduled entry time and paint volume flow prediction sequence to the CO2 chain controller. After the CO2 chain controller sends back the scheduled confirmation signal, the paint chain controller officially starts the paint pressurization metering pump to receive the paint delivery from the paint tank.

[0079] The aforementioned entry permit period refers to the permitted time slots for the safe introduction of coatings after an enforceability determination and access agreement.

[0080] During the permitted entry period, the coating chain controller sends the scheduled entry time and the predicted coating volume flow rate sequence (based on the millisecond-level forecast trajectory generated by the limited slope ramp and the device dynamic model) to the CO2 controller. Upon receiving the data, the CO2 controller analyzes and verifies its own back pressure, mass flow rate, temperature, and pulsation phase (including whether it is compatible with the dominant frequency / phase at that time point, whether it will trigger inlet total pressure fluctuations or phase window outages, and whether it conflicts with the kinematic correction table). If all constraints are met, a reservation confirmation signal is sent back, indicating that resources and phase balancing have been reserved for the ramp at that time point, allowing entry according to the reported sequence. If the conditions are not met, a rejection is marked or a rescheduling / reduction suggestion is given, so that the entry action is coordinated at the physical and control levels before execution.

[0081] Obtain the mechanical cycle time of the paint pressurization metering pump, which can be extracted from the observer of the paint pressurization metering pump and compared with the predefined initial mechanical cycle time interval: If the mechanical cycle time of the paint pressurizing metering pump does not fall within the initial mechanical cycle time range, the difference between the mechanical cycle time of the paint pressurizing metering pump and the minimum value of the initial mechanical cycle time range is processed to obtain the mechanical cycle time deviation of the paint pressurizing metering pump. This deviation is then correlated to obtain the throttle valve opening adjustment factor, which is used to correct the current opening of the inlet electronic throttle valve. Specifically, the throttle valve opening adjustment factor is multiplied by the current opening of the inlet electronic throttle valve to obtain the corrected opening of the inlet electronic throttle valve.

[0082] The system compares and calculates the volumetric flow rate of paint flowing into the paint pressurizing metering pump at the current time point with that at the previous time point. Specifically, it performs a ratio calculation to obtain the paint volumetric flow rate fluctuation coefficient of the paint pressurizing metering pump. When the paint volumetric flow rate fluctuation coefficient of the paint pressurizing metering pump is greater than the predefined paint volumetric flow rate fluctuation threshold coefficient at a certain time point, an opening reduction amount is matched based on the difference between the paint volumetric flow rate fluctuation coefficient and the paint volumetric flow rate fluctuation threshold coefficient. This reduction amount is used to correct the current opening of the inlet electronic throttle valve. It is expressed as the opening reduction amount multiplied by the current opening of the inlet electronic throttle valve to obtain the feasible opening of the inlet electronic throttle valve.

[0083] The paint pressurization metering pump delivers a metered amount of paint into the paint buffer tank, which absorbs the fluctuation margin of the paint chain through effective volume self-adaptation.

[0084] Specifically, the paint buffer tank, through effective volume self-adaptation, absorbs the fluctuation margin of the paint chain. The specific analysis process is as follows: Obtain the inlet pressure of the paint buffer tank at the current time point, compare it with the inlet pressure of the paint buffer tank at the previous time point, calculate the inlet pressure fluctuation coefficient of the paint buffer tank, and compare it with the predefined inlet pressure fluctuation boundary coefficient: If the inlet pressure fluctuation coefficient of the coating buffer tank is greater than the inlet pressure fluctuation threshold coefficient at a certain time point, then based on the difference between the inlet pressure fluctuation coefficient and the inlet pressure fluctuation threshold coefficient, an airbag volume compensation factor is obtained, which increases the current airbag volume of the coating buffer tank.

[0085] If the inlet pressure fluctuation coefficient of the coating buffer tank is less than or equal to the inlet pressure fluctuation boundary coefficient, the coating chain will maintain a stable operating state.

[0086] Simultaneously, based on the difference between the inlet pressure fluctuation coefficient and the inlet pressure fluctuation boundary coefficient, an opening reduction factor is obtained to reduce the current opening of the outlet electronic throttle valve of the paint pressurization metering pump.

[0087] The spray gun coating task execution module is used by the coordinator to prioritize the release of supercritical CO2 into the jet mixer based on CO2 chain stabilization. Subsequently, the coating is released into the jet mixer for shaping and mixing, and then enters the spray gun to execute the current round of coating task.

[0088] Furthermore, based on the stable operating state of the CO2 chain, the coordinator prioritizes allowing supercritical CO2 to enter the jet mixer, followed by allowing the coating material to enter the jet mixer for shaping and mixing. The specific analysis process is as follows: The CO2 chain controller publishes the CO2 pressure pulsation frequency and phase of the CO2 buffer tank in real time, and the paint chain controller schedules the inlet electronic throttle valve of the paint buffer tank to automatically align the anti-pulsation modulation phase to the opposite phase.

[0089] During the input and stabilization process, the coating chain controller reads the main frequency and phase of CO2 and the transient flow rate deviation at the jet mixer inlet in real time. Using this as a reference, it applies small-amplitude, same-frequency, and opposite-phase opening modulation to the electronic throttle valve at the coating buffer tank inlet: first, the main frequency is captured by a phase-locked tracker and the phase difference is calculated, and then an amplitude-adaptive micro-opening wave (typically ±0.5% to 2% of the stroke) is generated, so that the valve is slightly closed when CO2 and pump cycle form a peak and slightly opened when the wave is trough, thereby offsetting the synthetic pulsation; during the modulation process, rate and amplitude limiting, health deweighting and saturation protection are superimposed to avoid conflict with the back pressure valve, and the amplitude and phase are finely adjusted online with the minimum transient flow rate deviation at the jet mixer inlet as the closed-loop target. Finally, the phase is automatically aligned to the opposite phase, reducing the pressure / flow rate fluctuation at the mixing chamber inlet to below the preset percentage, thereby improving atomization stability and film thickness consistency.

[0090] The paint chain controller publishes the residual amplitude of the paint flow rate cycle in the paint buffer tank in real time, and the CO2 chain controller schedules the back pressure valve of the CO2 buffer tank to perform phase bias.

[0091] The residual amplitude of the paint flow rate beat refers to the equivalent amplitude of the paint volume flow rate signal in the pump / valve nominal beat and its first harmonic frequency band after removing the set ramp and low frequency trend. It is used to quantify the residual pulsation intensity caused by the mechanical beat during the cut-in / stabilization spraying stage.

[0092] During the steady-state spraying and entry phases, the paint chain controller calculates and publishes the residual amplitude of the paint flow rate cycle in the paint buffer tank in milliseconds (carrying timestamps, main frequency and phase, and health status) as a quantitative reference for inlet disturbances. After subscribing to this indicator, the CO2 chain controller, in comparison with the main frequency and phase of its own CO2 stabilizing tank outlet pressure, schedules the CO2 back pressure valve to perform a small phase offset: without changing the average back pressure and mass flow rate, a small lead / lag component is superimposed on the back pressure valve command, so that the vibration suppression action on the CO2 side cancels out the residual on the paint side, thereby reducing the synthetic pulsation at the jet mixer inlet to below the threshold. The entire process is equipped with amplitude and rate limiting, health status deweighting, and saturation protection. If valve violation or over-limit tendency occurs, it automatically reverts to the conservative bandwidth and requests a recalibration.

[0093] After receiving the CO2 chain stabilization signal, the coordinator opens the CO2 inlet valve of the jet mixer to receive CO2, extracts the real-time CO2 mass flow rate of the jet mixer, and compares it with the target CO2 mass flow rate of the jet mixer. If the real-time CO2 mass flow rate of the jet mixer is less than the target CO2 mass flow rate, the first speed correction factor is obtained based on the difference between the real-time CO2 mass flow rate and the target CO2 mass flow rate.

[0094] The known resistance of the CO2 pipeline is extracted and correlated to obtain the second speed correction factor. Combined with the first and second speed correction factors, the current speed of the CO2 booster is increased. Specifically, the first and second speed correction factors are multiplied by the current speed of the CO2 booster to obtain the appropriate speed of the CO2 booster, where the appropriate speed is greater than the current speed.

[0095] The real-time CO2 back pressure of the jet mixer is extracted and compared with the target CO2 back pressure of the jet mixer. If the real-time CO2 back pressure is less than the target CO2 back pressure, the opening degree of the current outlet electronic throttle valve of the CO2 buffer tank is limited based on the difference between the real-time CO2 back pressure and the target CO2 back pressure. Specifically, the difference between the real-time CO2 back pressure and the target CO2 back pressure is recorded as the CO2 back pressure deviation, and the throttle valve opening influence element is matched and multiplied with the current outlet electronic throttle valve opening degree of the CO2 buffer tank to obtain the corrected opening degree of the outlet electronic throttle valve of the CO2 buffer tank.

[0096] The jet mixer maintains the target back pressure and target mass flow rate. A stable main jet is formed at the throat of the jet mixer, and then the coating is released into the jet mixer for shaping and mixing.

[0097] The specific characteristics of the jet mixer are as follows: Figure 4 As shown, Figure 4 This is a prototype of a supercritical CO2 spraying device without a paint pump, which automatically introduces the paint and achieves full mixing by relying on a high-pressure CO2 jet.

[0098] Specifically, the coating is allowed to enter the jet mixer for shaping and mixing. The detailed analysis process is as follows: Maintain the paint return proportional valve of the paint buffer tank at the baseline opening, where the baseline opening represents the minimum opening of the paint return proportional valve. Open the paint inlet valve of the jet mixer to receive paint. The initial setting of the paint inlet valve opening is the defined opening.

[0099] Monitor the cumulative volumetric flow rate of the coating and the cumulative volume fraction of CO2 to obtain the real-time estimated ratio of CO2 to coating. The specific ratio is the sum of the cumulative CO2 volume and the cumulative coating volume. When the real-time estimated ratio of CO2 to coating is within the allowable ratio range, the stable ratio duration is calculated. The stable ratio duration is expressed as the duration during which the real-time estimated ratio is within the allowable ratio range. The stable ratio duration is compared with the predefined stable ratio benchmark duration. If the stable ratio duration is greater than or equal to the stable ratio benchmark duration, the mixing inlet alignment is completed, and the coating enters the spray gun to perform the current round of spraying.

[0100] If the stable mixing time is less than the stable mixing reference time, a temperature correction factor is obtained based on the difference between the stable mixing time and the stable mixing reference time. This factor is used to increase the current temperature of the jet mixer, and the stable mixing time is calculated again.

[0101] When the spray gun performs this round of spraying, it first monitors whether the paint volumetric flow rate and carbon dioxide volume fraction at the inlet of the jet mixer have entered the preset target window, and gradually reaches the target ratio in small steps; when both are stable and meet the target, the spray gun outflow is released and the formal spraying begins according to the kinematic-medium coupling setting. The shutdown cycle is as follows: first, the paint is cut off (paint side backflow), then CO2 is maintained for several seconds under pressure to purge and clean the jet mixer and nozzle, and finally the back pressure is slowly reduced according to the limited slope and vented / recovered to bring the system back to the safe baseline, avoiding end gelation, backflow, and frosting.

[0102] The aforementioned kinematic-medium coupling setting establishes a strongly linked integrated control system that links the kinematic parameters of the spray gun (gun speed, acceleration, trajectory curvature, distance and attitude between the spray gun and the workpiece, etc.) with the parameters of the spraying medium (coating volume flow rate, carbon dioxide mass flow rate and back pressure, spray width and pre-baking temperature, etc.). The spraying system uses trajectory preview to map the upcoming speed and curvature changes to the target settings on the medium side in real time, and is subject to hard constraints from the inlet targets of the jet mixer mixing chamber (inlet total pressure, channel velocity balance, and allowable carbon dioxide volume fraction range within the coating phase). In actual execution, feedforward is the main method, supplemented by micro-feedback (online wet film thickness, median droplet size, and visual defect signals), and is combined with the phase alignment of CO2 / coating vibration suppression to synchronize the medium supply and atomization state when kinematics changes. This achieves the optimal synergy of constant film thickness, defect suppression, and energy consumption / emissions while satisfying appearance and adhesion requirements.

[0103] It should be explained that when the spray gun is performing this round of spraying, once the online vision or film thickness / particle size monitoring determines that a whitening trend is detected, the controller immediately performs conservative correction: without changing the current spray gun cycle, it reduces the CO2 volume fraction in the coating phase (such as adjusting the carbon dioxide mass flow rate or increasing the back pressure to suppress excessive dissolution), moderately increases the pre-baking temperature (shortening solvent residence and promoting film densification), and increases the distance between the spray gun and the workpiece (weakening local cooling and instantaneous scouring, and opening the atomization cold end phase change window); the three adjustments are executed in a millisecond-level limiting and priority order until the whitening criterion is removed and the online quality is allowed to return to stability.

[0104] It should be noted that the spraying system automatically generates a pre-defined package for the next task transition cycle during the final stage: summarizing the steady-state settings of this cycle (target back pressure, carbon dioxide mass flow rate, paint volume flow rate, spray gun spacing / spray width, pre-baking temperature), dynamic fingerprints (dominant frequency and phase, pulsation residual, local pressure gradient in the mixing chamber and nozzle channel velocity balance curve), stable window and boundary crossing, and combining fault and correction records to form an access contract (including allowed setting range, limited slope, maximum change in a single cycle, and permitted entry time period); simultaneously solidifying the cavity ramp curve and coupling feedforward table (mapping the gun speed / curvature to the medium) The system updates the initial amplitude and phase values ​​and compensation allocation ratio of the two-chain vibration suppression, and issues recommended settings for the CO2 buffer tank airbag pre-charge pressure / liquid column height and inlet orifice plate. It calculates the termination zero point value in the shutdown sequence and marks it with a health status, archives the historical sequences of key observation channels as similarity variance benchmarks (to be used for the observer's executability determination in the next round). Finally, it generates the self-check / preheating plan for the next round (purifier and booster preheating, sensor zeroing / clock alignment, filter and recovery valve position preset), so that the next round of connection cycle starts with the converged parameter corridor combined with the reproducible inlet target, shortening the start-up time and reducing the risk of entry.

[0105] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A supercritical CO2 energy-saving spraying process control system for ship components, characterized in that, include: The equipment is equipped with an executable determination module, which is used to determine the executable status of the supercritical CO2 spraying equipment group and its associated observers when they enter the mission transition period. The CO2 chain control stabilization module is used to officially start operation after the CO2 chain controller receives the executable status signal. The internal CO2 is output from the storage tank, purified by the purifier, and pressurized to the supercritical state by the CO2 booster, and finally enters the CO2 buffer tank to maintain the stable operation of the CO2 chain. The CO2 chain controller sends the access contract. The coating chain control stabilization module is used by the coating chain controller to receive the access contract execution control feedforward, and then the coating pressurization metering pump delivers the coating from the coating tank into the coating buffer tank in a quantitative manner to maintain the stable operation of the coating chain. The spray gun coating task execution module is used by the coordinator to prioritize the release of supercritical CO2 into the jet mixer based on the stable operating state of the CO2 chain, and then release the coating into the jet mixer for shaping and mixing. After mixing, the coating enters the spray gun to perform the current round of spraying task.

2. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 1, characterized in that: The supercritical CO2 spraying equipment group and its associated observers complete the executable status determination within the task transition period. The specific analysis process is as follows: The executable state determination includes the executable state determination of the observer and the executable state determination of the CO2 spraying equipment group; The execution state determination of the observer is specifically performed as follows: Extract the historical observation data sequence corresponding to each observer of each CO2 spraying equipment in the CO2 spraying equipment group to obtain the historical observation data sequence of each observer of the CO2 spraying equipment. Based on the historical observation data sequence, the termination zero point value of each observer of the CO2 spraying equipment is extracted to determine whether it is zero. If it is zero, the data similarity variance is determined. If it is not zero, the observer data weighting factor is matched based on the termination zero point value of each observer of the CO2 spraying equipment. The data similarity variance determination process is as follows: The historical observation data sequences of each observer of the CO2 spraying equipment are sequentially processed to calculate the data sequence similarity. The variance of the similarity calculation results is then processed to obtain the data similarity variance of the CO2 spraying equipment observers. This variance is then compared with a predefined data similarity variance threshold. If the data similarity variance of the CO2 spraying equipment observers is less than the data similarity variance threshold, the executable status of the observers is determined to be passed. The executable status of the CO2 spraying equipment group is then determined sequentially. Otherwise, it is determined to be failed, and an early warning is issued for this round of spraying tasks.

3. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 2, characterized in that: The specific analysis process for determining the executable status of the CO2 spraying equipment group is as follows: The CO2 spraying equipment group includes a CO2 booster, a CO2 buffer tank, a paint pressurizing and metering pump, a paint buffer tank, a jet mixer, and a spray gun. When the executable status of the observer is determined to be passed, the historical observation data sequence of each observer of the CO2 spraying equipment is fused. If a certain observer has an observer data weighting factor, the data fusion process is configured based on the observer data weighting factor to obtain and record the historical execution dataset of the CO2 spraying equipment. Based on the historical execution dataset of the CO2 spraying equipment, the controllable status indicators of the CO2 spraying equipment are evaluated and compared with the predefined controllable status threshold indicators. If all the controllable status indicators of the CO2 spraying equipment are greater than or equal to the controllable status threshold indicators, the executable status of the CO2 spraying equipment group is judged as passed; otherwise, it is judged as failed, and an early warning is issued for this round of spraying tasks.

4. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 3, characterized in that: The specific analysis process for evaluating the controllable state indicators of the CO2 spraying equipment is as follows: The historical execution dataset of the CO2 spraying equipment includes the historical delay duration of the CO2 spraying equipment, the historical time-series CO2 back pressure of the CO2 spraying equipment, the historical time-series pipeline pressure difference of the CO2 spraying equipment, and the historical dwell time of the CO2 spraying equipment. The historical time-series CO2 back pressure and historical time-series pipeline pressure difference of the CO2 spraying equipment were processed by standard deviation to obtain the historical CO2 back pressure fluctuation coefficient and the historical pipeline pressure difference fluctuation coefficient of the CO2 spraying equipment. Extract the delay definition duration and baseline dwell time from the spraying control information database; The historical CO2 back pressure fluctuation coefficient and the historical pipeline pressure difference fluctuation coefficient of the CO2 spraying equipment are preprocessed separately. The data preprocessing includes normalization and de-unitization to obtain the data preprocessing results. The historical delay time of the CO2 spraying equipment is compared with the delay definition time to obtain the ratio processing result. The historical dwell time of the CO2 spraying equipment was compared with the baseline dwell time to obtain the deviation processing result. By introducing an influencing factor, the data preprocessing results, ratio processing results, and deviation processing results are correlated and merged to obtain the controllable state index of the CO2 spraying equipment.

5. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 1, characterized in that: The specific analysis process for maintaining the stable operation of the CO2 chain is as follows: The CO2 booster pressurizes the initial CO2 into supercritical CO2 and introduces it into the CO2 buffer tank. The observer of the CO2 booster monitors the outlet pressure of the CO2 booster in real time and uploads it to the edge analysis terminal. The edge analysis terminal analyzes the mechanical cycle frequency of the CO2 booster based on the real-time outlet pressure. The mechanical cycle frequency of the CO2 booster is verified in real time and compared with the predefined mechanical cycle reference frequency. Based on the difference between the mechanical cycle frequency of the CO2 booster and the mechanical cycle reference frequency, the compensation allocation ratio is mapped to obtain the compensation allocation ratio. Based on the compensation allocation ratio, configure the correction and compensation process of the active back pressure valve and proportional vent valve of the CO2 booster; The flow rate of CO2 in the pipeline and the inlet pressure of CO2 buffer tank are monitored in real time. The deviation between the flow rate of CO2 in the pipeline at the current time point and the flow rate of CO2 in the pipeline at the previous time point is processed and recorded as the CO2 flow rate fluctuation amplitude. The inlet pressure of CO2 buffer tank is compared with the outlet pressure of CO2 booster to calculate the CO2 pipeline pressure difference. The CO2 flow rate fluctuation amplitude is compared with the predefined flow rate fluctuation benchmark amplitude in real time to obtain the flow rate fluctuation comparison result. The CO2 pipeline pressure difference is compared with the predefined pipeline reference pressure difference to obtain the pressure difference comparison result. Based on the flow rate fluctuation comparison result and the pressure difference comparison result, the CO2 chain is kept stable. The CO2 chain controller sends an access contract to the coating chain control stabilization module.

6. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 5, characterized in that: The CO2 chain controller sends the admission contract, and the specific analysis process is as follows: The timing CO2 back pressure of the CO2 booster is obtained during the CO2 booster cycle, and the ratio of the CO2 booster cycle is processed to obtain the back pressure change rate of the CO2 booster. Based on the mapping relationship between the back pressure change rate and the upper limit of the back pressure slope, the upper limit of the back pressure slope is obtained and denoted as the upper limit of the coating back pressure slope. The timing CO2 back pressure of the CO2 booster is compared with the predefined target back pressure to obtain the average error of the CO2 back pressure of the CO2 booster. Based on the mapping relationship table between the average error of the CO2 back pressure and the upper limit of the initial inlet mass flow rate, the upper limit of the initial inlet mass flow rate is obtained and denoted as the upper limit of the initial inlet mass flow rate of the coating. Obtain the mass flow rate fluctuation coefficient of the CO2 pipeline. Based on the mapping relationship table between the mass flow rate fluctuation coefficient and the upper limit of the growth rate, match the upper limit of the growth rate and record it as the maximum growth rate of the coating entering the cavity. Based on the upper limit of the coating back pressure slope, the upper limit of the initial mass flow rate of the coating entering the cavity, and the maximum growth rate of the coating entering the cavity, an access contract is formed to enter the coating chain control stabilization module.

7. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 1, characterized in that: The specific analysis process for maintaining the stable operation of the coating chain is as follows: The coating chain controller generates a coating volumetric flow rate inlet ramp curve based on the admission contract, and obtains a coating volumetric flow rate prediction sequence by parsing the inlet ramp curve. During the permitted entry period, the paint chain controller sends the scheduled entry time and paint volume flow prediction sequence to the CO2 chain controller. After the CO2 chain controller sends back the scheduled confirmation signal, the paint chain controller starts the paint pressurization metering pump to receive the paint delivery from the paint tank. Obtain the mechanical cycle time of the paint pressurization metering pump and compare it with the predefined initial mechanical cycle time range: If the mechanical cycle time of the paint pressurizing metering pump does not belong to the mechanical initial set cycle time range, the difference between the mechanical cycle time of the paint pressurizing metering pump and the minimum value of the mechanical initial set cycle time range is processed to obtain the mechanical cycle time deviation of the paint pressurizing metering pump. This deviation is then correlated to obtain the throttle valve opening adjustment factor, which is used to correct the current opening of the inlet electronic throttle valve. The volumetric flow rate of paint flowing into the paint pressurizing metering pump at the current time point is compared and calculated with that at the previous time point to obtain the paint volumetric flow rate fluctuation coefficient of the paint pressurizing metering pump. When the paint volumetric flow rate fluctuation coefficient of the paint pressurizing metering pump is greater than the predefined paint volumetric flow rate fluctuation threshold coefficient at a certain time point, the opening reduction amount is matched based on the difference between the paint volumetric flow rate fluctuation coefficient and the paint volumetric flow rate fluctuation threshold coefficient, which is used to correct the current opening of the inlet electronic throttle valve. A paint pressurization metering pump delivers paint into a paint buffer tank, which adaptively absorbs paint chain fluctuations through its effective volume.

8. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 7, characterized in that: The paint buffer tank adaptively absorbs the paint chain fluctuation margin through its effective volume. The specific analysis process is as follows: Obtain the inlet pressure of the paint buffer tank at the current time point, compare it with the inlet pressure of the paint buffer tank at the previous time point, calculate the inlet pressure fluctuation coefficient of the paint buffer tank, and compare it with the predefined inlet pressure fluctuation boundary coefficient: If the inlet pressure fluctuation coefficient of the coating buffer tank is greater than the inlet pressure fluctuation threshold coefficient at a certain time point, then based on the difference between the inlet pressure fluctuation coefficient and the inlet pressure fluctuation threshold coefficient, an airbag volume compensation factor is obtained, which increases the current airbag volume of the coating buffer tank. Simultaneously, based on the difference between the inlet pressure fluctuation coefficient and the inlet pressure fluctuation boundary coefficient, an opening reduction factor is obtained to reduce the current opening of the outlet electronic throttle valve of the paint pressurization metering pump.

9. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 1, characterized in that: Based on the stable operating state of the CO2 chain, the coordinator prioritizes allowing supercritical CO2 to enter the jet mixer, and then allows the coating to enter the jet mixer for shaping and mixing. The specific analysis process is as follows: The CO2 chain controller publishes the CO2 pressure pulsation frequency and phase of the CO2 buffer tank in real time, and the paint chain controller schedules the inlet electronic throttle valve of the paint buffer tank to automatically align the anti-pulsation modulation phase to the opposite phase. The paint chain controller publishes the residual amplitude of the paint flow rate cycle in the paint buffer tank in real time, and the CO2 chain controller schedules the back pressure valve of the CO2 buffer tank to perform phase offset. After receiving the CO2 chain stabilization signal, the coordinator opens the CO2 inlet valve of the jet mixer to receive CO2, extracts the real-time CO2 mass flow rate of the jet mixer, and compares it with the target CO2 mass flow rate of the jet mixer. If the real-time CO2 mass flow rate of the jet mixer is less than the target CO2 mass flow rate, the first speed correction factor is obtained based on the difference between the real-time CO2 mass flow rate and the target CO2 mass flow rate. Extract the known resistance of the CO2 pipeline, correlate it to obtain the second speed correction factor, and combine the first speed correction factor and the second speed correction factor to increase the current speed of the CO2 booster. Extract the real-time CO2 back pressure of the jet mixer and compare it with the target CO2 back pressure of the jet mixer. If the real-time CO2 back pressure is less than the target CO2 back pressure, limit the opening of the current outlet electronic throttle valve of the CO2 buffer tank based on the difference between the real-time CO2 back pressure and the target CO2 back pressure. The jet mixer maintains the target back pressure and target mass flow rate. A stable main jet is formed at the throat of the jet mixer, and then the coating is released into the jet mixer for shaping and mixing.

10. The supercritical CO2 energy-saving spraying process control system for ship components according to claim 9, characterized in that: The released coating enters the jet mixer for shaping and mixing. The specific analysis process is as follows: Keep the paint return proportional valve of the paint buffer tank at the baseline opening, open the paint inlet valve of the jet mixer to receive paint, and initially set the paint inlet valve opening to the defined opening; Monitor the cumulative volumetric flow rate of the coating and the cumulative volume fraction of CO2 to obtain the real-time estimated ratio of CO2 to coating. When the real-time estimated ratio of CO2 to coating is within the allowable ratio range, the ratio stabilization time is calculated. The ratio stabilization time is compared with the predefined ratio stabilization benchmark time. If the ratio stabilization time is greater than or equal to the ratio stabilization benchmark time, the mixing inlet alignment is completed, and the spray gun is used to perform the current round of spraying. If the stable mixing time is less than the stable mixing reference time, a temperature correction factor is obtained based on the difference between the stable mixing time and the stable mixing reference time. This factor is used to increase the current temperature of the jet mixer, and the stable mixing time is calculated again.

Images