Intelligent inspection and detection method and system for unmanned aerial vehicle in hearth

Abstract

The invention discloses an unmanned aerial vehicle intelligent inspection detection method and system in a hearth, and relates to the technical field of industrial equipment state monitoring, and the method comprises the following steps: transmitting laser wavelength to a target area through a dual-wavelength laser radar carried by an unmanned aerial vehicle, and collecting reflected laser point cloud data and flue gas component data; according to the laser point cloud data, establishing a correlation mechanism of particle size distribution and scattering angles, and calculating the thickness of deposited dust after suspended particle interference is removed; and based on the flue gas component data, constructing a dynamic corrosion rate prediction model in combination with the ash deposition thickness, and outputting a corrosion risk level. And dynamically adjusting the laser scanning frequency of the corresponding area according to the corrosion risk level. Through the constructed dynamic corrosion rate prediction model, three types of core driving factors, namely the sulfate radical concentration, the chloride ion concentration and the ash deposition thickness, in the corrosion process are coupled in a linear addition form, and the change trend of the corrosion risk along with the operation condition can be accurately reflected.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment status monitoring, and in particular to a method and system for intelligent inspection of a drone in a furnace. Background Art

[0002] In recent years, with the widespread adoption of large-scale thermal energy systems such as waste incineration power plants and industrial waste heat boilers, the extreme operating conditions within the furnace, such as high temperature, high corrosion, and high dust levels, have increasingly exacerbated the wear and tear of furnace components (such as superheaters, economizers, and furnace walls). Traditional visual inspections, which rely on manual post-shutdown inspections, suffer from significant shortcomings such as long inspection cycles, low efficiency, high error rates, and the inability to conduct online dynamic assessments.

[0003] In recent years, the application of drones, combining laser scanning, multispectral detection, and high-precision sensors, has become increasingly mature. These drones are capable of performing short, low-altitude flights within high-temperature furnaces and collecting data. However, existing drone inspection systems commonly suffer from the following technical issues: They are unable to accurately detect particle size distribution and deposition characteristics under conditions of high ash concentration and high interference, leading to inaccurate assessments of ash accumulation levels. Furthermore, they lack dynamic corrosion risk prediction models linked to the corrosive components of flue gas, making truly intelligent risk prediction impossible. Summary of the Invention

[0004] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0005] In view of the above problems in the prior art, the present invention is proposed.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a method for intelligent inspection and detection of a drone in a furnace, characterized by comprising the following steps:

[0007] Step 1: Use the dual-wavelength laser radar onboard the drone to emit a pulse sequence of the first laser wavelength λ1 and the second laser wavelength λ2 to the target area, and simultaneously collect reflected laser point cloud data and smoke composition data;

[0008] Step 2: Based on the scattering intensity ratio of λ1 to λ2 in the laser point cloud data, a correlation mechanism between particle size distribution and scattering angle is established, and the dust accumulation thickness h after removing the interference of suspended particles is calculated;

[0009] Step 3: Based on the sulfur oxide concentration and chloride ion concentration in the flue gas composition data and combined with the ash accumulation thickness h, a dynamic corrosion rate prediction model is constructed to output the corrosion risk level R.

[0010] Step 4: Dynamically adjust the laser scanning frequency f of the corresponding area according to the corrosion risk level R, and feed back the adjusted scanning parameters to the laser pulse emission control module in step S1.

[0011] As a preferred solution of the method for intelligent inspection and detection of a drone in a furnace according to the present invention, the correlation mechanism between the particle size distribution and the scattering angle is:

[0012] S101: Based on Mie scattering theory, the particle size distribution is inferred from the laser point cloud data, that is, the calculation formula for the corrected scattering angle is constructed:

[0013]

[0014] Where d represents the median particle size of the suspended particles in the target area, n(λ) represents the refractive index of the particles at wavelength λ, and λ represents the laser wavelength. represents the laser incident angle, θ(λ) represents the scattering angle at wavelength λ;

[0015] S102: Calculate the ratio of θ(λ1) / θ(λ2) based on the scattering angle data extracted from the laser point cloud. Derived particle size distribution: d~N(μ,σ 2 ), μ represents the median particle size, σ 2 represents the variance of the particle size distribution;

[0016] S103: Based on the obtained particle size distribution, when the real-time collected particle concentration C is greater than the threshold concentration C1, it indicates that the point cloud is affected by suspended particles. At this time, the penetration algorithm is automatically triggered:

[0017] h=h meas ·[1+η·(μ·σ 2 / C ref )·(C / C1) -k ];

[0018] Among them, h meas represents the uncorrected dust thickness, h represents the corrected dust thickness, k represents the particle optical response attenuation coefficient, η represents the material characteristic parameter, C ref represents the reference particle concentration.

[0019] As a preferred solution of the method for intelligent inspection of furnaces by drones described in the present invention, the threshold concentration C1 is a dynamic threshold value, which is updated as the detection results are updated, and its update formula is:

[0020]

[0021] where N alarmN is the number of alarms within the current update period. total Indicates the total number of detections within an update period.

[0022] As a preferred embodiment of the method for intelligent inspection and detection of furnaces by drones described in the present invention, the dynamic corrosion rate prediction model is constructed based on the chemical corrosion rate term involving sulfur oxides, pitting corrosion caused by chloride ions, crevice corrosion rate term, and corrosion catalysis term caused by dust accumulation. The model expression is:

[0023]

[0024] Where K represents the predicted corrosion rate; represents the sulfate concentration, [Cl - ] represents the chloride ion concentration, α(T) and β(T) represent the corrosion reaction kinetic factors of sulfate and chloride ions related to temperature, respectively, and γ represents the accelerating factor of dust accumulation thickness on corrosion;

[0025] After the model outputs the conclusion, it is determined whether the acidic conditions are met. That is, if the local pH is less than D1, the risk level R assessment is performed, where D1 represents the acidic pH threshold.

[0026] As a preferred solution of the method for intelligent inspection and detection of furnaces by drones described in the present invention, the risk level R is evaluated as follows:

[0027] K < the first threshold, the risk level is R1 safe, indicating that the corrosion rate is low and no treatment is required for the time being;

[0028] The first threshold ≤ K < the second threshold, the risk level is R2, slight corrosion, there is initial corrosion, and continuous monitoring is required;

[0029] The second threshold ≤ K < the third threshold, and the risk level is R3 moderate corrosion, indicating that the corrosion is tending to worsen and the operating conditions need to be adjusted;

[0030] K ≥ the third threshold, the risk level is R4 severe corrosion, indicating high risk and the need for urgent anti-corrosion measures.

[0031] As a preferred solution of the method for intelligent inspection and detection of a drone in a furnace according to the present invention, the laser scanning frequency f is adjusted as follows:

[0032] Build a mapping model that adapts laser scanning strategies to regional differences in corrosion risk:

[0033] f = f0 + Δf(R); where f0 represents the default basic scanning frequency and Δf(R) represents the frequency increment based on the corrosion risk level R;

[0034]

[0035] Among them, f1<f2<f3.

[0036] As a preferred solution of the method for intelligent inspection and detection by drones in furnaces described in the present invention, the pH threshold D1 is determined based on the ash thickness h. The greater the ash thickness h, the easier it is for the local environment to form a humid acidic gas accumulation area, that is, the pH threshold D1 needs to be reduced.

[0037] A detection system applied to the above-mentioned method for intelligent inspection by drones in furnaces, the system comprising:

[0038] Laser pulse emission control module, used to control the frequency, wavelength, and direction of the laser emission pulse; dynamically adjust scanning parameters in response to corrosion level feedback;

[0039] The multi-wavelength scattering detection module captures the scattering signal intensity at different wavelengths; establishes a particle size-scattering angle correlation mechanism, implements multi-wavelength angle ratio inversion, and obtains particle size distribution data;

[0040] The dust accumulation thickness estimation module is used to calculate the thickness of the sediment structure by combining the particle size inversion results, providing key input for the corrosion rate model and dynamic pH threshold determination;

[0041] Flue gas composition sensing module collects sulfur oxide and chloride ion concentrations; it can integrate temperature data for dynamic corrosion rate modeling;

[0042] Corrosion rate modeling and risk assessment module, used to set pH thresholds and determine the corrosion risk level R based on the current acidity;

[0043] The laser scanning frequency adjustment module is used to receive the risk level R and adjust the laser scanning frequency. If R is high, the laser scanning frequency is increased and the adjustment result is fed back to the laser pulse emission control module to complete the control closed loop.

[0044] The central control and path planning module is used to control the UAV's flight path; coordinate the signal input and output of each module to achieve data fusion and strategy execution.

[0045] The present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the steps of the above-mentioned method and system for intelligent inspection and detection of unmanned aerial vehicles in a furnace.

[0046] The present invention also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method and system for intelligent inspection and detection of a drone in a furnace are implemented.

[0047] Beneficial effects of the present invention:

[0048] 1. The present invention estimates the original dust accumulation thickness from the initial point cloud, inverts the particle size through the scattering angle ratio, and determines whether to activate the penetration algorithm based on the concentration condition. Finally, the corrected thickness is obtained, and the concentration threshold is dynamically adjusted for the next cycle detection, which significantly improves the particle size recognition and dust accumulation thickness estimation accuracy in high-concentration interference environments.

[0049] 2. The dynamic corrosion rate prediction model constructed in this paper couples the three core driving factors of the corrosion process—sulfate concentration, chloride ion concentration, and ash deposit thickness—in a linear additive manner. This model has a clear physical basis and engineering feasibility. This model not only reflects the thermodynamic and kinetic behavior of the corrosion reaction but also considers the catalytic effects of external factors such as ash deposits, accurately reflecting the changing trends in corrosion risk as operating conditions change.

[0050] 3. The present invention not only introduces a multi-physics coupling mechanism in corrosion rate prediction, but also realizes a highly robust monitoring system with adaptive feedback function through dynamic linkage with scanning control parameters, which greatly improves the real-time and reliability of flue metal corrosion monitoring and meets the intelligent monitoring needs under complex working conditions.

[0051] 4. The present invention adopts a thickness-driven dynamic pH threshold model, which not only better fits the coupling relationship between flue gas ash accumulation and corrosion behavior in terms of physical mechanism, but also improves the adaptability and accuracy of corrosion activation judgment, which helps to achieve accurate early warning and regional differentiated control. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0053] Figure 1 This is a schematic diagram of the overall structure of an intelligent inspection method and system for drones inside a furnace proposed by the present invention. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0056] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0057] Reference Figure 1 , as one embodiment of the present invention, provides a method and system for intelligent inspection of a furnace by drone, the method comprising the following steps:

[0058] Step 1: Use the dual-wavelength laser radar carried by the drone to emit a pulse sequence of the first laser wavelength λ1 and the second laser wavelength λ2 to the target area, and synchronously collect the reflected laser point cloud data (this is because these laser beams undergo Mie scattering with the flue gas particles in the furnace, forming wavelength-sensitive backscattered point cloud data) and flue gas composition data; In summary, it can be seen that in order to obtain optical scattering information with distinguishable particle size characteristics, the present invention uses a dual-wavelength laser radar system in step S1 to respectively emit laser pulses with wavelengths of λ1 and λ2. The two wavelengths must have obvious light scattering differences (such as 850nm and 1550nm) to ensure that they produce separable scattering characteristics at the same particle size. These laser beams undergo Mie scattering with the flue gas particles in the furnace to form wavelength-sensitive backscattered point cloud data (including intensity information), which are synchronously collected as raw input.

[0059] Step 2: Based on the scattering intensity ratio of λ1 to λ2 in the laser point cloud data, a correlation mechanism between particle size distribution and scattering angle is established to calculate the dust accumulation thickness h after removing the interference of suspended particles.

[0060] Specifically, the correlation mechanism between the particle size distribution and the scattering angle is as follows:

[0061] S101: Based on Mie scattering theory, the particle size distribution is inferred from the laser point cloud data, that is, the calculation formula for the corrected scattering angle is constructed:

[0062]

[0063] Where d represents the median particle size of the suspended particles in the target area, n(λ) represents the refractive index of the particles at wavelength λ, and λ represents the laser wavelength. represents the laser incident angle, θ(λ) represents the scattering angle at wavelength λ, and this formula is used to calculate the angular characteristics at each wavelength from the scattering point cloud;

[0064] S102: Calculate the ratio of θ(λ1) / θ(λ2) based on the scattering angle data extracted from the laser point cloud. Derived particle size distribution: d~N(μ,σ 2 ), μ represents the median particle size (the angle ratios under a series of particle sizes are calculated in advance using Mie scattering theory simulation tools (such as MatScat or self-built MATLAB model) and organized into a reference corresponding table, which can be obtained by looking up the table), σ 2 Indicates the variance of the particle size distribution (obtained by normal fitting a set of particle size data inverted after multiple consecutive laser scans);

[0065] S103: Based on the obtained particle size distribution, when the real-time collected particle concentration C is greater than the threshold concentration C1, it indicates that the point cloud is affected by suspended particles. At this time, the penetration algorithm is automatically triggered:

[0066] h=h meas ·[1+η·(μ·σ 2 / C ref )·(C / C1) -k ];

[0067] Among them, h meas represents the uncorrected dust thickness, h represents the corrected dust thickness, k represents the particle optical response attenuation coefficient, η represents the material characteristic parameter, C ref represents the reference particle concentration.

[0068] Specifically: in step 1, the laser emission module emits pulsed lasers at multiple fixed angles (e.g., 15°, 30°, 45°);

[0069] By measuring the laser pulse echo time difference Δt and the scattering intensity attenuation ΔI, the distance L from the mirror layer to the dust accumulation surface is preliminarily calculated; compared with the reference clean surface distance L0, the thickness h is obtained. meas =L-L0.

[0070] It should also be noted that the threshold concentration C1 is a dynamic threshold, which is updated as the test results are updated. The update formula is:

[0071]

[0072] where N alarm N is the number of alarms within the current update period. totalRepresents the total number of detections within an update cycle. The 0.1 in the formula represents an adjustment factor that controls the speed or magnitude of the threshold drop with each update. 0.1 is an empirical value, a compromise determined through extensive experimentation and simulation, and is suitable for the feedback adjustment mechanisms of most industrial detection systems. The update mechanism expressed by this formula as a whole ensures that the algorithm is continuously optimized with actual usage, reducing the false positive rate.

[0073] Step 3: Based on the sulfur oxide concentration and chloride ion concentration in the flue gas composition data and combined with the ash accumulation thickness h, a dynamic corrosion rate prediction model is constructed to output the corrosion risk level R.

[0074] The dynamic corrosion rate prediction model is constructed based on the chemical corrosion rate term involving sulfur oxides, pitting corrosion caused by chloride ions, crevice corrosion rate term, and corrosion catalysis term caused by dust accumulation. The model expression is:

[0075]

[0076] To expand on this, the first term of the model This item indicates that sulfate in the flue gas reacts with the wall to form an acidic liquid film that corrodes the metal. It means that the reaction is controlled by temperature, reflecting thermal excitation behavior; the higher the temperature, the larger the item is. The corrosion effect is enhanced.

[0077] The second term of the model, β(T)·[Cl - ], which represents the chloride ion corrosion component, due to Cl - It will destroy the oxide film on the metal surface and cause pitting or crevice corrosion.

[0078] The third term in the model, γ·h, represents the physical impact of dust accumulation thickness. It mainly considers that dust accumulation can easily lead to: retention of acidic components; heat accumulation, which intensifies the reaction; and long-term low local pH.

[0079] Where K represents the predicted corrosion rate; represents the sulfate concentration, [Cl - ] represents the chloride ion concentration, α(T) and β(T) represent the corrosion reaction kinetic factors of sulfate and chloride ions related to temperature, respectively, and γ represents the accelerating factor of dust accumulation thickness on corrosion;

[0080] After the model outputs the conclusion, it is determined whether the acidic conditions are met. That is, if the local pH is less than D1, the risk level R assessment is performed, where D1 represents the acidic pH threshold.

[0081] It should be noted that the pH threshold D1 is determined based on the dust accumulation thickness h. The greater the dust accumulation thickness h, the more likely it is that a moist acidic gas accumulation area will form in the local environment, resulting in a lower pH and more severe corrosion, that is, the pH threshold D1 decreases.

[0082] Specifically, the design approach is as follows: corrosion reactions are inherently controlled by an acidic environment, and the hydrophilicity and acid absorption of the ash layer in flue gas significantly enhance local acidity. The thicker the ash layer, the more flue gas and water vapor are trapped, the stronger the acid accumulation, and the higher the corrosion risk. If a fixed pH threshold is used for risk assessment, it is easy to distort the judgment in cases of "light ash accumulation but strong acidity" or "heavy ash accumulation but critical pH".

[0083] A specific example is: when the dust accumulation thickness is 0 (i.e., the flue gas environment is clean), the system defaults to a basic corrosion trigger pH threshold of μ, which is usually derived from the boundary of the steel's corrosion resistance, such as carbon steel at Cl - The common corrosion starting pH in the presence of dust is ≈ 5.5. As the dust thickness h increases, the pH threshold should gradually decrease, indicating that corrosion is more easily activated. For every 1 mm increase in dust accumulation, the pH threshold decreases by ν. This coefficient can be fitted experimentally, and the common range is ν = 0.2 ~ 0.5 mm-1; the following linear equation can be used for adjustment:

[0084] D1 = μ-ν·h; μ represents the basic pH threshold under no-ash deposition conditions. Therefore, the thickness-driven dynamic pH threshold model is adopted. It not only better fits the coupling relationship between flue gas ash deposition and corrosion behavior in terms of physical mechanism, but also improves the adaptability and accuracy of corrosion activation judgment, which helps to achieve accurate early warning and regional differentiated control.

[0085] The risk level R is assessed as follows:

[0086] K is less than the first threshold, and the risk level is R1, which means the corrosion rate is low and no treatment is required.

[0087] The first threshold ≤ K < the second threshold, the risk level is R2 slight corrosion, there is initial corrosion, and continuous monitoring is required.

[0088] The second threshold ≤ K < the third threshold, and the risk level is R3 moderate corrosion, indicating that the corrosion tends to worsen and the operating conditions need to be adjusted. Generally speaking, the adjustment methods are mainly: fuel blending optimization (source control) or dosing / spraying control (neutralization and relief).

[0089] K ≥ the third threshold, the risk level is R4 severe corrosion, indicating high risk and the need for urgent anti-corrosion measures.

[0090] Step 4: Dynamically adjust the laser scanning frequency f of the corresponding area according to the corrosion risk level R, and feed back the adjusted scanning parameters to the laser pulse emission control module in step S1.

[0091] The laser scanning frequency f is adjusted as follows:

[0092] First, a mapping model is constructed that enables the laser scanning strategy to adapt to regional differences in corrosion risk:

[0093] f = f0 + Δf(R); where f0 represents the default basic scanning frequency and Δf(R) represents the frequency increment based on the corrosion risk level R;

[0094]

[0095] Among them, f1<f2<f3. In summary, the frequency f output in step 4 is updated in real time into the control instructions in the module to complete the following logic chain:

[0096] Obtain raw thickness data → determine high-concentration interference → predict corrosion rate and assess risk level R → adjust scanning frequency based on R → feedback adjustment to step 1 → real-time control of emission frequency → form an adaptive dynamic scanning network.

[0097] This achieves the following system-level advantages:

[0098] Resource concentration: Frequent scanning of high-risk areas and less scanning of low-risk areas to optimize system power consumption;

[0099] Improved data quality: higher-resolution change trajectories can be obtained for areas with significant corrosion trends;

[0100] Feedback control closed loop: The laser control module serves not only as a starting unit but also as a dynamic update unit, embodying the concept of intelligent adaptive control.

[0101] This embodiment further provides a furnace drone intelligent inspection and detection system, which is applied to the above-mentioned furnace drone intelligent inspection and detection method. The system includes:

[0102] The laser pulse emission control module is used to control the frequency, wavelength, and direction of the laser emission pulse; dynamically adjusts the scanning parameters in response to corrosion level feedback; the multi-wavelength scattering detection module captures the scattered signal intensity at different wavelengths; establishes a particle size-scattering angle correlation mechanism, implements multi-wavelength angle ratio inversion, and obtains particle size distribution data; and the dust accumulation thickness estimation module is used to combine the particle size inversion results to calculate the thickness of the deposited structure, providing key input for the corrosion rate model and dynamic pH threshold determination.

[0103] The flue gas composition sensing module collects sulfur oxide and chloride ion concentrations; it can integrate temperature data for dynamic corrosion rate modeling; the corrosion rate modeling and risk assessment module is used to set the pH threshold and determine the corrosion risk level R based on the current acidity; the laser scanning frequency adjustment module is used to receive the risk level R and adjust the laser scanning frequency; if R is high, the laser scanning frequency is increased to enhance the detection density and frequency; and the adjustment results are fed back to the laser pulse emission control module to complete the control closed loop; the central control and path planning module is used to control the UAV flight path; coordinate the signal input and output of each module to achieve data fusion and strategy execution.

[0104] This embodiment also provides a computer device, which is suitable for a method and system for intelligent inspection and detection of drones in a furnace, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a method and system for intelligent inspection and detection of drones in a furnace as proposed in the above embodiment.

[0105] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0106] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements a method and system for intelligent inspection and detection of a drone in a furnace as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0107] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for intelligent inspection and detection of a UAV in a furnace, characterized in that: The following steps are involved: Step 1: Use the dual-wavelength laser radar onboard the drone to emit a pulse sequence of the first laser wavelength λ1 and the second laser wavelength λ2 to the target area, and simultaneously collect reflected laser point cloud data and smoke composition data; Step 2: Based on the scattering intensity ratio of λ1 to λ2 in the laser point cloud data, a correlation mechanism between particle size distribution and scattering angle is established, and the dust accumulation thickness h after removing the interference of suspended particles is calculated; Step 3: Based on the sulfur oxide concentration and chloride ion concentration in the flue gas composition data and combined with the ash accumulation thickness h, a dynamic corrosion rate prediction model is constructed to output the corrosion risk level R. Step 4: Dynamically adjust the laser scanning frequency f of the corresponding area according to the corrosion risk level R, and feed back the adjusted scanning parameters to the laser pulse emission control module in step S1.

2. The method for intelligent inspection and detection of a furnace by using a drone according to claim 1, characterized in that: The correlation mechanism between the particle size distribution and the scattering angle is: S101: Based on Mie scattering theory, the particle size distribution is inferred from the laser point cloud data, that is, the calculation formula for the corrected scattering angle is constructed: Where d represents the median particle size of the suspended particles in the target area, n(λ) represents the refractive index of the particles at wavelength λ, and λ represents the laser wavelength. represents the laser incident angle, θ(λ) represents the scattering angle at wavelength λ; S102: Calculate the ratio based on the scattering angle data extracted from the laser point cloud: d~N(μ,σ 2 ), μ represents the median particle size, σ 2 represents the variance of the particle size distribution; S103: Based on the obtained particle size distribution, when the real-time collected particle concentration C is greater than the threshold concentration C1, it indicates that the point cloud is affected by suspended particles. At this time, the penetration algorithm is automatically triggered: h=h meas ·[1+η·(μ·σ 2 / C ref )·(C / C1) -k ]; Among them, h meas represents the uncorrected dust thickness, h represents the corrected dust thickness, k represents the particle optical response attenuation coefficient, η represents the material characteristic parameter, C ref represents the reference particle concentration.

3. The method for intelligent inspection and detection of a furnace by using a drone according to claim 2, characterized in that: The threshold concentration C1 is a dynamic threshold, which is updated as the test results are updated. The update formula is: where N alarm N is the number of alarms within the current update period. total Indicates the total number of detections within an update period.

4. The method for intelligent inspection and detection of a furnace by using a drone according to claim 3, characterized in that: The dynamic corrosion rate prediction model is constructed based on the chemical corrosion rate term involving sulfur oxides, pitting corrosion caused by chloride ions, crevice corrosion rate term, and corrosion catalysis term caused by dust accumulation. The model expression is: Where K represents the predicted corrosion rate; represents the sulfate concentration, [Cl - ] represents the chloride ion concentration, α(T) and β(T) represent the corrosion reaction kinetic factors of sulfate and chloride ions related to temperature, respectively, and γ represents the accelerating factor of dust accumulation thickness on corrosion; After the model outputs the conclusion, it is determined whether the acidic conditions are met. That is, if the local pH is less than D1, the risk level R assessment is performed, where D1 represents the acidic pH threshold.

5. The method for intelligent inspection and detection of a furnace by using a drone according to claim 4, characterized in that: The risk level R is assessed as follows: K < the first threshold, the risk level is R1 safe, indicating that the corrosion rate is low and no treatment is required for the time being; The first threshold ≤ K < the second threshold, the risk level is R2 slight corrosion, there is initial corrosion, and continuous monitoring is required; The second threshold ≤ K < the third threshold, and the risk level is R3 moderate corrosion, indicating that the corrosion is tending to worsen and the operating conditions need to be adjusted; K ≥ the third threshold, the risk level is R4 severe corrosion, indicating high risk and the need for urgent anti-corrosion measures.

6. The method for intelligent inspection and detection of a drone in a furnace according to claim 5, characterized in that: The laser scanning frequency f is adjusted as follows: Build a mapping model that adapts laser scanning strategies to regional corrosion risk differences: f = f0 + Δf(R); where f0 represents the default basic scanning frequency and Δf(R) represents the frequency increment based on the corrosion risk level R; Among them, f1<f2<f3.

7. The method for intelligent inspection and detection of a furnace by using a drone according to claim 6, characterized in that: The pH threshold D1 is determined based on the dust accumulation thickness h. The greater the dust accumulation thickness h, the more likely it is that a moist acidic gas accumulation area will form in the local environment, that is, the pH threshold D1 will decrease.

8. A furnace interior drone intelligent inspection and detection system, applied to a furnace interior drone intelligent inspection and detection method according to any one of claims 1 to 7, characterized in that: The system comprises: Laser pulse emission control module, used to control the frequency, wavelength, and direction of the laser emission pulse; dynamically adjust scanning parameters in response to corrosion level feedback; The multi-wavelength scattering detection module captures the scattering signal intensity at different wavelengths; establishes a particle size-scattering angle correlation mechanism, implements multi-wavelength angle ratio inversion, and obtains particle size distribution data; The dust accumulation thickness estimation module is used to calculate the thickness of the sediment structure by combining the particle size inversion results, providing key input for the corrosion rate model and dynamic pH threshold determination; Flue gas composition sensing module collects sulfur oxide and chloride ion concentrations; it can integrate temperature data for dynamic corrosion rate modeling; Corrosion rate modeling and risk assessment module, used to set pH thresholds and determine the corrosion risk level R based on the current acidity; The laser scanning frequency adjustment module is used to receive the risk level R and adjust the laser scanning frequency. If R is high, the laser scanning frequency is increased and the adjustment result is fed back to the laser pulse emission control module to complete the control closed loop. The central control and path planning module is used to control the UAV's flight path; coordinate the signal input and output of each module to achieve data fusion and strategy execution.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for intelligent inspection and detection of a drone in a furnace as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for intelligent inspection and detection of a drone in a furnace according to any one of claims 1 to 7 are implemented.

Images