A method and system for calculating the center and roundness of a guide bearing of a large Francis hydroelectric generator set during disc turning

Abstract

This invention discloses a method and system for calculating the center and roundness of the turning guide bearing of a large mixed-flow hydro-generator unit, relating to hydro-generator unit installation and maintenance technology. The method includes: collecting swing data in the X and Y directions from four sections: the water guide bearing, the rotor lower flange, the lower guide bearing, and the upper guide bearing; based on the collected swing data, each section is equivalent to a cam, and the coordinates of the contour points of each section are restored using a reversal method; an improved minimum swing method is used to locate the optimal center, defining the roundness evaluation function as the difference between the maximum and minimum distances from all contour points to the undetermined center, and determining the center coordinates that minimize this score through a three-stage "coarse-medium-fine" grid search strategy; calculating the X and Y direction data separately and taking the average as the final center of the section; based on the characteristic that the base circle radius is much larger than the swing amplitude, the design radius on the drawing is used as the base circle radius for contour point restoration; and an N / e optimal stopping strategy is introduced to optimize the selection of maintenance schemes.

Description

Technical Field

[0001] This invention relates to the field of hydro-generator installation and maintenance technology, specifically to a method and system for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator. Background Technology

[0002] The coaxiality of the turbine generator set's shaft is a key indicator determining the equipment's operational stability. Excessive shaft runout can easily lead to bearing wear, coupling deformation, and even rotor rubbing accidents, severely impacting power generation efficiency and equipment lifespan. Turning the rotor slowly, as the core method for detecting shaft coaxiality, collects runout data from four key sections: the water guide bearing, the rotor's lower flange, the lower guide bearing, and the upper guide bearing. This data is then used to determine the coaxial relationship of each section and guide shaft adjustment.

[0003] Traditional turning gear data processing uses the average geometric center method, which determines the center coordinates of the cross-section by decomposing the horizontal and vertical components of the dial gauge readings in the X and Y directions and taking the average value. However, this method has significant limitations: first, it is only applicable to ideal cross-sections that are centrally symmetrical; when the cross-section has asymmetrical wear or deformation, the deviation of the symmetrical point data will lead to inaccurate center positioning; second, it ignores the differences in the actual radii of each cross-section, simply equating the swing data with the eccentricity, and does not consider the influence of the actual radius of the cross-section on the roundness error. With the increase in the capacity and size of large hydropower station units, the accuracy requirements for turning gear data processing have increased from 0.01mm to 0.001mm, and traditional methods can no longer meet the engineering requirements.

[0004] Furthermore, traditional rotary cranking maintenance faces two major decision-making challenges: First, the selection of the base circle radius lacks theoretical basis, as the actual base circle cannot be accurately measured, leading to doubts about the reliability of contour point reconstruction; second, the selection of maintenance schemes relies on experience, resulting in problems of "blind trial and error" or "over-testing"—either missing the optimal solution or making premature decisions leading to excessive costs. Meanwhile, the swing adjustment lacks scientific limit references, and blindly pursuing extremely small swing angles easily leads to resource waste.

[0005] To address the aforementioned issues, this patent proposes a dual-optimization calculation method based on cross-sectional profile reconstruction: First, the coordinates of the actual profile points of each cross-section are reconstructed using sway data, and the independence of the base circle radius is verified; then, the center coordinates are calculated using both the least squares fitting method and the improved minimum sway method; finally, the scientific adjustment target is clarified by combining the sway limit characteristics, and the N / e optimal stopping strategy is introduced to optimize the selection of maintenance schemes, providing insights for high-precision turning gear data processing and decision-making. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit includes the following steps: During the turning process, the swing data of each section is collected. The above sections include the water guide bearing, the rotor lower flange, the lower guide bearing and the upper guide bearing. Measuring devices are set up along the X and Y directions at each section. The swing data is recorded once every 45° rotation. Data at a total of 8 angular positions are collected. Based on the collected swing data, each section is equivalent to a cam, and the coordinates of the contour points of each section are restored by the inversion method. The base circle radius of the section is taken as the design radius of the drawing. The coordinates of the contour points are determined according to the swing value of each measured angle and the cosine and sine values ​​of the measured angle, and the coordinates of the eight contour points in the X direction and the eight contour points in the Y direction are obtained. For each cross section, the improved minimum swing method is used to locate the optimal center. The improved minimum swing method aims at the optimal roundness. The roundness evaluation function is defined as the difference between the maximum and minimum distances from all contour points to the center to be determined. The larger of the differences between the contour points in the X direction and the contour points in the Y direction is taken as the comprehensive score. The center coordinates that minimize this score are determined by a three-stage ladder grid search strategy of "coarse-medium-fine". After calculating the X-direction profile point data and the Y-direction profile point data respectively, the average value of the two center coordinates is taken as the final center coordinate of the cross section.

[0009] Furthermore, in the steps described above for restoring the coordinates of each section profile point using the inversion method, the coordinates of the profile points in the X direction are determined using the following formula: The coordinates of the contour points in the Y direction are determined using the following formula: Where R is the radius of the base circle of the cross section, which is taken from the design radius on the drawing; , For measuring angles, i and j are the measurement point numbers, with values ​​ranging from 1 to 8; For the X direction in the measurement angle The swing value, For the Y direction in the measurement angle The swing value.

[0010] Furthermore, when defining the roundness evaluation function above, for contour points in the X direction, the difference Δ1 between the maximum and minimum distances from each contour point to the undetermined center (Cx, Cy) is calculated: For contour points in the Y direction, calculate the difference Δ2 between the maximum and minimum distances from each contour point to the undetermined center (Cx, Cy): The overall score is calculated as: score = max(Δ1, Δ2).

[0011] Furthermore, the aforementioned three-stage tiered grid search strategy of "coarse-medium-fine" includes: In the coarse search phase, the step size h1 is set to 0.1 mm, and the initial search range is [-1,1]×[-1,1] mm. All grid nodes within this range are traversed, the comprehensive score of each node is calculated, the center coordinates corresponding to the minimum comprehensive score are recorded, and the search range is shrunk to a square area with a side length of 2h1 centered on this center coordinate. In the intermediate search phase, the step size h2 = 0.01 mm is set, and the search range is the area shrunk in the coarse search phase. All grid nodes in this range are traversed, the comprehensive score of each node is calculated, the center coordinates corresponding to the minimum comprehensive score are recorded, and the search range is shrunk to a square area with a side length of 2h2 centered on this center coordinate. In the fine search phase, the step size h3 is set to 0.001 mm, and the search range is the area shrunk in the middle search phase. All grid nodes within this range are traversed, and the comprehensive score of each node is calculated. The center coordinates corresponding to the minimum comprehensive score are the optimal center coordinates calculated from the cross-sectional data.

[0012] Furthermore, the number of nodes in the horizontal direction during the coarse search phase Number of nodes in the vertical direction Total number of nodes ; Number of nodes in the horizontal direction during the search phase Number of nodes in the vertical direction Total number of nodes Number of nodes in the horizontal direction during the fine search phase Number of nodes in the vertical direction Total number of nodes .

[0013] Furthermore, the above-mentioned calculation of the X-direction contour point data and the Y-direction contour point data, followed by taking the average of the two center coordinates, is as follows: The improved minimum sway method is applied to the 8 contour points in the X-direction and the 8 contour points in the Y-direction respectively to obtain the centers (Cx1, Cy1) and (Cx2, Cy2). Finally, the center coordinates of the cross-section are taken as the average of the two.

[0014] Furthermore, the base circle radius mentioned above is taken from the design radius on the drawing. Based on the characteristic that the base circle radius is much larger than the swing amplitude, the influence of the base circle radius value on the center positioning and roundness calculation results can be ignored.

[0015] Furthermore, it also includes the maintenance plan decision-making steps: quantify the comprehensive optimality index of each maintenance plan into a numerical value, determine the observation sample size, use the optimal index in the preceding plan as a benchmark, evaluate the subsequent plans in turn, and adopt a plan when the index of a plan exceeds the benchmark; otherwise, adopt the last plan.

[0016] Furthermore, in the above steps for determining the observation sample size, the number of observation samples... , where N is the total number of maintenance plans, e is a natural constant, the first r-1 plans are used as observation samples, and the optimal index M among them is recorded.

[0017] A system for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit includes: The data acquisition module is used to collect the swing data of each section during the turning process. The sections include the water guide bearing, the rotor lower flange, the lower guide bearing and the upper guide bearing. Measuring devices are set up along the X and Y directions at each section. The swing data is recorded once every 45° rotation, and a total of 8 angular positions are collected. The contour restoration module is used to convert each cross section into an equivalent cam based on the collected swing data, and restore the contour point coordinates of each cross section using the inversion method. Here, the base circle radius of the cross section is taken as the design radius of the drawing. The contour point coordinates are determined according to the swing value of each measured angle and the cosine and sine values ​​of the measured angle, and the coordinates of 8 contour points in the X direction and 8 contour points in the Y direction are obtained. The center positioning module is used to locate the optimal center for each cross section using an improved minimum swing method. The improved minimum swing method aims at optimal roundness and defines the roundness evaluation function as the difference between the maximum and minimum distances from all contour points to the center to be determined. The larger of the differences between the contour points in the X direction and the contour points in the Y direction is taken as the comprehensive score. The center coordinates that minimize this score are determined through a three-stage ladder grid search strategy of "coarse-medium-fine". The data fusion module is used to calculate the X-direction contour point data and the Y-direction contour point data separately, and then take the average of the two center coordinates as the final center coordinates of the cross section.

[0018] Furthermore, the aforementioned central positioning module includes: The coarse search unit is used to set the step size h1=0.1mm, traverse all grid nodes within the initial search range of [-1,1]×[-1,1]mm, calculate the comprehensive score of each node, record the center coordinates corresponding to the minimum comprehensive score, and shrink the search range to a square area with the center coordinates as the center and a side length of 2h1. The intermediate search unit is used to set the step size h2=0.01mm, traverse all grid nodes in the area after the coarse search unit shrinks, calculate the comprehensive score of each node, record the center coordinates corresponding to the minimum comprehensive score, and shrink the search range to a square area with a side length of 2h2 centered on the center coordinates. The fine search unit is used to set the step size h3=0.001mm. It traverses all grid nodes in the area after the shrinking of the middle search unit, calculates the comprehensive score of each node, and the center coordinates corresponding to the minimum comprehensive score are the optimal center coordinates calculated from the cross-sectional data.

[0019] Furthermore, it also includes a maintenance decision module, which quantifies the comprehensive optimality index of each maintenance plan into a numerical value, determines the observation sample size, uses the optimal index in the preceding plan as a benchmark, and evaluates the subsequent plans in turn. When a plan's index exceeds the benchmark, it is adopted; otherwise, the last plan is adopted.

[0020] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention establishes a quantitative mapping relationship between runout data and cross-sectional geometry by equating each section of the turntable to a cam and using a reversal method to reconstruct the coordinates of the contour points. This overcomes the shortcomings of the traditional average geometric center method, which simply equates runout data to the projection of eccentricity. The improved minimum runout method, aimed at achieving optimal roundness, can effectively separate roundness error from eccentricity, avoiding the center positioning distortion problem of traditional methods when there is asymmetric wear or local deformation of the cross-section. Experimental data shows that the roundness index Δ = 0.105 mm of the water guide bearing calculated using the method of this invention is better than 0.142 mm of the average geometric center method and 0.128 mm of the least squares method, improving the center positioning accuracy to the 0.001 mm level.

[0021] 2. The "coarse-medium-fine" three-stage tiered grid search strategy proposed in this invention, through step-size progressive design and region shrinkage mechanism, controls the total number of computation nodes to 1323 times while ensuring a computational accuracy of 0.001mm. This improves computational efficiency by about 3000 times compared to direct fine grid traversal (about 4 million times), and solves the sharp contradiction between computational efficiency and accuracy under high precision requirements.

[0022] 3. This invention demonstrates through theoretical analysis that when R / δ≥3000, the variation of the base circle radius within ±10% of the design radius has no more than 0.001mm of influence on the center positioning and roundness calculation. It clarifies that the design radius of the drawing can be directly used as the base circle radius for contour point reconstruction, solving the technical problem of the unreliability of contour point reconstruction caused by the difficulty in accurately measuring the base circle radius in traditional methods.

[0023] 4. This invention introduces the swing limit theory, revealing the physical law that minimizes the swing when the optimal center of each cross-section coincides with the center of the lower guide, clarifying the scientific goal of axis adjustment and avoiding resource waste caused by blindly pursuing minimal swing. Simultaneously, the N / e optimal stopping strategy is applied to maintenance scheme selection, achieving an optimal balance between the "probability of missing the optimal solution" and the "cost loss of premature decision-making." The probability of selecting the optimal solution is approximately 37%, far higher than the 1 / N of random selection, providing a quantitative basis for maintenance decisions. Attached Figure Description

[0024] Figure 1 The main flowchart of a method for calculating the center and roundness of the turning guide bearing of a large mixed-flow hydro-generator unit provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a large mixed-flow hydro-generator unit turning bearing center and roundness calculation system provided in an embodiment of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] The method in this embodiment is executed by a terminal, which can be a mobile phone, computer, PDA, laptop or desktop computer, etc. Of course, it can also be other devices with similar functions, and this embodiment does not limit them.

[0027] Example 1 Please see Figure 1 This embodiment provides a method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit, including the following steps.

[0028] 1. Barrel turning data acquisition With the unit center as the origin, the upstream direction is defined as the +Y direction, and the right-hand direction facing upstream is defined as the +X direction. The azimuth angle is defined as follows: 0° in the +X direction, increasing counterclockwise, i.e., 90° in the +Y direction. Near the guide bearing on the main shaft, starting from the +X direction, eight turning points are marked at equal intervals counterclockwise, from turning point 1 to turning point 8. The azimuth angles corresponding to each turning point are shown in Table 1.

[0029] Table 1 Correspondence between turning point and azimuth angle

[0030] Dial gauges with an accuracy of 0.01 mm were installed at four sections: the water guide bearing, the rotor lower flange, the lower guide bearing, and the upper guide bearing, along the X-direction (horizontal) and Y-direction (vertical), respectively. The turning gear was started to slowly rotate the rotor, and the dial gauge reading was recorded every 45° of rotation, collecting runout data at a total of eight angular positions. A total of 16 sets of raw data were obtained for each section (8 sets in the X-direction and 8 sets in the Y-direction). Taking the turning gear experiment data of a 170MW hydro-generator unit as an example, the runout data of the four sections are shown in Table 2 (base circle radius R = 1200 mm, measurement error ≤ 0.01 mm).

[0031] Table 2. Runout data for each cross section (unit: mm)

[0032] 2. Contour point coordinate restoration Each section of the turning gear is equivalent to a cam, and the coordinates of the contour points of each section are reconstructed using the inversion method. Let the base circle radius of the section be the design radius R (R = 1200 mm in this embodiment), a certain measurement angle be θ, and the swing value in the X direction be δ. x (θ i The y-direction sway value is δ. (θ j Then, the coordinates of the 8 contour points corresponding to the X direction are determined by equation (1), and the coordinates of the 8 contour points corresponding to the Y direction are determined by equation (2):

[0033]

[0034] Wherein, θ1=0°, θ2=45°, θ3=90°, θ4=135°, θ5=180°, θ6=225°, θ7=270°, and θ8=315°. This yields 16 contour point coordinates (8 in the X direction and 8 in the Y direction). These two sets of data constitute a repeated measurement sample to improve the reliability of center positioning.

[0035] 3. Improved minimum swing method for locating the optimal center This embodiment employs an improved minimum sway method to locate the optimal center of each cross-section with the goal of achieving optimal roundness. The roundness evaluation index is defined as follows: For contour points in the X direction, the difference between the maximum and minimum distances from all contour points to the undetermined center (Cx, Cy) is calculated and denoted as Δ1; for contour points in the Y direction, the corresponding difference is calculated and denoted as Δ2. The comprehensive evaluation function is score = max(Δ1, Δ2). The optimal center is the (Cx, Cy) that minimizes the score.

[0036] The optimal center is determined by a three-stage tiered grid search strategy of "coarse-medium-fine".

[0037] 3.1 Coarse Search Phase (Phase 1) Step size h1 = 0.1 mm, search range mm (based on actual turning experience, the center offset is usually no more than 1 mm). Number of nodes in the horizontal direction N x1 = floor((1-(-1)) / 0.1) + 1 = 21, where N is the number of nodes in the vertical direction. 1 = 21, total number of nodes N1 = 21 × 21 = 441. Traverse all nodes (Cx k Cy k For each node ∈ S1, calculate its score and record the center (Cx) corresponding to the minimum score. 1opt Cy 1opt The contraction region S2 = [Cx] 1opt -0.1, Cx 1opt +0.1] ×[Cy 1opt -0.1, Cy 1opt +0.1] mm.

[0038] 3.2 Search Phase (Phase 2) Step size h2 = 0.01 mm, search range S2 = [-0.1, 0.1] × [-0.1, 0.1] mm (with the coarse search optimal center as the origin). Number of nodes N x2 = floor(0.2 / 0.01)+1=21, N =21, total number of nodes N2=441. Traverse all nodes and record the center (Cx) corresponding to the minimum score. 2opt Cy 2opt The contraction region S3 = [Cx] 2opt -0.01, Cx 2opt +0.01] × [Cy 2opt -0.01, Cy 2opt +0.01] mm.

[0039] 3.3 Fine-grained search phase (phase 3) Step size h3 = ε = 0.001 mm, search range S3 = [-0.01, 0.01] × [-0.01, 0.01] mm. Number of nodes N x3 =21, N =3=21, total number of nodes N3=441. Traverse all nodes, find the center (Cx) corresponding to the minimum score. 3opt Cy 3opt The center of this set of data is the optimal center.

[0040] The total number of nodes calculated is 441+441+441=1323 times, while if a fine mesh traversal is used directly, the number of nodes is approximately (2 / 0.001+1)²≈4,004,001 times. The computational efficiency of this method is improved by about 3000 times.

[0041] 4. Data fusion processing Applying the minimum sway method described above to the eight contour points in the X direction and the eight contour points in the Y direction respectively, we obtain the centers (Cx1, Cy1) and (Cx2, Cy2), and take the average value as the final center of the cross section:

[0042] 5. Analysis of the independence of base circle values In practical engineering, accurately measuring the base circle radius R is difficult, but the design radius R0 on the drawing is known. Since the design radius of the shaft system section is typically on the order of several meters, and the runout δ generally does not exceed 0.3 mm, satisfying R / δ ≥ 3000, theoretical analysis shows that when the base circle radius varies within ±10% of the design radius, the impact on center positioning and roundness calculation does not exceed 0.001 mm. Therefore, directly using the design radius given on the drawing as the base circle radius for contour point reconstruction has sufficient calculation accuracy and engineering reliability, eliminating the need for precise measurement of the actual base circle radius.

[0043] 6. Swing Limit Theory and Maintenance Scheme Decision-Making According to the sway limit theory, the sway is minimized when the optimal centers of each cross-section coincide with the center of the lower guide. Therefore, the goal of axis adjustment is not to blindly pursue the minimum absolute value of the sway of each cross-section, but to make the centers of each cross-section as close to coincide as possible.

[0044] In the selection of maintenance schemes, an N / e optimal stopping strategy is introduced. The comprehensive optimality index of N maintenance schemes (such as bearing clearance adjustment, stator-rotor air gap correction, guide vane opening optimization, etc.) is quantified into a numerical value S (for example, using the analytic hierarchy process, with weights allocated as cost 30%, efficiency 30%, reliability 30%, and maintenance cycle 10%, calculating a comprehensive score S∈[0,1], where a higher score indicates a better scheme). The optimal number of observation samples r is determined. N / e ≈0.37N. The first r-1 solutions are used as observation samples and are not adopted; only the optimal index M (i.e., the highest comprehensive score among the first r-1 solutions) is recorded. Starting from the r-th solution, the comprehensive score S of each solution is evaluated sequentially. k If S kIf the value exceeds M, the proposed solution is immediately adopted and subsequent evaluations cease; if none of the proposed solutions exceed M, the last proposed solution is adopted. This strategy guarantees that the probability of selecting the optimal solution is approximately 37%, which is much higher than random selection (1 / N), while avoiding the costly waste of "premature decision-making" or "overtesting".

[0045] 7. Calculation Results and Effect Verification The data in Table 2 were calculated using the method of this embodiment and compared with the average geometric center method and the least squares method. The results are shown in Table 3. Taking the water guide bearing as an example, the roundness index Δ=0.105mm of the improved minimum runout method is better than 0.142mm of the average geometric center method and 0.128mm of the least squares method, and the center positioning accuracy is significantly improved.

[0046] Table 3 Comparison of calculation results from the three methods (unit: mm)

[0047] This embodiment, while maintaining a calculation accuracy of 0.001mm, improves the calculation speed by approximately 360 times compared to direct fine mesh traversal (theoretically, it improves by approximately 3000 times; considering factors such as programming implementation, the measured improvement is more than 360 times), and the full oscillation index is significantly better than the traditional method.

[0048] In one optional embodiment, this embodiment provides an extended method for different numbers of manual check points (N=12) based on embodiment 1, to illustrate the adaptability of the present invention to the number of sampling points.

[0049] 3.1 Data Acquisition Unlike Example 1, this example increases the number of turning points to 12, meaning that swing data is recorded every 30°, with the measured angles being 0°, 30°, 60°, 90°, 120°, 150°, 180°, 210°, 240°, 270°, 300°, and 330° respectively. The remaining data acquisition methods are the same as in Example 1.

[0050] 3.2 Contour Point Reconstruction and Center Positioning The formula for restoring the contour point coordinates is the same as in Example 1, only θ in equations (1) and (2) needs to be changed. k Replace with the corresponding 12 angle values. The step size for the coarse search stage remains 0.1 mm, and the range remains unchanged; the step size for the medium search is 0.01 mm; and the step size for the fine search is 0.001 mm. Due to the increase in sampling points, the number of contour points increases from 16 to 24. The roundness evaluation function still uses score = max(Δ1,Δ2), where Δ1 is the range of distances between the 12 contour points in the X direction, and Δ2 is the range of distances between the 12 contour points in the Y direction. The total number of nodes in the ladder grid search remains unchanged at 1323.

[0051] 3.3 Comparison of Results When using 12-point sampling, the increased sampling density further improves the contour reconstruction accuracy, and the roundness calculation error can be reduced to below 0.0005mm. This embodiment demonstrates that the method of the present invention has good scalability, and the number of sampling points can be flexibly adjusted according to actual measurement accuracy requirements, while the core algorithm structure remains unchanged.

[0052] In one optional embodiment, this embodiment, based on embodiment 1, further incorporates the swing limit theory to provide a closed-loop control method for axis adjustment.

[0053] 4.1 Calculation of the swing limit value The optimal center (Cx) of each section was calculated according to Example 1. Cy (Water guide bearing, rotor lower flange, lower guide bearing, upper guide bearing), using the center of the lower guide bearing as a reference (limit bearing), calculate the offset of other sections relative to the center of the lower guide bearing. When the centers of all sections coincide with the center of the lower guide bearing, the runout reaches the theoretical minimum value. In actual adjustment, this theoretical minimum runout is used as the optimization target, rather than blindly pursuing the absolute value of the runout of each section to approach zero.

[0054] 4.2 Predicted Adjustment Amount Based on the center offset of each cross-section, the amount of shimming or scraping by the thrust head is calculated using geometric relationships. Specifically, for bends caused by non-perpendicularity of the flange assembly surface, the formula for calculating the amount of shimming is:

[0055] in, The inclination value at the water guide is... 'a' represents the inclination value at the flange. For flange diameter, The distance between the two measuring points is denoted as .

[0056] 4.3 Adjustment and Verification Adjustments are made on-site according to the calculated padding amount. After adjustment, the machine is rotated again, and the newly measured runout data is input into this method to recalculate the center and roundness of each section to verify whether it is close to the theoretical minimum value. If the deviation is still within the allowable range, the adjustment is completed; otherwise, the iteration continues.

[0057] This embodiment extends the invention from "single data processing" to "closed-loop adjustment decision-making," further improving the accuracy and efficiency of maintenance.

[0058] Example 2 See Figure 2This embodiment provides a system for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit, used to execute the method described in Embodiment 1. The system includes a data acquisition module, a contour coordinate reconstruction module, a center positioning module, a data fusion module, a theoretical analysis module, and a decision optimization module.

[0059] 1. System Overall Architecture The system includes: Data acquisition module: used to collect swing data along the X and Y directions at four sections of the water guide bearing, rotor lower flange, lower guide bearing and upper guide bearing during the turning process. Data is recorded once every 45° rotation, for a total of 8 angular positions. Contour coordinate restoration module: Based on the collected sway data, each cross section is equivalent to a cam, and the contour coordinates of each cross section are restored using the inversion method; Center positioning module: used to locate the optimal center of each cross section using the improved minimum sway method. The center positioning module includes a three-stage ladder grid search unit of "coarse-medium-fine". Data fusion module: used to calculate the X and Y direction data separately, and take the average of the two center coordinates as the final center of the cross section; Theoretical analysis module: used to determine the base circle radius as the base circle radius for contour point reconstruction based on the characteristic that the base circle radius is much larger than the swing amplitude; Decision optimization module: It is used to quantify the comprehensive optimality index of each maintenance plan into a numerical value, determine the observation sample size, use the optimal index of the previous plan as the benchmark, and evaluate the subsequent plans in turn. When the index of a plan exceeds the benchmark, it is adopted; otherwise, the last plan is adopted.

[0060] 2. Data Acquisition Module The data acquisition module includes dial indicators (accuracy 0.01mm) deployed at each cross-section, a turning gear drive device, and a data recording unit. With the unit center as the coordinate origin, the upstream direction is defined as the +Y direction, and the right-hand direction facing upstream is defined as the +X direction. The azimuth angle is defined as follows: +X direction is 0°, and the angle increases counterclockwise, i.e., +Y direction is 90°. Near the guide bearing on the main shaft, starting from the +X direction, eight turning gear points are marked at equal intervals counterclockwise.

[0061] The data acquisition module collects dial gauge readings every 45° rotation according to the correspondence between the turning points and azimuth angles shown in Table 1. It obtains 8 sets of data in the X direction and 8 sets of data in the Y direction for each cross section, and stores the collected data in a structured format.

[0062] 3. Contour Coordinate Restoration Module The contour coordinate reconstruction module receives the sway data output by the data acquisition module, treats each section of the turnout as an equivalent cam, and uses the inversion method to reconstruct the coordinates of the contour points of each section. This module is configured as follows: Obtain the base circle radius R of the cross section, and take the design radius from the drawing; For the X-direction data, the coordinates of the 8 contour points are calculated according to equation (1); For the Y-direction data, the coordinates of the 8 contour points are calculated according to equation (2); Output the coordinates of 16 contour points to the center positioning module.

[0063] 4. Central Positioning Module The center positioning module employs an improved minimum sway method to locate the optimal center of each cross-section with the goal of achieving optimal roundness. This module includes: Roundness evaluation unit: configured to calculate the distance range Δ1 of the contour points in the X direction, the distance range Δ2 of the contour points in the Y direction, and the comprehensive evaluation function score = max(Δ1,Δ2).

[0064] Tiered grid search unit: configured to perform the following three-stage search: Coarse search sub-unit: step size h1=0.1mm, search range S1=[-1,1]×[-1,1]mm, traverse all nodes to calculate the score value, record the center corresponding to the minimum score, and shrink the region to S2; Search sub-unit: step size h2=0.01mm, search range S2=[-0.1,0.1]×[-0.1,0.1]mm, traverse all nodes, record the center corresponding to the minimum score, and shrink the region to S3; Fine search sub-unit: step size h3=0.001mm, search range S3=[-0.01,0.01]×[-0.01,0.01]mm, traverse all nodes to determine the final optimal center.

[0065] 5. Data Fusion Module The data fusion module receives the first center (Cx1, Cy1) calculated by the center positioning module for the contour points in the X direction and the second center (Cx2, Cy2) calculated for the contour points in the Y direction, and is configured to calculate the average value according to equation (3) as the final center of the cross section:

[0066] 6. Theoretical Analysis Module The theoretical analysis module is configured to: acquire the design radius R0 and sway data δ of the shaft system section, and calculate the R0 / δ ratio. When R0 / δ ≥ 3000, the design radius on the drawing is used as the base circle radius for contour point reconstruction, and the error in center positioning and roundness calculation does not exceed 0.001mm. This module outputs the base circle radius parameter to the contour coordinate reconstruction module.

[0067] 7. Decision Optimization Module The decision optimization module is configured to execute the N / e optimal stopping strategy. Specifically, it includes: Scheme quantification unit: quantifies the comprehensive optimality index of N maintenance schemes into a numerical value S (S∈[0,1], the higher the score, the better the scheme). Observation sample determination unit: Calculate the optimal number of observation samples r= N / e ≈0.37N, take the first r-1 schemes as the observation sample, and record the optimal index M among them; Solution adoption unit: Starting from the r-th solution, evaluate them sequentially. If the current solution has a comprehensive score S... k If the value of the proposed solution exceeds M, then the proposed solution should be adopted immediately and the evaluation should be stopped; if none of the proposed solutions exceed M, then the last proposed solution should be adopted.

[0068] 8. System Workflow When the system is working, it will follow the procedure below: 1. The data acquisition module is equipped with dial gauges at each cross section, which drive the turning device to rotate and collect swing data every 45° to obtain swing data in the X and Y directions at 8 angular positions; 2. The contour coordinate restoration module receives the sway data, uses the design radius of the drawing as the base circle radius, and uses the inversion method to restore the coordinates of each section contour point, outputting the coordinates of 16 contour points; 3. The center positioning module aims for optimal roundness and determines the optimal center through a "coarse-medium-fine" tiered grid search unit; 4. The data fusion module processes the X-direction data and Y-direction data respectively, and takes the average value of the two centers as the final center of the cross section; 5. The theoretical analysis module verifies the rationality of the base circle radius value; 6. The decision optimization module recommends the optimal maintenance plan based on the comprehensive score of each maintenance plan and the N / e optimal stop strategy.

[0069] 9. System Performance Verification The system of this embodiment is used to process the swing data in Table 2 of Embodiment 1, and the calculation results are the same as those in Table 3 of Embodiment 1. While ensuring a calculation accuracy of 0.001mm, the system takes less than 0.5 seconds (based on a regular PC) to calculate the center positioning of a single section, achieving a balance between high precision and high efficiency.

[0070] In one optional embodiment, this embodiment further defines the decision optimization module based on embodiment 2.

[0071] 3.1 Extended Configuration of the Decision Optimization Module The decision optimization module in this embodiment, based on embodiment 2, further includes: The comprehensive score calculation unit is configured to use the analytic hierarchy process (AHP) to calculate the comprehensive score S for each maintenance plan. This unit obtains the cost index C, efficiency index E, reliability index R, and maintenance cycle index Maint for each plan, and calculates a weighted score according to preset weights (cost 30%, efficiency 30%, reliability 30%, maintenance cycle 10%).

[0072] The observation sample dynamic adjustment unit is configured to acquire historical maintenance record data and dynamically adjust the observation sample size r based on the distribution of the location of the historical optimal solution. When historical data shows that the optimal solution tends to appear in the early stages, the value of r is appropriately decreased; when it tends to appear in the later stages, the value of r is appropriately increased.

[0073] Adoption threshold adaptive unit: When the comprehensive score of multiple schemes exceeds the benchmark M and they are close to each other, secondary evaluation indicators (such as implementation difficulty and spare parts availability) are introduced for secondary screening.

[0074] 3.2 Closed-loop control extension The system in this embodiment also includes: Axis adjustment calculation module: Configured to calculate the thrust head shim addition or scraping amount based on the offset of the optimal center of each section relative to the center of the lower guide bearing. This module obtains the flange diameter. Distance between two measuring points Inclination value at water guide Inclination value at flange Calculate the padding amount using the following formula:

[0075] Adjustment effect verification module: After on-site adjustment according to the results of the axis adjustment calculation module, the data acquisition module, contour coordinate restoration module, center positioning module and data fusion module are triggered again to perform verification calculation. The center of each section obtained by verification is compared with the expected center. If the deviation exceeds the preset threshold (such as 0.005mm), the adjustment calculation is re-executed.

[0076] 3.3 System Output The system outputs the following results in this embodiment: Optimal center coordinates of each section (Cx) Cy and roundness value Δ; Recommended repair plan and overall score; The amount of shims or scraping required for axis adjustment; Comparison curves of swing angle before and after adjustment.

[0077] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0078] Based on the above embodiments, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described functionality.

[0079] Those skilled in the art will recognize that the modules and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0080] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, equipment, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0081] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0082] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program instructions, such as USB flash drives, portable hard drives, read-only storage servers, random access storage servers, magnetic disks, or optical disks.

[0083] Furthermore, it should be noted that the combination of the various technical features in this case is not limited to the combination methods described in the claims of this case or the combination methods described in the specific embodiments. All technical features described in this case can be freely combined or combined in any way, unless they contradict each other.

[0084] It should be noted that the above examples are merely specific embodiments of the present invention, and the present invention is obviously not limited to the above embodiments, with many similar variations. All modifications that can be directly derived or conceived by those skilled in the art from the content disclosed in this invention should fall within the protection scope of this invention.

[0085] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit, characterized in that, Includes the following steps: During the turning process, the swing data of each section is collected. Each section includes the water guide bearing, the rotor lower flange, the lower guide bearing and the upper guide bearing. Measuring devices are set up along the X and Y directions at each section. The swing data is recorded once every 45° rotation, and data at a total of 8 angular positions are collected. Based on the collected swing data, each section is equivalent to a cam, and the coordinates of the contour points of each section are restored by the inversion method. The base circle radius of the section is taken as the design radius of the drawing. The coordinates of the contour points are determined according to the swing value of each measured angle and the cosine and sine values ​​of the measured angle, and the coordinates of the eight contour points in the X direction and the eight contour points in the Y direction are obtained. For each cross section, the improved minimum sway method is used to locate the optimal center. The improved minimum sway method aims at the optimal roundness. The roundness evaluation function is defined as the difference between the maximum and minimum distances from all contour points to the center to be determined. The larger of the differences between the contour points in the X direction and the contour points in the Y direction is taken as the comprehensive score. The center coordinates that minimize this score are determined by a three-stage ladder grid search strategy of "coarse-medium-fine". After calculating the X-direction profile point data and the Y-direction profile point data respectively, the average value of the two center coordinates is taken as the final center coordinate of the cross section.

2. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 1, characterized in that, In the step of restoring the coordinates of each section contour point using the inversion method, the coordinates of the contour points in the X direction are determined using the following formula: The coordinates of the contour points in the Y direction are determined using the following formula: Where R is the radius of the base circle of the cross section, which is taken from the design radius on the drawing; , For measuring angles, i and j are the measurement point numbers, with values ​​ranging from 1 to 8; For the X direction in the measurement angle The swing value, For the Y direction in the measurement angle The swing value.

3. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 1, characterized in that, When defining the roundness evaluation function, for contour points in the X direction, the difference Δ1 between the maximum and minimum distances from each contour point to the undetermined center (Cx, Cy) is calculated: For contour points in the Y direction, calculate the difference Δ2 between the maximum and minimum distances from each contour point to the undetermined center (Cx, Cy): The overall score is calculated as: score = max(Δ1, Δ2).

4. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 1, characterized in that, The "coarse-medium-fine" three-stage tiered grid search strategy includes: In the coarse search phase, the step size h1 is set to 0.1 mm, and the initial search range is [-1,1]×[-1,1] mm. All grid nodes within this range are traversed, the comprehensive score of each node is calculated, the center coordinates corresponding to the minimum comprehensive score are recorded, and the search range is shrunk to a square area with a side length of 2h1 centered on this center coordinate. In the intermediate search phase, the step size h2 = 0.01 mm is set, and the search range is the area shrunk in the coarse search phase. All grid nodes in this range are traversed, the comprehensive score of each node is calculated, the center coordinates corresponding to the minimum comprehensive score are recorded, and the search range is shrunk to a square area with a side length of 2h2 centered on this center coordinate. In the fine search phase, the step size h3 is set to 0.001 mm, and the search range is the area shrunk in the middle search phase. All grid nodes within this range are traversed, and the comprehensive score of each node is calculated. The center coordinates corresponding to the minimum comprehensive score are the optimal center coordinates calculated from the cross-sectional data.

5. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 4, characterized in that, Number of nodes in the horizontal direction during the coarse search phase Number of nodes in the vertical direction Total number of nodes ; Number of horizontal nodes in the middle search phase Number of nodes in the vertical direction Total number of nodes ; Number of nodes in the horizontal direction during the fine search phase Number of nodes in the vertical direction Total number of nodes .

6. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 1, characterized in that, The process of calculating the X-direction and Y-direction contour point data separately and then taking the average of the two center coordinates involves: applying the improved minimum sway method to the 8 contour points in the X-direction and 8 contour points in the Y-direction respectively to obtain the centers (Cx1, Cy1) and (Cx2, Cy2), and finally taking the average of the two center coordinates for the cross-section. 。 7. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 1, characterized in that, The base circle radius is taken from the design radius on the drawing. Based on the characteristic that the base circle radius is much larger than the swing amplitude, the influence of the base circle radius value on the center positioning and roundness calculation results can be ignored.

8. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 1, characterized in that, It also includes the maintenance plan decision-making steps: quantify the comprehensive optimality index of each maintenance plan into a numerical value, determine the observation sample size, use the optimal index in the previous plan as a benchmark, evaluate the subsequent plans in turn, and adopt the plan when the index of a plan exceeds the benchmark; otherwise, adopt the last plan.

9. The method for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 8, characterized in that, In the step of determining the observation sample size, the number of observation samples... , where N is the total number of maintenance plans, e is a natural constant, the first r-1 plans are used as observation samples, and the optimal index M among them is recorded.

10. A system for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit, characterized in that, include: The data acquisition module is used to collect the swing data of each section during the turning process. Each section includes a water guide bearing, a rotor lower flange, a lower guide bearing, and an upper guide bearing. Measuring devices are set up along the X and Y directions at each section. Swing data is recorded once every 45° rotation, and data at a total of 8 angular positions are collected. The contour restoration module is used to convert each cross section into an equivalent cam based on the collected swing data, and restore the contour point coordinates of each cross section using the inversion method. Here, the base circle radius of the cross section is taken as the design radius of the drawing. The contour point coordinates are determined according to the swing value of each measured angle and the cosine and sine values ​​of the measured angle, and the coordinates of 8 contour points in the X direction and 8 contour points in the Y direction are obtained. The center positioning module is used to locate the optimal center for each cross section using an improved minimum sway method. The improved minimum sway method aims at optimal roundness and defines the roundness evaluation function as the difference between the maximum and minimum distances from all contour points to the center to be determined. The larger of the differences between the contour points in the X direction and the contour points in the Y direction is taken as the comprehensive score. The center coordinates that minimize this score are determined through a three-stage ladder grid search strategy of "coarse-medium-fine". The data fusion module is used to calculate the X-direction contour point data and the Y-direction contour point data separately, and then take the average of the two center coordinates as the final center coordinates of the cross section.

11. The system for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 10, characterized in that, The central positioning module includes: The coarse search unit is used to set the step size h1=0.1mm, traverse all grid nodes within the initial search range of [-1,1]×[-1,1]mm, calculate the comprehensive score of each node, record the center coordinates corresponding to the minimum comprehensive score, and shrink the search range to a square area with the center coordinates as the center and a side length of 2h1. The intermediate search unit is used to set the step size h2=0.01mm, traverse all grid nodes in the area after the coarse search unit shrinks, calculate the comprehensive score of each node, record the center coordinates corresponding to the minimum comprehensive score, and shrink the search range to a square area with a side length of 2h2 centered on the center coordinates. The fine search unit is used to set the step size h3=0.001mm. It traverses all grid nodes in the area after the shrinking of the middle search unit, calculates the comprehensive score of each node, and the center coordinates corresponding to the minimum comprehensive score are the optimal center coordinates calculated from the cross-sectional data.

12. The system for calculating the center and roundness of the turning bearing of a large mixed-flow hydro-generator unit according to claim 10, characterized in that, It also includes a maintenance decision module, which quantifies the comprehensive optimality index of each maintenance plan into a numerical value, determines the observation sample size, uses the optimal index in the preceding plan as a benchmark, and evaluates the subsequent plans in turn. When a plan's index exceeds the benchmark, it is adopted; otherwise, the last plan is adopted.

Images