A method for arranging welding points of a cast aluminum alloy side drive case

Abstract

The present application relates to a kind of welding point arrangement methods of casting aluminum alloy side transmission box, the method uses laser welding equipment, along the preset biomimetic reticular trajectory local melt depth controllable welding is carried out to gear chamber and shell casting blank, as a whole, it is the spiderweb distribution of inside dense and outside sparse;During welding, melt pool state is monitored in real time and laser power and welding speed are dynamically adjusted.The method realizes the local melt depth controllable of weld area by accurately controlling laser power, pulse width and weld point spacing, avoids excessive thermal influence on casting body, ensures that the microstructure of welded joint is dense, pore-free and crack-free, effectively improves the anti-cracking performance and fatigue life of weld area.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of aluminum alloy casting and heat treatment, and particularly relates to a welding point arrangement method for casting an aluminum alloy side transmission case. BACKGROUND

[0002] The side transmission system is a key component of special vehicles, and its main function is to reduce the output speed of the engine and increase the torque to meet the driving and maneuvering needs of the vehicle. The system plays a crucial role in the maneuverability and driving stability of the vehicle. In order to improve the overall performance of special vehicles, lightweight design of the side transmission system is particularly important. Lightweight not only reduces the weight of the vehicle, but also reduces energy loss, thereby improving the maneuverability, acceleration performance and transmission efficiency of the vehicle.

[0003] In traditional design, the side transmission case is usually made of cast steel. In order to achieve the goal of lightweight, ZL205A high-strength cast aluminum alloy is used. However, this material has some challenges in the casting process. Specifically, the casting performance of ZL205A alloy is poor, and it is prone to defects such as hot cracking, segregation, shrinkage and shrinkage porosity. These problems lead to microsegregation, cracks, deformation and insufficient mechanical properties, making the performance and stability of the part unable to meet the design requirements, and the yield rate is as low as less than 30%. These problems seriously limit the implementation and application of lightweight design of the side transmission system.

[0004] Therefore, it is crucial to develop a welding method that can effectively solve the casting defects of ZL205A alloy, in order to improve the performance and stability of the side transmission case and achieve the lightweight design of special vehicles. SUMMARY

[0005] The purpose of the present application is to overcome the shortcomings of the prior art, and to provide a welding point arrangement method for casting an aluminum alloy side transmission case.

[0006] The present application adopts the following technical solutions:

[0007] The application discloses a welding point arrangement method for a cast aluminum alloy side transmission case, characterized by adopting a laser welding device to perform local depth-controllable welding on a gear chamber and an outer shell cast blank along a preset bionic reticular track, wherein the overall welding points are distributed in a cobweb shape, and the density of the welding points gradually increases from inside to outside; during the welding process, laser power and welding speed are dynamically adjusted by monitoring the molten pool state in real time, and the steps include the following: S1: determining welding parameters by a reverse double-BP neural network model according to the size information of the case weld; S2: obtaining information for arranging local unit welding points in a diamond shape according to the welding parameters; constructing a local diamond unit by determining four welding points to form four straight lines of edges, and distributing the diamond welding points in a horizontal and vertical diamond shape along the diamond diagonal lines; and S3: sequentially forming a cobweb structure distribution from local to overall, and completing the whole welding process.

[0008] Preferably, the welding parameters include laser power P, laser pulse width w and welding point spacing d, the reverse double-BP neural network model includes model A and model B, and the reverse double-BP neural network model is used to determine the case weld width and the penetration depth; first, the corresponding predicted welding parameters are obtained according to the preset weld width and penetration depth by model B, the predicted welding parameters are adjusted according to actual requirements and then input into model A, and it is verified whether the weld width and penetration depth obtained under the welding parameter condition meet the requirements; if yes, the process is ended; and if not, the weld width and penetration depth obtained by model A are input into model B for further adjustment.

[0009] Preferably, a comprehensive relationship model is established by a regression analysis method: a comprehensive model of the weld width W is W=a0+a1P+bd, and a comprehensive model of the penetration depth H is H=2.15xP-4.06+cw, wherein a0, a1, b and c are coefficients obtained by regression analysis, P is laser power, d is welding point spacing, and w is laser pulse width.

[0010] Preferably, during the welding process, the diamond side length d and the internal angle θ are adaptively adjusted by controlling laser power, pulse width and welding point spacing; in the dense area, the diamond side length is d1 and the internal angle is θ1; in the loose area, the diamond side length is d2 and the internal angle is θ2, and d1

[0011] Preferably, the monitoring of the welding process is realized by a visual real-time monitoring system, the visual real-time monitoring system includes an image acquisition module, an image processing and analysis module, a control decision module and an execution module; the image acquisition module is used to capture the image of a welding object; the image processing and analysis module is used to process the acquired image and extract relevant features; the control decision module is used to obtain the welding speed according to the extracted relevant features; and the execution module is used to adjust the welding motor speed according to the welding speed.

[0012] Preferably, the relevant features extracted by the image processing and analysis module include the groove temperature, the signal-to-noise ratio (SNR) of the welding object temperature image, the groove surface area, the weld width, the weld height, and the weld surface roughness.

[0013] Preferably, the control decision module first determines whether the SNR is within a qualified range. If yes, the welding speed is obtained according to scheme one; if no, the welding speed is obtained according to scheme two. Scheme one is to obtain the welding speed based on the groove surface area, the groove temperature, the weld width, the weld height, and the weld surface roughness. Scheme two is to obtain the welding speed based on the weld point density at the current position and the groove surface area, the weld width, the weld height, and the weld surface roughness.

[0014] The present application has the following beneficial effects:

[0015] 1) The reverse double BP neural network model is used to accurately predict and adjust the welding parameters (laser power P, laser pulse width w, and weld point spacing d). Through the bidirectional prediction of model A and model B, the optimal parameter combination is quickly matched, which not only improves the welding efficiency and reliability, but also reduces the manual intervention and improves the production efficiency.

[0016] 2) Based on the principle of bionics, the high-strength mechanical distribution characteristics of spider webs are simulated to design a spiderweb-shaped weld point arrangement structure with dense inside and sparse outside, which effectively disperses stress concentration and significantly improves the crack resistance and fatigue life of the weld area. At the same time, the welding parameters are accurately controlled to avoid excessive thermal influence on the casting body, ensuring that the welding joint organization is dense and defect-free.

[0017] 3) High-precision laser welding equipment is used in combination with a visual real-time monitoring system to dynamically adjust the laser power and welding speed, ensuring that the weld is uniformly formed without pores or cracks, and the welding deformation is controlled within ±0.1 mm, which is significantly better than traditional welding methods and greatly improves the welding quality, the dimensional accuracy of the box body, and the structural stability.

[0018] To further understand the features and technical content of the present application, please refer to the following detailed description and drawings of the present application. However, the provided drawings are only used for reference and illustration, and are not used to limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The flowchart of the present application is shown in the figure;

[0020] Figure 2 The neural network diagram of the present application is shown in the figure;

[0021] Figure 3 The rhombus weld point arrangement diagram of the present application is shown in the figure;

[0022] Figure 3 (a) is a diamond cell of the center area of the diamond-shaped solder point arrangement of the present application;

[0023] Figure 3 (b) is a diamond cell of the whole area after the present application is welded into a mesh;

[0024] Figure 3 (c) is a diamond cell of the outer area of the diamond-shaped solder point arrangement of the present application;

[0025] Figure 4 is a schematic diagram of the three-dimensional model of the box body of the present application;

[0026] Figure 4 (a) is a schematic diagram of the three-dimensional model of the gear chamber of the box body of the present application;

[0027] Figure 4 (b) is a schematic diagram of the three-dimensional model of the outer shell of the box body of the present application;

[0028] Figure 5 is a schematic diagram of the porosity simulation results of the outer shell of the present application by gravity casting and low-pressure casting;

[0029] Figure 5 (a) is a schematic diagram of the porosity simulation results of the outer shell of the present application by gravity casting;

[0030] Figure 5 (b) is a schematic diagram of the porosity simulation results of the outer shell of the present application by low-pressure casting. DETAILED DESCRIPTION

[0031] The following is to illustrate the embodiments of the present application by specific specific embodiments, and those skilled in the art can understand the advantages and effects of the present application from the disclosure of the present application. The present application can be implemented or applied by other different specific embodiments, and the details in the present application can be modified and changed based on different viewpoints and applications without departing from the spirit of the present application. In addition, the drawings of the present application are only simple schematic illustrations, not actual size drawings, and it is declared in advance. The following embodiments will further illustrate the related technical content of the present application, but the disclosed content is not used to limit the protection scope of the present application.

[0032] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present patent. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0033] Example 1: As Figure 1 As shown, a method for arranging weld points in a cast aluminum alloy side transmission housing is characterized by employing laser welding equipment to perform locally controlled-penetration welding on the gear chamber and outer casing castings along a preset biomimetic mesh trajectory. The overall weld points are distributed in a spiderweb pattern, with the weld point density gradually increasing from the inside to the outside. During the welding process, the state of the molten pool is monitored in real time, and the laser power and welding speed are dynamically adjusted to ensure uniform weld formation, absence of porosity and cracks, and welding deformation controlled within ±0.1mm. The steps include: S1: Determining welding parameters based on the dimensions of the housing weld using a reverse double BP neural network model; S2: Determining that the local unit weld points are arranged in a rhombus shape based on the welding parameters, constructing a local rhombus unit by determining the linear equations of the four edges using four weld points, and distributing the rhombus weld points laterally and longitudinally along the diagonal of the rhombus; S3: Sequentially forming a spiderweb-like structure distribution from local to overall, with sparse outer edges and dense inner edges, completing the entire welding process.

[0034] Preferably, the welding parameters include laser power P, laser pulse width w, and weld spot spacing d. The inverse double BP neural network model includes model A and model B. The inverse double BP neural network model is used to determine the weld width and penetration depth of the box body: first, model B obtains the corresponding predicted welding parameters based on the preset weld width and penetration depth. After adjusting the predicted welding parameters according to actual needs, the parameters are then input into model A to check whether the weld width and penetration depth obtained under the welding parameters meet the requirements. If they do, the process ends; otherwise, the weld width and penetration depth obtained from model A are input into model B for further adjustment.

[0035] Preferably, the laser power P is 1000 W to 3000 W, the laser pulse width w is 1 ms to 5 ms, and the solder joint spacing is 2 mm to 8 mm.

[0036] Preferred, such as Figure 3 As shown, the equation of the rhombus quadrilateral can be expressed as follows: Let point A be at the origin (0,0), and another vertex B be at (d, 0). θis the angle between adjacent sides AC and AB of the rhombus. The coordinates of vertex C are obtained by rotating point B around point A θ d-cosθ d-sinθ ); the rotation is preferably clockwise, and vertex D is the symmetric point of vertex A with respect to BC, and the coordinates of D are obtained according to the feature that the diagonals of a rhombus are perpendicular to each other and bisect each other: d· (1 + -cosθ), -d-sinθ ， The equation of the straight line on which side AB lies is: y = 0, the equation of the straight line on which side BC lies is: , the equation of the straight line on which side CD lies is: y =-d-sinθ、 The equation of the straight line on which side DA lies is: x.

[0037] The embodiment utilizes the inverse double BP neural network model to accurately predict and adjust the welding parameters (laser power P, laser pulse width w, and weld point spacing d). Through the bidirectional prediction of model A and model B, the optimal parameter combination is quickly matched, which not only improves the welding efficiency and reliability, but also reduces manual intervention and improves production efficiency.

[0038] In addition, based on the principle of bionics, the embodiment simulates the high-strength mechanical distribution characteristics of spider webs and designs a spider web-shaped weld point arrangement structure with dense inside and sparse outside, effectively disperses stress concentration, and significantly improves the crack resistance and fatigue life of the weld area. At the same time, the welding parameters are accurately controlled to avoid excessive heat effect on the casting body, ensuring that the welded joint structure is dense and defect-free.

[0039] Embodiment Two: This embodiment includes all the contents of Embodiment One, as shown in Figure 2 Step S1 of the weld point arrangement method includes:

[0040] (1) By setting the experimental data of weld width W and penetration depth H under different laser power P, laser pulse width w, and weld point spacing d. These data should cover a variety of parameter combinations to ensure the generalization ability of the model. Then, all input parameters and output results are normalized to the interval [0, 1] to improve the training effect of the neural network. The normalization formula is as follows:

[0041] ,

[0042] where x norm is the normalized result, x min is the minimum value of the same type of parameter to be normalized, and x max is the maximum value of the same type of parameter to be normalized.

[0043] ​​​(2) Two BP neural network models are constructed: Model A and Model B. Model A is used to predict the weld width and penetration from the welding parameters, with 3 neurons in the input layer corresponding to P, w, and d, respectively; 2 layers in the hidden layer with 64 neurons in each layer and ReLU as the activation function; and 2 neurons in the output layer corresponding to W and D with linear activation function. Model B is used to predict the welding parameters from the weld width and penetration, with 2 neurons in the input layer corresponding to W and H; the same hidden layer structure as Model A; and 3 neurons in the output layer corresponding to P, w, and d with linear activation function.

[0044] (3) The collected experimental data are used to train Model A and Model B. For Model A, the inputs are P, w, and d, and the outputs are W and H; for Model B, the inputs are W and H, and the outputs are P, w, and d. During the training process, mean squared error (MSE) is used as the loss function, Adam optimizer is used for optimization, and the training period is set to 100 epochs.

[0045] (4) First, input the preset weld width W and penetration H into Model B to obtain the predicted welding parameters P pred , w pred , and d pred . Then, adjust these predicted parameters appropriately according to actual requirements and experience to obtain the adjusted parameters P adj , w adj , and d adj . Next, input the adjusted parameters into Model A to check whether the output weld width W adj and penetration H adj meet the preset requirements. If not, return to Model B for further adjustment until the requirements are met.

[0046] (5) The relationship models between the weld width W and the laser power P and the spot spacing d are established by the least squares method: W= a 0 + a 1 · P+b · d; the relationship model between the penetration H and the laser power P and the laser pulse width w is established as H=2.15·P-4.06+c·w, where a0, a1, b, and c are coefficients obtained by regression analysis, P is the laser power, d is the spot spacing, and w is the laser pulse width. The laser power P is set to 1000 W to 3000 W, the laser pulse width w is set to 1 ms to 5 ms, and the spot spacing d is set to 2 mm to 8 mm. The regression analysis method is used to establish the mathematical model based on the experimental data in Table 1.

[0047] Table 1 Experimental data for establishing the mathematical model

[0048]

[0049] The step S2 of the welding point arrangement method comprises:

[0050] (1) Local diamond-shaped welding point construction, selecting a suitable diamond side length d, selecting an internal angle θ of the diamond, vertex A is located at the origin (0, 0), vertex B is located at (d, 0), θ is the included angle of the diamond adjacent sides AC and AB. The coordinates of vertex C are obtained by rotating the B point around the A point θ at an angle of θ: d · cosθ , - d · sinθ ); vertex D is the symmetric point of vertex A about BC, and the coordinates of D are obtained according to the feature that the diagonals of a diamond are perpendicular to each other and bisect each other: d · (1 + cosθ), -d · sin θ ) ， The straight line equation of side AB is: y = 0, the straight line equation of side BC is: , the straight line equation of side CD is: y =-d · sinθ、 The straight line equation of side DA is: x.

[0051] (2) Forming a spider web structure, distributing welding points horizontally and vertically along the diagonals of the diamond. The specific steps are as follows: horizontally distribute in the diagonal direction of each diamond, sequentially translate the diamond to form a row of diamond units. Vertically distribute on the basis of each horizontal distribution, translate the diamond along the vertical direction to form a column of diamond units.

[0052] As shown in Figure 3 , the step S3 of the welding point arrangement method comprises:

[0053] (1) Outer sparse and inner dense distribution: in the central region, select a diamond unit with a side length of d1 and an internal angle of θ 1, in the outer region, select a diamond unit with a side length of d2 and an internal angle of θ 2, and satisfy d1 < d2, θ 1< θ 2, realize that the local unit welding points of the welding seam area are arranged in a diamond shape, and the diamond welding points are distributed horizontally and vertically along the diagonals of the diamond. Wherein Figure 3 (a) is the diamond unit in the central region, Figure 3 (b) is the diamond unit in the whole region after welding into a net shape, Figure 3 (c) is the diamond unit in the outer region.

[0054] (2) Using high-precision laser welding equipment, welding along the preset diamond trajectory. Through the visual real-time monitoring system, dynamically adjust the laser power and welding speed to ensure that the weld is uniformly formed, free of pores and cracks, and the welding deformation is controlled within ±0.1 mm. Verify the uniformity, absence of pores and cracks of the weld through non-destructive testing (such as ultrasonic testing, X-ray testing). Check whether the welding deformation is controlled within ±0.1 mm.

[0055] As shown in Figure 4 , Figure 4 (a) is a schematic diagram of the gear chamber three-dimensional model of the box body of the present application, Figure 4 (b) is a schematic diagram of the outer shell three-dimensional model of the box body of the present application; as Figure 5 shown, Figure 5 (a) is a schematic diagram of the porosity simulation results of the outer shell gravity casting of the present application, Figure 5 (b) is a schematic diagram of the porosity simulation results of the outer shell low-pressure casting of the present application. That is Figure 5 (a) is a schematic diagram of the porosity of gravity casting using the welding point arrangement method of the present application, Figure 5 (b) is a schematic diagram of the porosity of low-pressure casting using the welding point arrangement method of the present application.

[0056] This embodiment uses high-precision laser welding equipment combined with a visual real-time monitoring system to dynamically adjust the laser power and welding speed, ensuring that the weld is uniformly formed, free of pores and cracks, and the welding deformation is controlled within ±0.1 mm, which is significantly better than traditional welding methods, greatly improving the welding quality and the dimensional accuracy and structural stability of the box body. High-precision laser welding technology is used to improve the welding quality and efficiency.

[0057] Example Three: This example includes all the contents of the above-mentioned Examples One and / or Two. The dynamically adjusting the welding speed through the visual real-time monitoring system includes:

[0058] The visual real-time monitoring system includes an image acquisition module, an image processing and analysis module, a control decision module, and an execution module. The image acquisition module is used to capture images of the welding object, including the relevant images of the groove, the weld, and the temperature. The image processing and analysis module is used to process the acquired images and extract relevant features. The control decision module is used to obtain the welding speed according to the extracted relevant features. The execution module is used to adjust the welding motor speed according to the welding speed.

[0059] Preferably, the relevant features extracted by the image processing and analysis module include the welding object temperature, the signal-to-noise ratio SNR of the welding object temperature image, the groove surface area, the weld width, the weld height, and the weld surface roughness.

[0060] Preferably, the control decision module first determines whether the SNR is in the qualified range, and if yes, the welding speed is obtained according to scheme one, and if no, the welding speed is obtained according to scheme two; scheme one is to obtain the welding speed based on the groove surface area S, the groove temperature PW (i.e. the welding object temperature described above), the weld width W, the weld height G, and the weld surface roughness GH; scheme two is to obtain the weld point density MD at the current position based on the welding track, and to obtain the welding speed according to the weld point density, the groove surface area, the weld width, the weld height, and the weld surface roughness.

[0061] Preferably, the scheme one implementation includes: when , , , and are all greater than or equal to zero and less than the corresponding threshold values, the welding speed is adjusted according to the following formula, otherwise the system alarms an abnormality;

[0062] ;

[0063] wherein is the difference between the groove surface area and the standard groove surface area (cm²) (S-Sa), is the difference between the groove temperature and the standard temperature (℃) (PW-PWa), is the difference between the weld width and the standard width (mm) (W-Wa), is the difference between the weld height and the standard height (mm) (G-Ga), is the difference between the weld surface roughness and the standard roughness (GH’=GH-GHa), and V is the adjusted welding speed (mm / s), is the initial welding speed (mm / s). Wherein, Sa is the standard groove surface area (cm²), HDa is the standard thickness (mm), PWa is the standard temperature (℃), Wa is the standard width (mm), Ga is the standard height (mm), and GHa is the standard smoothness.

[0064] Preferably, the GH can be realized by any one of the following ways:

[0065] Contrast: measures the degree of local variation in an image. The higher the value, the rougher the surface (uneven, strong light and dark contrast).

[0066] Entropy: measures the randomness of the information contained in an image. The higher the value, the rougher the surface (complex and disordered texture).

[0067] ​​​​Standard deviation of weld reinforcement fluctuation: The degree of fluctuation in weld reinforcement height along the length of the weld joint, deviating from its average value. The smaller the standard deviation of weld reinforcement fluctuation, the smoother the weld profile.

[0068] Transition zone slope: Calculate the slope at the junction of the weld reinforcement and the base metal. The smaller the slope, the smoother the transition.

[0069] Preferably, the implementation method of Scheme 2 includes: when the conditions are met , , , If all values ​​are greater than or equal to zero and less than the corresponding threshold, the welding speed is adjusted using the following formula; otherwise, the system alarms to indicate an abnormality.

[0070] ;

[0071] ;

[0072] in The difference between the current solder joint density and the standard density ( =MD-MDa, where MDa is the standard density (the number of solder joints per square centimeter), and T0 is the ambient temperature (degrees Celsius).

[0073] Preferably, after running for a preset time... The welding speed (mm / s) at the previous detection time.

[0074] The embodiment selects the welding speed adjustment mode according to the signal-to-noise ratio (SNR) of the welding object temperature image, avoids the problem that the power and speed adjustment are wrong due to the temperature judgment error of the welding object caused by excessive noise, and simultaneously, when the welding object temperature cannot be detected by visual detection, the welding track of the system can calculate the welding spot (welding spot density) per unit area at the current position, the welding spot density is large, and the current position is greatly affected by the surrounding welding spot temperature, since the present application is for aluminum alloy material, the present application is limited to the improvement based on the material, the weld width and the weld height directly reflect the geometric characteristics of the molten pool, the required laser power and welding speed are calculated through these parameters, the weld forming is ensured to be good, and the weld surface roughness is an important index for measuring the welding quality, which is affected by the power density, the cooling speed and the molten pool stability, the welding speed is obtained based on the welding spot density, the groove surface area, the weld width, the weld height and the weld surface roughness, the optimal power and speed combination can be quickly derived based on the multi-parameter modeling or data-driven method, and the process development cycle is significantly shortened. The ln() function is used to appropriately reduce the influence of roughness on speed, so as to ensure the efficiency and guarantee that the weld quality is within an acceptable range, the exponential function is used to convert the density into the temperature influence on speed, so that the influence is not significantly weakened, the present application reasonably matches the power and the speed, the welding efficiency is maximized under the premise of ensuring the quality, and the unit energy consumption is reduced.

[0075] The above disclosed content is only the preferred feasible embodiment of the present application, and does not limit the protection scope of the present application, so that any equivalent technical change made by applying the content of the present application specification and drawings is included in the protection scope of the present application, and furthermore, the elements can be updated as the technology develops.

Claims

1. A method for arranging weld points on a cast aluminum alloy side transmission housing, characterized in that, Using laser welding equipment, the gear chamber and outer shell castings are locally welded with controllable penetration along a preset biomimetic mesh trajectory. The overall weld points are distributed in a mesh pattern, and the density of the weld points gradually increases from the inside to the outside. The welding process involves real-time monitoring of the molten pool and dynamic adjustment of laser power and welding speed. The steps include: S1: Determining welding parameters using a reverse double BP neural network model based on the dimensions of the weld seam; S2: Obtaining information that makes the local unit weld points arranged in a rhombus shape based on the welding parameters: constructing local rhombus units by determining the linear equations of the four edges through four weld points, and distributing rhombus weld points laterally and longitudinally along the diagonal of the rhombus; S3: Forming a mesh structure distribution from local to overall with a denser outer layer and a sparser inner layer, thus completing the entire welding process.

2. The method for arranging weld points of a cast aluminum alloy side transmission housing according to claim 1, characterized in that, The welding parameters include laser power P, laser pulse width w, and weld spot spacing d. The inverse double BP neural network model includes model A and model B. The inverse double BP neural network model is used to determine the weld width and penetration depth of the box body: first, model B obtains the corresponding predicted welding parameters based on the preset weld width and penetration depth. After adjusting the predicted welding parameters according to actual needs, they are then input into model A to check whether the weld width and penetration depth obtained under the welding parameter conditions meet the requirements. If they do, the process ends; otherwise, the weld width and penetration depth obtained from model A are input into model B for further adjustment.

3. The method for arranging weld points of a cast aluminum alloy side transmission housing according to claim 1, characterized in that, A comprehensive relationship model was established using regression analysis: the comprehensive model for weld width W is W=a0+a1P+bd, and the comprehensive model for weld penetration H is H=2.15×P-4.06+cw, where a0, a1, b, and c are coefficients obtained from regression analysis, P is laser power, d is weld point spacing, and w is laser pulse width.

4. The method for arranging weld points of a cast aluminum alloy side transmission housing according to claim 1, characterized in that, During the welding process, the side length d and interior angle of the rhombus are adaptively adjusted by controlling the laser power, pulse width, and weld point spacing. θ In dense regions, the rhombus has a side length of d1 and interior angles of d1. θ 1. In the loose region, the rhombus has a side length of d2 and interior angles of d2. θ 2, and satisfy d1 < d2, θ 1< θ 2. To achieve a diamond-shaped arrangement of local unit weld points in the weld area, the diamond weld points are distributed horizontally and vertically along the diagonal of the diamond.

5. The method for arranging weld points of a cast aluminum alloy side transmission housing according to claim 4, characterized in that, The welding process is monitored by a real-time visual monitoring system, which includes an image acquisition module, an image processing and analysis module, a control decision module, and an execution module. The image acquisition module is used to capture images of the welding object; the image processing and analysis module is used to process the acquired images and extract relevant features. The control decision module is used to obtain the welding speed based on the extracted relevant features; the execution module is used to adjust the welding motor speed according to the welding speed.

6. The method for arranging weld points of a cast aluminum alloy side transmission housing according to claim 5, characterized in that, The relevant features extracted by the image processing and analysis module include bevel temperature, signal-to-noise ratio (SNR) of the weld object temperature image, bevel surface area, weld width, weld height, and weld surface roughness.

7. The method for arranging weld points of a cast aluminum alloy side transmission housing according to claim 6, characterized in that, The control decision module first determines whether the SNR is within the acceptable range. If so, the welding speed is obtained according to Scheme 1; otherwise, the welding speed is obtained according to Scheme 2. Scheme 1 is to obtain the welding speed based on the bevel surface area, bevel temperature, weld width, weld height, and weld surface roughness. Scheme 2 is to obtain the weld point density at the current position based on the welding trajectory, and obtain the welding speed according to the weld point density, bevel surface area, weld width, weld height, and weld surface roughness.

Images