Cold-rolled pickled steel coil coiling centering method based on 3D point cloud

Abstract

The invention belongs to the technical field of coiling of steel coils, and particularly discloses a cold-rolled pickled steel coil coiling centering method based on 3D point clouds, which comprises the following steps of: acquiring point cloud data of a steel coil end surface on a coiling trolley and a mandrel of an uncoiling machine through a laser radar sensor; and splicing the point cloud data into a complete steel coil and mandrel model through a point cloud reconstruction technology, obtaining circle center coordinates of the steel coil and the mandrel for the first time through a circle center fitting algorithm, and calculating moving coordinates required by centering of a coil feeding trolley according to the circle center coordinates, so that the coil feeding trolley is adjusted through the moving coordinates, and the circle centers of the steel coil and the mandrel are centered. The problems that in the manual adjusting process, the situation of centering deviation is prone to occurring, so that steel coil core pulling is caused, and even equipment is damaged due to steel coil overturning are solved. According to the invention, point cloud data is collected in real time through the laser radar, and full-automatic centering of the steel coil and the mandrel is realized in combination with a circle center fitting algorithm and closed-loop control.

Description

Technical Field

[0001] The present invention relates to the technical field of steel coil winding, and in particular to a cold-rolled pickled steel coil winding centering method based on 3D point cloud. Background Art

[0002] When the cold-rolled steel coil is coiled up, the operator controls the crane to place the steel coil onto the coiling trolley, which then moves to the front of the uncoiler via the track. At this time, there is a certain deviation between the core shaft of the uncoiler and the center of the steel coil, but the center of the core shaft can be seen through the center of the center of the steel coil. The operator then controls the coiling trolley to adjust up and down to align the center of the steel coil with the core shaft. After the alignment is completed, the steel coil is inserted into the core shaft of the uncoiler to complete the coiling work.

[0003] However, during the manual adjustment process, the alignment of the steel coil and the mandrel is still prone to deviation, which may cause the steel coil to pull out of the core when passing through the uncoiler mandrel, or even cause the steel coil and the mandrel to collide and the steel coil to overturn, causing damage to the equipment. Summary of the Invention

[0004] The purpose of the present invention is to provide a coil centering method for cold-rolled pickled steel coils based on 3D point clouds to solve the problem that centering deviations are likely to occur during manual adjustment, resulting in coil core pulling or even coil tipping over, causing equipment damage.

[0005] To achieve the above-mentioned object, the present invention provides a basic solution: a method for centering a cold-rolled pickled steel coil based on a 3D point cloud, comprising the following steps:

[0006] S1, first transport the steel coil to the uncoiler mandrel via the coiling trolley;

[0007] S2, using a laser radar sensor to obtain 3D point cloud data of the end surface of the steel coil on the coiling trolley;

[0008] S3. Splicing the acquired 3D point cloud data into a complete steel coil end face model using point cloud reconstruction technology;

[0009] S4. After obtaining the steel coil end face model, the center coordinates of the steel coil are obtained for the first time through the circle center fitting algorithm;

[0010] S5. Calculate the movement coordinates required for centering the coil trolley using the center coordinates of the coil and the mandrel obtained for the first time;

[0011] S6. The moving coordinates are transmitted to the PLC controller in real time through the control room. The PLC controller drives the coiling trolley to move up and down according to the moving coordinates to align the center of the steel coil with the center of the mandrel.

[0012] S7. After the coiling trolley is adjusted, the 3D point cloud data of the coil end face is obtained again through the laser radar sensor, and the center coordinates of the coil are obtained for the second time through point cloud reconstruction technology and circle center fitting algorithm. The deviation of the center coordinates of the coil is calculated. If the deviation is less than or equal to the threshold, the coil is automatically loaded; if the deviation is greater than the threshold, the automatic loading is stopped and manual confirmation is performed.

[0013] The beneficial effects of this invention include: through real-time data acquisition by lidar, combined with a center-of-circle fitting algorithm and closed-loop control, fully automatic alignment of the coil and mandrel is achieved, increasing the automatic coiling rate at the cold rolling entrance from 62.5% to 95.4%, significantly shortening downtime for adjustment, and reducing coil core pulling, tipping, and equipment collision accidents caused by manual operation errors, thereby ensuring production safety and equipment stability. At the same time, the vertical coordinate deviation is reduced from 8.89mm in manual operation to 2.45mm, and both horizontal and vertical deviations are controlled within the 20mm threshold, significantly improving alignment accuracy and avoiding production accidents caused by excessive deviations.

[0014] Option 2 is the preferred option of the basic option. In S2, the lidar sensor is an area array lidar that uses TOF ranging technology, and the point cloud density of a single scan is 10,000 points. The steel coil is quickly scanned using TOF ranging technology through the area array lidar. At the same time, TOF technology adapts to complex industrial environments and can prevent dust, light and other factors from interfering with the scanning results, ensuring the integrity of the model.

[0015] Solution 3, which is the preferred solution of the basic solution, in S3, the method of stitching the model using the point cloud reconstruction technology includes the following steps:

[0016] Step 1: Align the acquired 3D point cloud data into the same coordinate system;

[0017] Step 2: Remove noise and abnormal points through Gaussian filtering and statistical outlier removal;

[0018] Step 3: Splice the processed point cloud data into a complete steel coil end face model and core shaft model; effectively remove environmental noise through Gaussian filtering and statistical outlier removal (SOR); align multiple frames of point cloud data to the same coordinate system and then splice them to fully restore the three-dimensional geometric features of the steel coil end face and core shaft, avoiding fitting errors caused by local missing data.

[0019] Solution 4 is the preferred solution for the basic solution. In S4, the circle center fitting algorithm mainly includes the following steps:

[0020] Step 1: Plane fitting: Construct a three-dimensional discrete point coordinate system (x i ,y i ,z i ), where x i 、yi 、z i Respectively represent the horizontal, vertical, and front-back coordinates of the discrete points;

[0021] The discrete point cloud is fitted by the least squares method, that is, all discrete points are located in the same plane, and the plane equation is expressed as:

[0022] ax+by+cz=1

[0023] It is presented in matrix form as follows:

[0024]

[0025] If the order (a,b,c) T =A,(1,1,...1) T =N

[0026] but

[0027] M·A=N

[0028] Further solve the plane vector A by minimizing the residual:

[0029] A=(M T M) -1 ·M T ·N

[0030] Among them, a represents the horizontal component of the plane normal vector, b represents the vertical component of the plane normal vector, z represents the forward and backward component of the plane normal vector, and x represents the n Represents the horizontal coordinate of the discrete point, y n Represents the vertical coordinate of the discrete point, z n Represents the coordinates of the discrete point in the forward and backward directions, and T represents the symbol of matrix transpose in linear algebra;

[0031] Step 2. Find the center of the circle: Let the coordinates of the center of the circle be (x0, y0, z0). Take any two discrete points P1 (x1, y1, z1) and P2 (x2, y2, z2). The center of the circle C must satisfy the following condition: the perpendicular bisector of the line connecting P1 and P2 passes through the center of the circle. Then the vector connecting P1 and P2 is:

[0032] V1=(x2-x1,y2-y1,z2-z1)

[0033] The coordinates of the midpoint of the line connecting P1 and P2 are:

[0034]

[0035] Center C and P 中点 The connection vector is

[0036]

[0037] According to the perpendicular bisector property, vector V1 is perpendicular to V2, so the dot product is zero:

[0038] V1·V2=0

[0039] Substituting the specific vectors of V1 and V2 into the above formula, we get the linear equation:

[0040]

[0041] Simplifying the above equation:

[0042]

[0043] Among them, (x0, y0, z0) represents the three-dimensional coordinates of the center C, (x1, y1, z1) represents the three-dimensional coordinates of the discrete point P1, (x2, y2, z2) represents the three-dimensional coordinates of the discrete point P2, V1 represents the line vector connecting P1 and P2, and V2 represents the line vector between the center C and P 中点 Connection vector, P 中点 Indicates the coordinates of the midpoint of the line connecting P1 and P2;

[0044] The above formula is expressed in matrix form as follows:

[0045]

[0046] Where Δx (n-1)n =x n -x (n-1) ,

[0047] If the order

[0048] but

[0049] B·C=L

[0050] Among them, x n Indicates the horizontal coordinate of the nth discrete point, y n Represents the vertical coordinate of the nth discrete point, z n represents the forward and backward coordinates of the nth discrete point, and T represents the standard symbol for matrix transposition in linear algebra;

[0051] Since the center C is found to be in the plane in step 1, then:

[0052] A·C=1

[0053] Combining the above two equations, and presenting them in matrix form is as follows:

[0054]

[0055] Using the least squares method to solve, the closed solution of the circle center coordinate C is:

[0056]

[0057] Express the above formula in the form of three-dimensional coordinates:

[0058] Let the merge matrix The coordinates of the circle center are:

[0059] The discrete point cloud is fitted to the plane equation through the least squares method. At the same time, the extended matrix equation is constructed based on the properties of the perpendicular bisector, and the coordinates of the circle center are solved in combination with the plane constraints.

[0060] Solution 5, which is the preferred solution of the basic solution, in S5, the calculation method of the moving coordinates required for the centering of the winding trolley is:

[0061] Δx=x 钢卷1 -x 芯轴1

[0062] Δy=y 钢卷1 -y 芯轴1

[0063] Among them, Δx represents the horizontal movement of the winding trolley, Δy represents the vertical movement of the winding trolley, and x 钢卷1 Indicates the horizontal coordinate of the center of the steel coil obtained for the first time, y 钢卷1 Indicates the vertical coordinate of the center of the steel coil obtained for the first time, x 芯轴1 Indicates the horizontal coordinate of the center of the mandrel obtained for the first time, y 芯轴1 Indicates the vertical coordinate of the center of the mandrel obtained for the first time.

[0064] Solution 6, which is the preferred solution of the basic solution, in S7, the calculation method of the center coordinate deviation of the steel coil and the core shaft is:

[0065] x ε =x 钢卷2 -x 芯轴2

[0066] y ε =y 钢卷2 -y 芯轴2

[0067] Among them, x ε Indicates the horizontal deviation of the center coordinates of the steel coil and the core shaft, y ε Indicates the vertical deviation of the center coordinates of the steel coil and the core shaft, x 钢卷2 Indicates the horizontal coordinate of the center of the steel coil obtained for the second time, y 钢卷2Indicates the vertical coordinate of the center of the steel coil obtained for the second time, x 芯轴2 Indicates the horizontal coordinate of the center of the mandrel obtained for the second time, y 芯轴2 Indicates the vertical coordinate of the center of the mandrel obtained for the second time; through secondary scanning and center verification, the horizontal and vertical deviations are calculated, and the centering status is dynamically evaluated. When the deviation is greater than 20mm, the automatic process is immediately stopped and manual intervention is triggered to avoid equipment damage and production accidents.

[0068] Option seven is the preferred option of the basic option. In S7, the threshold is 6 mm. The deviation threshold is set to 6 mm to clarify the switching conditions between automation and manual intervention to prevent major risks caused by errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a structural schematic diagram of a cold-rolled pickled steel coil centering method based on 3D point cloud according to the present invention. DETAILED DESCRIPTION

[0070] The present invention will be further described in detail below through specific embodiments:

[0071] The figure marks in the drawings of the specification include: 1. steel coil; 2. uncoiler; 3. mandrel; 4. coiling trolley; 5. lidar sensor; 6. track.

[0072] Example 1

[0073] A method for centering a cold-rolled pickled steel coil based on a 3D point cloud comprises the following steps:

[0074] S1. First, the control room issues a command to make the PLC controller drive the coil trolley to transport the steel coil to 1m in front of the uncoiler mandrel;

[0075] S2. Obtain 3D point cloud data of the end surface of the steel coil on the coil loading trolley through the LiDAR sensor. The LiDAR sensor is an area array LiDAR that uses TOF ranging technology. The point cloud density of a single scan is 10,000 points.

[0076] S3. The acquired 3D point cloud data is spliced ​​into a complete steel coil end face model using point cloud reconstruction technology. The method for splicing the model using point cloud reconstruction technology includes the following steps:

[0077] Step 1: Align the acquired 3D point cloud data into the same coordinate system;

[0078] Step 2: Remove noise and abnormal points through Gaussian filtering and statistical outlier removal;

[0079] Step 3: Splice the processed point cloud data into a complete steel coil end face model and mandrel model;

[0080] S4. After obtaining the coil end face model, the center coordinates of the coil are obtained for the first time through the center fitting algorithm. The center fitting algorithm mainly includes the following steps:

[0081] Step 1: Plane fitting: Construct a three-dimensional discrete point coordinate system (x i ,y i ,z i ), where x i 、y i 、z i Respectively represent the horizontal, vertical, and front-back coordinates of the discrete points;

[0082] The discrete point cloud is fitted by the least squares method, that is, all discrete points are located in the same plane, and the plane equation is expressed as:

[0083] ax+by+cz=1

[0084] It is presented in matrix form as follows:

[0085]

[0086] If the order (a,b,c) T =A,(1,1,...1) T =N

[0087] but

[0088] M·A=N

[0089] Further solve the plane vector A by minimizing the residual:

[0090] A=(M T M) -1 ·M T ·N

[0091] Among them, a represents the horizontal component of the plane normal vector, b represents the vertical component of the plane normal vector, z represents the forward and backward component of the plane normal vector, and x represents the n Represents the horizontal coordinate of the discrete point, y n Represents the vertical coordinate of the discrete point, z n Represents the coordinates of the discrete point in the forward and backward directions, and T represents the symbol of matrix transpose in linear algebra;

[0092] Step 2. Find the center of the circle: Let the coordinates of the center of the circle be (x0, y0, z0). Take any two discrete points P1 (x1, y1, z1) and P2 (x2, y2, z2). The center of the circle C must satisfy the following condition: the perpendicular bisector of the line connecting P1 and P2 passes through the center of the circle. Then the vector connecting P1 and P2 is:

[0093] V1=(x2-x1,y2-y1,z2-z1)

[0094] The coordinates of the midpoint of the line connecting P1 and P2 are:

[0095]

[0096] Center C and P 中点 The connection vector is

[0097]

[0098] According to the perpendicular bisector property, vector V1 is perpendicular to V2, so the dot product is zero:

[0099] V1·V2=0

[0100] Substituting the specific vectors of V1 and V2 into the above formula, we get the linear equation:

[0101]

[0102] Simplifying the above equation:

[0103]

[0104] Among them, (x0, y0, z0) represents the three-dimensional coordinates of the center C, (x1, y1, z1) represents the three-dimensional coordinates of the discrete point P1, (x2, y2, z2) represents the three-dimensional coordinates of the discrete point P2, V1 represents the line vector connecting P1 and P2, and V2 represents the line vector between the center C and P 中点 Connection vector, P 中点 Indicates the coordinates of the midpoint of the line connecting P1 and P2;

[0105] The above formula is expressed in matrix form as follows:

[0106]

[0107] Where Δx (n-1)n =x n -x (n-1) ,

[0108] If the order

[0109] but

[0110] B·C=L

[0111] Among them, x n Indicates the horizontal coordinate of the nth discrete point, y n Represents the vertical coordinate of the nth discrete point, z n represents the forward and backward coordinates of the nth discrete point, and T represents the standard symbol for matrix transposition in linear algebra;

[0112] Since the center C is found to be in the plane in step 1, then:

[0113] A·C=1

[0114] Combining the above two equations, and presenting them in matrix form is as follows:

[0115]

[0116] Using the least squares method to solve, the closed solution of the circle center coordinate C is:

[0117]

[0118] Express the above formula in the form of three-dimensional coordinates:

[0119] Let the merge matrix The coordinates of the circle center are:

[0120]

[0121] S5. Calculate the moving coordinates required for centering the coil trolley using the center coordinates of the coil and the mandrel obtained for the first time. The calculation method for the moving coordinates required for centering the coil trolley is:

[0122] Δx=x 钢卷1 -x 芯轴1

[0123] Δy=y 钢卷1 -y 芯轴1

[0124] Among them, Δx represents the horizontal movement of the winding trolley, Δy represents the vertical movement of the winding trolley, and x 钢卷1 Indicates the horizontal coordinate of the center of the steel coil obtained for the first time, y 钢卷1 Indicates the vertical coordinate of the center of the steel coil obtained for the first time, x 芯轴1 Indicates the horizontal coordinate of the center of the mandrel obtained for the first time, y 芯轴1 Indicates the vertical coordinate of the center of the mandrel obtained for the first time;

[0125] S6. The moving coordinates are transmitted to the PLC controller in real time through the control room. The PLC controller drives the coiling trolley to move up and down according to the moving coordinates to align the center of the steel coil with the center of the mandrel.

[0126] S7. After the coiling trolley is adjusted, the 3D point cloud data of the coil end face and the mandrel is acquired again through the LiDAR sensor. The center coordinates of the coil and mandrel are obtained again through point cloud reconstruction technology and the circle center fitting algorithm. The deviation of the center coordinates of the coil and mandrel is calculated. If the deviation is less than or equal to 6mm, the coil is automatically loaded. If the deviation is greater than 6mm, the automatic loading is stopped and manual confirmation is performed. The deviation of the center coordinates of the coil and mandrel is calculated as follows:

[0127] x ε =x 钢卷2 -x 芯轴2

[0128] y ε =y 钢卷2 -y 芯轴2

[0129] Among them, x ε Indicates the horizontal deviation of the center coordinates of the steel coil and the core shaft, y ε Indicates the vertical deviation of the center coordinates of the steel coil and the core shaft, x 钢卷2 Indicates the horizontal coordinate of the center of the steel coil obtained for the second time, y 钢卷2 Indicates the vertical coordinate of the center of the steel coil obtained for the second time, x 芯轴2 Indicates the horizontal coordinate of the center of the mandrel obtained for the second time, y 芯轴2 Indicates the vertical coordinate of the center of the mandrel obtained for the second time.

[0130] Table 1 - Comparison of data before and after implementation of this method

[0131]

[0132] By implementing this method, the automatic coiling rate in the cold rolling entrance section increased from 62.5% to 93.75%, and the vertical coordinate deviation was reduced from 8.89mm to 2.45mm. As a result, the automation rate was significantly improved. The system was able to independently complete most coiling operations, and the need for manual intervention was reduced from 37.5% to 6.25%, greatly shortening the downtime adjustment time. At the same time, the vertical coordinate deviation was reduced, effectively avoiding steel coil core pulling and mandrel collision accidents caused by excessive deviation.

[0133] Example 2

[0134] A method for centering a cold-rolled pickled steel coil based on a 3D point cloud comprises the following steps:

[0135] S1. First, the control room issues a command to make the PLC controller drive the coil trolley to transport the steel coil to 1m in front of the uncoiler mandrel;

[0136] S2. Obtain 3D point cloud data of the end surface of the steel coil on the coil loading trolley through the LiDAR sensor. The LiDAR sensor is an area array LiDAR that uses TOF ranging technology. The point cloud density of a single scan is 10,000 points.

[0137] S3. The acquired 3D point cloud data is spliced ​​into a complete steel coil end face model using point cloud reconstruction technology. The method for splicing the model using point cloud reconstruction technology includes the following steps:

[0138] Step 1: Align the acquired 3D point cloud data into the same coordinate system;

[0139] Step 2: Remove noise and abnormal points through Gaussian filtering and statistical outlier removal;

[0140] Step 3: Splice the processed point cloud data into a complete steel coil end face model and mandrel model;

[0141] S4. After obtaining the coil end face model, the center coordinates of the coil are obtained for the first time through the center fitting algorithm. The center fitting algorithm mainly includes the following steps:

[0142] Step 1: Plane fitting: Construct a three-dimensional discrete point coordinate system (x i ,y i ,z i ), where x i 、y i 、z i Respectively represent the horizontal, vertical, and front-back coordinates of the discrete points;

[0143] The discrete point cloud is fitted by the least squares method, that is, all discrete points are located in the same plane, and the plane equation is expressed as:

[0144] ax+by+cz=1

[0145] It is presented in matrix form as follows:

[0146]

[0147] If the order (a,b,c) T =A,(1,1,...1) T =N

[0148] but

[0149] M·A=N

[0150] Further solve the plane vector A by minimizing the residual:

[0151] A=(M T M) -1 ·M T ·N

[0152] Among them, a represents the horizontal component of the plane normal vector, b represents the vertical component of the plane normal vector, z represents the forward and backward component of the plane normal vector, and x represents the n Represents the horizontal coordinate of the discrete point, y n Represents the vertical coordinate of the discrete point, z n Represents the coordinates of the discrete point in the forward and backward directions, and T represents the symbol of matrix transpose in linear algebra;

[0153] Step 2. Find the center of the circle: Let the coordinates of the center of the circle be (x0, y0, z0). Take any two discrete points P1 (x1, y1, z1) and P2 (x2, y2, z2). The center of the circle C must satisfy the following condition: the perpendicular bisector of the line connecting P1 and P2 passes through the center of the circle. Then the vector connecting P1 and P2 is:

[0154] V1=(x2-x1,y2-y1,z2-z1)

[0155] The coordinates of the midpoint of the line connecting P1 and P2 are:

[0156]

[0157] Center C and P 中点 The connection vector is

[0158]

[0159] According to the perpendicular bisector property, vector V1 is perpendicular to V2, so the dot product is zero:

[0160] V1·V2=0

[0161] Substituting the specific vectors of V1 and V2 into the above formula, we get the linear equation:

[0162]

[0163] Simplifying the above equation:

[0164]

[0165] Among them, (x0, y0, z0) represents the three-dimensional coordinates of the center C, (x1, y1, z1) represents the three-dimensional coordinates of the discrete point P1, (x2, y2, z2) represents the three-dimensional coordinates of the discrete point P2, V1 represents the line vector connecting P1 and P2, and V2 represents the line vector between the center C and P 中点 Connection vector, P 中点 Indicates the coordinates of the midpoint of the line connecting P1 and P2;

[0166] The above formula is expressed in matrix form as follows:

[0167]

[0168] Where Δx (n-1)n =x n -x (n-1) ,

[0169] If the order

[0170] but

[0171] B·C=L

[0172] Among them, x n Indicates the horizontal coordinate of the nth discrete point, y n Represents the vertical coordinate of the nth discrete point, z n represents the forward and backward coordinates of the nth discrete point, and T represents the standard symbol for matrix transposition in linear algebra;

[0173] Since the center C is found to be in the plane in step 1, then:

[0174] A·C=1

[0175] Combining the above two equations, and presenting them in matrix form is as follows:

[0176]

[0177] Using the least squares method to solve, the closed solution of the circle center coordinate C is:

[0178]

[0179] Express the above formula in the form of three-dimensional coordinates:

[0180] Let the merge matrix The coordinates of the circle center are:

[0181]

[0182] The center coordinates of the steel coil are obtained through the above algorithm model as (63, 732, 4851), and the center coordinates of the core shaft are known to be (55, 786, 8088);

[0183] S5. Calculate the moving coordinates required for centering the coil trolley using the center coordinates of the coil and the mandrel obtained for the first time. The calculation method for the moving coordinates is:

[0184] Δx=x 钢卷1 -x 芯轴1

[0185] Δx=63-55=8mm

[0186] Δy=y 钢卷1 -y芯轴1

[0187] Δy=786-732=54mm

[0188] Among them, Δx represents the horizontal movement of the winding trolley, Δy represents the vertical movement of the winding trolley, and x 钢卷1 Indicates the horizontal coordinate of the center of the steel coil obtained for the first time, y 钢卷1 Indicates the vertical coordinate of the center of the steel coil obtained for the first time, x 芯轴1 Indicates the horizontal coordinate of the center of the mandrel obtained for the first time, y 芯轴1 Indicates the vertical coordinate of the center of the mandrel obtained for the first time;

[0189] S6. The moving coordinates are transmitted to the PLC controller in real time through the control room. The PLC controller drives the coiling trolley to move 54 mm according to the moving coordinates to align the center of the steel coil with the center of the mandrel.

[0190] S7. After the coil trolley is adjusted, the 3D point cloud data of the coil end face and the mandrel is obtained again using the LiDAR sensor. The center coordinates of the coil and mandrel are obtained again using point cloud reconstruction technology and a circle center fitting algorithm. The center coordinates of the coil are (58, 788, 4855), and the center coordinates of the mandrel are (55, 786, 8088). The center coordinate deviation of the coil and mandrel is calculated as:

[0191] x ε =x 钢卷2 -x 芯轴2

[0192] x ε =61-58=3mm

[0193] y ε =y 钢卷2 -y 芯轴2

[0194] y ε =788-786=2mm

[0195] Since the deviation x ε and y ε All are less than 6mm, so the steel coil is automatically wound;

[0196] Among them, x ε Indicates the horizontal deviation of the center coordinates of the steel coil and the core shaft, y ε Indicates the vertical deviation of the center coordinates of the steel coil and the core shaft, x 钢卷2 Indicates the horizontal coordinate of the center of the steel coil obtained for the second time, y 钢卷2 Indicates the vertical coordinate of the center of the steel coil obtained for the second time, x芯轴2 Indicates the horizontal coordinate of the center of the mandrel obtained for the second time, y 芯轴2 Indicates the vertical coordinate of the center of the mandrel obtained for the second time.

[0197] The above is only an embodiment of the present invention, and the common knowledge such as the specific structure and characteristics of the scheme is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for centering cold-rolled pickled steel coils based on 3D point clouds, characterized in that: The following steps are involved: S1, first transport the steel coil to the uncoiler mandrel via the coiling trolley; S2, using a laser radar sensor to obtain 3D point cloud data of the end surface of the steel coil on the coiling trolley; S3. Splicing the acquired 3D point cloud data into a complete steel coil end face model using point cloud reconstruction technology; S4. After obtaining the steel coil end face model, the center coordinates of the steel coil are obtained for the first time through the circle center fitting algorithm; S5. Calculate the movement coordinates required for centering the coil trolley using the center coordinates of the coil and the mandrel obtained for the first time; S6. The moving coordinates are transmitted to the PLC controller in real time through the control room. The PLC controller drives the coiling trolley to move up and down according to the moving coordinates to align the center of the steel coil with the center of the mandrel. S7. After the coiling trolley is adjusted, the 3D point cloud data of the coil end face is obtained again through the laser radar sensor, and the center coordinates of the coil are obtained for the second time through point cloud reconstruction technology and circle center fitting algorithm. The deviation of the center coordinates of the coil is calculated. If the deviation is less than or equal to the threshold, the coil is automatically loaded; if the deviation is greater than the threshold, the automatic loading is stopped and manual confirmation is performed.

2. The method for centering a cold-rolled pickled steel coil based on 3D point cloud according to claim 1, characterized in that: In S2, the laser radar sensor is an array laser radar that uses TOF ranging technology, and the point cloud density of a single scan is 10,000 points.

3. The method for centering a cold-rolled pickled steel coil based on 3D point cloud according to claim 1, characterized in that: In S3, the method for stitching models using point cloud reconstruction technology includes the following steps: Step 1: Align the acquired 3D point cloud data into the same coordinate system; Step 2: Remove noise and abnormal points through Gaussian filtering and statistical outlier removal; Step 3: Splice the processed point cloud data into a complete steel coil end face model and core shaft model.

4. The method for centering a cold-rolled pickled steel coil based on 3D point cloud according to claim 1, characterized in that: In S4, the circle center fitting algorithm mainly includes the following steps: Step 1: Plane fitting: Construct a three-dimensional discrete point coordinate system (x i ,y i ,z i ), where x i 、y i 、z i Respectively represent the horizontal, vertical, and front-back coordinates of the discrete points; The discrete point cloud is fitted by the least squares method, that is, all discrete points are located in the same plane, and the plane equation is expressed as: ax+by+cz=1 It is presented in matrix form as follows: but M·A=N Further solve the plane vector A by minimizing the residual: A=(M T ·M) -1 ·M T ·N Among them, a represents the horizontal component of the plane normal vector, b represents the vertical component of the plane normal vector, z represents the forward and backward component of the plane normal vector, and x represents the n Indicates the horizontal coordinate of the discrete point, y n Represents the vertical coordinate of the discrete point, z n Represents the coordinates of the discrete point in the forward and backward directions, and T represents the symbol of matrix transpose in linear algebra; Step 2. Find the center of the circle: Let the coordinates of the center of the circle be (x0, y0, z0). Take any two discrete points P1 (x1, y1, z1) and P2 (x2, y2, z2). The center of the circle C must satisfy the following condition: the perpendicular bisector of the line connecting P1 and P2 passes through the center of the circle. Then the vector connecting P1 and P2 is: V1=(x2-x1,y2-y1,z2-z1) The coordinates of the midpoint of the line connecting P1 and P2 are: Center C and P 中点 The connection vector is According to the perpendicular bisector property, vector V1 is perpendicular to V2, so the dot product is zero: V1·V2=0 Substituting the specific vectors of V1 and V2 into the above formula, we get the linear equation: Simplifying the above equation: Among them, (x0, y0, z0) represents the three-dimensional coordinates of the center C, (x1, y1, z1) represents the three-dimensional coordinates of the discrete point P1, (x2, y2, z2) represents the three-dimensional coordinates of the discrete point P2, V1 represents the line vector connecting P1 and P2, and V2 represents the line vector between the center C and P 中点 Connection vector, P 中点 Indicates the coordinates of the midpoint of the line connecting P1 and P2; The above formula is expressed in matrix form as follows: in, If the order but B·C=L Among them, x n Indicates the horizontal coordinate of the nth discrete point, y n Represents the vertical coordinate of the nth discrete point, z n represents the forward and backward coordinates of the nth discrete point, and T represents the standard symbol for matrix transposition in linear algebra; Since the center C is found to be in the plane in step 1, then: A·C=1 Combining the above two equations, and presenting them in matrix form is as follows: Using the least squares method to solve, the closed solution of the circle center coordinate C is: Express the above formula in the form of three-dimensional coordinates: Let the merge matrix The coordinates of the circle center are:

5. The method for centering cold-rolled pickled steel coil based on 3D point cloud according to claim 1, characterized in that: In S5, the calculation method of the moving coordinates required for the centering of the winding trolley is: Δx=x 钢卷1 -x 芯轴1 Δy=y 钢卷1 -y 芯轴1 Among them, Δx represents the horizontal movement of the winding trolley, Δy represents the vertical movement of the winding trolley, and x 钢卷1 Indicates the horizontal coordinate of the center of the steel coil obtained for the first time, y 钢卷1 Indicates the vertical coordinate of the center of the steel coil obtained for the first time, x 芯轴1 Indicates the horizontal coordinate of the center of the mandrel obtained for the first time, y 芯轴1 Indicates the vertical coordinate of the center of the mandrel obtained for the first time.

6. The method for centering cold-rolled pickled steel coils based on 3D point cloud according to claim 1, characterized in that: In S7, the calculation method of the center coordinate deviation of the steel coil and the core shaft is: x ε =x 钢卷2 -x 芯轴2 and ε =and 钢卷2 -and 芯轴2 Among them, x ε Indicates the horizontal deviation of the center coordinates of the steel coil and the core shaft, y ε Indicates the vertical deviation of the center coordinates of the steel coil and the core shaft, x 钢卷2 Indicates the horizontal coordinate of the center of the steel coil obtained for the second time, y 钢卷2 Indicates the vertical coordinate of the center of the steel coil obtained for the second time, x 芯轴2 Indicates the horizontal coordinate of the center of the mandrel obtained for the second time, y 芯轴2 Indicates the vertical coordinate of the center of the mandrel obtained for the second time.

7. The method for centering cold-rolled pickled steel coils based on 3D point cloud according to claim 1, characterized in that: In S7, the threshold is 6 mm.

Images