A method for pipeline root pass welding process planning based on the collaboration of laser vision scanner and process knowledge base

Abstract

The application discloses a pipe backing welding process planning method based on laser vision scanner and process knowledge base cooperation, and core contents comprise the following steps: firstly, an excellent backing welding forming size model and a process knowledge base are established based on an excellent weld evaluation standard; then, a laser vision scanner is used to scan a groove and acquire characteristics of each section of the groove; finally, welding process parameters of each section are acquired based on a process parameter calculation model. In the method, the excellent weld size model can calculate a required welding wire melting area, a high side base metal melting area of the groove, and an area of more melting welding wire than the high side of the low side of the groove according to a scanning obtained gap and a misalignment amount of each section. The process knowledge base can automatically match parameters in the welding process parameter calculation model according to pipe material, a pipe diameter and a wall thickness to be welded. The characteristics extracted by the vision scanner are converted into variables in the calculation model through the excellent weld forming size model, and parameters in the model are matched in cooperation with the process knowledge base, so that accurate backing welding process parameter planning under actual working conditions is realized.

Description

Technical Field

[0001] This invention relates to the field of robotic pipe welding, and more particularly to a method for planning the pipe root pass welding process based on the collaboration of a laser vision scanner and a process knowledge base. Background Technology

[0002] Pressure pipeline welding is widely used in industries such as petrochemicals, nuclear power, and thermal power. Promoting robotic welding can help alleviate current problems such as high labor costs and a shortage of welders. Currently, due to the challenge of consistent precision in beveling for medium and thick-walled pressure pipelines, the root pass welding is mostly done manually, which has long resulted in high labor intensity and insufficient process adaptability.

[0003] Existing technologies primarily rely on process knowledge bases for process parameter planning, employing reasoning methods based on cases, rules, and models. Case-based reasoning, as the name suggests, uses analogical reasoning based on past cases, while rule-based reasoning condenses expert knowledge within the domain and expresses it in rule form by knowledge engineers. The reasoning power and accuracy of both case-based and rule-based reasoning depend on the completeness and accuracy of the cases and rules in the knowledge base. However, due to the random variations in pipeline beveling and the complexity of inter-factor constraints, these reasoning methods suffer from problems such as large amounts of case data and difficulties in rule formulation. Model-based reasoning methods combining knowledge bases with neural networks or other machine learning methods are also used; however, because the neural network reasoning process is a "black box," users lack confidence in the results.

[0004] In recent years, research on visual sensing technology in bevel feature recognition has become relatively mature. In process planning, it can realize precise welding path planning and multi-layer and multi-pass planning for medium and thick plates. However, for process parameter planning, especially root pass welding, it is generally obtained through process experiments, and there is a lack of ideal process parameter planning strategies.

[0005] To address the shortcomings of relying solely on visual guidance and a single process knowledge base strategy, this paper proposes an intelligent process planning strategy that integrates visual guidance and a process knowledge base. Furthermore, a mathematical model mapping actual bevel dimensions to welding process parameters is established. This model effectively improves the efficiency and accuracy of welding process parameter planning under actual working conditions. Summary of the Invention

[0006] The purpose of this invention is to provide a pipeline root pass welding process planning method based on the collaboration of a laser vision scanner and a process knowledge base, for planning the welding process parameters of petrochemical pipeline robot root pass welding.

[0007] The technical solution to achieve the purpose of this invention is as follows:

[0008] A method for planning the pipe root pass welding process based on the collaboration of a laser vision scanner and a process knowledge base includes the following steps:

[0009] Step 1: Build a smart pipeline welding hardware platform, including an argon arc welding robot, a laser vision scanner, a main control computer, a rotary positioner, a robot control cabinet, a wire feeding mechanism, a shielding gas mechanism, and a welding power source.

[0010] Step 2: Drive the welding robot guide rail to move the welding torch to the bevel position to be welded. This position is 10-20° perpendicular to the pipe. Adjust the welding torch posture so that the welding torch is perpendicular to the tangent of the bevel to be welded. Adjust the distance between the end of the welding wire and the bottom of the bevel to 2-3mm.

[0011] Step 3: Use the constructed platform to conduct welding experiments to obtain welding process parameters and corresponding forming size data. At the same time, establish an "excellent weld evaluation standard" based on the forming size and stress concentration of the pipe weld back to select excellent welds. Then, establish an excellent root pass forming size model and corresponding process knowledge base based on the excellent welds.

[0012] Furthermore, the "Excellent Weld Evaluation Criteria" include:

[0013] First, based on the forming dimensions of the root pass weld, a "forming dimension rating standard" is established. This standard sets four levels for the forming dimension rating, including back reinforcement height, bottom weld width, and weld height. Among them, back reinforcement height is the distance from the lowest point of the weld to the bottom of the lower bevel; bottom weld width is the horizontal distance between the bottom welds of the two bevels; and weld height is the vertical distance from the lowest point of the upper surface weld of the root pass weld to the lowest point of the lower surface weld.

[0014] For the backfill height, A, B, C, and D are 0-1mm, 1-2mm, 2-3mm, and over 3mm, respectively; the bottom melt width rating is as follows.

[0015]

[0016] Where d is the bevel gap;

[0017] The weld height rating is as follows:

[0018]

[0019] Where η is the bevel misalignment amount;

[0020] Excellent weld formation dimensions must meet the following requirements: the score must contain an A and cannot contain a D. When the score contains one A, the other two scores must be C or above. When the score contains two A's, the other score can be C or above.

[0021] Secondly, superior welds are further selected based on the stress concentration degree of the weld. Due to the asymmetry of the back weld of the root pass, the stress concentration factor (SCF) on both sides of the back weld of the root pass is calculated, and the maximum value is taken as the stress concentration factor of the root pass weld.

[0022] The formula for calculating the stress concentration factor is as follows:

[0023]

[0024] Where r1 and r2 are the arc transition radii at the weld toes on both sides of the back side; θ1 and θ2 are the weld toe angles on the lower and higher sides of the back side; δ is the pipe wall thickness; W b h is the width of the back side of the weld, which is the distance between two weld roots; b The weld reinforcement on the back side, perpendicular to W. b h b1 h b2 The values ​​of k0, K1, K2 and K3 are 0.3, 0.3, 0.3 and 0.33, respectively, representing the back heights of the low side and high side of the bevel.

[0025] If the calculated stress concentration factor satisfies SCF≤2.0, it is finally defined as an excellent weld.

[0026] Furthermore, the establishment of an excellent weld forming size model based on excellent welds includes:

[0027] Firstly, the upper surface of a superior weld is parabolic, conforming to y = ax². 2 The lower surface consists of two asymmetrical right trapezoids, and the angle of the trapezoid on the higher side of the bevel is defined as θ. p2 The trapezoidal angle on the lower side of the bevel is θ. p1 The height of the weld on the higher side of the bevel is H, the weld width on the back side is D, and the reinforcement height on the lower side of the weld back side is h. b1 The height of the weld back side is h. b2 .

[0028] Among them, H, D, h b1 h b2 The constants are determined by the rating criteria; a, θ p2 θ p1 This is a constant and depends on the material of the base material. For carbon steel, a = 0.113, θ p2 =32.68°, θ p1 =25.91°.

[0029] Secondly, a coordinate system was established for the model, and a calculation model was established for the welding wire melting area, the base metal melting area, and the area of ​​welding wire melting more on the lower side of the groove than on the higher side of the groove.

[0030] The formula for calculating the melting area of ​​the welding wire is as follows:

[0031]

[0032] S H =S H1 +S H2

[0033] Where S H S H1 S H2 n1, n2, and n1 represent the areas of wire melting, the area of ​​wire melting above the bevel, and the area of ​​wire melting below the bevel, respectively; n2 and n1 represent the x-coordinates of the intersection points of the upper surface of the weld and the two sides of the bevel, respectively; m is the y-coordinate of the intersection points of the upper surface of the weld and the two sides of the bevel; the specific values ​​of n2, n1, and m can be obtained by solving the curve function expression of the upper surface of the weld and the straight line function expression of the two sides of the bevel.

[0034] Then, based on the unequal areas of molten welding wire on both sides due to the asymmetry of weld size, the following formula is established to calculate the area of ​​more molten welding wire on the lower side of the bevel than on the higher side:

[0035]

[0036] S Hd =S Hd2 -S Hd1

[0037] Where S Hd S Hd2 S Hd1 These represent the area of ​​the lower bevel that requires more molten welding wire than the higher bevel, the area enclosed by the upper surface of the weld and the lower bevel, and the area enclosed by the upper surface of the weld and the parallel dashed line of the lower bevel, respectively.

[0038] Finally, the base metal melting area results of the excellent weld need to be fitted. From the fitting results of the base metal melting area on the low and high sides of the excellent weld bevel, it can be seen that the base metal melting area on the high side of the bevel is a constant value S. M Furthermore, the melting area of ​​the lower bevel is more than 70% of the melting area of ​​the base material on the higher bevel. For carbon steel pipes S... M =12.31mm 2 .

[0039] Furthermore, the welding process knowledge base is presented in tabular form, and can automatically match the parameters in the calculation model of the following three process parameters—bevel low-side dwell time, welding current, and wire feed speed—based on the pipe material, pipe diameter, and wall thickness, as well as the bevel high-side dwell time t0 and welding speed V. WThe parameters include b1, b2, and a2 / a1. The specific values ​​of these parameters are obtained by fitting the process parameters of excellent welds into the welding current calculation model and the wire feed speed calculation model using weld size data. The bevel high-side dwell time and welding speed are adjusted only according to the base material during process experiments. Therefore, the bevel high-side dwell time and welding speed in the process knowledge base are only related to the material. For carbon steel pipes, t0 = 0.2S, V W =0.14 m / min.

[0040] The fitting of b1 and b2 can be done as follows: Select 10-20 sets of excellent welds for pipes of different materials, diameters, and wall thicknesses, preferably in even numbers. Substitute the process parameters corresponding to these excellent welds, the melting area of ​​the base material on the high side of the bevel, the area of ​​the welding wire that needs to be melted more on the low side of the bevel than on the high side, and the welding wire melting area into the welding current calculation model below. After solving the equations in pairs, obtain 5-10 sets of numerical solutions for b1 and b2. Calculate the average of the 5-10 sets of b1 and b2 respectively, which are the values ​​of b1 and b2 in the process knowledge base corresponding to the pipe of this material, diameter, and wall thickness.

[0041] The fitting of a2 / a1 can be done as follows: Select 10-20 sets of excellent welds for pipes of different materials, substitute the process parameters and wire melting area values ​​corresponding to the excellent welds into the wire feeding speed calculation model below, and obtain 10-20 sets of numerical solutions for a2 / a1. Calculate the average of the 10-20 sets of a2 / a1, which is the value of a2 / a1 in the process knowledge base corresponding to the pipe material.

[0042] Step 4: Adjust the welding platform as in Step 2. Then start the laser vision scanner. The rotary positioner will rotate the pipe one revolution from 0° at a speed of 1-2° / s to scan the bevel features. After the scan is completed, bevel feature parameters will be fitted in segments of 5-10°.

[0043] The bevel characteristic parameters include misalignment (η) and clearance (d).

[0044] Step 5: Using the base material as input, match the welding speed and bevel high-side dwell time from the process knowledge base.

[0045] Step 6: Using the triangular swing mode, substitute the gap of each bevel segment into the swing amplitude and swing length calculation model to obtain the swing amplitude and swing length parameters.

[0046] Furthermore, the calculation model for the swing amplitude and swing length includes the following formulas for calculating the swing length and swing amplitude of each bevel segment:

[0047]

[0048] l = l w +0.1

[0049] Among them l w Let l be the swing amplitude and l be the swing length.

[0050] Step 7: Substitute the bevel gap and misalignment of each segment into the excellent root pass welding forming size model to calculate the area of ​​additional molten welding wire required on the lower side of the bevel compared to the higher side of the bevel. Based on the size of the area of ​​additional molten welding wire required, select the corresponding bevel lower side dwell time calculation model.

[0051] Furthermore, the mathematical expression of the calculation model for the dwell time on the lower side of the bevel is as follows:

[0052]

[0053] Where η0 represents the thermal efficiency of tungsten inert gas welding, S Hd0 This is the area of ​​additional welding wire required to melt on the lower side of the bevel compared to the higher side when the melting area of ​​the base metal on the lower side of the bevel is 70% of the melting area of ​​the base metal on the higher side of the bevel. The calculation formula is as follows:

[0054] b2S Hd0 =0.3b1S M

[0055] Step 8: Substitute the bevel gap and misalignment of each segment into the excellent root pass welding forming size model to calculate the required molten wire area S for each segment. H Using the base material, model, and pipe thickness as input conditions, the process knowledge base is used to match b1 and b2. The values ​​of b1 and b2 are substituted into the welding current calculation model, and the value of b2 is substituted into the groove bottom dwell time calculation model. The welding current calculation model and the groove bottom dwell time calculation model are combined to calculate the welding current and groove bottom dwell time for each segment.

[0056] Furthermore, the welding current calculation model is based on the principle of conservation of heat input energy, and its mathematical expression formula is as follows:

[0057] When S hd ≤S Hd0 At that time, the energy conservation equation for heat input is:

[0058]

[0059] When S hd >S Hd0 At that time, the energy conservation equation for heat input is:

[0060] η0UIT=(1.7b1S M +b2S H )l

[0061] Where U is the welding voltage and T is the period of one oscillation, and its calculation formula is as follows:

[0062]

[0063] Among them, T L and T R These represent the dwell time on the left side of the bevel and the dwell time on the right side of the bevel, respectively, i.e., the dwell time on the lower side of the bevel and the dwell time on the higher side of the bevel mentioned above; when S Hd ≤S Hd0 At this time, t L =t R =t0 is directly substituted into the above welding current calculation model to calculate the welding current. When S Hd >S Hd0 When calculating the bottom dwell time of the bevel and the welding current, it is necessary to combine the above-mentioned calculation model of the bottom dwell time of the bevel with the calculation model of the welding current to solve for the bottom dwell time of the bevel and the welding current.

[0064] The mathematical expression of the combined computational model is as follows:

[0065]

[0066] Step 9: Substitute the bevel gap and misalignment of each section into the excellent root pass welding forming size model to calculate the required molten wire area S. H Using the material of the parent material as input, and with the help of the process knowledge base, a2 / a1 is matched, and the values ​​of a2 / a1 are substituted into the wire feeding speed calculation model to calculate the wire feeding speed.

[0067] The wire feeding speed calculation model is established based on the mass conservation equation, and its mathematical expression formula is as follows:

[0068] a1V f T = a2S H l

[0069] Compared with the prior art, the significant advantages of this invention are: 1) the establishment of an "excellent weld evaluation standard" based on the pipe back forming size rating and stress concentration ensures the rationality of sample selection; 2) the establishment of a welding process parameter calculation model based on the heat input energy conservation equation and the mass conservation equation ensures the accuracy of process parameter calculation; 3) the pipe root pass welding process planning strategy based on the collaboration of laser vision scanner and process knowledge base can realize the automatic generation of process parameters under actual working conditions. Attached Figure Description

[0070] Figure 1 This is a flowchart of a pipeline root pass welding process planning method based on the collaboration of a laser vision scanner and a process knowledge base, according to the present invention.

[0071] Figure 2 This is a diagram of a device for a pipeline root pass welding process planning method based on the collaboration of a laser vision scanner and a process knowledge base, according to the present invention.

[0072] Figure 3 This is a schematic diagram of the superior weld formation size model corresponding to a DN300, 11mm thick carbon steel pipe of grade Q235.

[0073] Figure 4 This is a partial screenshot of the process knowledge base established by the pipeline root pass welding process planning method based on the collaboration of a laser vision scanner and a process knowledge base according to the present invention. Detailed Implementation

[0074] The invention will be further described below with reference to the accompanying drawings and using a DN300, 11mm thick carbon steel pipe of grade Q235. The specific process is as follows: Figure 1 As shown.

[0075] The specific steps are as follows:

[0076] Step 1: Build a smart pipeline welding hardware platform, including 1. Argon arc welding robot, 2. Laser vision scanner, 3. Main controller, 4. Rotary positioner, 5. Robot control cabinet, 6. Wire feeding mechanism, 7. Shielding gas mechanism, and 8. Welding power source.

[0077] The laser vision scanner 2 is mounted on the end of the argon arc welding robot 1 to scan and extract the bevel features of each section of the pipeline. The scanned bevel features are transmitted to the process knowledge base in the main control computer 3 to obtain the process parameters of each section. The main control computer can then transmit the process parameters of each section to the robot control cabinet 5 to control the argon arc welding robot 1, which integrates a wire feeding mechanism 6, a shielding gas mechanism 7, and a welding power source 8, to complete the actual pipeline welding work.

[0078] Step 2: Move the welding gun to the bevel position on the guide rail of the welding robot. The position is 20° perpendicular to the pipe. Adjust the posture of the welding gun to A: 179.09 B: 68.97 C: -0.46. Adjust the distance between the end of the welding wire and the bottom of the bevel to 2mm.

[0079] Step 3: Conduct numerous welding experiments and establish an "excellent weld evaluation standard" based on the back-side forming dimensions and stress concentration of the pipe weld to select excellent welds. Then, based on these excellent welds, establish... Figure 3 The excellent weld formation size model and corresponding process knowledge base are shown.

[0080] Figure 4 A screenshot of the process knowledge base is provided.

[0081] Furthermore, the "Excellent Weld Evaluation Criteria" include:

[0082] First, based on the forming dimensions of the root pass weld, a "forming dimension rating standard" is established. This standard includes forming dimension ratings for back reinforcement height, bottom weld width, and weld height. Specifically, back reinforcement height is the distance from the lowest point of the weld to the bottom of the lower bevel; bottom weld width is the horizontal distance between the bottom welds of the two bevels; and weld height is the vertical distance from the lowest point of the upper surface weld of the root pass weld to the lowest point of the lower surface weld.

[0083] The rating standard for the root pass weld shape size includes four levels. For the back fillet height, A, B, C, and D are 0-1mm, 1-2mm, 2-3mm, and 3mm or more, respectively; the rating for the bottom weld width is as follows.

[0084]

[0085] Where d is the bevel gap.

[0086] The weld height rating is as follows:

[0087]

[0088] Where η is the bevel misalignment amount.

[0089] Excellent weld formation dimensions must meet the following requirements: the score must contain an A and cannot contain a D. When the score contains one A, the other two scores must be C or above. When the score contains two A, the other score can be C or above.

[0090] Secondly, superior welds are further selected based on the stress concentration degree of the weld. Due to the asymmetry of the back weld of the root pass, the stress concentration factor (SCF) on both sides of the back weld of the root pass is calculated, and the maximum value is taken as the stress concentration factor of the root pass weld.

[0091] The formula for calculating the stress concentration factor is as follows:

[0092]

[0093] Where r1 and r2 are the arc transition radii at the weld toes on both sides of the back side; θ1 and θ2 are the weld toe angles on the lower and higher sides of the back side; δ is the pipe wall thickness; W b h is the width of the weld back side, which is the distance between the two weld toes; b The weld reinforcement on the back side, perpendicular to W. b h b1 h b2 The values ​​of k0, K1, K2 and K3 are 0.3, 0.3, 0.3 and 0.33, respectively, representing the back heights of the low side and high side of the bevel.

[0094] A weld is defined as excellent if the stress concentration factor SCF ≤ 2.0.

[0095] Furthermore, the establishment of an excellent weld forming size model based on excellent welds includes:

[0096] First, the upper surface of the excellent weld seam of carbon steel pipe is fitted with a parabolic shape, conforming to y = 0.113x. 2 The lower surface consists of two asymmetrical right-angled trapezoids, with the angle θ of the trapezoid on the higher side of the bevel. p2 The trapezoidal angle θ on the lower side of the bevel is 32.68°. p1 The bevel angle is 25.91°, the height of the weld on the high side of the bevel is (H), the weld width on the back side is (D), and the weld reinforcement height on the low side of the back side is (h). b1 The height of the weld back side is (h) b2 ).

[0097] Among them, H, D, h b1 h b2 The constant is determined by the rating criteria;

[0098] Secondly, a coordinate system was established for the model, and a calculation model was established for the welding wire melting area, the base metal melting area, and the area of ​​welding wire melting more on the lower side of the groove than on the higher side of the groove.

[0099] The formula for calculating the melting area of ​​the welding wire is as follows:

[0100]

[0101] S H =S H1 +S H2

[0102] Where S H S H1 S H2 n1, n2, and n1 represent the areas of wire melting, the area of ​​wire melting above the bevel, and the area of ​​wire melting below the bevel, respectively; n2 and n1 represent the x-coordinates of the intersection points of the upper surface of the weld and the two sides of the bevel, respectively; m is the y-coordinate of the intersection points of the upper surface of the weld and the two sides of the bevel; the specific values ​​of n2, n1, and m can be obtained by solving the curve function expression of the upper surface of the weld and the straight line function expression of the two sides of the bevel.

[0103] Then, based on the unequal areas of molten welding wire on both sides due to the asymmetry of weld size, the following formula is established to calculate the area of ​​more molten welding wire on the lower side of the bevel than on the higher side:

[0104]

[0105] S Hd =S Hd2 -S Hd1

[0106] Where S Hd SHd2 S Hd1 These represent the area of ​​the lower bevel that requires more molten welding wire than the higher bevel, the area enclosed by the upper surface of the weld and the lower bevel, and the area enclosed by the upper surface of the weld and the parallel dashed line of the lower bevel, respectively.

[0107] Finally, based on the fitting results of the base metal melting area of ​​the excellent weld seam of carbon steel pipe, the base metal melting area S on the high side of the bevel can be determined. M =12.31mm 2 Furthermore, the melting area of ​​the lower bevel is more than 70% of the melting area of ​​the parent material on the higher bevel.

[0108] Furthermore, the welding process knowledge base is presented in tabular form, and can automatically match the parameters in the calculation model of the following three process parameters—bevel low-side dwell time, welding current, and wire feed speed—based on the pipe material, pipe diameter, and wall thickness, as well as the bevel high-side dwell time t0 and welding speed V. W The parameters include b1, b2, and a2 / a. 1， The specific values ​​of these parameters are obtained by substituting the process parameters of excellent welds into the welding current calculation model and the wire feed speed calculation model using weld size data; while the high-side dwell time and welding speed of the bevel are adjusted only according to the base material during process experiments. For carbon steel pipes, t0 = 0.2S, V W =0.14 m / min.

[0109] Matching the process knowledge base yielded the following values ​​for a DN300, 11mm thick carbon steel pipe with grade Q235: b1 = 29.52957, b2 = 30.75997, and a2 / a1 = 1.5323.

[0110] Step 4: Set up the welding platform as in Step 2. Then start the laser vision scanner. The rotary positioner will rotate the pipe one revolution from 0° at a speed of 1-2° / s to scan the bevel features. After the scan is completed, bevel feature parameters will be fitted in segments of 5-10°.

[0111] The bevel characteristic parameters include misalignment (η) and clearance (d).

[0112] Step 5: Using carbon steel as the base material as input, match the welding speed V from the process knowledge base. W =0.14m / min, dwell time on the high side of the slope t0 =0.2s.

[0113] Step 6: Using the triangular swing mode, substitute the gap of each bevel segment into the swing amplitude and swing length calculation model to obtain the swing amplitude and swing length parameters.

[0114] Furthermore, the calculation model for the swing amplitude and swing length includes the following formulas for calculating the swing length and swing amplitude of each bevel segment:

[0115]

[0116] l = l w +0.1

[0117] Among them l w Let l be the swing amplitude and l be the swing length.

[0118] Step 7: Substitute the bevel gap and misalignment of each segment into the excellent root pass welding forming size model to calculate the area of ​​additional molten welding wire required on the lower side of the bevel compared to the higher side of the bevel. Based on the size of the area of ​​additional molten welding wire required, select the corresponding bevel lower side dwell time calculation model.

[0119] Furthermore, the mathematical expression of the calculation model for the dwell time on the lower side of the bevel is as follows:

[0120]

[0121] Where η0 represents the thermal efficiency of tungsten inert gas welding, and 3.55 is the area of ​​welding wire that needs to be melted more on the lower side of the bevel than on the higher side when the melting area of ​​the base material on the lower side of the bevel is 70% of the melting area of ​​the base material on the higher side of the bevel for a DN300, 11mm thick carbon steel pipe of grade Q235.

[0122] Step 8: Substitute the bevel gap and misalignment of each segment into the excellent root pass welding forming size model to calculate the required molten wire area S for each segment. H Using Q235 carbon steel as the base material, DN300 as the model, and 11mm as the pipe thickness as the input conditions, the process knowledge base is used to match b1=29.52957 and b2=30.75997. The values ​​of b1 and b2 are substituted into the welding current calculation model, and the value of b2 is substituted into the groove bottom dwell time calculation model. The welding current calculation model and the groove bottom dwell time calculation model are combined to calculate the welding current and groove bottom dwell time for each segment.

[0123] Furthermore, the welding current calculation model is based on the principle of conservation of heat input energy, and its mathematical expression formula is as follows:

[0124] When S hd When ≤3.55, the heat input energy conservation equation is:

[0125]

[0126] When S hd When >3.55, the heat input energy conservation equation is:

[0127] η0UIT=(50.20027*S M +30.75997*S H )l

[0128] Where U is the welding voltage and T is the period of one oscillation, and its calculation formula is as follows:

[0129]

[0130] Among them, T L and T R These represent the dwell time on the left side of the bevel and the dwell time on the right side of the bevel, respectively, i.e., the dwell time on the lower side of the bevel and the dwell time on the higher side of the bevel mentioned above; when S Hd ≤S Hd0 At this time, t L =t R =0.2S, directly substitute into the above welding current calculation model to calculate the welding current, when S Hd >S Hd0 When calculating the bottom dwell time of the bevel and the welding current, it is necessary to combine the above-mentioned calculation model of the bottom dwell time of the bevel with the calculation model of the welding current to solve for the bottom dwell time of the bevel and the welding current.

[0131] The mathematical expression of the simultaneous computational model is as follows:

[0132]

[0133] Step 7: Substitute the bevel gap and misalignment of each section into the excellent root pass welding forming size model to calculate the required molten wire area (S). H Using the material of the parent material as input, and with the help of the process knowledge base, a2 / a1 = 1.5323 is matched, and the value of a2 / a1 is substituted into the wire feeding speed calculation model to calculate the wire feeding speed.

[0134] The wire feeding speed calculation model is established based on the mass conservation equation, and its mathematical expression formula is as follows:

[0135] V f T = 1.5323 * S H l.

Claims

1. A pipeline preparation welding process planning method based on laser vision scanner and process knowledge base cooperation, characterized in that, The method comprises the following steps: Step 1: build a pipeline intelligent welding hardware platform, including argon arc welding robot, laser vision scanner, main control machine, rotary positioner, robot control cabinet, wire feeding mechanism, protective gas mechanism, welding power supply; Step 2: drive the welding robot guide rail to move the welding torch to the position to be welded, which is 10-20° to the vertical direction of the pipeline, and adjust the welding torch attitude to make the welding torch perpendicular to the tangent position of the to-be-welded groove, and adjust the distance between the welding wire end and the groove bottom to 2-3mm; Step 3: use the built platform to perform welding experiments to obtain welding process parameters and corresponding forming size data, and further select excellent welds based on the forming size and stress concentration of the pipeline weld back, and then establish an excellent backing weld forming size model and a corresponding process knowledge base based on the excellent welds; Step 4: adjust the welding platform according to step 2; then start the laser vision scanner and the rotary positioner to drive the pipeline to rotate a circle at 1-2° / s from 0° to scan the groove features, and obtain the groove feature parameters at 5-10° per section after the scanning is completed; Step 5: Match the welding speed V from the process knowledge base with the input condition of base material W and the high side of the groove dwell time t0; Step 6: adopt a triangular swing mode, and put each groove gap into a swing amplitude and swing length calculation model to obtain the swing amplitude and swing length parameters; Step 7: put each groove gap and the misalignment into the excellent backing weld forming size model, calculate the area of the extra melting welding wire required by the low side of the groove compared with the high side of the groove, and select a corresponding groove low side dwell time calculation model based on the size of the required extra melting welding wire area; Step 8: calculate the required welding wire melting area S of each section by substituting the gap and the amount of misalignment of each section into the excellent backing welding forming size model H and taking the base material material, model, and pipe thickness as input conditions, matching b1 and b2 with the help of the process knowledge base, substituting the b1 and b2 values into the welding current calculation model, substituting the b2 value into the low side of the groove dwell time calculation model, and simultaneously calculating the welding current and the low side of the groove dwell time of each section by the welding current calculation model and the low side of the groove dwell time calculation model. Step 9: Calculate the required welding wire melting area S by substituting the gap and the amount of misalignment of each section of the bevel into the excellent backing welding forming size model H And with the base material as the input condition, the a2 / a1 value is substituted into the wire feeding speed calculation model to calculate the wire feeding speed by means of the process knowledge base matching a2 / a1.

2. The method of pipe preparation welding process planning based on laser vision scanner and process knowledge base collaboration according to claim 1, characterized in that, The excellent weld evaluation standard in step 3 comprises: Firstly, based on the forming size of the backing weld, a forming size rating standard is established, which sets four levels A, B, C and D for the back excess height, back width and weld height respectively; wherein the back excess height is the distance from the low side of the groove to the lowest part of the weld, the bottom width is the horizontal distance of the weld on both sides of the groove bottom, and the weld height is the vertical distance from the lowest part of the weld on the upper surface to the lowest part on the lower surface of the backing weld; For back relief, A, B, C, and D are 0 < h b1 < 1 mm, 1 < h b1 < 2 mm, 2 < h b1 < 3 mm, and h b1 ≥ 3 mm; and for back fusion width, the ratings are as follows: Wherein, d is the groove gap, unit: mm; For the weld height rating, the following is shown: Wherein, η is the groove misalignment, unit: mm; The excellent weld forming size needs to meet: the score must contain A and cannot appear D, when the score contains one A, the other two scores need to be above C, and when the score contains two A, the other score can be C or above C; Secondly, further select excellent welds based on the stress concentration degree of the weld; based on the asymmetry of the backing weld, the stress concentration coefficients SCF of both sides of the backing weld are calculated respectively, and the maximum value is taken as the stress concentration coefficient of the backing weld; The stress concentration coefficient calculation formula is as follows Wherein r1 and r2 are the circular arc transition radius at the back two sides of the weld toe; θ1 and θ2 are the back low side and high side weld toe angle; δ is the pipe wall thickness; W b is the weld back width, the distance between the two weld toes; h b is the weld back reinforcement, perpendicular to W b ; h b1 , h b2 are the back reinforcement of the low side and high side of the groove respectively, and the values of k0, K1, K2 and K3 are 0.3, 0.3, 0.3 and 0.33 respectively. The calculated stress concentration coefficient satisfies SCF≤2.0, and is finally defined as an excellent weld.

3. The method of pipe preparation welding process planning based on laser vision scanner and process knowledge base collaboration according to claim 2, characterized in that, The excellent backing weld forming size model established based on the excellent weld in step 3 comprises: First, the superior surface of the weld is parabolic, consistent with y=ax 2 , the inferior surface is two asymmetric right trapezoids, defining the high-side trapezoid angle as θ p2 , the low-side trapezoid angle as θ p1 , the distance from the lowest point of the parabolic superior surface of the backing weld to the bottom surface of the high-side groove as H, and the weld back melt width as W; where H, W, h b1 , h b2 are constants determined by the rating criteria; a, θ p2 , θ p1 are constants related to the base material; for carbon steel a = 0.113, θ p2 = 32.68°, θ p1 = 25.91°; Secondly, a coordinate system is established for the model, and a welding wire melting area, base material melting area and groove low side more melting welding wire area calculation model are established; The welding wire melting area calculation formula is as follows: S H = S H1 + S H2 wherein S H , S H1 , S H2 respectively represent the welding wire melting area, the welding wire melting area above the groove, and the welding wire melting area below the groove; n2, n1 respectively represent the horizontal coordinates of the intersection points of the upper surface of the weld and the two sides of the groove, and m represents the vertical coordinates of the intersection points of the upper surface of the weld and the two sides of the groove; the specific values of n2, n1, and m can be obtained by solving the function expression of the curve where the upper surface of the weld is located and the function expression of the straight line where the two sides of the groove are located. Then, based on the asymmetry of the weld size leading to the difference of the area of the melted welding wire on both sides, the formula for calculating the area of the low side of the groove being more than that of the high side is as follows: S Hd = S Hd2 - S Hd1 wherein S Hd , S Hd2 , S Hd1 respectively represent the area of the groove low side requiring more melting welding wire than the groove high side, the area surrounded by the weld upper surface and the groove low side, and the area surrounded by the weld upper surface and the groove low side parallel dotted line. Finally, the base metal melting area results of the superior welds also need to be fitted; from the base metal melting area fitting results of the low side and high side of the superior weld groove, the base metal melting area of the high side of the groove is a constant value S M , and the melting area of the low side of the groove is more than 70% of the base metal melting area of the high side of the groove; for carbon steel pipelines S M = 12.31 mm 2 .

4. The method of pipe preparation welding process planning based on laser vision scanner and process knowledge base collaboration according to claim 3, characterized in that, The process knowledge base established based on the excellent weld in step 3 comprises: The welding process knowledge base includes pipe material, pipe diameter, wall thickness, and corresponding high-side dwell time t0 and welding speed V W It also includes parameters b1, b2 and a2 / a1 in the calculation model of three process parameters of low-side dwell time, welding current and wire feeding speed, which can automatically match the corresponding high-side dwell time t0 and welding speed V based on pipe material, pipe diameter and wall thickness W and parameters in the calculation model of three process parameters of low-side dwell time, welding current and wire feeding speed.

5. The method of pipe preparation welding process planning based on laser vision scanner and process knowledge base collaboration according to claim 4, characterized in that, The swing length and swing amplitude calculation model in step 6 comprises: The swing length and swing amplitude calculation formula of each groove is as follows: l＝l w +0.1 where l w is the swing, l is the swing length.

6. The method of pipe preparation welding process planning based on laser vision scanner and process knowledge base collaboration according to claim 5, characterized in that, The low side dwell time calculation model of the groove selected based on the required area of the melted welding wire in step 7 comprises: The mathematical expression formula of the low side dwell time calculation model of the groove is as follows: wherein η0represents the heat efficiency of tungsten argon arc welding, U represents the welding voltage, I represents the welding current, S Hd0 is the area of the low side of the groove that is required to melt more welding wire than the high side of the groove when the low side of the groove has 70% of the area of the high side of the groove that is melted, and the calculation formula is as follows: b2S Hd0 = 0.3b1S M .

7. The method of pipe preparation welding process planning based on laser vision scanner and process knowledge base collaboration according to claim 6, characterized in that, The calculation of the welding current and the low side dwell time of the groove in step 8 comprises: The welding current calculation model is established based on the heat input energy conservation principle, and the mathematical expression formula is as follows: When S hd ≤ S Hd0 , the heat input energy conservation equation is When S hd > S Hd0 , the heat input energy conservation equation is η0UIT = (1.7b1S M +b2S H )l Wherein T is the period of one swing, and the calculation formula is as follows: wherein t L and t R respectively represent the left side of the groove residence time and the right side of the groove residence time, that is, the low side of the groove residence time and the high side of the groove residence time of the welding groove; when S Hd ≤ S Hd0 , at this time, t L =t R =t0 is directly substituted into the above welding current calculation model to calculate the welding current, when S Hd >S Hd0 , the low side of the groove residence time and the welding current need to be solved by simultaneously solving the above low side of the groove residence time calculation model and the welding current calculation model. The mathematical expression formula of the calculation model after being combined is as follows:

8. The method of pipe preparation welding process planning based on laser vision scanner and process knowledge base collaboration according to claim 7, characterized in that, The wire feeding speed calculation model in step 9 comprises: The wire feeding speed calculation model is established by the mass conservation equation, and the mathematical expression formula is as follows: a1V f T = a2S H l where V f represents the wire feed speed.

Images