Wind generating set tower drum welding seam optimization method based on fluid dynamics

Abstract

The invention provides a wind generating set tower tube weld joint optimization method based on hydrodynamics, and relates to the technical field of hydrodynamics, the wind generating set tower tube weld joint optimization method comprises the following steps: obtaining an image of a junction of to-be-welded positions of a wind generating set tower tube, and extracting a junction center line in the image of the junction; according to the extension direction of the boundary center line, a constraint condition for keeping energy continuity of the molten metal at the inflection point of the extension direction is constructed, and the heat flux density formed by the molten metal at the inflection point of the center line is calculated; according to the heat flow density needing to be met in the step S2, the welding heat needing to be absorbed by the to-be-welded material when the welding seam is formed at each inflection point of the tower barrel welding seam is calculated; and a welding parameter-inflection point heat mapping model is constructed, optimal welding parameters are output through a global optimization algorithm, the fatigue strength of a welding seam under the wind load effect is improved, and the service life of a tower drum of the wind generating set is prolonged.

Description

Technical Field

[0001] The invention proposes a wind turbine tower weld optimization method based on fluid dynamics, and relates to the technical field of fluid dynamics. Background Art

[0002] Wind turbine towers have unique structural characteristics. They are usually welded from several sections and need to withstand complex dynamic loads during operation. Therefore, the welded structure of the tower is prone to fatigue damage under long-term random dynamic loads, especially the welded section at the section welds.

[0003] In existing technologies, tower weld optimization methods typically use equivalent fatigue loads to measure the magnitude of fatigue loads and assess fatigue strength. However, due to the limitations of the algorithm itself, the equivalent load may not fully reflect the impact of various fatigue load characteristics on structural strength, which may lead to inaccurate calculation results. Furthermore, the distribution of wind resources in a wind farm varies in time and space. When operating in a wind farm, wind turbines will actively yaw to control the wind. Changes in wind direction can cause cracks in the same direction to experience changes in the force they bear after the rotor is loaded. When optimizing tower welds, it is necessary to consider changes in the inflection point of the crack and optimize it based on the form of the crack inflection point and the heat flow pattern.

[0004] At the same time, the molten metal is subject to a variety of forces during the welding process, including but not limited to gravity, electromagnetic force, surface tension, and thermal convection caused by the welding heat source. These forces cause the molten metal to flow, which in turn affects the formation and quality of the weld. Therefore, fully considering the fluid dynamics of the molten metal during welding to study the flow behavior of the molten metal during welding and the related heat and mass transfer phenomena is beneficial to the optimization of the tower weld. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention proposes a wind turbine tower weld optimization method based on fluid dynamics, comprising the following steps: S1. Obtain an image of the junction of the wind turbine tower to be welded, and extract the junction centerline in the junction image; S2. Based on the extension direction of the boundary centerline, a constraint condition is established for the molten metal to maintain energy continuity at the inflection point of the extension direction, and the heat flux density formed by the molten metal at the inflection point of the centerline is calculated; S3, according to the heat flux density that needs to be met in step S2, calculating the welding heat that the material to be welded needs to absorb when forming a weld at each inflection point of the tower weld; S4, build welding parameter-inflection point heat mapping model, and output the optimal welding parameters through global optimization algorithm. In the preferred embodiment, in step S2, the heat flux density q of the molten metal at the inflection point of the welding process is xy The following constraints need to be met: ; ; Where: u xy is the flow velocity of the molten metal at the inflection point (x, y), including u x ,u y Quantity; is the viscous dissipation function, is the density of the molten metal; k is the thermal conductivity of the material to be welded; is the viscosity of the molten metal; H is the latent heat of phase change of the material to be welded at the inflection point; q xy is the heat flux density of the molten metal at the inflection point (x, y), is the temperature gradient.

[0006] In a preferred embodiment, in step S3, the heat flux density q of the molten metal at the i-th inflection point is set to i and The welding heat Q absorbed by the welding material during the dwell time i The expression is as follows: ; Among them, S i is the effective heat transfer area at the inflection point i.

[0007] In a preferred embodiment, in step S4, the parameter space x of the i-th inflection point is defined as i =[I i ,U i ,t i ], I i is the welding current, U i is the voltage, t i is the residence time at the i-th inflection point, and M is the total number of inflection points; Construct the objective function J(x): ; Q i =ηI i U i t i ; Where η is the thermal efficiency, w i is the inflection point weight, Weld area , is the heat deviation threshold of the i-th inflection point, is the actual heat input at the i-th inflection point, ω is the balance coefficient; Minimize the objective function J(x) until J(x) converges and output the optimal welding parameter space , including the optimized welding current I, voltage U and dwell time t.

[0008] In a preferred embodiment, in step S3, the heat correction value Q is calculated. f , heat correction value Q f is the welding heat Q i The average heat value Q at the inflection point i of the historical big data C The ratio of: Q f =Q i / Q C .

[0009] In a preferred embodiment, the heat correction value Q f Adjust welding parameters to obtain adjusted current ,Voltage and residence time : ; ; ; is the current sensitivity coefficient, is the voltage sensitivity coefficient, is the residence time sensitivity coefficient.

[0010] In a preferred embodiment, The range is 0.6~0.8, The range is 0.2~0.4, The range is -0.6~-0.4, The range is 0.3~0.5.

[0011] In a preferred embodiment, the NSGA-II multi-objective genetic algorithm is used to implement the frontier search of the objective function J(x) and output the welding parameter space that satisfies the optimal coordination between heat deviation and residual stress.

[0012] Compared with the prior art, the present invention has the following beneficial technical effects: 1. Fluid dynamics modeling accurately calculates the heat flux density distribution of the molten metal at the inflection point, resolving the energy discontinuity problem caused by geometric abrupt changes in traditional welding. Energy continuity constraints are established at the inflection point of the tower weld to ensure stable molten metal flow and effectively avoid defects such as porosity and lack of fusion. Experimental results show that this method reduces the internal defect rate of the weld by over 40% and increases fatigue life by 25%, significantly enhancing the structural reliability of the wind turbine under complex wind loads.

[0013] 2. A spatial energy transfer model is established based on the geometric centerline extracted from the grayscale image. The heat absorption of the material is inferred through the heat flux density, and the heat input requirement at each inflection point is quantified. This model breaks through traditional empirical parameter settings and achieves dynamic energy matching in local areas at the millimeter level. The width of the welding heat-affected zone is reduced by 30%, the thermal deformation is reduced by 50%, and dimensional accuracy is significantly improved.

[0014] 3. A parameter-heat mapping model was constructed and combined with a global optimization algorithm to automatically determine the optimal combination of parameters such as current, voltage, and welding speed. This reduced average heat input by 15% while ensuring quality, shortened the welding time of a single tower weld by 20%, reduced argon consumption by 18%, and increased annual production capacity by approximately 35%.

[0015] 4. Realize digital extraction of weld features, replace manual process debugging with fluid constraint models, and globally optimize and output executable process parameters, providing welding robots with autonomous decision-making capabilities and reducing dependence on senior welders. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 Schematic diagram of the main steps of the wind turbine tower weld optimization method based on fluid dynamics of the present invention; Figure 2 Schematic diagram of the entire process of the wind turbine tower weld optimization method based on fluid dynamics of the present invention; Figure 3 This is a schematic diagram of a crack inflection point image of the present invention; Figure 4 Schematic diagram showing the crack centerline as an elongated white area; Figure 5 The effective heat transfer area is the specific area covered by the welding heat; Figure 6 This is a schematic diagram of welding at the right-angle turning point of the wind turbine tower flange; Figure 7 This is a picture of the on-site welding process of the wind turbine tower weld; Figure 8 Schematic diagram comparing welding qualification rate data of the present invention and the prior art; Figure 9 Schematic diagram comparing weld strength data of the present invention and the prior art. DETAILED DESCRIPTION

[0018] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] In the drawings of the specific embodiments of the present invention, in order to better and more clearly describe the working principles of the various components in the system, the connection relationship of the various parts in the device is shown, which only clearly distinguishes the relative position relationship between the various components, and does not constitute a limitation on the signal transmission direction, connection sequence and structural size, size and shape of each part within the component or structure.

[0020] Example 1 like Figure 1-2 As shown, the wind turbine tower weld optimization method based on fluid dynamics of the present invention includes the following steps: S1. Obtain an image of the junction of the wind turbine tower to be welded, and extract the junction centerline in the junction image.

[0021] Acquiring images of the junction of the wind turbine tower where welding is to be performed requires on-site inspection or photography. First, an on-site inspection of the wind turbine tower where welding is to be performed is required to determine the specific location and area where welding is to be performed.

[0022] Use appropriate photographic equipment to photograph these areas. Ensure the images are clear and accurately reflect the details of the area to be welded.

[0023] In a preferred embodiment, a color image needs to be converted into a grayscale image. The conversion can be achieved by calculating the color temperature of the color image and calculating the gradient distribution characteristics of the grayscale values ​​of the pixels of the input image. The gradient is a vector whose direction points to the direction in which the grayscale of the image changes fastest. The gradient magnitude is proportional to the rate of change of the grayscale value. The convolution and gradient calculation operations are applied to each grayscale pixel block in the image.

[0024] Define a threshold to determine which grayscale pixel blocks have gradients high enough to be considered target blocks. Iterate over all pixels marked as target blocks and calculate their center positions. Use curve fitting techniques to determine an approximate centerline for these center points. If needed, optimize the fitted centerline, for example, by smoothing or removing outliers.

[0025] Finally, the centers of all target blocks are fitted into center lines, and the fitted center lines are visually superimposed on the original image for easy observation and verification.

[0026] S2. Based on the extension direction of the boundary centerline, a constraint condition is established for the molten metal to maintain energy continuity at the inflection point of the extension direction, and the heat flux density formed by the molten metal at the inflection point of the centerline is calculated.

[0027] The control of welding heat source and the energy continuity formed at the inflection point are crucial for weld optimization. During welding, the heat source heats the material, causing it to melt and form a weld, such as Figure 4 As shown, it is very important to maintain the energy continuity of the weld at the inflection point during the welding process to ensure the quality and consistency of the weld joint.

[0028] The inflection point is a turning point in the weld path, i.e. the centerline extension path. Figure 3 As shown, the weld may need to be made along a complex path. At these inflection points, the energy provided by the heat source will change due to the change in welding direction, which requires adjustment of the heat source to ensure that the energy input of the weld at the inflection point is continuous and uniform.

[0029] In this embodiment, according to the extension direction of the central block, when the direction changes greatly, an inflection point is generated in the extension direction, so a continuous constraint condition for the molten metal energy needs to be established at the inflection point.

[0030] In a preferred embodiment, the curve of the crack path can be fitted based on curve fitting, and the curvature change can be analyzed to identify the inflection point. The crack edge can also be converted into a chain code representation based on chain code analysis, and sharp changes can be detected in the chain code representation. A machine learning model can also be trained to identify the inflection point of the crack, and the model can be trained based on image features and known inflection points.

[0031] In a preferred embodiment, the inflection points of the crack curve are extracted based on the curvature scale space. Based on the curvature scale space, the crack curve is smoothed using a Gaussian convolution kernel at a large scale to calculate the crack curvature g(u,b), and the points exceeding the set curvature threshold are taken as inflection points.

[0032] The following Gaussian kernel function is used as the smoothing function: ; Where u is the distance from the point on the crack curve to the crack starting point, and b is the standard deviation of the Gaussian function.

[0033] During the welding process, the density affects the mass and volume of the molten metal, which in turn affects the physical and mechanical properties of the weld joint; the flow rate of the molten metal at the inflection point directly affects the mixing of the molten metal and the formation of the weld, which has an important impact on the welding quality; and the thermal conductivity of the material to be welded affects the distribution and transfer of heat during the welding process, thereby affecting the temperature distribution of the weld and the welding quality.

[0034] The constant-pressure specific heat capacity of the material being welded is the amount of heat absorbed per unit mass of the material for each 1K increase in temperature when the pressure remains constant. It determines the material's ability to absorb heat during heating, which in turn influences the rate and distribution of temperature rise during welding. The enthalpy of the material being welded is a thermodynamic state function of the system, representing the sum of the material's internal energy and its pressure potential energy. During welding, changes in enthalpy reflect changes in the system's energy and are a key parameter for analyzing energy conversion and transfer during welding. The viscosity of the molten metal influences its fluidity and uniformity during welding, significantly impacting weld formation and quality.

[0035] Since the molten metal flow is also a fluid, based on fluid dynamics, in order to ensure the continuity of the weld energy at the inflection point, the heat flux density of the molten metal at the inflection point of the centerline during welding needs to meet the following constraints: ; ; Where: u xy is the flow velocity of the molten metal at the inflection point (x, y), including u x ,u y Quantity; is the viscous dissipation function, is the density of the molten metal; k is the thermal conductivity of the material to be welded; is the viscosity of the molten metal; H is the latent heat of phase change of the material to be welded at the inflection point; q xy is the heat flux density of the molten metal at the inflection point (x, y), is the temperature gradient.

[0036] Since the molten metal is approximately an incompressible fluid , then the viscous dissipation function is simplified to:

[0037] According to the above constraints, the heat flux density q of the molten metal at the inflection point (x, y) can be obtained xy .

[0038] Taking S355ML steel for wind turbine tower as an example, the typical values ​​of key parameters in welding are shown in Table 1 below. Table 1 Typical values ​​of key parameters in welding

[0039] In the actual welding process, the melt flow velocity can be measured by high-speed camera and the velocity gradient can be calculated. , for example, at a straight weld: =120s -1 , and at the inflection point: =180s -1 , an increase of 50%.

[0040] The turning point =180s -1 Substituting into the dissipation function: =2×[(180) 2 +(180) 2 ]=12.96×10 4 s -2 .

[0041] In the actual welding process, the inflection point At the moment of solidification, it can reach 10 7 W / m 3 , is the core component of heat flux density; during high-speed welding, The contribution to the heat flux density of the molten metal at the inflection point is 15%~20%.

[0042] S3. Calculate the welding heat that the material to be welded needs to absorb when forming a weld at each inflection point of the tower weld according to the heat flux density that needs to be met in step S2.

[0043] Let the heat flux density q of the molten metal at the i-th inflection point be i and The welding heat Q absorbed by the welding material during the dwell time i The expression is as follows: ; Among them, S i is the effective heat transfer area at the inflection point i. Since the heat to be absorbed by the material to be welded is calculated here, the thermal conductivity of the material to be welded must be considered. The effective heat transfer area is the specific area covered by the welding heat. The effective heat transfer area is as follows: Figure 5 The shaded portion is shown.

[0044] like Figure 6As shown in the figure, in the actual welding process of the right-angle inflection point of the wind turbine tower flange, the material at the inflection point is S355ML steel, the thickness at the inflection point is 25mm, and the effective heat transfer area at the inflection point is S i =15mm 2 ; The heat flux density measured by infrared thermal imager is q i =4.2×10 6 W / m 2 , residence time =0.8s.

[0045] Calorie calculation: =(4.2×10 6 )×(1.5×10 -5 )×0.8=50.4J.

[0046] S4. Construct a welding parameter-inflection point heat mapping model and output the optimal welding parameters through a global optimization algorithm.

[0047] This step ensures that the heat input at each inflection point not only satisfies local optimization and reduces stress concentration, but also meets the mechanical performance requirements of the overall tower structure, such as weld strength and fatigue life.

[0048] S41. Construct a welding parameter-inflection point heat mapping model.

[0049] Define the parameter space x of the i-th inflection point i =[I i ,U i ,t i ], I i is the welding current, U i is the voltage, t i is the residence time at the i-th inflection point, and M is the total number of inflection points.

[0050] Based on the heat conduction formula Q i =ηI i U i t i ; η is the thermal efficiency, and the welding heat Q at each inflection point is established i The mapping of parameters to quantify the contribution of each parameter to the inflection point heat is shown in Table 2.

[0051] Table 2 Welding parameters

[0052] S42, output the optimal welding parameters through global optimization algorithm .

[0053] Construct the objective function J(x): ; Among them, w i is the weight of the i-th inflection point, preferably, the key inflection point w i =0.8, normal inflection point w i =0.2, is the maximum residual stress value in the weld area, is the heat deviation threshold of the i-th inflection point, that is, the maximum relative deviation allowed, such as the key inflection point , ordinary inflection point .

[0054] is the actual heat input at the i-th inflection point, through the actual welding parameter space x, based on Q=ηIUt i Calculated; η is the thermal efficiency, ω is the balance coefficient, ω ∈[0.1,1.0], adjusted according to actual needs.

[0055] Specifically, the thermal efficiency η is determined by the arc thermal efficiency calibration experiment: a known electric power P = I × U is applied to the same welding material S355ML steel, and the actual heat input to the workpiece is measured by calorimetry. , calculate η= / (IUt), the experimental calibration thermal efficiency η ranges from 0.75 to 0.85.

[0056] By minimizing the objective function J(x), the coordinated optimization of multi-inflection point heat deviation and residual stress is achieved, and the heat of each local inflection point and the overall mechanical performance target of the global weld are balanced until J(x) converges and the optimal welding parameter space is output. , including the optimized welding current I, voltage U and dwell time t, as shown in Table 3.

[0057] Table 3 Welding parameter output after optimization

[0058] In a preferred embodiment, the NSGA-II multi-objective genetic algorithm is used to implement the Pareto frontier search of the objective function J(x) and output the welding parameter space that satisfies the optimal coordination between heat deviation and residual stress.

[0059] For each individual X, calculate: Thermal deviation term: f1(X)=

[0060] Residual stress term: f2(X)= Select the closest ideal point f1=0,f2= The welding parameter combination is used as the optimal welding parameter space , is the maximum allowable residual stress value.

[0061] Example 2 In a preferred embodiment, due to the inflection point identification error, the calculated value of the heat required to be absorbed per unit volume of the heat source at each inflection point may have an uncertain deviation. Therefore, it is necessary to calculate a heat correction value. The heat correction value can eliminate the uncertainty of the inflection point identification error. Specifically: Calorie correction value Q f is the welding heat Q i The average heat value Q at the inflection point of historical big data C The ratio of: Q f =Q i / Q C .

[0062] Among them, it is necessary to collect the heat absorbed by the material at the inflection point of historical big data. These heat data include the actual heat input values ​​of the inflection point i at different positions during the welding process.

[0063] Calorie correction value Q f The heat correction value Q is used to adjust and optimize the heat input during welding to ensure welding quality. f Adjust welding parameters to obtain adjusted current ,Voltage and residence time , to ensure that the heat input during welding matches the weld optimization result.

[0064] ; ; ; is the current sensitivity coefficient, which is used to control the response intensity of the current I to the heat correction. The range is 0.6~0.8. The larger the current sensitivity coefficient index is, the more significant the effect of current adjustment on heat input is.

[0065] is the voltage sensitivity coefficient, which is used to control the response intensity of voltage U to heat correction. The range is 0.2~0.4. The smaller the voltage sensitivity coefficient index is, the more conservative the voltage adjustment is.

[0066] is the residence time sensitivity coefficient, which controls the inflection point residence time t i The strength of the response to the heat correction, The range is 0.3~0.5. The larger the residence time sensitivity coefficient is, the more radical the residence time adjustment is.

[0067] In some preferred embodiments, by analyzing the heat correction value Qf The distribution and trend of heat energy can be used to identify possible inflection point problem areas and optimize the welding process; at the same time, the new welding data is fed back to the optimization system to continuously update the heat correction value Q f , to achieve continuous weld optimization and improvement.

[0068] In some specific embodiments, Figure 7 In the welding process shown, the power of the heat source is adjusted according to the thickness, type and welding speed of the welding material to ensure sufficient heat input; for the control of the welding speed, the welding speed needs to be slowed down at the inflection point to ensure that enough heat can be absorbed by the material to avoid discontinuity of the weld.

[0069] In some specific embodiments, the position of the heat source can be further precisely controlled. At the inflection point, the position of the heat source relative to the weld can be adjusted to ensure uniform heat distribution.

[0070] Example 3 As shown in the data comparison in Table 4, the present invention accurately simulates the heat flow at the inflection point through the energy continuity constraint condition, with an error much lower than the fatigue load optimization technology that relies on empirical formulas, reducing weld defects caused by uneven heat input. The present invention's precise calculation of heat absorption makes the weld microstructure more uniform, with higher strength and less fluctuation, thereby improving the tower's ability to resist wind loads. After optimizing the heat flow, the residual stress distribution of the weld is more reasonable, the fatigue crack initiation cycle is extended, the service life is significantly improved, and the operation and maintenance costs are reduced. The precise control of the heat flux density reduces temperature mutations during welding, reduces defects such as undercuts and lack of fusion, and greatly improves the quality of weld formation.

[0071] Table 4 Data comparison

[0072] like Figure 8 As shown, the vertical axis represents the welding forming qualification rate, and the vertical axis represents the number of optimizations. With the increase of optimization iterations, the final forming qualification rate of the present invention can reach 98%, which is 15.3% higher than the final forming qualification rate of 85% of the fatigue load optimization scheme of the traditional technology.

[0073] like Figure 9 As shown, the vertical axis represents the weld strength and the horizontal axis represents time. As time increases, the weld strength of the present invention is always higher than that of the fatigue load optimization solution of the traditional technology.

[0074] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media integrated therein. The available medium can be a magnetic medium or a semiconductor medium, etc.

[0075] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A wind turbine tower weld optimization method based on fluid dynamics, characterized in that: The steps include: S1. Obtain an image of the junction of the wind turbine tower to be welded, and extract the junction centerline in the junction image; S2. Based on the extension direction of the boundary centerline, a constraint condition is established for the molten metal to maintain energy continuity at the inflection point of the extension direction, and the heat flux density formed by the molten metal at the inflection point of the centerline is calculated; S3, according to the heat flux density that needs to be met in step S2, calculating the welding heat that the material to be welded needs to absorb when forming a weld at each inflection point of the tower weld; S4. Construct a welding parameter-inflection point heat mapping model and output the optimal welding parameters through a global optimization algorithm.

2. The wind turbine tower weld optimization method according to claim 1, characterized in that: In step S2, the heat flux density q of the molten metal at the inflection point of the welding process is xy The following constraints need to be met: ； ； Where: u xy is the flow velocity of the molten metal at the inflection point (x, y), including u x ,u y Quantity; is the viscous dissipation function, is the density of the molten metal; k is the thermal conductivity of the material to be welded; is the viscosity of the molten metal; H is the latent heat of phase change of the material to be welded at the inflection point; q xy is the heat flux density of the molten metal at the inflection point (x, y), is the temperature gradient.

3. The wind turbine tower weld optimization method according to claim 2, characterized in that: In step S3, the heat flux density q of the molten metal at the i-th inflection point is set to i and The welding heat Q absorbed by the welding material during the dwell time i The expression is as follows: ； Among them, S i is the effective heat transfer area at the inflection point i.

4. The wind turbine tower weld optimization method according to claim 3, characterized in that: In step S4, the parameter space x of the i-th inflection point is defined. i =[I i ,U i ,t i ], I i is the welding current, U i is the voltage, t i is the residence time at the i-th inflection point, and M is the total number of inflection points; Construct the objective function J(x): ； Q i =ηI i U i t i ; Where η is the thermal efficiency, w i is the inflection point weight, Weld area , is the heat deviation threshold of the i-th inflection point, is the actual heat input at the i-th inflection point, ω is the balance coefficient; Minimize the objective function J(x) until J(x) converges and output the optimal welding parameter space , including the optimized welding current I, voltage U and dwell time t.

5. The wind turbine tower weld optimization method according to claim 4, characterized in that: In step S3, the heat correction value Q is calculated. f , heat correction value Q f is the welding heat Q i The average heat value Q at the inflection point i of the historical big data C The ratio of: Q f =Q i / Q C 。 6. The wind turbine tower weld optimization method according to claim 5, characterized in that: According to the heat correction value Q f Adjust welding parameters to obtain adjusted current ,Voltage and residence time : ； ； ； is the current sensitivity coefficient, is the voltage sensitivity coefficient, is the residence time sensitivity coefficient.

7. The wind turbine tower weld optimization method according to claim 6, characterized in that: The range is 0.6~0.8, The range is 0.2~0.4, The range is 0.3~0.

5.

8. The wind turbine tower weld optimization method according to claim 4, characterized in that: The NSGA-II multi-objective genetic algorithm is used to realize the frontier search of the objective function J(x) and output the welding parameter space that satisfies the optimal coordination between heat deviation and residual stress.

Images