Self-adaptive titanium alloy laser cladding system and method based on multi-sensor fusion

Abstract

The invention discloses a self-adaptive titanium alloy laser cladding system and method based on multi-sensor fusion, and belongs to the technical field of laser repair. The self-adaptive titanium alloy laser cladding system based on multi-sensor fusion comprises a cladding control module (1), a laser module (2), a powder feeding module (3), a protective gas module (4), a light path reflection module (5), an image acquisition module (6), a temperature acquisition module (7) and a repair workbench (9), the self-adaptive titanium alloy laser cladding method based on multi-sensor fusion comprises the steps of installing a to-be-repaired part, starting cladding operation, collecting image and temperature information and the like. A multi-sensor data fusion method is adopted, operation is easy, safety is high, the cladding repair process is automatically controlled by the controller, and the operation difficulty is lowered.

Description

Technical Field

[0001] The present invention belongs to the technical field of laser repair, and in particular relates to an adaptive titanium alloy laser cladding system and method based on multi-sensor fusion. Background Art

[0002] With the emergence and continuous development of laser technology, laser cladding technology has been widely used in the field of alloy material damage repair. Titanium is known as a smart metal that will surely rise in this century after iron and aluminum, and has good prospects in industrial applications. Titanium alloy has a series of advantages. Due to its low density, high specific strength, and good high-temperature creep resistance, it is used in aviation, aerospace and other fields; due to its non-magnetic and good low-temperature resistance, it can be used in marine engineering structures; due to its non-toxicity, good corrosion resistance and biocompatibility, it can be used in the biomedical field.

[0003] During the production and use of titanium alloy devices, a large number of processing tolerances are caused due to various unavoidable reasons, or cracks, corrosion, wear and other damages occur during use, which requires the titanium alloy material to be repaired. Compared with traditional repair methods, laser cladding repair technology has no thermal deformation, high strength and good controllability of the repair process after processing, so the application of laser cladding technology in the field of titanium alloy material repair is becoming more and more extensive. The existing laser cladding technology has fixed parameters during the device repair process, and it is impossible to adjust the cladding parameters in time according to the actual situation, resulting in the repair effect failing to meet expectations, and even causing unnecessary damage to the normal tissues around the injury. In view of the above problems, an adaptive titanium alloy laser cladding system and method based on multi-sensor fusion is designed. Summary of the invention

[0004] In order to solve the problem that the laser cladding repair effect cannot achieve the expected effect and unnecessary damage is caused to normal tissues around the injury during the cladding process, the present invention provides an adaptive laser cladding system and method based on multi-sensor fusion to achieve the purpose of solving the above problems.

[0005] In order to solve the above technical problems, the technical solution provided by the present invention is:

[0006] An adaptive titanium alloy laser cladding system based on multi-sensor fusion, comprising a cladding control module 1, a laser module 2, a powder feeding module 3, a protective gas module 4, an optical path reflection module 5, an image acquisition module 6, a temperature acquisition module 7, and a repair workbench 9; the laser module 2 comprises a laser mainframe box 20, a water cooler 21, a laser output head 22, a support frame 23, a mainframe box cooling water output pipeline 24, a mainframe box cooling water return pipeline 25, a laser head cooling water output pipeline 26, and a laser head cooling water return pipeline 27; the laser output head 22 and the laser mainframe box 20 are connected by an external optical fiber ; The laser main box 20 and the water cooler 21 are connected through the main box cooling water output pipe 24 and the main box cooling water return pipe 25; the laser output head 22 and the water cooler 21 are connected through the laser head cooling water output pipe 26 and the laser head cooling water return pipe 27; the optical path reflection module 5 includes a primary one-way reflector 50 and a secondary one-way reflector 51; the powder feeding module 3 includes a powder source device 30, a powder nozzle 31 and a carrier gas source 32; the protective gas module 4 includes a protective gas source device 40 and a protective gas nozzle 41; the repair workbench 9 is composed of a loading platform 90 and a clamp 91.

[0007] An adaptive titanium alloy laser cladding method based on multi-sensor fusion has the following steps:

[0008] Step 1: Place the workpiece to be repaired on the repair table in the inert gas protection room, and use a mechanical clamp to firmly fix the accessories to be repaired on the repair table to prevent the workpiece from moving during the repair process. Adjust the angle of the reflector so that the laser is vertically irradiated to the surface of the workpiece to be repaired in the initial state;

[0009] Step 2: The controller plans the cladding path from left to right and from top to bottom, and controls the movement of the laser, reflector, powder feeding nozzle, and shielding gas nozzle according to the set path, so that the laser is always irradiated vertically to the surface of the repaired part during the repair process, and the angle between the two nozzles and the repaired part on the left and right sides of the laser is 65°, and the vertical distance from the repaired part is 20mm;

[0010] Step 3: Start the laser, powder feeding device, jet device, image acquisition device and temperature sensor, and perform cladding operation on the repaired part according to the set path;

[0011] Step 4: The controller processes the image collected by the image sensor to extract the molten pool information. The image processing includes grayscale, filtering and denoising, and image segmentation.

[0012] Step 5: Integrate the extracted molten pool information with the temperature information collected by the temperature sensor, adjust the laser power, powder feeding speed, and gas feeding speed, and complete the laser cladding repair operation.

[0013] Furthermore, the grayscale in step 4 adopts the weighted average method, through the formula G r =0.229×R+0.587×G+0.114×B to calculate the image grayscale, and convert the collected color image into a grayscale image;

[0014] Among them G r is the calculated grayscale value, where R, G, and B are the red, green, and blue grayscale values ​​of the pixel, respectively.

[0015] Furthermore, the filtering and denoising described in step 4 uses the median filtering method to perform median filtering on all pixels in the neighborhood centered at (x, y) to reduce the impact of noise on the processing results. The result is G r (x,y)=med[G r (i,j)], where G r (i, j) is the grayscale value of all pixels in the neighborhood centered at point (x, y).

[0016] Furthermore, the image segmentation in step 4 uses a K-means algorithm to accurately extract feature information of the molten pool.

[0017] Furthermore, in step 5, the relationship between the laser power and various parameters in the cladding process is constructed according to the following formula:

[0018]

[0019] Where T S is the temperature collected by the temperature sensor, T0 is the original temperature of the powder, C is the specific heat capacity of the powder, L is the latent heat of melting of the powder, ρ is the specific heat capacity of the powder, d is the diameter of the molten pool, and v is the cladding speed;

[0020] Set the cladding device operation space set function A: A = [P a ,P b ,M a ,M b ,L a ,L b ]

[0021] The set of operation space functions includes all actions that the controller will take after receiving the real-time temperature Ts and image information, and their meanings are:

[0022] P a :Laser output power increased by 100W

[0023] P b :Laser output power reduced by 100W

[0024] M a :Carrier gas flow rate increased by 0.1L / min

[0025] M b :Carrier gas flow rate decreased by 0.1L / min

[0026] L a : Shielding gas flow rate increased by 0.1L / min

[0027] L b :The protective gas flow rate is reduced by 0.1L / min

[0028] The specific action instructions of the controller are determined by the PID algorithm:

[0029]

[0030] Where u(t) is the controlled variable at the tth moment, K p is the proportional gain, K i is the integral gain, K d is the differential gain, e(t) is the deviation between the expected value and the feedback value at the tth moment;

[0031] The controller controls the laser module, powder feeding module, and shielding gas module to select and execute instructions from the operation function set in real time to achieve the best cladding repair effect.

[0032] Beneficial effects:

[0033] 1. The present invention adopts a reflective lens to change the position of the laser spot, so that the laser can always be vertically irradiated on the surface of the titanium alloy device to be repaired without moving the laser, making the structure of the laser cladding system simpler, reducing the price of the entire cladding system, and at the same time making the system more adaptable to working conditions and improving the practicality of the system.

[0034] 2. The present invention adopts a multi-sensor data fusion method, which collects surface image data of laser cladding effect through an infrared camera and temperature data of cladding working point through a temperature sensor for fusion analysis and processing, effectively reducing the data error caused by a single type of sensor and improving data acquisition accuracy and work efficiency.

[0035] 3. The present invention analyzes the real-time cladding effect and the controller adaptively adjusts the cladding parameters, which can effectively improve the cladding quality and avoid unnecessary damage to normal tissues.

[0036] 4. The present invention is easy to operate and highly safe. The controller automatically controls the cladding repair process, which reduces the difficulty of operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a front view of the structure of the present invention.

[0038] Figure 2 It is a back view of the structure of the present invention. DETAILED DESCRIPTION

[0039] Example 1

[0040] Combined with the accompanying drawings, the specific structure of each part of the present invention is described:

[0041] The hardware structure block diagram and connection diagram used in the present invention are as follows Figure 1 , Figure 2 As shown: the cladding control module 1 is used to receive the image information collected and uploaded by the image acquisition module 6 and the temperature information collected and uploaded by the temperature acquisition module 7, and analyze and process them, and decide the next action according to the processing result; the laser main box 20 is used to provide a laser light source; the water cooler 21 is used to cool the laser main box 20 and the laser output head 22; the laser output head 22 is used for laser output; the powder nozzle 31 is used to spray titanium alloy powder; the powder source device 30 is used to provide titanium alloy powder; the carrier gas source 32 is used to provide a carrier gas flow to transport the titanium alloy powder; the protective gas nozzle 41 is used to output protective gas; the protective gas source device 40 is used to provide protective gas; the primary one-way reflector 50 and the secondary one-way reflector 51 are used to reflect the laser so that the laser remains vertical to the surface of the device to be repaired; the image acquisition module 6 is used to collect and upload the obtained image data; the temperature acquisition module 7 is used to collect and upload the obtained temperature data in real time; the loading platform 90 is used to place the part to be repaired 8; the fixture 91 is used to fix the device to be repaired 8.

[0042] Example 2

[0043] An adaptive titanium alloy laser cladding method based on multi-sensor fusion of the present invention comprises the following steps:

[0044] Step 1: Place the workpiece to be repaired on the repair table in the inert gas protection room, and use a mechanical clamp to firmly fix the accessories to be repaired on the repair table to prevent the workpiece from moving during the repair process. Adjust the angle of the reflector so that the laser is vertically irradiated to the surface of the workpiece to be repaired in the initial state;

[0045] Step 2: The controller plans the cladding path from left to right and from top to bottom, and controls the movement of the laser, reflector, powder feeding nozzle, and shielding gas nozzle according to the set path, so that the laser is always irradiated vertically to the surface of the repaired part during the repair process, and the angle between the two nozzles and the repaired part on the left and right sides of the laser is 65°, and the vertical distance from the repaired part is 20mm;

[0046] Step 3: Start the laser, powder feeding device, jet device, image acquisition device and temperature sensor, and perform cladding operation on the repaired part according to the set path;

[0047] Step 4: grayscale, filter, denoise and segment the image uploaded by the image acquisition device to extract the appearance and morphology information of the molten pool;

[0048] Image grayscale processing uses weighted average method, through the formula G r =0.229×R+0.587×G+0.114×B to calculate the image grayscale and convert the collected color image into a grayscale image, where G r is the calculated grayscale value, where R, G, and B are the red, green, and blue grayscale values ​​of the pixel, respectively.

[0049] Grayscale image filtering and denoising uses the median filtering method to perform median filtering on all pixels in the neighborhood centered at (x, y) to reduce the impact of noise on the processing results. The result is G r (x,y)=med[G r (i,j)], where G r (i, j) is the grayscale value of all pixels in the neighborhood centered at point (x, y).

[0050] The K-means algorithm is used in image segmentation processing to accurately extract the characteristic information of the molten pool.

[0051] Step 5: The extracted molten pool information is integrated with the temperature information collected by the temperature sensor, and the laser power, powder feeding speed, and gas feeding speed are adjusted to complete the laser cladding repair operation;

[0052] Construct the relationship between laser power and various parameters in the cladding process: Where T s is the temperature collected by the temperature sensor, T0 is the original temperature of the powder, C is the specific heat capacity of the powder, L is the latent heat of melting of the powder, ρ is the specific heat capacity of the powder, d is the set melt channel width, and v is the set cladding speed.

[0053] Assume that there is a relationship between the cladding effect R and the laser output power P, carrier gas flow rate M, and shielding gas flow rate L: R = X·P a +Y·M b +Z·L c , where R is the cladding effect index obtained by the controller analyzing the image information uploaded by the image acquisition device, X, Y, and Z are constants, and a, b, and c are the influence coefficients of P, M, and L respectively;

[0054] The cladding effects of different parameter combinations are imported into the deep convolutional neural network model for analysis and calculation training, and the deep convolutional neural network model is continuously optimized to obtain the relationship between the cladding effect R and the laser output power P, carrier gas flow rate M, and shielding gas flow rate L;

[0055] Set the cladding device operation space set function A: A = [P a ,P b ,M a ,M b,L a ,L b ]

[0056] The set of operation space functions includes all actions that the controller will take after receiving the real-time temperature Ts and image information, and their meanings are:

[0057] P a :Laser output power increased by 100W

[0058] P b :Laser output power reduced by 100W

[0059] M a :Carrier gas flow rate increased by 0.1L / min

[0060] M b :Carrier gas flow rate decreased by 0.1L / min

[0061] L a : Shielding gas flow rate increased by 0.1L / min

[0062] L b :The protective gas flow rate is reduced by 0.1L / min

[0063] The specific action instructions of the controller are determined by the PID algorithm: Where u(t) is the controlled variable at the tth moment, K p is the proportional gain, K i is the integral gain, K d is the differential gain, and e(t) is the deviation between the expected value and the feedback value at the tth moment.

[0064] During the laser cladding repair operation, the controller controls the laser module, powder feeding module, and shielding gas module to select and execute instructions from the operation function set in real time to achieve the best cladding repair effect.

Claims

1. An adaptive titanium alloy laser cladding system based on multi-sensor fusion, comprising a cladding control module (1), a laser module (2), a powder feeding module (3), a protective gas module (4), an optical path reflection module (5), an image acquisition module (6), a temperature acquisition module (7), and a repair workbench (9); the laser module (2) comprises a laser mainframe box (20), a water cooler (21), a laser output head (22), a support frame (23), a mainframe box cooling water output pipeline (24), a mainframe box cooling water return pipeline (25), a laser head cooling water output pipeline (26), and a laser head cooling water return pipeline (27); the laser output head (22) and the laser mainframe box (20) are connected by an external optical fiber; The laser mainframe box (20) and the water cooler (21) are connected via a mainframe box cooling water output pipe (24) and a mainframe box cooling water return pipe (25); the laser output head (22) and the water cooler (21) are connected via a laser head cooling water output pipe (26) and a laser head cooling water return pipe (27); the optical path reflection module (5) comprises a primary one-way reflector (50) and a secondary one-way reflector (51); the powder delivery module (3) comprises a powder source device (30), a powder nozzle (31) and a carrier gas source (32); the protective gas module (4) comprises a protective gas source device (40) and a protective gas nozzle (41); and the repair workbench (9) comprises a material loading platform (90) and a fixture (91).

2. An adaptive titanium alloy laser cladding method based on multi-sensor fusion, comprising the following steps: Step 1: Place the workpiece to be repaired on the repair table in the inert gas protection room, and use a mechanical clamp to firmly fix the accessories to be repaired on the repair table to prevent the workpiece from moving during the repair process. Adjust the angle of the reflector so that the laser is vertically irradiated to the surface of the workpiece to be repaired in the initial state; Step 2: The controller plans the cladding path from left to right and from top to bottom, and controls the movement of the laser, reflector, powder feeding nozzle, and shielding gas nozzle according to the set path, so that the laser is always irradiated vertically to the surface of the repaired part during the repair process, and the angle between the two nozzles and the repaired part on the left and right sides of the laser is 65°, and the vertical distance from the repaired part is 20mm; Step 3: Start the laser, powder feeding device, jet device, image acquisition device and temperature sensor, and perform cladding operation on the repaired part according to the set path; Step 4: The controller processes the image collected by the image sensor to extract the molten pool information. The image processing includes grayscale, filtering and denoising, and image segmentation. Step 5: Integrate the extracted molten pool information with the temperature information collected by the temperature sensor, adjust the laser power, powder feeding speed, and gas feeding speed, and complete the laser cladding repair operation.

3. The adaptive titanium alloy laser cladding method based on multi-sensor fusion according to claim 2 is characterized in that: The grayscale in step 4 adopts the weighted average method, through the formula G r =0.229×R+0.587×G+0.114×B to calculate the image grayscale, and convert the collected color image into a grayscale image; Among them G r is the calculated grayscale value, where R, G, and B are the red, green, and blue grayscale values ​​of the pixel, respectively.

4. According to the adaptive titanium alloy laser cladding method based on multi-sensor fusion described in claim 2, the filtering and denoising in step 4 uses the median filtering method to perform median filtering on all pixel points in its neighborhood centered at (x, y) to reduce the impact of noise on the processing result, and the result is G r (x,y)=med[G r (i,j)], where G r (i, j) is the grayscale value of all pixels in the neighborhood centered at point (x, y).

5. According to the adaptive titanium alloy laser cladding method based on multi-sensor fusion described in claim 2, the image segmentation in step 4 uses a K-means algorithm to accurately extract the characteristic information of the molten pool.

6. According to the adaptive titanium alloy laser cladding method based on multi-sensor fusion according to claim 2, in step 5, the relationship between the laser power and various parameters in the cladding process is constructed according to the following formula: Where T S is the temperature collected by the temperature sensor, T0 is the original temperature of the powder, C is the specific heat capacity of the powder, L is the latent heat of melting of the powder, ρ is the specific heat capacity of the powder, d is the diameter of the molten pool, and v is the cladding speed; Set the cladding device operation space set function A: A = [P a ,P b ,M a ,M b ,L a ,L b ] The set of operation space functions includes all actions that the controller will take after receiving the real-time temperature Ts and image information, and their meanings are: P a :Laser output power increased by 100W P b :Laser output power reduced by 100W M a :Carrier gas flow rate increased by 0.1L / min M b :Carrier gas flow rate decreased by 0.1L / min L a : Shielding gas flow rate increased by 0.1L / min L b :The protective gas flow rate is reduced by 0.1L / min The specific action instructions of the controller are determined by the PID algorithm: Where u(t) is the controlled variable at the tth moment, K p is the proportional gain, K i is the integral gain, K d is the differential gain, e(t) is the deviation between the expected value and the feedback value at the tth moment; The controller controls the laser module, powder feeding module, and shielding gas module to select and execute instructions from the operation function set in real time to achieve the best cladding repair effect.

Images