A multi-bolt tightening method based on rolling equalization control and related device

Abstract

The application discloses a multi-bolt tightening method based on rolling balance control and a related device, and belongs to the technical field of bolt tightening processes; the method obtains an initial axial force vector by measuring the current axial force of all N bolts; and gathers the measurement quality information of each bolt; M tentative point bolts are selected from the N bolts, and an angle increment is sequentially applied, and after each application, the full-circle axial force is immediately re-measured to obtain a change amount Af, and an influence matrix is constructed and updated online; the confidence weight of each bolt is calculated according to the measurement quality information to form a confidence matrix; and the bolt state is divided and marked through a weight threshold; a quadratic programming (QP) problem is constructed, and the angle increment vector Du to be executed in the current round is solved; the angle tightening operation is executed in combination with the angle increment vector Du to be executed in the current round and the bolt state; the axial force of all N bolts is measured again, and the influence matrix is updated; and the cycle is repeated for multiple times until a preset termination condition is met.

Description

Technical Field

[0001] This invention belongs to the field of bolt tightening technology, and relates to a multi-bolt tightening method and related device based on rolling balance control. Background Technology

[0002] As a crucial component of clean and renewable energy, the reliability and safety of wind power equipment are paramount. Key mechanical connections in wind turbines, such as the tower and hub, main shaft and hub, and main shaft and gearbox, commonly employ high-strength ring flange bolt connections. The core of this type of connection lies in applying precise axial preload through the bolts to generate sufficient clamping force at the mating surfaces, enabling the transmission of enormous moments, bending moments, and shear loads. Therefore, it is essential not only that the preload of each individual bolt reaches the design target range, but also that the preload distribution across the entire ring of bolts is highly uniform, ensuring even stress distribution on the flange mating surfaces and preventing localized stress concentrations.

[0003] However, due to the elastic interactions between flanges, gaskets, and connected components under stress, complex coupling effects (also known as "bolt interactions") exist among multiple bolts. This means that tightening or adjusting a single bolt significantly alters the existing axial force distribution of its adjacent bolts and even the entire ring of bolts. This can lead to uneven preload even if each individual bolt meets the requirements, potentially causing warping of the connection surface, bolt loosening, or even fatigue fracture. This seriously threatens the safe and stable operation of wind turbine units.

[0004] Currently, automated tightening systems (such as industrial robots or smart wrenches) are increasingly widely used in the assembly of large structures. However, the control modes of existing technologies have significant limitations. Most systems still employ fixed-sequence tightening strategies (such as cross-tightening sequence), supplemented by the detection and control of the axial force of individual bolts. This mode is essentially an "open-loop" or "partially closed-loop" control, which focuses only on the single objective of the currently operated bolt, and fails to establish a global closed-loop control mechanism for "balanced axial force distribution across the entire bolt ring".

[0005] In summary, there is an urgent need to develop a closed-loop tightening method for wind turbine flange bolts, which controls the multi-bolt robot sequence tightening strategy to achieve convergence of the axial force vector to the target equilibrium state throughout the entire rotation; in order to improve the accuracy of operation and ensure the safe and stable operation of wind turbine units. Summary of the Invention

[0006] The purpose of this invention is to provide a multi-bolt tightening method and related device based on rolling balance control, so as to solve the technical problem that the bolt tightening method of the prior art is affected by the bolt coupling effect, resulting in poor reliability and safety.

[0007] To achieve the above objectives, the present invention employs the following technical solution: In a first aspect, the present invention provides a multi-bolt tightening method based on rolling balance control, comprising the following steps: Measure the current axial force of all N bolts to obtain the initial axial force vector. And collect the measurement quality information of each bolt; Select M test bolts from N bolts and apply angle increments sequentially. After each application, the change in axial force across the entire circle is immediately measured to obtain the change Δf, and the influence matrix is ​​constructed and updated online. ; The confidence weight of each bolt is calculated based on the measured quality information, forming a confidence matrix. ; and the bolt status is divided and marked using weighted thresholds; According to the influence matrix and confidence matrix Construct a quadratic programming (QP) problem and solve it to obtain the angle increment vector Δu to be executed in this round; Based on the angle increment vector Δu to be executed in this round and the bolt status, perform the angle tightening operation; measure the axial force of all N bolts again and update the influence matrix. Repeat the process multiple times until the preset termination condition is met.

[0008] Furthermore, the measurement quality information includes at least the ultrasonic echo signal-to-noise ratio (SNR), temperature stability, and torque-angle fitting residual.

[0009] Furthermore, the step involves selecting M test bolts from N bolts and sequentially applying minute angular increments. After each application, the change in axial force across the entire circle is immediately measured to obtain the change Δf, and the influence matrix is ​​constructed and updated online. The steps specifically include: Select M test bolts from N bolts, and apply a preset angle increment to each of the selected test bolts in turn. , In each application Then, immediately remeasure the axial force of all N bolts to obtain the change in axial force. ; The influence matrix is ​​constructed and updated online using a recursive least squares (RLS) algorithm with a forgetting factor. The specific recursive formula is as follows:

[0010] In the formula, This is the gain matrix; Forgetting factor; It is the covariance matrix; is the column vector of the influence matrix.

[0011] Furthermore, the confidence weight of each bolt is calculated based on the measured quality information to form a confidence matrix. The steps of classifying and marking bolt states using weighted thresholds specifically include: Calculate the confidence weight of each bolt based on the measured quality information. Forming a confidence matrix ; Confidence weight The bolts are marked as frozen and will not participate in the tightening action in this round.

[0012] Furthermore, the statement based on the influence matrix and confidence matrix The steps for constructing a quadratic programming (QP) problem and solving it to obtain the angle increment vector Δu to be executed in the current round include: According to the influence matrix and confidence matrix Constructing a quadratic programming (QP) problem; Solving the quadratic programming (QP) problem yields the angle increment vector Δu to be executed in this round; the solution process satisfies the constraints. ; This represents the single-step angle increment for a single bolt. This represents the upper limit of the angle increment for a single bolt in a single step. This represents the cumulative angle increment of the bolt. This is the preset threshold for the cumulative angle increment of the bolt.

[0013] Furthermore, the objective function of the quadratic programming (QP) problem is:

[0014] In the formula, For the influence matrix; This is the confidence matrix; This is the current axial force vector; The target axial force vector.

[0015] Furthermore, the preset termination conditions include: a. ; The variance of the axial force of the entire bolt ring; The target variance; b. ; This indicates extremely poor axial force across the entire bolt ring; This indicates a very poor target. c. Improvement rate ; A preset threshold representing the improvement rate; d. ; Indicates the iteration round; Indicates the maximum number of rounds; The tightening process will terminate if any one of the above termination conditions a~d is met.

[0016] Secondly, the present invention provides a multi-bolt tightening system based on rolling equalization control, comprising: The information acquisition module is used to measure the current axial force of all N bolts and obtain the initial axial force vector. And collect the measurement quality information of each bolt; The matrix update module is used to select M test point bolts from N bolts and apply angle increments sequentially. After each application, the change in axial force across the entire circle is immediately measured to obtain the change Δf, and the influence matrix is ​​constructed and updated online. ; The bolt status management module is used to calculate the confidence weight of each bolt based on the measured quality information, forming a confidence matrix. ; and the bolt status is divided and marked using weighted thresholds; The rolling optimization solution module is used to solve problems based on the influence matrix. and confidence matrix Construct a quadratic programming (QP) problem and solve it to obtain the angle increment vector Δu to be executed in this round; The multi-round loop module is used to combine the angle increment vector Δu to be executed in this round with the bolt status to perform the angle tightening operation; it then measures the axial force of all N bolts again and updates the influence matrix. Repeat the process multiple times until the preset termination condition is met.

[0017] Thirdly, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the multi-bolt tightening method based on rolling balance control as described above.

[0018] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the multi-bolt tightening method based on rolling balance control described above.

[0019] Compared with the prior art, the present invention has the following beneficial effects: This invention discloses a multi-bolt tightening method and related apparatus based on rolling equilibrium control. Through initial axial force measurement and full-circle re-measurement mechanisms, it acquires real-time global information on the force state of the bolt group, overcoming the limitation of traditional single-point control in failing to perceive coupling effects, thus laying a data foundation for overall equilibrium control. Secondly, by introducing a strategy of online construction and dynamic updating of the influence matrix G, it can accurately characterize the mechanical coupling relationship between bolts, enabling the system to have adaptive learning and environmental response capabilities. This effectively eliminates model errors caused by flange deformation, material creep, or changes in operating conditions, significantly improving the accuracy and robustness of control. Furthermore, by introducing a confidence weight matrix W and bolt state classification based on measurement quality, the system can identify and isolate interference caused by sensor anomalies, data lag, or local faults, preventing unreliable data from participating in decision-making, further enhancing the system's anti-interference capability and reliability. This method transforms the tightening operation into a quadratic programming problem with engineering constraints, solving for the optimal adjustment strategy using rolling optimization, ensuring both operational safety and feasibility while significantly improving convergence efficiency. Ultimately, through multiple iterations, retests, and model updates, this invention enables the axial force distribution of the entire bolt ring to converge quickly and stably to a uniform target state, greatly reducing the risk of connection surface warping, bolt loosening, and fatigue fracture. It is particularly suitable for high-value critical connection structures such as wind turbine towers and heavy equipment. It has low hardware dependence, strong versatility, and can be implemented on electric wrenches or maintenance robot controllers. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the system of the present invention; Figure 3 This is a flowchart of a multi-bolt tightening method based on rolling balance control, according to an embodiment of the present invention. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0023] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0024] See Figure 1 This invention discloses a multi-bolt tightening method based on rolling balance control, comprising the following steps: S1, Measure the current axial force of all N bolts to obtain the initial axial force vector. And collect the measurement quality information of each bolt; The measured quality information includes at least the ultrasonic echo signal-to-noise ratio (SNR), temperature stability, and torque-angle fitting residual.

[0025] S2, select M (8~12) test point bolts from N bolts, and test the nth bolt. Apply a very small angle (suggestion (The test result must not exceed the safety threshold). After each test, immediately retest the entire lap to obtain the change over the entire lap. After each trial, the change in axial force over the entire circumference was remeasured. Build and update the influence matrix online ; Select M test bolts from N bolts, and apply a preset angle increment to each of the selected test bolts in turn. , In each application Then, immediately remeasure the axial force of all N bolts to obtain the change in axial force. ; The influence matrix is ​​constructed and updated online using a recursive least squares (RLS) algorithm with a forgetting factor. The specific recursive formula is as follows:

[0026] In the formula, This is the gain matrix; Forgetting factor; It is the covariance matrix; is the column vector of the influence matrix. This table represents the first-order mapping relationship between a small angular increment and the change in axial force over the entire revolution. Multiple probes can be tested simultaneously using a block RLS, with the principle being small probe angles and uniform point distribution.

[0027] S3. Calculate the confidence weight of each bolt based on the measured quality information to form a confidence matrix. ; and the bolt status is divided and marked using weighted thresholds; Calculate the confidence weight of each bolt based on the measured quality information. Forming a confidence matrix ; Confidence weight The bolts are marked as frozen and will not participate in the tightening process in this round. Redundant retesting or the use of neighboring templates should be arranged to avoid bad data affecting the overall process.

[0028] S4, based on the influence matrix and confidence matrix Construct a quadratic programming (QP) problem and solve it to obtain the angle increment vector Δu to be executed in this round; According to the influence matrix and confidence matrix Construct a quadratic programming (QP) problem; the objective function of the quadratic programming (QP) problem is:

[0029] In the formula, For the influence matrix; This is the confidence matrix; This is the current axial force vector; The target axial force vector.

[0030] Solving the quadratic programming (QP) problem, based on the principles of approaching the objective and sparse actions, yields the angle increment vector Δu to be executed in the current round; the solution process satisfies the constraints. ; This represents a single-step angle increment for a single bolt. This represents the upper limit of the angle increment for a single bolt in a single step. This represents the cumulative angle increment of the bolt. This is a preset threshold for the cumulative angle increment of bolts. It avoids consecutive adjacent bolts and disrupts symmetry, ensuring the cumulative angle does not exceed the limit. This step yields a list of bolt numbers to be tightened in this round, along with their angle increments (e.g., #3: +8°, ​​#12: +5°, ... a total of 6 bolts).

[0031] S5, combining the angle increment vector Δu to be executed in this round and the bolt status, execute the angle tightening operation; measure the axial force of all N bolts again and update the influence matrix. Repeat the process multiple times until the preset termination condition is met.

[0032] Perform this round of tightening according to the established checklist, and immediately measure another rotation after completion. Then, re-measure the axial force for the entire rotation to obtain the new result. .

[0033] This round of actual Corresponding Re-enter and update the recursive formula of S2. This will improve the accuracy of the next round of decision-making.

[0034] The preset termination conditions include: a. ; The variance of the axial force of the entire bolt ring; The target variance; b. ; This indicates extremely poor axial force across the entire bolt ring; This indicates a very poor target; c. Improvement rate ; A preset threshold representing the improvement rate; d. ; Indicates the iteration round; Indicates the maximum number of rounds; The tightening process terminates when any one of the termination conditions a~d is met. A report containing information on the bolt tightening sequence, increments, measurements, and criteria is ultimately generated.

[0035] See Figure 2 This invention discloses a multi-bolt tightening system based on rolling equalization control, including an information acquisition module, a matrix update module, a bolt state management module, a rolling optimization solution module, and a multi-round loop module. The information acquisition module measures the current axial force of all N bolts to obtain an initial axial force vector. It also collects the measurement quality information of each bolt; the matrix update module is used to select M test point bolts from N bolts and apply angle increments sequentially. After each application, the change in axial force across the entire circle is immediately measured to obtain the change Δf, and the influence matrix is ​​constructed and updated online. The bolt status management module is used to calculate the confidence weight of each bolt based on the measured quality information, forming a confidence matrix. The bolt states are classified and marked using weighted thresholds; a rolling optimization solution module is used to solve the problem based on the influence matrix. and confidence matrix A quadratic programming (QP) problem is constructed, and the angle increment vector Δu to be executed in the current round is obtained by solving it. A multi-round loop module is used to combine the angle increment vector Δu to be executed in the current round with the bolt status to execute the angle tightening operation. The axial force of all N bolts is measured again and the influence matrix is ​​updated. Repeat the process multiple times until the preset termination condition is met.

[0036] Example: See Figure 3This embodiment focuses on 24 M30 high-strength bolts around a tower flange. The target axial force is uniformly set at 120 kN. The tool used is an electric wrench with angle and torque measurements, and the measurement employs ultrasonic TOF and visual fusion with an indicating washer. The maximum concurrent measurement is set at 4 bolts, the maximum single-step angle for the same bolt is 10 degrees, and the minimum trial angle is 2 degrees. (The judgment is based on two lines: variance not exceeding 7.2 squared kN). 2 The range is no higher than 21.6 kN. Here, variance refers to the dispersion of each point within a loop relative to 120 kN, and range refers to the difference between the maximum and minimum values ​​within a loop. Following step 1: After measuring the axial force in numerical order for one lap, an initial list is obtained. The first eight values ​​are 102, 110, 116, 130, 138, 129, 121, and 112 kN, and the remaining values ​​are obtained in the same way. The data from this lap are then statistically analyzed to obtain Table 1.

[0037] Table 1 Initial Axial Force Statistics

[0038] Conclusion: The average value is close to 120 kN, but the distribution is uneven, with both variance and range exceeding the judgment threshold, requiring balancing.

[0039] According to step 2: Select 8 evenly distributed points from the 24 as test points, numbered 1, 4, 7, 10, 13, 16, 19, and 22. The procedure is as follows: add 2 degrees to one of the points, immediately measure another lap, and use the difference between the two measurements as the influence curve of that point on the entire lap. Repeat this method to test all 8 points.

[0040] Here's an example of how to record a point. After adding 2 degrees to point 4, the neighboring points generally increase, the opposite edge increases slightly, and the far end remains almost unchanged. This change vector is used as the fourth column of the influence curve. The remaining seven points also have their own influence curves. Combining these eight columns yields the first version of the influence matrix G (intuitively explained as: if a certain number of points are added a certain degree in the next step, how many kN will the entire circle increase? The system uses this matrix to calculate the angle allocation for the next step). According to step 3: the measured quality of each bolt is assigned a score from 0 to 1, taking into account three factors: ultrasonic echo signal-to-noise ratio, temperature stability, and torque angle fitting residual. Bolts with low scores are considered unreliable for this round and are frozen until retesting confirms their quality. This round of evaluation shows that bolt number 17 has a score of 0.28, below the threshold of 0.35, and is therefore frozen. The scores of the remaining points are between 0.70 and 0.95, which is normal.

[0041] According to step 4: The goal is to push the axial force of one revolution towards the target of 120 kN, while ensuring each bead does not exceed 10 degrees per step, and a maximum of 4 beads are screwed per revolution, with symmetry as much as possible and avoiding consecutive screwing of adjacent beads. The influence matrix from step 2 and the difference between the current revolution and the target are then used to calculate which beads should be screwed in this revolution and by how many degrees.

[0042] The results of the first round of calculations are as follows: Add 8.5 degrees to number 1, 6.0 degrees to number 9, 5.5 degrees to number 13, and 7.0 degrees to number 21. These four are basically evenly spaced, not adjacent to each other, and also avoid number 17, which is currently freezing. According to the linear prediction of the influence matrix, after this round, the variance of one cycle is expected to decrease to approximately 8.9 squared. The range is estimated to be approximately 28 kN.

[0043] Follow-up step 5: Execute the steps concurrently, pairwise according to the checklist. Immediately after completion, retest the entire cycle to obtain new data after the first round. Statistical results are shown in the first row of Table 2. The measured variance is... The measured range was 29.0 kN, which is very close to the prediction in step 4, indicating that the influence matrix can reflect the on-site coupling relationship well. The overall error of this round has decreased compared to the initial value. The convergence is obvious.

[0044] According to step 6: Fill the influence matrix with the first round of distribution angle and retest changes, so that the matrix will better fit the current flange ring. At this time, perform a confidence assessment. After the retest on the 17th, the echo recovered and the score rose to 0.62. The freeze was lifted, and it can participate normally in the future.

[0045] The second round of the list was recalculated using the updated matrix, resulting in increments of 4.5 degrees for item 3, 5.0 degrees for item 11, 3.5 degrees for item 17, and 4.0 degrees for item 23. After execution and retesting, the statistics are shown in the second row of Table 2, with a variance of [variable value missing]. The range is 22.0 kN, and the overall error has decreased further. .

[0046] By the third round, the lap was relatively even, requiring only minor adjustments: adding 3.0 degrees to track number 6 and 2.5 degrees to track number 14. The variance after retesting was... The range was 19.5 kN, both of which were below the judgment threshold. The process was then completed and a report was issued. The report included which chips were tightened in each round, how many were tightened, one revolution curve, statistical indicators, freeze / thaw records, and exception codes, facilitating acceptance and traceability.

[0047] Table 2. Actions and Effects of Each Round

[0048] At this time, variance kN 2 If the square of the voltage is not higher than 7.2 and the range is not higher than 19.5 kN and not higher than 21.6 kN, the process terminates.

[0049] In one embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions from the computer storage medium to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of a multi-bolt tightening method based on rolling equalization control.

[0050] This invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the multi-bolt tightening method based on rolling equalization control in the above embodiments.

[0051] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0052] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0053] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0054] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A multi-bolt tightening method based on rolling balance control, characterized in that, Includes the following steps: Measure the current axial force of all N bolts to obtain the initial axial force vector. And collect the measurement quality information of each bolt; Select M test bolts from N bolts and apply angle increments sequentially. After each application, the change in axial force across the entire circle is immediately measured to obtain the change Δf, and the influence matrix is ​​constructed and updated online. ; The confidence weight of each bolt is calculated based on the measured quality information, forming a confidence matrix. ; and the bolt status is divided and marked using weighted thresholds; According to the influence matrix and confidence matrix Construct a quadratic programming (QP) problem and solve it to obtain the angle increment vector Δu to be executed in this round, specifically including: According to the influence matrix and confidence matrix Construct a quadratic programming (QP) problem; the objective function of the quadratic programming (QP) problem is: In the formula, For the influence matrix; This is the confidence matrix; This is the current axial force vector; The target axial force vector; The regularization coefficient is used to control the angle increment vector. sparsity; for The L1 norm represents the sum of the absolute values ​​of the angle increments of each bolt; Solving the quadratic programming (QP) problem yields the angle increment vector Δu to be executed in this round; the solution process satisfies the constraints. ; This represents the single-step angle increment for a single bolt. This represents the upper limit of the angle increment for a single bolt in a single step. Δu represents the number of non-zero elements, i.e., the number of bolts actually tightened in this round; K is the preset upper limit for the number of bolts involved in tightening in this round. Based on the angle increment vector Δu to be executed in this round and the bolt status, perform the angle tightening operation; measure the axial force of all N bolts again and update the influence matrix. Repeat the process multiple times until the preset termination condition is met.

2. The multi-bolt tightening method based on rolling balance control according to claim 1, characterized in that, The measured quality information includes at least the ultrasonic echo signal-to-noise ratio (SNR), temperature stability, and torque-angle fitting residual.

3. The multi-bolt tightening method based on rolling balance control according to claim 1, characterized in that, The process involves selecting M test bolts from N bolts and sequentially applying angle increments. After each application, the change in axial force across the entire circle is immediately measured to obtain the change Δf, and the influence matrix is ​​constructed and updated online. The steps specifically include: Select M test bolts from N bolts, and apply a preset angle increment to each of the selected test bolts in turn. , In each application Then, immediately remeasure the axial force of all N bolts to obtain the change in axial force. ; The influence matrix is ​​constructed and updated online using a recursive least squares (RLS) algorithm with a forgetting factor. The specific recursive formula is as follows: In the formula, This is the gain matrix; Forgetting factor; It is the covariance matrix; These are the column vectors of the influence matrix.

4. The multi-bolt tightening method based on rolling balance control according to claim 1, characterized in that, The confidence weight of each bolt is calculated based on the measured quality information to form a confidence matrix. ; The steps of classifying and marking bolt states using weighted thresholds specifically include: Calculate the confidence weight of each bolt based on the measured quality information. Forming a confidence matrix ; Confidence weight The bolts are marked as frozen and will not participate in the tightening action in this round.

5. A multi-bolt tightening method based on rolling balance control according to claim 1, characterized in that, The preset termination conditions include: a. ; The variance of the axial force of the entire bolt ring; The target variance; b. ; This indicates extremely poor axial force across the entire bolt ring; This indicates a very poor target. c. Improvement rate In the formula, Let be the axial force vector of the bolt after the t-th round of tightening, measured again. Let be the axial force vector of the bolt after the (t+1)th tightening cycle, where t is the iteration cycle and t=0,1,2,… A preset threshold representing the improvement rate; d. ; Indicates the iteration round; Indicates the maximum number of rounds; The tightening process will terminate if any one of the above termination conditions a~d is met.

6. A multi-bolt tightening system based on rolling equalization control, characterized in that, A multi-bolt tightening method based on rolling balance control according to any one of claims 1 to 5 includes: The information acquisition module is used to measure the current axial force of all N bolts and obtain the initial axial force vector. And collect the measurement quality information of each bolt; The matrix update module is used to select M test point bolts from N bolts and apply angle increments sequentially. After each application, the change in axial force across the entire circle is immediately measured to obtain the change Δf, and the influence matrix is ​​constructed and updated online. ; The bolt status management module is used to calculate the confidence weight of each bolt based on the measured quality information, forming a confidence matrix. ; and the bolt status is divided and marked using weighted thresholds; The rolling optimization solution module is used to solve problems based on the influence matrix. and confidence matrix Construct a quadratic programming (QP) problem and solve it to obtain the angle increment vector Δu to be executed in this round; The multi-round loop module is used to combine the angle increment vector Δu to be executed in this round with the bolt status to perform the angle tightening operation; it then measures the axial force of all N bolts again and updates the influence matrix. Repeat the process multiple times until the preset termination condition is met.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the multi-bolt tightening method based on rolling balance control as described in any one of claims 1-5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-bolt tightening method based on rolling balance control as described in any one of claims 1-5.

Images