Double-light-path pulse laser cleaning process for ship segmentation

Abstract

The invention relates to a double-light-path pulse laser cleaning process for secondary rust removal of a ship section. The double-light-path pulse laser cleaning process comprises the following steps: a, scene and pollutant identification: determining structural characteristics and pollutant states of a to-be-cleaned area of the ship section; b, a cleaning strategy is formulated, specifically, based on the recognition result of the step a, the carrying and moving mode of a laser cleaning head is determined according to structural characteristics, and a cleaning scheme of double-light-path pulse laser is determined according to the pollutant state; c, double-pulse cleaning is executed, specifically, a laser cleaning head is controlled to move in a set mode, and the double-light-path pulse laser cleaning scheme is executed; and d, the cleaning effect is verified, specifically, the rust removal grade, the surface roughness Rz and the cleaning efficiency of the cleaned surface are measured and evaluated. According to the optimized combination scheme, various complex cleaning working conditions on the ship sections can be flexibly dealt with, accurate and efficient treatment on different pollutants and geometrical characteristics is achieved, the adaptability is wide, and the treatment process is environmentally friendly.

Description

Technical Field

[0001] This invention relates to the technical field of shipbuilding and surface treatment, specifically to a dual-path pulsed laser cleaning process for ship sections, and more particularly to a laser cleaning process for secondary rust removal of section surfaces before painting. Background Technology

[0002] During the construction of ship sections, steel plates that have undergone pretreatment and been coated with shop primer may experience localized damage to the shop primer after welding, handling, and stacking, accompanied by rust, welding slag, oil stains, and other contaminants. Before applying high-performance anti-corrosion coatings, these areas (i.e., areas requiring secondary rust removal) must be thoroughly cleaned to meet the required cleanliness standards (e.g., Sa2). 1 / Level 2).

[0003] Currently, secondary rust removal for ship sections almost entirely relies on sandblasting (shot blasting). While this method is technically mature and has strong processing capabilities, it suffers from the following inherent and insurmountable drawbacks: 1. Extremely serious environmental pollution: It generates a large amount of dust (mainly inhalable particulate matter PM10, PM2.5 and heavy metal dust), which seriously endangers the respiratory health of operators (can lead to silicosis) and pollutes the factory area and surrounding environment.

[0004] 2. High risk of matrix damage: High-speed abrasive impact can easily cause segmental deformation of thin plates, change the stress state of critical weld areas, and excessively cut the steel matrix, affecting structural strength. Abrasive particles may sometimes embed into the matrix, causing subsequent corrosion.

[0005] 3. Poor adaptability: It does not handle complex geometric features such as welds, corners, and edges evenly, and is prone to leaving cleaning dead spots or causing excessive wear.

[0006] 4. High overall costs: These include continuous abrasive consumption, rapid wear and tear on equipment and nozzles, high investment and maintenance costs for dust recovery and treatment systems, and high occupational health protection costs for workers.

[0007] 5. High noise pollution: The noise level at the work site often exceeds the safety threshold, causing damage to workers' hearing.

[0008] In recent years, laser cleaning has been explored as an emerging technology for rust removal from metal surfaces. However, the single pulsed laser mode presents a contradiction between cleaning efficiency and quality when dealing with complex conditions on ship section surfaces, such as intact shop primer, surface rust, and the coexistence of primer and rust: using high-energy, low-frequency pulses to remove hard shop primer may cause thermal damage to the substrate and generate a large amount of dust; while using lower-energy, high-frequency pulses cannot effectively peel off the strong-adhesion, intact primer layer, leading to repetitive processes, low efficiency, and failing to meet the intelligent requirements of modern shipbuilding. Existing technologies lack an efficient laser cleaning process that can intelligently adapt to this complex combination of contaminants.

[0009] Therefore, there is an urgent need in this field for a new process that can completely replace sandblasting for secondary rust removal of ship sections. Summary of the Invention

[0010] The purpose of this invention is to provide an improved dual-path pulsed laser cleaning process for ship sections, replacing the existing sandblasting method, and achieving high efficiency, non-destructive and environmentally friendly results.

[0011] To achieve the above objectives, the technical solution of the present invention is: a dual-path pulsed laser cleaning process for ship sections, characterized in that the laser cleaning process includes the following steps: a) Scene and contaminant identification: determining the structural characteristics and contaminant state of the area to be cleaned in the ship section; b) Developing a cleaning strategy: based on the identification results of step a), determining the mounting and movement mode of the laser cleaning head according to the structural characteristics, and determining the dual-path pulsed laser cleaning scheme according to the contaminant state; c) Executing dual-pulse cleaning: controlling the laser cleaning head to move in a predetermined manner and executing the dual-path pulsed laser cleaning scheme; d) Verifying the cleaning effect: measuring and evaluating the rust removal grade, surface roughness Rz, and cleaning efficiency of the cleaned surface.

[0012] Preferably, in step a, the structural features are identified by visual observation, three-dimensional laser scanning, or a visual sensor, and the identified structural features are classified into at least one of the following features: flat surface, curved surface, weld area, local repair, or hard-to-reach corner edge; the contaminant state is identified by visual observation, multispectral imaging, or laser-induced breakdown spectroscopy, and the identified contaminant state is classified into at least one of the following states: intact shop primer, rust, or a mixture of shop primer and rust, dust, or oil.

[0013] Furthermore, in step b, the mounting and movement of the laser cleaning head are as follows: 1) For areas with complex three-dimensional curved surfaces, the laser cleaning head is mounted on the end of an industrial robot and moves in multiple degrees of freedom; 2) For flat surfaces with a length ≥ 8m or a width ≥ 3m, the laser cleaning head is mounted on a wall-climbing robot and moves in two-dimensional or three-dimensional linear motion; 3) For areas requiring local repair or difficult to reach, the laser cleaning head is moved by hand by the operator.

[0014] Furthermore, in step b, the cleaning scheme is as follows: 1) For stubborn contaminants in weld areas, curved surfaces, local or hard-to-reach corner edges, the first and second pulse lasers use low single-pulse energy, combined with a high overlap rate; 2) For large-area flat surfaces, the first and second pulse lasers use high single-pulse energy, combined with a medium overlap rate; 3) For areas requiring aesthetically pleasing surfaces, the first and second pulse lasers use low single-pulse energy, combined with a high overlap rate.

[0015] Further, in step c, the dual-path pulsed laser includes a first pulsed laser and a second pulsed laser. The first pulsed laser has a single pulse energy range of 0.5-50 mJ, an overlap rate of 30%-70%, and a galvanometer scanning speed of 1000-40000 mm / s, and is used to break up and peel off the main contaminant layer. The second pulsed laser has a single pulse energy range of 0.5-30 mJ, an overlap rate of 40%-80%, and a galvanometer scanning speed of 1000-50000 mm / s, and is used to remove residual contaminants and beautify the surface of the substrate. The cleaning speed of the laser cleaning head is 5-50 mm / s.

[0016] Furthermore, in step d, the surface rust removal grade is verified by comparing with the standard chart of GB / T8923.1-2011 to determine whether it reaches Sa 2½ or higher, and the cleaning efficiency is verified by calculating the cleaning area per unit time; the surface roughness Rz is in the range of 30-75μm; the cleaning efficiency requirement is: ≥10 μm for flat surfaces. 2 / h, curved surface ≥8 m 2 / h, weld seam ≥20 m / h.

[0017] Furthermore, in the first pulse laser, the single pulse energy E1 is considered low when 0.5mJ≤E1<25mJ, and high when 25mJ≤E1≤50mJ. The overlap rate R1 is considered medium when 30%≤R1<45%, and high when 45%≤R1≤70%. In the second pulse laser, the single pulse energy E2 is considered low when 0.5mJ≤E2<15mJ, and high when 15mJ≤E2≤30mJ. The overlap rate R2 is considered low when 40%≤R2<55%, and high when 55%≤R2≤80%.

[0018] Furthermore, the first pulse laser is a flat-top beam, and the second pulse laser is a flat-top beam or a Gaussian beam, with the first pulse laser in front and the second pulse laser behind, and the two are 2-20mm apart.

[0019] Furthermore, the energy distribution uniformity of the flat-top light spot is ≥80%, and the energy density at the center of the Gaussian light spot is 2-5 times that at the edge.

[0020] Acquire scene and contaminant identification information, and establish a 3D model data mapping between cleaning strategy and cleaning effect; based on the optimization of the indicators of the mounting and movement mode of the laser cleaning head and the cleaning scheme indicators of dual-path pulsed laser, solve for the laser cleaning strategy; optimize the laser cleaning strategy to obtain the optimal combination of process parameters that meet the requirements of different contaminant states and workpieces.

[0021] Compared with the prior art, the technical solution of the present invention not only improves the overall technical solution, but also includes many improvements in details. Specifically, it has the following beneficial effects: 1. The improved solution described in this invention formulates a cleaning strategy based on the scene and pollutant identification, including the mounting and movement mode of the top-mounted laser cleaning head and the cleaning scheme of dual-path pulsed laser, and then performs dual-pulse cleaning. Through the above optimized combination scheme, it can flexibly cope with various complex cleaning conditions on ship sections, and achieve accurate and efficient treatment of different pollutants and geometric features. It has wide adaptability and the treatment process is green and environmentally friendly. 2. In the technical solution of the present invention, the combination of key parameters such as the energy distribution form of the dual-pulse laser, the single-pulse energy, and the overlap rate are intelligently optimized for different cleaning scenarios, which solves the contradiction between efficiency and quality of a single laser mode, improves the efficiency of dual-pulse synergy, enhances the cleaning effect, and greatly reduces the risk of damage to the substrate being cleaned. 3. In the technical solution of the present invention, the single pulse energy range of the first pulse laser is 0.5-50mJ, which is mainly responsible for the breaking and peeling of contaminants; the single pulse energy range of the second pulse laser is 0.5-30mJ, which is used to remove residual contaminants and beautify the surface of the substrate; the dual pulses work together to achieve the best cleaning efficiency and effect. 4. The process of this invention is a dry, dust-free, and low-noise treatment throughout. The non-contact cleaning method ensures that the steel substrate is free from deformation, overcutting, and embedding, perfectly maintaining the structural integrity of the segments. At the same time, it is green, environmentally friendly, energy-saving, and efficient. 5. The process method of the present invention has self-learning and optimization capabilities through the continuous updating mechanism of the RSM model, which can adapt to different equipment states and operating conditions, greatly improving the adaptability and reliability of the system and facilitating its promotion and application. Attached Figure Description

[0022] Figure 1 This is a process flow diagram of the present invention.

[0023] Figure 2 This is a diagram showing the effect after cleaning in the second embodiment of the present invention.

[0024] Figure 3 This is a diagram showing the effect after cleaning in the third embodiment of the present invention.

[0025] Figure 4 This is a diagram showing the effect after cleaning in the fourth embodiment of the present invention. Detailed Implementation

[0026] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] This invention provides a dual-path pulsed laser cleaning process for ship sections, see details below. Figure 1 The difference between this technology and existing technologies lies in the following steps: a) Scene and contaminant identification: Determine the structural characteristics and contaminant status of the ship section to be cleaned; b) Develop a cleaning strategy: Based on the identification results of step a), determine the mounting and movement mode of the laser cleaning head according to the structural characteristics, and determine the dual-path pulsed laser cleaning scheme according to the contaminant status; c) Execute dual-pulse cleaning: Control the laser cleaning head to move in a predetermined manner and execute the dual-path pulsed laser cleaning scheme; d) Verify the cleaning effect: Measure and evaluate the rust removal grade, surface roughness Rz, and cleaning efficiency of the cleaned surface.

[0028] During implementation, cleaning strategies are formulated based on the scenario and pollutant identification. These strategies include the mounting and movement of the top-mounted laser cleaning head, and the dual-path pulsed laser cleaning scheme. Dual-pulse cleaning is then executed. This optimized combination of methods flexibly addresses various complex cleaning conditions on ship sections, achieving precise and efficient treatment of different pollutants and geometric features. It is highly adaptable and the process is environmentally friendly. Furthermore, the synergistic efficiency of the first and second pulses improves the cleaning effect and significantly reduces the risk of damage to the cleaned substrate, preventing deformation or impact on its structural strength.

[0029] Example 1 This embodiment describes a dual-path pulsed laser cleaning process for ship sections. The laser cleaning process includes the following steps: a) Scene and contaminant identification: determining the structural features and contaminant status of the area to be cleaned in the ship section; b) Cleaning strategy formulation: based on the identification results of step a), determining the mounting and movement mode of the laser cleaning head according to the structural features, and determining the dual-path pulsed laser cleaning scheme according to the contaminant status; c) Executing dual-pulse cleaning: controlling the laser cleaning head to move in a predetermined manner and executing the dual-path pulsed laser cleaning scheme; d) Verifying cleaning effect: testing and verifying the cleanliness and surface quality of the cleaned surface. If it does not meet the requirements, the process is repeated after optimization in step b, and cleaning is performed again until the requirements are met.

[0030] In step a, structural features are identified through visual observation, three-dimensional laser scanning, or visual sensors. The identified structural features are classified into at least one of the following: flat surface, curved surface, weld area, local repair, or hard-to-reach corner edge. The contaminant state is identified through visual observation, multispectral imaging, or laser-induced breakdown spectroscopy. The identified contaminant state is classified into at least one of the following: intact shop primer, rust, or a mixture of shop primer and rust, dust, or oil.

[0031] Specifically, structural features include the following: flat surfaces, weld areas, fillet welds and bend edges, as well as the geometric dimensions (such as weld width and reinforcement height), three-dimensional morphology (such as surface curvature), and material properties (such as the difference between the base material and the welding material). The identification methods employed include: using a 3D laser scanner to acquire high-precision point cloud data and construct a digital model of the area to be cleaned; or using a high-resolution vision sensor (such as a CCD camera) combined with image processing algorithms to automatically identify and locate various structural features.

[0032] Then, the state of contaminants was quantitatively analyzed and categorized into the following types: 1. Intact shop primer: its type (e.g., epoxy, inorganic zinc) and average thickness were identified. 2. Rust: surface rust, medium rust, and thick rust were differentiated, and their coverage area and thickness were assessed. 3. Mixed contaminants: the mixing ratio of shop primer and rust, the layering, and the types of surface deposits (e.g., salt, oil) were analyzed. The identification methods used to analyze the above contaminant states included: qualitative and semi-quantitative analysis of contaminant components using laser-induced breakdown spectroscopy; or differentiation and mapping of surface contaminant elemental distribution based on the differences in spectral reflectance characteristics of different substances; and measurement of coating thickness using photothermal non-contact thickness measurement technology.

[0033] Furthermore, in step b, the mounting and movement of the laser cleaning head are as follows: 1) For areas with complex three-dimensional curved surfaces, the laser cleaning head is mounted on the end of an industrial robot and moves in multiple degrees of freedom; 2) For flat surfaces with a length ≥ 8m or a width ≥ 3m, the laser cleaning head is mounted on a wall-climbing robot and moves in two-dimensional or three-dimensional linear motion; 3) For areas requiring local repair or difficult to reach, the laser cleaning head is moved by hand by the operator.

[0034] Specifically, the mounting and movement methods for laser cleaning heads are selected as follows: Industrial robot end-effector mounting: Suitable for complex 3D curved surfaces with good accessibility, such as those inside or at the edges of ship hull sections. This method offers high flexibility, precise path control, and can clean complex contours such as welds and corners. Wall-climbing robot end-effector mounting: Specifically designed for large-area vertical walls, inclined walls, and overhead walls of large ship hull sections. The robot needs strong adsorption capabilities (such as magnetic adsorption or vacuum adsorption) and autonomous navigation and obstacle avoidance functions to achieve automated, high-volume operations. Handheld movement: Suitable for localized areas difficult for robots to reach, small-batch operations, or rapid touch-ups. To ensure operational safety and quality, the handheld device should integrate safety locking, distance sensing, and real-time process monitoring functions.

[0035] Furthermore, in step b, the cleaning scheme is as follows: 1) For stubborn contaminants in weld areas, curved surfaces, local or hard-to-reach corner edges, the first and second pulse lasers use low single-pulse energy, combined with a high overlap rate; 2) For large-area flat surfaces, the first and second pulse lasers use high single-pulse energy, combined with a medium overlap rate; 3) For areas requiring aesthetically pleasing surfaces, the first and second pulse lasers use low single-pulse energy, combined with a high overlap rate.

[0036] Specifically, cleaning schemes are generally divided into three types: Scheme 1: First pulse (small core diameter flat-top beam): The beam diameter is small and the energy is concentrated, used for precise peeling of hard-to-reach fine areas such as weld roots and corners, or for preliminary treatment of thin-layer contaminants. Second pulse (large core diameter flat-top beam or Gaussian beam): The beam diameter is large (e.g., 2-5mm), used to quickly sweep the area treated by the first pulse or a large flat area to remove residues and achieve uniform surface treatment. The flat-top beam ensures uniformity, while the Gaussian beam has high central energy and can be used to treat localized stubborn contaminants. Scheme 2: First pulse (large core diameter flat-top beam): Performs large-area pretreatment to quickly remove most of the loose contaminants and improve overall efficiency. Second pulse (small core diameter flat-top beam): Then, performs precise cleaning of stubborn points and key areas such as welds that were not completely removed by the first pulse. Scheme 3: First pulse (flat-top beam): Utilizes its uniform energy distribution to perform overall and uniform ablation or vibration peeling of the contaminant layer, avoiding damage to the substrate or uneven cleaning caused by uneven energy. The second pulse (Gaussian light): Utilizing its extremely high central energy, it is used to concentrate on cleaning hard particles or contaminant cores that may remain after the first pulse cleaning and are more tightly bound to the substrate.

[0037] Further, in step c, the dual-path pulsed laser includes a first pulsed laser and a second pulsed laser. The first pulsed laser has a single pulse energy range of 0.5-50 mJ, an overlap rate of 30%-70%, and a galvanometer scanning speed of 1000-40000 mm / s, and is used to break up and peel off the main contaminant layer. The second pulsed laser has a single pulse energy range of 0.5-30 mJ, an overlap rate of 40%-80%, and a galvanometer scanning speed of 1000-50000 mm / s, and is used to remove residual contaminants and beautify the surface of the substrate. The cleaning speed of the laser cleaning head is 5-50 mm / s.

[0038] Furthermore, in step d, the surface rust removal grade is verified by comparing with the standard chart of GB / T8923.1-2011 to determine whether it reaches Sa 2½ or higher, and the cleaning efficiency is verified by calculating the cleaning area per unit time; the surface roughness Rz is in the range of 30-75μm; the cleaning efficiency requirement is: ≥10 μm for flat surfaces. 2 / h, curved surface ≥8 m 2 / h, weld seam ≥20 m / h.

[0039] In the first pulse laser, the single pulse energy E1 is considered low when 0.5mJ ≤ E1 < 25mJ, and high when 25mJ ≤ E1 ≤ 50mJ. The overlap rate R1 is considered medium when 30% ≤ R1 < 45%, and high when 45% ≤ R1 ≤ 70%. In the second pulse laser, the single pulse energy E2 is considered low when 0.5mJ ≤ E2 < 15mJ, and high when 15mJ ≤ E2 ≤ 30mJ. The overlap rate R2 is considered low when 40% ≤ R2 < 55%, and high when 55% ≤ R2 ≤ 80%.

[0040] The first laser pulse is a flat-top beam, and the second laser pulse is either a flat-top beam or a Gaussian beam. The first laser pulse is placed in front, followed by the second laser pulse, with a distance of 2-20 mm between them. The energy distribution uniformity of the flat-top beam is ≥80%, and the energy density at the center of the Gaussian beam is 2-5 times that at the edge.

[0041] Example 2 In this embodiment, a laser cleaning head mounted on the end effector of an industrial robot is used to process straight weld seams on complex curved surfaces. The specific process method is as follows: S10. Identification: The area to be cleaned was identified as a continuous, straight weld seam using a 3D laser scanner. LIBS (Liquid Surface Iron) technology determined the contaminant to be a mixture of shop primer and surface rust.

[0042] S20. Strategy Formulation: Based on structural characteristics, a six-axis industrial robot equipped with a laser cleaning head at its end effector is selected. Based on the state of the contaminants, a combination of a first pulse with a small flat-top and a second pulse with a Gaussian surface is adopted.

[0043] The first pulse has a single pulse energy of 25 mJ, an overlap rate of 45%, and a galvanometer scanning speed of 12000 mm / s; the second pulse has a single pulse energy of 2 mJ, an overlap rate of 60%, and a galvanometer scanning speed of 45000 mm / s; the cleaning speed is 20 mm / s.

[0044] S30, Perform cleaning: The robot moves along the weld at a speed of 20 mm / s. First pulse (power density = 3.5 × 10⁻⁶) 5 W / cm 2 The pulse width is 100ns, Vs1 is 12000mm / s, and W1 is 100mm. The second pulse (power density = 8.9 × 10⁻⁶) is then broken up. 6 W / cm 2A laser plasma spectrum, acoustic signal, or reflected light signal (pulse width = 500ns, Vs2 = 40000mm / s, W2 = 100mm) is used for cleaning and surface finishing. The first pulse, with its lower power density and shorter pulse width, is suitable for fine peeling and avoids heat accumulation. The second pulse, with its wider pulse width, sometimes helps to peel off thicker rust layers through thermal vibration. Optimizing the power and pulse width of both pulses in tandem achieves the best cleaning efficiency and effect. During the cleaning process, the laser plasma spectrum, acoustic signal, or reflected light signal are monitored in real time to determine the cleaning progress online and provide feedback adjustment capabilities to prevent over-cleaning or under-cleaning.

[0045] S40. Verification Results: See [link / reference] Figure 2 Visual inspection confirmed that there were no residual oxides or contaminants inside the weld seam, such as undercut and porosity; the roughness tester measured Rz=45.3μm, which is within the ideal range of 30-75μm; by comparing with standard charts, it was confirmed that the rust removal grade reached Sa 2½ level; the cleaning efficiency was calculated to be 25.2m / h, which meets the technical requirement of ≥20m / h for the weld area.

[0046] Example 3 In this embodiment, a wall-climbing robot equipped with a laser cleaning head is used to process the large, flat surfaces in the ship sections. The specific process is as follows: S10. Identification: Visually identify the area to be cleaned as a flat, straight board frame at the bottom of a ship, characterized by a large area of ​​flat surface, and the contaminant as a large area of ​​intact workshop primer.

[0047] S20. Strategy Formulation: Based on structural characteristics, a wall-climbing robot laser cleaning head is selected. Based on the state of the contaminants, a combination of a large flat-top first pulse and a small flat-top second pulse is adopted.

[0048] The first pulse has a single pulse energy of 33mJ, an overlap rate of 40%, and a galvanometer scanning speed of 10000mm / s; the second pulse has a single pulse energy of 20mJ, an overlap rate of 50%, and a galvanometer scanning speed of 15000mm / s; the cleaning speed is 25mm / s.

[0049] S30, Perform cleaning: First pulse (power density = 1.8 × 10⁻⁶) 5 W / cm 2 Rapid stripping is performed with a pulse width of 200ns, Vs1 = 10000mm / s, and W1 = 200mm. A second pulse (power density = 3.5 × 10⁻⁶) is then applied. 5 W / cm 2 Fine processing was performed on pulse width = 100ns, Vs2 = 15000mm / s, W2 = 200mm.

[0050] S40. Verification Results: See [link / reference] Figure 3 The rust removal grade was confirmed to be Sa 3.0 through standard spectrum comparison; the surface roughness Rz was measured to be 39.2 μm; and the cleaning efficiency reached 13.5 m. 2 / h, satisfying a flat surface ≥10m 2 / h technical requirements. In summary, the present invention has the following advantages: Multi-mode adaptive cleaning: Through three optimized beam and core diameter combinations, it can flexibly cope with various complex cleaning conditions on ship sections (flat surfaces, welds, intact shop primer, floating rust, primer + rust), achieving precise and efficient treatment of different pollutants and geometric features.

[0051] Synergistic efficiency and high quality: The first pulse flat-top light ensures uniform and controllable macroscopic stripping, avoiding local overburning and plasma shielding effects that may be caused by Gaussian light; the multiple options of the second pulse are optimized for different objectives such as "maximizing efficiency" (large flat-top), "precise processing" (small flat-top), and "optimizing quality" (Gaussian light), which solves the contradiction between efficiency and quality of a single laser mode.

[0052] Green and environmentally friendly with zero damage to the substrate: The entire process is dry, dust-free, and low-noise. Non-contact processing ensures that the steel substrate is free from deformation, overcutting, and embedding, perfectly maintaining the structural integrity of the segments and fundamentally solving the inherent defects of the sandblasting process.

[0053] Process versatility and intelligent potential: This process defines a complete parameter system that is applicable to various mounting methods such as industrial six-axis robots, wall-climbing robots, and handheld devices. It is easy to integrate with machine vision and automated control systems, providing core process support for the full automation and intelligentization of secondary rust removal in ship sections.

[0054] Example 4 (1) RSM model construction stage: Acquire scene and contaminant identification information, and establish a 3D model data mapping between cleaning strategies and cleaning effects. Based on the optimization of indicators of laser cleaning head mounting and movement, and the cleaning scheme indicators of dual-path pulsed laser, the laser cleaning strategy is solved. For example, for marine steel AH36, a central composite design and planning experimental scheme is adopted, using five sets of process parameters—energy distribution, single pulse energy, overlap rate, galvanometer scanning speed, and cleaning speed—as input variables. Quality indicators under different combinations of process parameters are collected, and an RSM response surface model is constructed. Y = β0 + ΣβᵢXᵢ + ΣβᵢᵢXᵢ 2 + ΣβᵢⱼXᵢXⱼ + ε Where Xᵢ represents process parameters; and “energy distribution form” is a categorical variable used to define the energy combination of the dual optical paths, such as a flat-top distribution coded as 1 and a Gaussian distribution coded as 2. Y represents quality indicators such as surface rust removal grade (refer to GB / T8923.1-2011), surface roughness Rz, and cleaning efficiency; β is the regression coefficient; and ε is the error term. The surface roughness Rz is evaluated using a utility function based on the target interval: U(Rz) = 1 - |Rz -52.5| / 22.5, when 30 ≤ Rz ≤ 75, otherwise U(Rz) = 0; The cleaning efficiency is set with minimum requirements based on the workpiece structure: for flat surfaces, the minimum cleaning efficiency is ≥10m. 2 / h, curved surface ≥8 m 2 / h, weld seam ≥20m / h.

[0055] Subsequently, the least squares method was used to perform multiple regression analysis on the experimental data to obtain the specific values ​​of the regression coefficients β in the RSM model. This set of coefficients β is the optimal solution that minimizes the sum of squared residuals between the model's predicted values ​​and the experimentally measured values. The error term ε represents the random measurement errors and experimental noise that the model cannot explain.

[0056] The model's average coefficient of determination R has been verified. 2 > 0.91, indicating reliable prediction accuracy.

[0057] (2) Model application stage: Based on the identified pollutant state and structural characteristics, and according to the priority of quality indicators under different application scenarios, the system sets corresponding weight coefficients w1, w2, and w3, and obtains the optimal process parameters by solving the optimization function Maximize [w1·NS(S) + w2·U(Rz) + w3·Nη(η)] (where NS, U, and Nη are the normalization functions of each indicator). This optimization problem is solved using a genetic algorithm.

[0058] Wherein, NS(S) is the normalized value of the surface rust removal grade, Nη(η) is the normalized value of the cleaning efficiency, w1, w2, and w3 are weighting coefficients, and w1 + w2 + w3 = 1. The normalization treatment of the surface rust removal grade adopts the grade mapping method: NS(S) = 0.2, when S = Sa1 level 0.5, when S = Sa2 level 0.8, when S = Sa2.5 level 1.0, when S = Sa3 level The normalization of the cleaning efficiency was performed using the min-max normalization method: Nn(n) = (n - n_min) / (nn_max - n_min) Wherein, η_min is the set minimum cleaning efficiency requirement, and the value is taken as follows according to the workpiece structure: 10m for flat surfaces. 2 / h, curved surface 8 m 2 / h, weld seam 20 m / h; η_max is the set maximum cleaning efficiency target, specifically: flat surface 25 m 2 / h, curved surface 20 m 2 / h, weld seam 35 m / h; By solving this optimization function, the optimal combination of process parameters that meets the requirements of different contaminant states and workpiece structures can be obtained.

[0059] For example: For rust removal applications, set w1=0.5 (weight for high rust removal levels), w2=0.3, and w3=0.2. For applications other than shop primer, set w1=0.3, w2=0.6 (high roughness control weight), and w3=0.1. For applications requiring guaranteed cleaning efficiency: set w1=0.2, w2=0.3, w3=0.5 (weighting for ensuring high cleaning efficiency). (3) Verification and update mechanism: After each cleaning application, the measured data is fed back to the RSM model database. When the amount of new data accumulates to 20% of the original data, the system automatically recalculates the regression coefficient β, achieving iterative updates to the model and continuous improvement in prediction accuracy.

[0060] Example 5 In this embodiment, a handheld laser cleaning head is used to treat the fillet weld areas in the ship sections. The specific process is as follows: S10. Identification: Visually identify the area to be cleaned as a fillet weld in the deck-side area. The contaminant is a mixture of light rust and dust. The system calls up the RSM model specifically for fillet welds.

[0061] S20. Strategy Formulation: Based on structural characteristics, a handheld laser cleaning head was selected. Based on the state of the contaminants, a combination of a first pulse with a large flat top and a second pulse with Gaussian light was chosen.

[0062] With weights set to w1=0.6, w2=0.3, and w3=0.1, the following results were obtained through RSM optimization: The energy distribution combination of the first pulse flat top and the second pulse Gaussian light is adopted: First pulse: single pulse energy 15mJ, overlap rate 50%, galvanometer scanning speed 12000mm / s; Second pulse: single pulse energy 1mJ, overlap rate 62.5%, galvanometer scanning speed 30000mm / s; Cleaning speed: 12mm / s.

[0063] S30, Perform cleaning: First pulse (power density = 2.1 × 10⁻⁶) 5 W / cm 2 (Pulse width = 100ns, Vs1 = 12000mm / s, W1 = 80mm) to remove rust; second pulse (power density = 1.3 × 10⁻⁶) 7 W / cm 2 Surface laser cleaning was performed with pulse width = 500ns, Vs2 = 30000mm / s, W2 = 80mm.

[0064] S40. Verification Results: See [link / reference] Figure 4 Through comparison with standard drawings and visual inspection, it was confirmed that the rust at the root of the fillet weld and at all parts of the weld leg was completely removed and free of rust, and the overall surface rust removal grade reached Sa2. 1 / 2, roughness Rz=48.5μm, cleaning efficiency is 21.1m / h, no visible grease, dirt, scale, rust, coating and foreign impurities.

[0065] In summary, the present invention has the following advantages: (1) Scientific decision-making and precise control: By using RSM response surface analysis, a quantitative relationship between process parameters and cleaning effect is established, and data-driven process parameter optimization is achieved, overcoming the blindness of traditional empirical methods.

[0066] (2) Multi-objective adaptive optimization: A multi-objective optimization function including rust removal level, roughness and cleaning efficiency is established, which can intelligently adjust the weight coefficients according to different pollutant states and structural characteristics to achieve precise process customization.

[0067] (3) Quality consistency assurance: The surface roughness utility function U(Rz) ensures that the Rz value is always in the ideal range of 30-75μm, providing consistent surface quality for subsequent coating processes.

[0068] (4) Continuous learning and improvement: Establish a model update mechanism and continuously optimize the RSM model parameters through continuous data feedback, so that the system has the ability to learn and adapt on its own.

[0069] (5) High efficiency and environmental protection and substrate protection: Parameter optimization based on scientific models ensures the high efficiency and environmental protection of the cleaning process. Non-contact processing ensures no damage to the steel substrate and perfectly maintains the structural integrity of the segments.

[0070] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A dual-path pulsed laser cleaning process for ship sections, characterized in that: The laser cleaning process includes the following steps: a) Scene and contaminant identification: Determine the structural characteristics and contaminant status of the area to be cleaned in the ship section; b) Develop a cleaning strategy: Based on the identification results in step a), determine the mounting and movement mode of the laser cleaning head according to the structural characteristics, and determine the dual-path pulsed laser cleaning scheme according to the contaminant status; c) Execute dual-pulse cleaning: Control the laser cleaning head to move in a predetermined manner and execute the dual-path pulsed laser cleaning scheme; d) Verify the cleaning effect: Measure and evaluate the rust removal grade, surface roughness Rz, and cleaning efficiency of the cleaned surface.

2. The dual-path pulsed laser cleaning process for ship sections according to claim 1, characterized in that: In step a, structural features are identified through visual observation, three-dimensional laser scanning, or visual sensors. The identified structural features are classified into at least one of the following: flat surface, curved surface, weld area, local repair, or hard-to-reach corner edge. The contaminant state is identified through visual observation, multispectral imaging, or laser-induced breakdown spectroscopy. The identified contaminant state is classified into at least one of the following: intact shop primer, rust, or a mixture of shop primer and rust, dust, or oil.

3. The dual-path pulsed laser cleaning process for ship sections according to claim 1, characterized in that: In step b, the mounting and movement of the laser cleaning head are as follows: 1) For areas with complex three-dimensional curved surfaces, the laser cleaning head is mounted on the end of an industrial robot and moves in multiple degrees of freedom; 2) For flat surfaces with a length ≥ 8m or a width ≥ 3m, the laser cleaning head is mounted on a wall-climbing robot and moves in two-dimensional or three-dimensional straight lines; 3) For areas requiring local repair or difficult to reach, the laser cleaning head is moved by hand by the operator.

4. The dual-path pulsed laser cleaning process for ship sections according to claim 1, characterized in that: In step b, the cleaning scheme is as follows: 1) For stubborn contaminants in weld areas, curved surfaces, local or hard-to-reach corner edges, the first and second pulse lasers use low single-pulse energy with a high overlap rate; 2) For large-area flat surfaces, the first and second pulse lasers use high single-pulse energy with a medium overlap rate; 3) For areas requiring aesthetically pleasing surfaces, the first and second pulse lasers use low single-pulse energy with a high overlap rate.

5. The dual-path pulsed laser cleaning process for ship sections according to claim 1, characterized in that: In step c, the dual-path pulsed laser includes a first pulsed laser and a second pulsed laser. The first pulsed laser has a single pulse energy range of 0.5-50 mJ, an overlap rate of 30%-70%, and a galvanometer scanning speed of 1000-40000 mm / s, and is used to break up and peel off the main contaminant layer. The second pulsed laser has a single pulse energy range of 0.5-30 mJ, an overlap rate of 40%-80%, and a galvanometer scanning speed of 1000-50000 mm / s, and is used to remove residual contaminants and beautify the surface of the substrate. The cleaning speed of the laser cleaning head is 5-50 mm / s.

6. The dual-path pulsed laser cleaning process for ship sections according to claim 1, characterized in that: In step d, the surface rust removal grade is verified against the standard chart of GB / T8923.1-2011 to determine whether it reaches Sa 2½ or higher. The cleaning efficiency is verified by calculating the cleaning area per unit time. The surface roughness Rz is in the range of 30-75 μm. The cleaning efficiency requirement is: ≥10 μm for flat surfaces. 2 / h, curved surface ≥8 m 2 / h, weld seam ≥20 m / h.

7. The dual-path pulsed laser cleaning process for ship sections according to claim 4, characterized in that: In the first pulse laser, the single pulse energy E1 is considered low when 0.5mJ ≤ E1 < 25mJ, and high when 25mJ ≤ E1 ≤ 50mJ. The overlap rate R1 is considered medium when 30% ≤ R1 < 45%, and high when 45% ≤ R1 ≤ 70%. In the second pulse laser, the single pulse energy E2 is considered low when 0.5mJ ≤ E2 < 15mJ, and high when 15mJ ≤ E2 ≤ 30mJ. The overlap rate R2 is considered low when 40% ≤ R2 < 55%, and high when 55% ≤ R2 ≤ 80%.

8. A dual-path pulsed laser cleaning process for ship sections according to claim 5, characterized in that: The first pulse laser is a flat-top beam, and the second pulse laser is a flat-top beam or a Gaussian beam. The first pulse laser is in front, and the second pulse laser is behind, with a distance of 2-20mm between them.

9. A dual-path pulsed laser cleaning process for ship sections according to claim 8, characterized in that: The energy distribution uniformity of the flat-top light spot is ≥80%, and the energy density at the center of the Gaussian light spot is 2-5 times that at the edge.

10. A dual-path pulsed laser cleaning process for ship sections according to claim 1, characterized in that: Acquire scene and contaminant identification information, and establish a 3D model data mapping between cleaning strategy and cleaning effect; based on the optimization of the indicators of the mounting and movement mode of the laser cleaning head and the cleaning scheme indicators of dual-path pulsed laser, solve for the laser cleaning strategy; optimize the laser cleaning strategy to obtain the optimal combination of process parameters that meet the requirements of different contaminant states and workpieces.

Images