Intelligent laser de-icing method for aircraft wings

Abstract

The present application relates to the technical field of aircraft deicing, and particularly relates to a kind of intelligent laser deicing method for aircraft wing. First, environmental parameters are obtained by environmental monitoring device;Visual parameters are obtained by visual monitoring device, and airborne parameters are obtained by airborne monitoring unit;The environmental parameters, visual parameters and airborne parameters are transmitted to intelligent control system;Local icing area on the surface of the wing and the type of icing are identified, the wing surface coordinate system is established, and local ice type, local ice thickness, substrate temperature and wing surface normal are obtained;Local icing short-time growth rate model and grid priority function are established;Short-time predicted ice thickness is calculated and obtained, and the risk of leading edge area, slat near and local thick ice area is sorted, to determine the scanning order of the target grid to be deiced;According to the short-time predicted ice thickness and the type of icing, the laser deicing mode is determined, and the long-wave laser is used to heat the surface ice layer while the short-wave laser is used to heat the internal ice layer. The present application solves the core problem of accurate energy control and skin safety protection under complex and variable icing conditions.

Description

Technical Field

[0001] This invention relates to the technical field of aircraft de-icing, and more particularly to an intelligent laser de-icing method for aircraft wings. Background Technology

[0002] Aircraft are susceptible to icing environments such as supercooled water droplets, large supercooled water droplets, wet snow, and freezing rain during taxiing, takeoff, climb, holding, approach, and landing. Icing alters the actual aerodynamic shape of the wing's leading edge and upper surface, reducing lift, increasing drag, and causing a decrease in stall margin and deterioration in handling characteristics. Therefore, solving the problem of wing surface icing is of vital practical significance for ensuring flight safety.

[0003] Currently, active anti-icing and de-icing technologies applied to aircraft surfaces mainly include chemical, mechanical, thermal, and electrothermal anti-icing / de-icing systems.

[0004] Chemical anti-icing / de-icing systems achieve anti-icing or de-icing by pumping anti-icing fluid to the protected surface, but they suffer from problems such as the amount of liquid formulation affecting the effective load, limited protection time, easy leakage, and poor performance under severe icing conditions. Mechanical de-icing systems (such as airbags and electro-pulse de-icing) use mechanical force to break the ice layer, but they generally face challenges such as easily damaging surfaces, generating noise, and difficulty in removing ice layers with strong interfacial adhesion.

[0005] While widely used hot-gas anti-icing / de-icing systems are technologically mature, they suffer from high thermal inertia, low efficiency, heavy piping, and stringent requirements for insulation and leak prevention. They are particularly unsuitable for new aerospace structures with poor thermal conductivity, such as carbon fiber composites. Electrothermal anti-icing / de-icing systems offer fast response, ease of control, and integration with composite materials, but their high energy consumption (often tens to hundreds of kW per unit area) is a major bottleneck restricting their large-scale application.

[0006] In recent years, laser paint stripping and laser surface treatment technologies have demonstrated that lasers can selectively deposit energy based on the differences in absorption, reflection, and penetration of different materials at different wavelengths. Introducing this approach into wing de-icing has the potential for non-contact, directional, and fast-response processing. However, most current laser de-icing research remains at the stage of open-loop irradiation or fixed-parameter irradiation, making it difficult to maintain stable results under real-world operating conditions where wing curvature, attitude changes, ice shape changes, and skin safety boundaries all change simultaneously. Especially during airport taxiing, takeoff, and approach phases, the wing's local normal, relative incoming flow velocity, liquid water content, ice thickness distribution, and skin temperature rise margin are all dynamically changing. If fixed power, fixed spot size, and fixed scanning speed are still used, the following problems easily arise: first, energy cannot be concentrated on high-risk icing areas; second, there is a lack of division of labor between long-wave surface thermal melting and short-wave internal stress fracturing; third, it is difficult to simultaneously constrain skin temperature and thermal stress; and fourth, it is impossible to perform feedforward correction based on the short-term icing growth trend, leading to incomplete de-icing or skin overheating. For example, US patent application US4900891A discloses a "Laser ice removal system." This system removes ice formed on surfaces, such as aircraft wings. A heat-generating laser beam is directed onto the ice-covered surface, thereby evaporating the ice and snow formed thereon. By moving the laser generator along the frozen surface, the ice and snow formed on the entire surface are evaporated.

[0007] Studies on wing icing formation, growth, and its impact on flight behavior have shown that wing icing has significant three-dimensionality, transient nature, and regional variability. Therefore, de-icing systems must not only be able to identify the current icing state, but also have predictive and hierarchical control capabilities.

[0008] In practical applications, the de-icing effect of laser de-icing technology heavily depends on the absorption, penetration, and reflectivity of different materials to the laser. Only when the laser energy is fully absorbed by the ice layer and the skin reflectivity is sufficiently high can the laser energy selectively act on the ice layer and induce thermal stress, causing it to crack. Furthermore, the more precise the energy density matching and the more accurate the action time, the higher the de-icing efficiency. Therefore, the matching degree between laser parameters and material absorption characteristics is a key factor restricting the efficiency and safety of laser de-icing. However, in real-world flight icing environments, the type, thickness, and state of the ice layer change constantly, and the thermal safety threshold of the skin material is stringent. Under these conditions, fixed laser energy parameters are difficult to adapt to dynamically changing de-icing needs, leading to unstable de-icing effects and even skin overheating damage or de-icing failure. Currently, most related research involves open-loop irradiation, which cannot cope with the complex and variable icing conditions during flight. Moreover, due to the difficulty in precisely controlling the energy, it may lead to skin ablation, carbonization, or irreversible degradation of mechanical properties. Therefore, existing technologies cannot solve the core problems of precise energy control and skin safety protection under complex and variable icing conditions. Summary of the Invention

[0009] To address the aforementioned technical problems, the present invention aims to provide an intelligent laser de-icing method for aircraft wings. This method solves the problems of energy shift and local defocusing under the traditional planar assumption by introducing effective incident angle correction for the wing curved surface and high-precision positioning compensation. It combines current ice thickness identification with short-term growth prediction, performing responsive de-icing based on the existing ice layer state and feedforward correction based on short-term growth trends, improving adaptability to dynamic icing during taxiing, takeoff, and approach. A dual-band division of labor mechanism is established, involving long-wave surface thermal melting and short-wave internal stress / interface debonding, reducing energy consumption per unit area and minimizing the risk of localized thermal damage. Skin temperature threshold and thermal stress threshold are simultaneously incorporated into the control closed loop, improving the compatibility and safety of aluminum alloy and composite material skins. Prioritizing leading edge and highly aerodynamically sensitive areas and using residual ice layer termination criteria, differentiated power, dwell time, and scanning speed control are implemented for different grid cells, improving de-icing efficiency in critical areas and reducing invalid scans.

[0010] Addressing the heavy reliance on preset parameters in aircraft wing laser de-icing technology, this invention provides an intelligent laser de-icing method for rapid, safe, and adaptive removal of icing from aircraft wing surfaces. This method can be deployed at fixed airport de-icing stations, ground-based mobile de-icing platforms, or airborne accompanying de-icing platforms. Depending on the deployment method, environmental, visual, and airborne parameters can be acquired from the corresponding platform or the aircraft itself, enabling the identification, prediction, classification, and closed-loop de-icing control of wing surface icing. By introducing a dual-band composite laser source and combining environmental, visual, and airborne parameters with historical icing data to establish an icing growth prediction model, the wing surface icing state can be identified in real time, and a de-icing strategy can be formulated. This forms a complete de-icing technology system encompassing state perception, mode selection, and energy closed-loop control, thereby overcoming the technical bottlenecks of existing laser de-icing technologies and achieving high-efficiency, skin-free dynamic aircraft de-icing, ensuring flight safety. The core component is a dual-band composite laser source, which dynamically switches or combines the output of long-wavelength and short-wavelength lasers based on real-time monitoring of the ice type. This achieves a highly efficient and coordinated de-icing mode, primarily involving rapid surface melting while internal thermal stress assists in breaking up the ice. The state perception system senses the ice layer's state and temperature, identifying ice types (such as frost, clear ice, and mixed ice) online and accurately measuring ice thickness and base temperature, providing crucial feedback data for real-time adjustment of laser power and wavelength. Furthermore, the system includes an icing prediction module that, based on flight parameters and external meteorological data, predicts the icing speed and thickness at different locations on the wing in real time, providing preliminary decision input for the laser system. The mode selection system determines the spot size and position, and the automatic energy control system dynamically adjusts the laser's irradiation position, spot size, energy density, and duration based on predicted and measured data. This ensures maximum energy absorption by the ice layer while strictly controlling the skin temperature rise within safe thresholds, achieving a fundamental leap from "open-loop" to "closed-loop" operation.

[0011] The intelligent laser de-icing method for aircraft wings described in this invention includes the following steps: Step 1: Acquire environmental parameters, including ambient temperature and humidity, and liquid water content, through environmental monitoring devices; acquire visual parameters, including visible light images, infrared thermal images, and 3D point clouds, through visual monitoring devices; acquire airborne parameters, including aircraft attitude and relative incoming flow, through airborne monitoring units; and transmit the environmental parameters, visual parameters, and airborne parameters to the intelligent control system. Step 2: The intelligent control system identifies localized icing areas and icing types on the wing surface based on visual parameters. It establishes a wing surface coordinate system Ω based on environmental, visual, and airborne parameters, and obtains the local ice type ζ(x, y,t), local ice thickness hi(x, y, t), and base temperature T. b (x, y, t) and wing surface normal vector n(x, y, t); establish a local icing short-term growth rate model and grid priority function; Step 3: Obtain short-term predicted ice thickness using a local icing short-term growth rate model; perform risk ranking for the leading edge region, slat area, and locally thick ice area using a grid priority function, and determine the scanning order of the target grid to be de-iced according to the risk ranking results; Step 4: Determine the laser de-icing mode based on the short-term predicted ice thickness and ice type. If the target area is covered with clear ice or mixed ice, or the short-term predicted ice thickness is greater than the critical thickness, then enter the collaborative mode. In the collaborative mode, a long-wavelength laser is used to melt the surface ice layer, while a higher-power short-wavelength laser, allocated according to the collaborative mode power distribution, heats the interior of the ice layer to achieve thermal stress induction and interface debonding. If the target area is covered with thin frost or thin ice, and the short-term predicted ice thickness does not exceed the critical thickness, then enter the rapid melting mode. In the rapid melting mode, a long-wavelength laser is used to melt the surface ice layer, while a low-power short-wavelength laser heats the interior of the ice layer to eliminate local cold spots and uneven adhesion.

[0012] Preferably, in step two, the short-term growth rate model for icing is as follows: In the formula, h i (x, y, t) represents the local ice thickness at time t and coordinate (x, y); β(x, y, t) represents the local water collection efficiency of supercooled water droplets in the relative incoming flow at coordinate (x, y) under the current environment; LWC represents the mass of liquid water droplets per unit volume in the relative incoming air under the current environment; V ∞ n represents the relative inflow velocity. f The freezing fraction, or the fraction of the total mass of liquid water impacting the target area that freezes into ice; ρ i This refers to the ice density, measured in kg / m³, used to convert the increase in ice mass per unit area into a local increase in ice thickness; m melt The rate of reduction in equivalent mass per unit area caused by environmental heat exchange or natural melting during the short-term icing growth prediction phase, m shed It is the rate of reduction of equivalent mass per unit area caused by aerodynamic scouring, vibration, or natural shedding.

[0013] In step three, the short-time predicted ice thickness is obtained using the following formula. In the formula, For short-term ice thickness prediction; h i (x, y, t) represents the measured or identified local ice thickness at coordinate (x, y) at the current time t. Δt is the rate of change of ice thickness obtained from the short-term growth rate model of icing, used to characterize the growth or thinning trend of the ice layer at this location per unit time; Δt is the prediction time step; x and y are the local surface coordinates in the wing surface coordinate system Ω; t is the current time.

[0014] Preferably, in step two, the grid priority function is as follows: In the formula, I LE (x, y) is the leading edge indicator function, which takes the value 1 if the current grid is located at the leading edge, slat, wingtip, or other high-risk icing area, and 0 otherwise; C a (x, y) represents the local aerodynamic sensitivity coefficient; a1, a2, a3, and a4 are weighting coefficients. This is a normalized ice thickness prediction term, used to characterize the ratio of the current grid's predicted ice thickness to the current region's maximum predicted ice thickness or the preset maximum allowable ice thickness. The normalized ice thickness gradient term is used to characterize the degree of non-uniformity in ice thickness variation near the current grid; the intelligent control system determines the scanning order according to the S value from high to low.

[0015] Preferably, in step four, the long-wave laser is set to the 8-11 μm band, preferably 10.6 μm, for rapid surface thermal melting, weakening adhesion, and forming a boundary weakening zone; the short-wave laser is set to the 0.8-1.1 μm band, preferably 1064 nm, for depositing energy into the ice layer to induce local temperature differences, thermal stress, and interface debonding.

[0016] Preferably, in step four, for the current target grid to be de-iced, the effective incident angle of the laser is calculated based on the unit vector of the laser beam direction and the normal vector of the wing surface, and the equivalent spot area and equivalent absorbed heat flux density of the long-wave laser and short-wave laser in the current target grid are corrected based on the effective incident angle.

[0017] Preferably, in step four, the intelligent control system calculates the energy required for de-icing the current target grid based on the local ice thickness, ice layer temperature, base temperature, equivalent absorbed heat flux density of long-wave laser, equivalent absorbed heat flux density of short-wave laser, convective heat transfer loss, radiative heat transfer loss, and heat loss conducted to the skin, and determines the long-wave laser power, short-wave laser power, dwell time, and scanning speed.

[0018] Preferably, in step four, during the de-icing process, the surface temperature of the skin, the thermal stress of the skin, and the information on residual icing are acquired in real time. The allowable temperature threshold and the allowable thermal stress threshold of the skin are used as safety constraints to perform closed-loop adjustments on the long-wave laser power, short-wave laser power, equivalent spot area, dwell time, and scanning speed. When the current target grid meets the residual icing termination criterion, the laser output of the target grid is stopped, and the process moves to the next target grid.

[0019] Preferably, in step one, the environmental monitoring device, visual monitoring device, and laser execution device are installed at a fixed airport de-icing station, a ground-based mobile de-icing platform, or an airborne accompanying de-icing platform, and the airborne parameters are obtained by the aircraft itself or an airborne monitoring unit that is connected to the aircraft in communication.

[0020] Due to the adoption of the above technical solutions: 1. This invention, by introducing effective incident angle correction and high-precision positioning compensation for the wing curved surface, enables the laser to maintain a stable focal spot and effective energy density even when the aircraft's attitude and local normal change continuously, solving the problems of energy shift and local defocusing under the traditional planar assumption. 2. This invention combines current ice thickness identification with short-term growth prediction, not only performing responsive de-icing based on the existing ice layer state, but also performing feedforward correction based on the short-term growth trend, thereby improving the adaptability to dynamic icing during taxiing, takeoff, and approach. 3. This invention establishes a dual-band division of labor mechanism of long-wave surface thermal melting and short-wave internal stress / interface debonding, allowing different wavelengths to undertake different tasks on the surface and inside. Compared with a single-wavelength method of simply melting the entire block of ice, this can reduce energy consumption per unit area and reduce the risk of local thermal damage. 4. This invention incorporates both the aircraft skin temperature threshold and thermal stress threshold into the control closed loop, limiting both temperature rise and thermal stress accumulation, thus improving the adaptability and safety of aluminum alloy skin and composite material skin from the control logic perspective. 5. This invention employs a priority ranking system for leading edges and highly aerodynamically sensitive areas, along with a residual ice layer termination criterion, to implement differentiated power, dwell time, and scanning speed control for different grid cells. This significantly improves de-icing efficiency in critical areas and reduces invalid scans. 6. This invention can be deployed at fixed airport de-icing stations, ground-based mobile de-icing platforms, or airborne accompanying de-icing platforms. Under different deployment methods, environmental, visual, and airborne parameters can be acquired from the corresponding platform or aircraft itself. This achieves non-contact, zoned, and closed-loop de-icing without altering the aircraft's main structure, combining engineering feasibility, scalability, and economy. Attached Figure Description

[0021] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram illustrating the working principle of the intelligent laser de-icing method for aircraft wings according to the present invention. Detailed Implementation

[0023] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this patent, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this patent.

[0024] The intelligent laser de-icing method for aircraft wings described in this invention is characterized by comprising the following steps: Step 1: Obtain environmental parameters including ambient temperature and humidity, and liquid water content through an environmental monitoring device; obtain visual parameters including visible light images, infrared thermal images, and 3D point clouds through a visual monitoring device; obtain airborne parameters including aircraft attitude and relative incoming flow through an airborne monitoring unit; transmit the environmental parameters, visual parameters, and airborne parameters to the intelligent control system; the environmental monitoring device, visual monitoring device, and laser actuator are installed at a fixed airport de-icing station, a ground-based mobile de-icing platform, or an airborne accompanying de-icing platform, and the airborne parameters are obtained by the aircraft itself or by an airborne monitoring unit that is connected to the aircraft in communication.

[0025] Step Two: The intelligent control system identifies localized icing areas and icing types on the wing surface based on visual parameters. Icing types include frost ice, clear ice, and mixed ice. A wing surface coordinate system Ω is established based on environmental, visual, and airborne parameters, and the local ice type ζ(x, y, t) and local ice thickness h are obtained. i (x, y, t), base temperature T b (x, y, t) and wing surface normal vector n(x, y, t); establish a local icing short-term growth rate model and grid priority function.

[0026] The short-term growth rate model for icing is shown below: In the formula, h i (x, y, t) represents the local ice thickness at time t and coordinate (x, y); β(x, y, t) represents the local water collection efficiency of supercooled water droplets in the relative inflow at coordinate (x, y) under the current environment. It characterizes the proportion of supercooled water droplets in the relative inflow that are captured and deposited on the wing surface at coordinate (x, y), and is a dimensionless parameter. A database of local water collection efficiency, expressed by the following formula, can be established using CFD simulation and historical flight icing data.

[0027] Where: x, y are the coordinates of the wing surface; V ∞ α represents the relative incoming flow velocity; δ represents the angle of attack; δ represents the flap / slat deflection angle or airfoil configuration parameter; and MVD represents the median volume diameter of the water droplet. During actual operation, the system reads or interpolates the current mesh's β(x, y, t) from the database based on the current airborne and environmental parameters.

[0028] LWC is the mass of liquid water droplets per unit volume relative to the incoming airflow in the current environment. It mainly includes supercooled water droplets, cloud droplets, and freezing rain droplets, and can be obtained through liquid water content sensors, cloud water content probes, and water droplet spectrometers installed on environmental monitoring devices mounted on aircraft, ground de-icing platforms, or airborne accompanying platforms. V ∞ n represents the relative inflow velocity. f The freezing fraction, which is the fraction of the total mass of liquid water impacting the target area that freezes into ice, is a dimensionless parameter with a value range of 0 ≤ n. f ≤1. During the short-term icing growth prediction phase, n f The value is determined by ambient temperature, substrate temperature, liquid water content, relative inflow velocity, droplet size, and local ice type. During the closed-loop update phase following laser irradiation, n... f Further corrections can be made based on laser thermal input and local thermal equilibrium results; ρ i ρ is the ice density, measured in kg / m³, used to convert the increase in ice mass per unit area into the increase in local ice thickness. The ice density can be obtained by looking up a table in a pre-set ice density database based on the local ice type ζ(x, y, t), or by calibration using icing wind tunnel test data. In a simplified implementation, ρ i It can be taken as a preset constant. In a detailed implementation, ρ i Adjustments are made based on the type of frost ice, clear ice, or mixed ice; m melt The rate of reduction in equivalent mass per unit area caused by environmental heat exchange or natural melting during the short-term icing growth prediction phase, m shed The reduction rate of equivalent mass per unit area is caused by aerodynamic scouring, vibration, or natural shedding; the melting and shedding generated during the laser active de-icing stage are described by the subsequent energy deposition model, local thermal equilibrium equation, and termination criterion.

[0029] The grid priority function is as follows: In the formula, I LE (x, y) is the leading edge indicator function, which takes the value 1 if the current grid is located at the leading edge, slat, wingtip, or other high-risk icing area, and 0 otherwise; C a(x, y) represents the local aerodynamic sensitivity coefficient; a1, a2, a3, and a4 are weighting coefficients, corresponding to the predicted ice thickness weight, ice thickness gradient weight, leading edge / high-risk area weight, and aerodynamic sensitivity weight, respectively, and satisfying a1+a2+a3+a4=1, a1, a2, a3, a4≥0. These weighting coefficients can be preset, look up in a table, or adjusted online based on the aircraft's operational phase, local ice type, predicted ice thickness, ice thickness gradient, wing configuration, local aerodynamic sensitivity, skin safety margin, and energy consumption constraints. This is a normalized ice thickness prediction term, used to characterize the ratio of the current grid's predicted ice thickness to the current region's maximum predicted ice thickness or the preset maximum allowable ice thickness. The normalized ice thickness gradient term is used to characterize the degree of non-uniformity in ice thickness variation near the current grid; the intelligent control system determines the scanning order according to the S value from high to low.

[0030] Step 3: Obtain short-term predicted ice thickness using a local icing short-term growth rate model; rank the risks of the leading edge region, slat area, and locally thick ice area using a grid priority function.

[0031] The short-term predicted ice thickness is obtained from the following formula using the local icing short-term growth rate model. In the formula, For short-time ice thickness prediction, it is used to represent the ice thickness at the wing surface coordinates (x, y) at the prediction time t+Δt; h i (x, y, t) represents the measured or identified local ice thickness at coordinate (x, y) at the current time t. Δt is the ice thickness change rate obtained from the local icing short-term growth rate model, used to characterize the growth or thinning trend of the ice layer at that location per unit time; Δt is the prediction time step, which can be set according to the control system refresh cycle, sensor sampling cycle, or laser scanning cycle; x and y are the local surface coordinates in the wing surface coordinate system Ω; t is the current time.

[0032] Step 4: Determine the laser de-icing mode based on the short-term predicted ice thickness and ice type. If the target area is covered with clear ice or mixed ice, or the short-term predicted ice thickness is greater than the critical thickness, then enter the collaborative mode. In the collaborative mode, a long-wavelength laser is used to melt the surface ice layer, while a higher-power short-wavelength laser, allocated according to the collaborative mode power distribution, heats the interior of the ice layer to achieve thermal stress induction and interface debonding. If the target area is covered with thin frost or thin ice, and the short-term predicted ice thickness does not exceed the critical thickness, then enter the rapid melting mode. In the rapid melting mode, a long-wavelength laser is used to melt the surface ice layer, while a low-power short-wavelength laser heats the interior of the ice layer to eliminate local cold spots and uneven adhesion.

[0033] The long-wavelength laser is set in the 8–11 μm band, preferably 10.6 μm, for rapid surface thermal melting, weakening adhesion, and forming a boundary weakening zone; the short-wavelength laser is set in the 0.8–1.1 μm band, preferably 1064 nm, for depositing energy into the ice layer to induce local temperature differences, thermal stress, and interface debonding.

[0034] During the de-icing process, the intelligent control system calculates the energy required for de-icing the current target grid based on the local ice thickness, ice temperature, base temperature, equivalent absorbed heat flux density of long-wave laser, equivalent absorbed heat flux density of short-wave laser, convective heat transfer loss, radiative heat transfer loss, and heat loss conducted to the skin. It also determines the long-wave laser power, short-wave laser power, dwell time, and scanning speed. For the current target grid to be de-iced, the system calculates the effective incident angle of the laser based on the unit vector of the laser beam direction and the normal vector of the wing surface. Based on the effective incident angle, it corrects the equivalent spot area and equivalent absorbed heat flux density of the long-wave laser and short-wave laser within the current target grid.

[0035] In addition, during the de-icing process, the aircraft skin surface temperature, skin thermal stress, and residual icing information are acquired in real time. The allowable skin temperature threshold and allowable skin thermal stress threshold are used as safety constraints to perform closed-loop adjustments on long-wave laser power, short-wave laser power, equivalent spot area, dwell time, and scanning speed. When the current target grid meets the residual icing termination criterion, the laser output of that target grid is stopped, and the process moves to the next target grid until all target grids meet the residual icing termination criterion.

[0036] Based on the predicted ice thickness and ice type, this invention sets two operating modes: M=1 is the rapid surface melting mode for thin frost / thin ice, and M=2 is the long-wave surface melting + short-wave internal stress / interface debonding synergistic mode for clear ice or mixed ice. The decision rule is as follows: In the above formula, h c The critical thickness for mode switching.

[0037] To adapt to changes in wing curvature and aircraft attitude, this invention further calculates the effective incident angle of the laser: In the formula, d L (t) is the unit vector of the laser beam direction; n(x, y, t) is the unit normal vector of the wing surface at this location; (x,y) are the surface coordinates of the wing after unfolding or parameterization; t is the current time.

[0038] Considering the change in the focal spot projection after oblique incidence, the equivalent area of ​​the current mesh is: In the formula, r λ,1r λ,2 Let denot be the semi-axis length of the corresponding wavelength focal spot in the main axis and secondary axis directions.

[0039] The attenuation law of dual-band laser in ice is as follows: Let λ be the intensity of the laser light at a depth z in the ice layer. α represents the initial light intensity of the laser beam at that wavelength incident on the ice surface. λ The equivalent absorption coefficient of this wavelength in the ice layer; z is the depth coordinate of entering the ice layer along the local normal direction, z=0 represents the ice surface, z=h i This indicates the interface between the ice layer and the skin or substrate.

[0040] The corresponding volumetric heat source is: In the formula, I 0,λ η is the incident laser intensity; λ The overall energy utilization rate is represented by z, where z is the depth coordinate of the ice layer along the local normal direction, z=0 represents the ice surface, and z=h i This represents the interface between the ice layer and the skin or substrate. After obtaining the laser volumetric heat source, the equivalent absorbed heat flux density generated by the laser within the current target mesh is expressed as: The intelligent control system calculates the energy required for de-icing the current target grid based on the local ice thickness, ice temperature, base temperature, equivalent absorbed heat flux density of long-wave laser, equivalent absorbed heat flux density of short-wave laser, convective heat transfer loss, radiative heat transfer loss, and heat loss conducted to the skin. It also determines the long-wave laser power, short-wave laser power, dwell time, and scanning speed.

[0041] After obtaining the dual-band heat source term, this invention establishes the local thermal balance equation for the current target mesh: In the formula, ρ i c represents the ice density within the current target grid. i Specific heat capacity of ice, used to characterize the amount of heat required to raise the temperature of a unit mass of ice by a unit amount; h i T represents the local ice thickness at the current target grid location; i ∂T represents the equivalent temperature of the ice layer within the current target grid; t represents the current time; ∂T i / ∂t is the rate of change of the equivalent temperature of the ice layer with time; L fφ represents the latent heat of fusion of ice, used to characterize the heat absorbed by a unit mass of ice when it changes from a solid to a liquid state; φ is the local melting ratio of the ice layer within the current target grid, with a value ranging from 0 to 1; ∂φ / ∂t is the rate of change of the local melting ratio over time; q abs, sw q represents the equivalent absorbed heat flux density generated by the short-wavelength laser within the current target grid; abs, lw h represents the equivalent absorbed heat flux density generated by the long-wavelength laser within the current target grid. conv T is the convective heat transfer coefficient between the ice surface and the ambient air. s T represents the outer surface temperature of the ice layer. ∞ For ambient temperature; ε s σ is the surface emissivity; SB K is the Stefan-Boltzmann constant; s δ is the equivalent thermal conductivity of the skin; s For skin equivalent thickness; T b The base temperature is used to characterize the temperature near the interface between the ice layer and the skin or base.

[0042] According to the heat balance equation, the basic energy required for de-icing the current target grid can be expressed as: In the formula, E0(x, y, t) is the base de-icing energy of the current target grid; ρ i (x, y, t) represents the ice density within the current target grid; A g h represents the surface area of ​​the current target mesh on the wing surface. The current target mesh is a local discrete region of the target area to be de-iced, selected by the intelligent control system and subjected to laser irradiation after meshing. i (x, y, t) represents the local ice thickness at the current target grid; c i T is the specific heat capacity of ice; m is the melting temperature of the ice; T0(x, y, t) is the initial temperature of the ice layer in the current target mesh before laser irradiation; L f The latent heat of melting ice; φ ∗ The target melting ratio is used to characterize the preset degree of melting that the ice layer within the current target grid needs to achieve, and its value ranges from 0 to φ. ∗ ≤1.

[0043] This allows us to further determine the length of stay: In the formula, t d (x, y, t) represents the coordinates of the laser on the wing surface; (x, y) represents the dwell time on the current target grid, indicating the duration of the long-wavelength and short-wavelength lasers within that grid; E0(x, y, t) represents the base de-icing energy of the current target grid; ηeff The comprehensive effective energy coefficient, taking into account environmental heat dissipation, optical path loss, reflection loss, and system efficiency, is taken in the range of 0 < η. eff ≤1; A g q represents the surface area of ​​the current target mesh on the wing surface. abs, lw(x, y, t) q represents the equivalent absorbed heat flux density generated by the long-wavelength laser within the current target grid; abs, sw(x, y, t) denoted as , where is the equivalent absorbed heat flux density generated by the short-wave laser within the current target grid; x and y are the local surface coordinates in the wing surface coordinate system Ω; and t is the current time.

[0044] Therefore, the energy required for de-icing the current target grid can be expressed as: In the formula, E req(x, y, t) E represents the total energy required for the current target grid to reach the preset de-icing target at time t. 0(x, y, t) A represents the base de-icing energy for the current target grid. g The current target mesh is the surface area of ​​the target mesh on the wing surface. The current target mesh is a local discrete region of the target area to be de-iced, which is selected by the intelligent control system and subjected to laser irradiation after being meshed. conv(x, y, t) q represents the convective heat transfer loss heat flux density between the ice surface of the current target grid and the ambient air; rad(x, y, t) q represents the heat flux density of radiative heat transfer loss between the ice surface of the current target grid and the external environment; cond(x, y, t) The heat flux density transferred from the ice layer of the current target grid to the wing skin or base; t d The dwell time of the laser on the current target grid is initially estimated from the base de-icing energy and corrected during closed-loop control based on feedback from heat loss, residual ice thickness, and skin temperature. The E... req Used for energy consumption statistics, closed-loop correction, and subsequent optimization objective function evaluation, the residence time is primarily calculated from the basic de-icing energy E0.

[0045] The total power of the dual-band laser meets the following requirements: Power allocation between long-wavelength and short-wavelength lasers by mode: Where, γ M This represents the mode-dependent proportion of long-wavelength laser power. When M=1, γ M The value is taken as 0.80~0.95, preferably 0.85; when M=2, γ M The value is typically between 0.50 and 0.70, preferably 0.60. γ MFor the initial value or constraint range of the mode-dependent power allocation, the closed-loop optimization module can fine-tune P within this range. lw and P sw In coaxial synchronous scanning mode, long-wavelength lasers and short-wavelength lasers act on the same current target grid, and both share a dwell time. The scanning speed is: In the formula, l g,lw(x, y, t) and l g,sw(x, y, t) These are the equivalent scan lengths of the long-wavelength laser and the short-wavelength laser within the current target grid, respectively. Both are determined by the equivalent spot size, effective scan width, spot overlap rate, and current target grid area for the corresponding wavelength bands. Due to the differences in spot size, absorption depth, and intended use of the long-wavelength and short-wavelength lasers, l g,lw With l g,sw They can be the same or different.

[0046] The actual absorbed energies of long-wavelength lasers and short-wavelength lasers are respectively: Among them, A g Let q be the current target grid area. abs,lw and q abs,sw These represent the equivalent absorbed heat flux densities of the long and short wavelengths within the current target grid, respectively. Thus, a calculable distribution relationship is established between the power, energy, and scanning speed of the dual-band laser.

[0047] During the de-icing process, the surface temperature, thermal stress, and residual icing information of the skin are acquired in real time. The allowable temperature threshold and allowable thermal stress threshold of the skin are used as safety constraints to perform closed-loop adjustments on the long-wave laser power, short-wave laser power, equivalent spot area, dwell time, and scanning speed. When the current target mesh meets the residual icing termination criterion, the laser output of the target mesh is stopped, and the process moves to the next target mesh.

[0048] To ensure that the skin is not damaged during de-icing, this invention simultaneously defines a temperature threshold and a thermal stress threshold. The skin surface temperature should meet the following requirements: T skin (t) represents the temperature at the monitoring point of the skin material or the surface of the skin; T allow The allowable temperature threshold for the skin material.

[0049] Skin thermal stress satisfies: In the formula, T allow E represents the permissible temperature threshold for the skin. s α is the elastic modulus of the skin. s T is the coefficient of thermal expansion; ref Reference temperature; ν s σ is Poisson's ratio; allow The allowable thermal stress threshold for the material is defined. The skin temperature threshold and thermal stress threshold serve as hard constraints, while the temperature and thermal stress terms in the closed-loop optimization objective function act as safety margin penalty terms, used to reduce risk in advance before the hard thresholds are triggered.

[0050] This invention establishes a closed-loop optimization objective function with "de-icing efficiency - structural safety - energy consumption" as the core: in, u ∗ The optimal combination of control parameters is obtained for the intelligent control system; u is the set of control parameters to be optimized, which includes at least the long-wavelength laser power P. lw Shortwave laser power P sw Long-wavelength equivalent spot area A eff,lw Shortwave equivalent spot area A eff,sw Duration of stay t d and scanning speed v scan N represents the number of control steps or the number of target meshes to be processed in the current optimization time domain; k represents the control step or target mesh number; h res,k h represents the residual ice thickness corresponding to the k-th control step or target grid. tar The target residual ice thickness threshold; T skin,k T represents the skin surface temperature corresponding to the k-th control step or target mesh. allow The allowable temperature threshold for the skin material; σ th,k σ represents the skin thermal stress corresponding to the k-th control step or target mesh. allow P represents the permissible thermal stress threshold for the skin material. lw,k and P sw,k These represent the long-wavelength laser power and short-wavelength laser power corresponding to the k-th control step, respectively; Δt is the control time step; w h , w T , w σ , w E These are the weighting coefficients corresponding to residual ice thickness, skin temperature exceeding limits, skin thermal stress exceeding limits, and energy consumption, respectively.

[0051] The aforementioned objective function is used to comprehensively evaluate the merits of candidate laser control parameters within each control cycle. Specifically, the first term constrains the residual ice thickness, ensuring that the remaining ice layer of the current target grid is as close as possible to or below the target residual ice thickness threshold; the second term constrains the skin temperature, being zero when the skin surface temperature does not exceed the allowable temperature threshold, and incurring a penalty when the skin surface temperature exceeds the allowable temperature threshold; the third term constrains the skin thermal stress, being zero when the thermal stress does not exceed the allowable thermal stress threshold, and incurring a penalty when the thermal stress exceeds the allowable thermal stress threshold; and the fourth term constrains the dual-band laser energy consumption, reducing the total energy consumption of long-wavelength and short-wavelength lasers while meeting de-icing efficiency and structural safety requirements. The intelligent control system obtains the optimal laser power, spot parameters, dwell time, and scanning speed corresponding to the current target grid by minimizing this objective function.

[0052] To avoid overscanning and ineffective heating, a termination criterion for the current target region is set: When R≤R th When this happens, the system terminates laser output in that area and moves on to the next target grid. th This is the residual risk termination threshold for the current target mesh, representing the maximum residual icing risk the system is allowed to retain. The system does not require unlimited heating of the current target mesh to a completely ice-free state; instead, it calculates a comprehensive risk value R based on the residual ice area, residual ice thickness, and skin safety status. When R falls below the set threshold R0... th If the system continues to irradiate, it indicates that the de-icing benefits are minimal and may even increase energy consumption and the risk of thermal damage to the skin. Therefore, the system stops the laser output of the current grid and switches to the next target grid.

[0053] De-icing Mode 1: Closed-loop removal of mixed ice on the wing leading edge during taxiing to takeoff. In the low-temperature, high-humidity airport environment during winter, an aircraft awaiting de-icing taxis to the pre-takeoff support area. The multi-spectral visual monitoring system first acquires visible light images, thermal images, and point cloud information of the wing's leading edge and upper surface. Combining this with the aircraft's real-time attitude and environmental parameters, it identifies areas of exposed / mixed ice on the wing's leading edge. The system calculates the ice thickness growth rate and short-term predicted ice thickness for the current grid, then prioritizes the risk of the leading edge region, areas near the slats, and areas with locally thick ice. If any ice is detected... Greater than the critical thickness h cIf the ice type is clear ice or mixed ice, the system automatically determines to enter mode M=2. In mode M=2, the intelligent control system calculates the effective incident angle of the wing surface, corrects the equivalent spot area after oblique incidence, and then calculates the absorption attenuation and heat source distribution of the long-wavelength 10.6 μm laser and the short-wavelength 1064 nm laser in the ice layer. Subsequently, it calculates the de-icing energy required for the current grid and automatically obtains the power combination, dwell time, and scanning speed of the long-wavelength / short-wavelength lasers. The long-wavelength laser preferentially weakens the surface adhesion and forms a boundary weakening zone, while the short-wavelength laser induces local stress and interface debonding in the internal region, achieving low-damage peeling. During the de-icing process, the system continuously collects information on the skin surface temperature and residual ice in the region, and checks whether the skin temperature and thermal stress are close to the allowable upper limit; when any safety variable approaches the threshold, the closed-loop optimization module immediately reduces the power, increases the spot size, or increases the scanning speed. After the termination condition of the current region is met, the system automatically shuts down the laser in that region and moves to the next high-priority grid. This allows for zoned, quantitative, and adaptive de-icing of dynamically icing wings without modifying the main wing structure.

[0054] De-icing Mode 2: Quick switching operation in thin frost ice scenarios If the system identifies the target area as thin frost and ice and Not exceeding the critical thickness h c If the signal is not received, the system will automatically switch to mode M=1. In this mode, the control system reduces the proportion of shortwave power, focusing on rapid surface melting with longwave, while retaining shortwave low-energy assistance to eliminate local cold spots and uneven adhesion. Compared to thick ice scenarios, this mode can shorten the dwell time, increase the scanning speed, and significantly reduce ineffective energy input, making it suitable for light icing and preventative rapid treatment scenarios.

[0055] The aforementioned de-icing mode 1 and de-icing mode 2 are two typical operating modes triggered by the intelligent laser de-icing method for the same aircraft wing under different icing conditions. In actual operation, the intelligent control system can identify ice type, predict ice thickness, and determine mode for different target grids on the wing surface, enabling different target grids to dynamically switch between de-icing mode 1 and de-icing mode 2 according to the actual icing condition during the same de-icing process.

Claims

1. An intelligent laser de-icing method for aircraft wings, characterized in that, Includes the following steps, Step 1: Acquire environmental parameters, including ambient temperature and humidity, and liquid water content, through environmental monitoring devices; acquire visual parameters, including visible light images, infrared thermal images, and 3D point clouds, through visual monitoring devices; acquire airborne parameters, including aircraft attitude and relative incoming flow, through airborne monitoring units; and transmit the environmental parameters, visual parameters, and airborne parameters to the intelligent control system. Step 2: The intelligent control system identifies local icing areas and icing types on the wing surface based on visual parameters. It establishes a wing surface coordinate system Ω based on environmental, visual, and airborne parameters, and obtains the local icing type ζ(x, y, t) and local icing thickness h. i (x, y, t), base temperature T b (x, y, t) and wing surface normal vector n(x, y, t); establish a local icing short-term growth rate model and grid priority function; Step 3: Obtain short-term predicted ice thickness using a local icing short-term growth rate model; perform risk ranking for the leading edge region, slat area, and locally thick ice area using a grid priority function, and determine the scanning order of the target grid to be de-iced according to the risk ranking results; Step 4: Determine the laser de-icing mode based on the short-term predicted ice thickness and the type of icing. If the target area is covered with open ice or mixed ice, or if the short-term predicted ice thickness is greater than the critical thickness, then the system enters the collaborative mode. In the collaborative mode, a long-wavelength laser is used to melt the surface ice layer, while a higher-power short-wavelength laser, allocated according to the collaborative mode power distribution, heats the interior of the ice layer to achieve thermal stress induction and interface debonding. If the target area is covered with thin frost or thin ice, and the short-term predicted ice thickness does not exceed the critical thickness, then the system enters the rapid melting mode. In the rapid melting mode, a long-wavelength laser is used to melt the surface ice layer, while a low-power short-wavelength laser heats the interior of the ice layer to eliminate local cold spots and uneven adhesion.

2. The intelligent laser de-icing method for aircraft wings according to claim 1, characterized in that: In step two, the short-term growth rate model for icing is shown below. In the formula, h i (x, y, t) represents the local ice thickness at time t and coordinate (x, y); β(x, y, t) represents the local water collection efficiency of supercooled water droplets in the relative incoming flow at coordinate (x, y) under the current environment; LWC represents the mass of liquid water droplets per unit volume in the relative incoming air under the current environment; V ∞ n represents the relative inflow velocity. f The freezing fraction, or the fraction of the total mass of liquid water impacting the target area that freezes into ice; ρ i This refers to the ice density, measured in kg / m³, used to convert the increase in ice mass per unit area into a local increase in ice thickness; m melt The rate of reduction in equivalent mass per unit area caused by environmental heat exchange or natural melting during the short-term icing growth prediction phase, m shed It is the rate of reduction in equivalent mass per unit area caused by aerodynamic scouring, vibration, or natural shedding; In step three, the short-time predicted ice thickness is obtained using the following formula. In the formula, For short-term ice thickness prediction; h i (x, y, t) represents the measured or identified local ice thickness at coordinate (x, y) at the current time t. Δt is the rate of change of ice thickness obtained from the short-term growth rate model of icing, used to characterize the growth or thinning trend of the ice layer at this location per unit time; Δt is the prediction time step; x and y are the local surface coordinates in the wing surface coordinate system Ω; t is the current time.

3. The intelligent laser de-icing method for aircraft wings according to claim 1 or 2, characterized in that: In step two, the grid priority function is as follows: In the formula, I LE (x, y) is the leading edge indicator function, which takes the value 1 if the current grid is located at the leading edge, slat, wingtip, or other high-risk icing area, and 0 otherwise; C a (x, y) represents the local aerodynamic sensitivity coefficient; a1, a2, a3, and a4 are weighting coefficients. This is a normalized ice thickness prediction term, used to characterize the ratio of the current grid's predicted ice thickness to the current region's maximum predicted ice thickness or the preset maximum allowable ice thickness. The normalized ice thickness gradient term is used to characterize the degree of non-uniformity in ice thickness variation near the current grid; the intelligent control system determines the scanning order according to the S value from high to low.

4. The intelligent laser de-icing method for aircraft wings according to claim 1 or 2, characterized in that: In step four, the long-wave laser is set to the 8-11 μm band, preferably 10.6 μm, for rapid thermal melting of the surface layer, weakening adhesion and forming a boundary weakening zone; the short-wave laser is set to the 0.8-1.1 μm band, preferably 1064 nm, for depositing energy into the ice layer to induce local temperature difference, thermal stress and interface debonding.

5. The intelligent laser de-icing method for aircraft wings according to claim 1 or 2, characterized in that: In step four, for the current target grid to be de-iced, the effective incident angle of the laser is calculated based on the unit vector of the laser beam direction and the normal vector of the wing surface, and the equivalent spot area and equivalent absorbed heat flux density of the long-wave laser and short-wave laser in the current target grid are corrected based on the effective incident angle.

6. The intelligent laser de-icing method for aircraft wings according to claim 1 or 2, characterized in that: In step four, the intelligent control system calculates the energy required for de-icing the current target grid based on the local ice thickness, ice temperature, base temperature, equivalent absorbed heat flux density of long-wave laser, equivalent absorbed heat flux density of short-wave laser, convective heat transfer loss, radiative heat transfer loss, and heat loss conducted to the skin. It also determines the long-wave laser power, short-wave laser power, dwell time, and scanning speed.

7. The intelligent laser de-icing method for aircraft wings according to claim 1 or 2, characterized in that: In step four, during the de-icing process, the surface temperature, thermal stress, and residual icing information of the skin are acquired in real time. The allowable temperature threshold and allowable thermal stress threshold of the skin are used as safety constraints to perform closed-loop adjustments on the long-wave laser power, short-wave laser power, equivalent spot area, dwell time, and scanning speed. When the target mesh meets the residual icing termination criterion, the laser output of the target mesh is stopped, and the process moves to the next target mesh.

8. The intelligent laser de-icing method for aircraft wings according to claim 1 or 2, characterized in that: In step one, the environmental monitoring device, visual monitoring device, and laser execution device are installed at a fixed airport de-icing station, a ground-based mobile de-icing platform, or an airborne accompanying de-icing platform. The airborne parameters are obtained by the aircraft itself or by an airborne monitoring unit that is connected to the aircraft in communication.

Images