Multi-support shafting load compensation method based on strip-shaped symmetric deviation model

Abstract

The invention discloses a multi-support shafting load compensation method based on a strip-shaped symmetric deviation model, which comprises the following steps of: selecting an existing centering strategy to carry out centering on a current shafting for a plurality of times, and recording the total adjustment height and the bearing load of each bearing after each adjustment; determining the bandwidth, constructing a strip-shaped symmetric deviation matrix, and setting the size of a sliding window; constructing a historical measured data set; solving a strip-shaped symmetric deviation matrix; load deviation value prediction and compensation amount calculation; the height compensation amount of the bearing is calculated, and an actual shaft system is adjusted; and judging whether the load of each bearing meets a target requirement or not, and if not, repeating the steps until the load of each bearing meets the target requirement. According to the method, the deviation model between the design shaft system and the actual shaft system is established by utilizing the deviation data between the design shaft system and the actual shaft system before and after adjustment, load deviation prediction and compensation are performed on the adjustment amount given by the centering strategy, and the centering effect is improved.

Description

Technical Field

[0001] This invention relates to the field of alignment technology for multi-support shafting systems in ships, and in particular to a load compensation method for multi-support shafting systems based on a banded symmetric deviation model. Background Technology

[0002] When properly aligning a ship's multi-support shafting system, the coupling relationship between the bearings means that adjusting the height of one bearing will cause changes in the load of other bearings. Currently, shipyards usually use trial and error methods without scientific theoretical guidance, which makes the alignment work time-consuming and labor-intensive.

[0003] Based on the above problems, in existing research on reasonable alignment strategies for multi-support shaft systems, the neural network models built based on simulation models do not use measured data. When there is a deviation between the actual shaft system and the simulated shaft system, the alignment effect will decrease. The network model that combines measured data with transfer learning becomes more accurate as the measured data increases, but it only uses the bearing load and its change before and after the adjustment process. The alignment strategy based on full-state feedback and model predictive controller uses the load change before and after the adjustment and provides feedback on the deviation between the current bearing load and the target value.

[0004] In summary, the current shaft alignment strategies have the following drawbacks: 1. The data generated during the adjustment process was not fully utilized. For example, the deviation data between the designed shaft system and the actual shaft system before and after the adjustment.

[0005] 2. The above-mentioned alignment strategy becomes less effective when applied to actual shaft systems. The alignment strategy is based on the designed shaft system. When applied to actual shaft systems, deviations due to manufacturing, installation, and positioning issues between the actual and designed shaft systems lead to a decrease in practical effectiveness. Summary of the Invention

[0006] The main objective of this invention is to propose a load compensation method for multi-support shaft systems based on a banded symmetrical deviation model. By utilizing the deviation data between the designed shaft system and the actual shaft system before and after adjustment, a deviation model between the designed shaft system and the actual shaft system is established. The method then predicts and compensates for the load deviation given by the alignment strategy, thereby improving the alignment effect.

[0007] The technical solution adopted in this invention is: A load compensation method for multi-support shaft systems based on a banded symmetrical deviation model includes the following steps: S1. Select an existing alignment strategy to perform several alignments on the current shaft system, and record the total adjustment height and bearing load after each adjustment of each bearing. S2. Determine the bandwidth, construct the band-symmetric deviation matrix, and set the sliding window size; S3. Calculate the next adjustment amount of each bearing height by combining the centering strategy with the current shaft system status; S4. Construct a historical measured dataset: Select the current shaft system state as the base state, calculate the Euclidean distance between the total adjustment height vector of the remaining shaft system states and the total adjustment height vector of the base state, and sort the remaining shaft system states from farthest to nearest according to the Euclidean distance; calculate the bearing adjustment height of the remaining shaft system states from the base state and the corresponding actual bearing load change; calculate the corresponding design bearing load change by combining the design load influence matrix with the bearing adjustment height; subtract the design bearing load change from the actual bearing load change to obtain the load deviation value between the actual shaft system and the design shaft system at the bearing adjustment height, and combine it with the bearing adjustment height to form a historical measured data sample set; S5. Solve for the banded symmetric deviation matrix: Select several sets of historical measured data and solve for the banded symmetric deviation matrix by combining the internal elements of the banded symmetric deviation matrix. S6. Load Deviation Prediction and Compensation Calculation: Based on the next adjustment height of each bearing calculated in S3, the load deviation value caused by the adjustment height is calculated using a strip-shaped symmetrical deviation matrix; it is determined whether the deviation value has a positive effect. Deviations with a positive effect are not compensated, while deviations with a negative effect are compensated, thereby calculating the load compensation amount of each bearing. S7. Calculate the bearing height compensation amount and adjust the actual shaft system: After obtaining the bearing load that needs to be compensated, solve the bearing height compensation amount by designing the shaft system load influence matrix. The bearing height adjustment amount calculated by S3 plus the bearing height compensation amount is the actual shaft system bearing height adjustment amount. Adjust the bearings according to the actual shaft system bearing height adjustment amount. S8. Determine whether the load of each bearing meets the target requirements. If the target requirements are not met, repeat S3-S8 until the load of each bearing meets the target requirements.

[0008] In the above scheme, in step S1, during the calibration and adjustment process, the total adjustment height of each bearing after each adjustment and the bearing load sample set are recorded. Used to indicate the current state of the shaft system. This represents the total number of adjustments, and , For the first This adjustment ; For the first i After the adjustment Total adjustment height of each bearing For the number of bearings, For the first i After the first adjustment The total adjustment height of each bearing; for For the first i After the adjustment The load on each bearing, For the first i After the first adjustment The load on each bearing, upper right corner This represents the transpose of a vector.

[0009] In the above scheme, step S2 specifically includes: S2.1 Determine the number of adjacent bearings that significantly affect any bearing based on the shaft system structure, and use this to determine the bandwidth; S2.2 Construct a banded symmetric deviation matrix, determine the non-zero elements in the banded symmetric deviation matrix based on the bandwidth, and obtain a banded symmetric deviation matrix containing only the necessary unknowns; S2.3. Determine the sliding window size based on the total number of intermediate bearings, ensuring that the sliding window size multiplied by the total number of intermediate bearings is not less than the number of unknowns in the banded symmetric deviation matrix.

[0010] In the above scheme, the strip-shaped symmetric deviation matrix constructed in step S2.2 The structure is as follows: (1) in, The Middle Line 1 Column elements are , , , n For the number of bearings, Indicates adjustment Bearing unit height relative to Deviation caused by bearing load, due to If it is a symmetric matrix, then we have ; The specific method for determining the non-zero elements in the band-symmetric deviation matrix based on bandwidth is as follows: when the bandwidth is... At that time, In the case of .

[0011] In the above scheme, step S4 specifically includes: S4.1, in the After this adjustment, there is The state of the shaft system, through calculation before Total adjustment height of the shaft system With the Total adjustment height of the shaft system The Euclidean distance between them, for the former The states of the group axes are sorted from farthest to nearest to obtain a set of sample pairs. ,in, To be ranked in The total adjustment height of the shaft system under the specified condition, numbered From 1 to , This refers to the bearing load under the corresponding conditions; S4.2 Calculate the first row after permutation The state of the shaft system relative to the first The bearing adjustment height between different shaft system states and the corresponding actual bearing load changes are used to obtain a set of sample pairs. : (2) S4.3, Design the shaft system load influence matrix Adjusting the height in conjunction with the bearing The calculation shows that the bearing adjustment height is... The design bearing load variation under certain conditions; S4.4 Subtract the designed bearing load change from the actual bearing load change to obtain the final data set. ,in For each bearing at the adjusted height The deviation between the actual bearing load change and the designed bearing load change is given by: (3) S4.5, with As a collection of historical measured data samples.

[0012] In the above scheme, the shaft system load influence matrix The results are obtained from the shaft alignment calculation book, or calculated using the three bending moment method based on the parameters in the alignment calculation book.

[0013] In the above scheme, the specific method of step S5 is as follows: in the first... Before this adjustment, based on the size of the sliding window Choose the European style closest to you. Shaft system status When the amount of existing measured data is less than When using the data, all measured data were used; combined with the strip-shaped symmetrical deviation matrix. The internal element definition and measured data are used to solve the banded symmetric deviation matrix using the least squares method. The least-squares identification form of the banded symmetric deviation matrix is ​​as follows: (4).

[0014] In the above scheme, step S6 includes: S6.1, Obtaining the first through school-based strategies Adjustment amount of each bearing height Then, through the banded symmetric deviation matrix Calculate the load deviation under this set of height adjustment amounts. : (5) ,in For the first time under this set of height adjustment amounts Deviation of bearing load; S6.2 For each bearing, first determine whether the result of the deviation will make the bearing load more closely approach the target value. If so, the bearing load deviation is set to 0 during compensation; otherwise, the bearing deviation value is multiplied by a confidence factor to obtain the amount of compensation required. The calculation formula for the load compensation amount of each bearing is as follows: (6) In the formula, For the bearing height adjustment amount The bearing load compensation amount is as follows, among which For the first time under this set of height adjustment amounts The load compensation amount of each bearing; As a credibility factor; Represents a diagonal matrix, where the deviation has a positive effect. ,otherwise .

[0015] In the above scheme, the specific method of step S7 is as follows: obtain the inverse of the design shaft system load influence number matrix. Multiply by the load compensation amount The bearing height that needs compensation is obtained, plus the bearing height adjustment amount calculated by the calibration strategy. Get the final Second adjustment of height As shown in the following formula: (7) The present invention also proposes a computer storage medium storing a computer program that can be executed by a processor, the computer program executing the above-described multi-support shaft load compensation method based on the banded symmetric deviation model.

[0016] The beneficial effects of this invention are: This method employs a sliding modeling approach to the deviation between the designed and actual shaft systems. Based on a banded symmetric deviation model, it compensates for the deviation by adjusting the output of the alignment strategy. This results in an input-output relationship that more closely approximates the designed shaft system, enhancing the robustness and speed of the alignment strategy in actual shaft alignment. As alignment progresses, the deviation model updates with changes in the shaft system state, effectively utilizing measured data generated during alignment to more accurately represent the deviation between the designed and actual shaft systems under the current condition. The banded symmetric matrix retains the physical reciprocity and distance decay characteristics of simply supported beams while significantly reducing the number of parameters to be estimated and improving numerical conditions (significantly reducing the condition number and improving solution stability). Therefore, it is most suitable when data is scarce and most likely to yield stable and reliable compensation results.

[0017] This method was applied to the full-state feedback control strategy, with the stopping condition being that the load error of each bearing was within 1%. Comparing the calibration results with the original full-state feedback calibration strategy, it can be seen that after adding this method, the calibration strategy can achieve the control target with fewer adjustments, thus improving the robustness and speed of the calibration strategy. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the overall process of the multi-support shaft system load compensation method based on the banded symmetric deviation model of the present invention; Figure 2 This is a flowchart of the deviation compensation calculation in the method of the present invention; Figure 3 This is the finite element model of the shaft system in the embodiments of the present invention; Figure 4 This is the alignment effect of the full-state feedback strategy in the embodiments of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0021] It should be noted that the illustrations provided in the embodiments of the present invention are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0022] To fully utilize the deviation data between the designed and actual shaft systems before and after adjustment, and to improve the application effect of existing alignment strategies in actual shaft systems, this invention proposes a multi-support shaft system load compensation method based on a strip-shaped symmetric deviation model. This method, based on existing alignment strategies, constructs a deviation model using the deviation data between the actual and designed shaft systems during adjustment. It estimates the load deviation of the adjustment amount given by the alignment strategy and compensates for the deviation using the design shaft system load influence matrix, thereby improving alignment performance. The deviation model constructed by this method is a strip-shaped symmetric deviation matrix. To improve the stability and accuracy of solving the strip-shaped symmetric deviation matrix, after obtaining... After obtaining a set of measured data, a matrix solution is performed, that is, in the first set of measured data... During the adjustment, a matrix is ​​constructed to compensate for the deviation. For example... Figure 1 As shown, the method of the present invention specifically includes the following steps: S1. Select an existing alignment strategy, use this alignment strategy to perform several alignments based on the current shaft system, and record the total adjustment height and bearing load after each adjustment of each bearing.

[0023] During the adjustment process at the school, record the total adjustment height of each bearing after each adjustment and the bearing load sample set. This is used to indicate the current state of the shaft system. To determine the total number of adjustments, two or more sets of measured data are required to solve the banded symmetric deviation model. ; For the first This adjustment ; For the first i After the adjustment Total adjustment height of each bearing For the number of bearings, For the first i After the first adjustment Total adjustment height of each bearing; For the first i After the adjustment The load on each bearing, For the first i After the first adjustment The load on each bearing; upper right corner This represents the transpose of a vector.

[0024] S2. Determine the bandwidth, construct a band-symmetric deviation matrix, and set the sliding window size. The specific steps are as follows: S2.1 Determine the number of adjacent bearings that significantly affect any given bearing based on the shaft system structure, and use this number to determine the bandwidth. .

[0025] Based on the load influence matrix (which can be obtained from the shaft alignment calculation book, or calculated using the three bending moment method based on the shaft structure parameters in the alignment calculation book), it is known that for a given bearing, the farther away it is, the smaller the impact on that bearing during adjustment. Therefore, the deviation caused by more distant bearings can be ignored. The number of adjacent bearings to be considered is determined primarily by focusing on bearings in the middle, ignoring the farthest bearing, thus determining the bandwidth. Bandwidth is This indicates that for a single bearing, the corresponding values ​​on both sides of the bearing are considered. The adjustment of each bearing affects the deviation caused by the load on that bearing.

[0026] S2.2 Constructing a banded symmetric deviation matrix Based on the bandwidth, the non-zero elements in the banded symmetric deviation matrix are determined, resulting in a banded symmetric deviation matrix containing only the necessary unknowns.

[0027] Constructed banded symmetric deviation matrix The structure is as follows: (1) in, The Middle Line 1 Column elements are , , , n For the number of bearings, Indicates adjustment Bearing unit height relative to Deviation caused by bearing load, due to If it is a symmetric matrix, then we have .

[0028] When the bandwidth is At that time, In the case of .

[0029] S2.3. Determine the sliding window size based on the total number of intermediate bearings, ensuring that the sliding window size multiplied by the total number of intermediate bearings is not less than the number of unknowns in the banded symmetric deviation matrix.

[0030] To solve for the banded symmetric deviation matrix, it is necessary to define the sliding window size. In other words, the amount of measured data used, in order to achieve identifiability, must meet the following conditions: The number of non-zero elements inside the symmetrical deviation matrix is ​​greater than the total number of intermediate bearings. To improve the accuracy of the deviation matrix, the total number of intermediate bearings is usually reduced by 1.

[0031] S3. Calculate the next bearing height adjustment amount by combining the alignment strategy with the current shaft system status.

[0032] Assuming it has already been done During the second calibration, adjust and record the current shaft system status. Based on the current shaft system condition and the target load (given by the shaft system alignment calculation sheet), the first... Secondary bearing height adjustment amount , , For the first The bearing number The height that should be adjusted next.

[0033] S4. Constructing a Historical Measurement Dataset: After obtaining multiple sets of shaft system states, select the current shaft system state as the base state. Calculate the Euclidean distance between the total adjustment height vector of the remaining shaft system states and the total adjustment height vector of the base state. Sort the remaining shaft system states from farthest to closest according to the Euclidean distance. Calculate the bearing adjustment height of the remaining shaft system states from the base state and the corresponding actual bearing load change. Calculate the corresponding design bearing load change using the design load influence matrix combined with the bearing adjustment height. Subtract the design bearing load change from the actual bearing load change to obtain the load deviation value between the actual shaft system and the design shaft system at that bearing adjustment height. Combine this with the bearing adjustment height to form a historical measurement data sample set. The specific steps are as follows: S4.1, in the After this adjustment, there is The state of the shaft system (including the state after the 0th adjustment, i.e., the initial state). This is determined by calculation before... Total adjustment height of the shaft system With the Total adjustment height of the shaft system The Euclidean distance between them, for the former The state of the shaft system is sorted from farthest to nearest (numbered). From 1 to ), to obtain the sample pair set ,in, To be ranked in The total adjustment height of the shaft system under the specified condition. This represents the bearing load under the corresponding conditions.

[0034] S4.2 Calculate the first arrangement according to formula (2) The state of the shaft system relative to the first The bearing adjustment height between different shaft system states and the corresponding actual bearing load changes are used to obtain a set of sample pairs. : (2) S4.3, Design the shaft system load influence matrix Adjusting the height in conjunction with the bearing The calculation shows that the bearing adjustment height is... The design bearing load variation under certain conditions; S4.4 Subtract the designed bearing load change from the actual bearing load change to obtain the final data set. ,in For each bearing at the adjusted height The deviation between the actual bearing load change and the designed bearing load change is given by: (3) S4.5, with As a collection of historical measured data samples.

[0035] S5. Solve for the banded symmetric deviation matrix: Select the historical measured data with the closest Euclidean distance, and combine them with the internal elements of the banded symmetric deviation matrix to solve for the banded symmetric deviation matrix using the least squares method.

[0036] In the Before this adjustment, based on the size of the sliding window Choose the European style closest to you. Shaft system status When the amount of existing measured data is less than When using the group, all measured data are employed. This is combined with the strip-symmetric deviation matrix. The internal element definition and measured data are used to solve the banded symmetric deviation matrix using the least squares method. The least squares identification form of the banded symmetric deviation matrix is ​​shown in equation (4).

[0037] (4) S6. Load Deviation Prediction and Compensation Calculation: Based on the next adjustment height of each bearing calculated in S3, a banded symmetric deviation model (i.e., a banded symmetric deviation matrix) is used to calculate the load deviation value caused by this adjustment height. After obtaining the load deviation value on each bearing, it is first determined whether the deviation value has a positive effect, that is, whether the final result of the deviation value makes the load of the bearing closer to the target load. Deviations with a positive effect are not compensated, while deviations with a negative effect are compensated. The final compensation amount is obtained by multiplying the deviation with a negative effect by a confidence factor. Figure 2As shown, the specific steps are as follows: S6.1, Obtaining the first through school-based strategies Adjustment amount of each bearing height Then, according to equation (5), through the banded symmetric deviation matrix Calculate the load deviation under this set of height adjustment amounts. : (5) ,in For the first time under this set of height adjustment amounts Deviation of bearing load. S6.2 For each bearing, first determine whether the result of the deviation will make the bearing load more approach the target value. If so, the bearing load deviation is set to 0 during compensation (i.e., no compensation is made for the bearing deviation); if not, the bearing deviation value is multiplied by a confidence factor to obtain the final amount to be compensated (since the banded symmetrical deviation model is an approximation of the actual deviation relationship, a confidence factor is used to process the deviation; the confidence factor...). (Usually taken as 0 to 1).

[0038] After the above processing, the load compensation of each bearing is calculated according to formula (6): (6) In the formula, For bearing height adjustment amount The bearing load compensation amount is as follows, among which For the first time under this set of height adjustment amounts The final load compensation amount for each bearing; As a credibility factor; Represents a diagonal matrix, where the deviation has a positive effect. ,otherwise .

[0039] S7. Calculate the bearing height compensation amount and adjust the actual shaft system: After obtaining the bearing load that needs compensation, solve for the bearing height compensation amount by designing the shaft system load influence matrix. The actual shaft system bearing height adjustment amount is obtained by adding the bearing height adjustment amount calculated in S3 to the bearing height compensation amount. Adjust the bearings according to the actual shaft system bearing height adjustment amount. The specific method is as follows: The load influence matrix of the designed shaft system is obtained by inverting it. Multiply by the load compensation amount The bearing height that needs compensation is obtained, plus the bearing height adjustment amount calculated by the calibration strategy. Get the final Second adjustment of height As shown in the following formula: (7) S8. Determine whether the load of each bearing meets the target requirements. If the target requirements are not met, repeat S3-S8 until the load of each bearing meets the target requirements.

[0040] The present invention will be further described in detail below through a specific embodiment.

[0041] This implementation uses a shaft system with four intermediate bearings. Based on the data in its shaft system alignment calculation book, the load influence matrix is ​​obtained. As shown in equation (8), its design load is the same as the target load. Based on the load influence matrix, a state-space model of the shaft system is constructed, thereby designing a full-state feedback controller. The stopping condition for this calibration strategy is that the load error of each bearing is within 1%. A finite element model is constructed based on its structural parameters without simplification, establishing a solid bearing bush as the actual shaft system. Therefore, the actual shaft system has deviations from the designed shaft system due to factors such as equivalent bearing support displacement and long shaft deflection. The constructed model is as follows... Figure 3 As shown. A set of intermediate bearing heights was randomly selected as the initial state for alignment. The initial loads on the four intermediate bearings were... .

[0042] (8) The specific steps for implementing this example are as follows: S1. Selecting the full-state feedback alignment strategy, alignment was performed twice based on the axis system state, resulting in the following three states of the axis system: (9) S2. Since this shaft system has four intermediate bearings, we only consider the deviation caused by adjusting the two adjacent bearings to the bearing load, i.e., the bandwidth. The constructed band-symmetric deviation matrix As shown in equation (10), it can be seen that there are 7 unknowns in the banded symmetric deviation matrix. The sliding window size is set. .

[0043] (10) S3. Alignment strategy based on full-state feedback combined with current shaft system status. The height adjustment amount of each bearing in the third alignment was calculated as follows: .

[0044] S4. Construct a historical measured dataset. Set the current state as the base state, and calculate the total adjustment height vector by combining the base state with the previous two sets of axis system states. The Euclidean distance between the two sets of states is obtained by sorting them: (11) The results are obtained by calculation using equations (2) and (3): (12) S5. Solve for the banded symmetric deviation matrix The internal elements of the symmetrical deviation matrix were solved using the three sets of measured historical data with the closest Euclidean distance. During the third calibration, only two sets of measured historical data were available for calculation. Therefore, only two sets of historical measured data are used to solve the problem. According to equation (4), the symmetrical deviation matrix of the strip during the third calibration process is... The two sets of measured historical data can form eight equations. By solving for the seven element values ​​using the least squares method, we obtain... .

[0045] S6. Deviation prediction and compensation calculation. Before the third actual calibration, based on the calculated deviation... Multiply by the school's strategy. Obtain the predicted deviation value Multiply the design shaft system load influence matrix by The change in design shaft load under this set of height adjustments is calculated, and then the deviation prediction value is added to obtain the result. The actual load change on the shaft system is used to determine whether the bearing deviation is positive or negative, based on this load change and the target value. After calculation, All values ​​are negative deviations. Based on the magnitude of this deviation, we take... Therefore, according to equation (6), we can obtain .

[0046] S7. Calculate the bearing height adjustment amount. Calculate the amount of adjustment required for the third actual shaft alignment using formula (7), and obtain... Input this set of adjustment values ​​into the actual shaft system to obtain the bearing load. .

[0047] S8. Determine if the bearing load meets the target requirements. (This is achieved through comparison.) With target value It can be seen that the error is relatively large, therefore steps S3 to S8 are repeated. After the 8th adjustment, the bearing load... The stopping condition is met. It should be noted that in the subsequent calculation of the deviation value... The results of the school visit can be found here. Figure 4Figure (a) shows the alignment adjustment using only full-state feedback. It can be seen that the load influence matrix of the designed shaft system cannot accurately describe the relationship between the actual shaft system's height adjustment and load change, i.e., there is a deviation. This causes fluctuations in the percentage of load error for each bearing after the 7th and 9th alignments. Furthermore, the percentage of error for bearing #3 decreases slowly in the first 5 adjustments, finally reaching the stopping condition after 12 adjustments. This is because the deviation between the designed and actual shaft systems causes fluctuations in the result curve. Figure (b) shows the alignment results with load compensation using a banded symmetrical deviation model. The percentage of load error for each bearing does not fluctuate significantly, and the average load error for each bearing decreases rapidly with the increase in the number of alignments, finally reaching the stopping condition after the 8th adjustment. This is because the deviation model makes the input adjustment amount and the load output feedback relationship of the full-state feedback controller closer to the designed shaft system. Figure (c) compares the average bearing load error after calibration with and without the deviation model. It can be seen that the calibration efficiency and accuracy are improved with the deviation model. The average error drops below 1% to 0.87% by the 6th calibration, and finally reaches 0.35% after the 8th calibration. Without the deviation model, the error only reaches below 1% for the first time by the 10th calibration, then rises to 1.33% by the 11th calibration, and finally reaches 0.46% by the 12th calibration. In summary, the proposed sliding deviation model can effectively compensate for shaft bearing loads, significantly improving calibration efficiency and accuracy.

[0048] It should be noted that, depending on the implementation needs, the various steps / components described in this application can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0049] The order of the steps in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0050] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A load compensation method for multi-support shaft systems based on a banded symmetrical deviation model, characterized in that, Includes the following steps: S1. Select an existing alignment strategy to perform several alignments on the current shaft system, and record the total adjustment height and bearing load after each adjustment of each bearing. S2. Determine the bandwidth, construct the band-symmetric deviation matrix, and set the sliding window size; S3. Calculate the next adjustment amount of each bearing height by combining the centering strategy with the current shaft system status; S4. Construct a historical measured dataset: Select the current axis system state as the base state, calculate the Euclidean distance between the total adjustment height vector of the other axis system states and the total adjustment height vector of the base state, and sort the other axis system states from farthest to nearest according to the Euclidean distance; Calculate the bearing adjustment height of the remaining shaft system states from the basic state and the corresponding actual bearing load change; calculate the corresponding design bearing load change by combining the design load influence matrix with the bearing adjustment height; subtract the design bearing load change from the actual bearing load change to obtain the load deviation value between the actual shaft system and the design shaft system at the bearing adjustment height, and combine it with the bearing adjustment height to form a historical measured data sample set. S5. Solve for the banded symmetric deviation matrix: Select several sets of historical measured data and solve for the banded symmetric deviation matrix by combining the internal elements of the banded symmetric deviation matrix. S6. Load Deviation Prediction and Compensation Calculation: Based on the next adjustment height of each bearing calculated in S3, the load deviation value caused by the adjustment height is calculated using a strip-shaped symmetrical deviation matrix; it is determined whether the deviation value has a positive effect. Deviations with a positive effect are not compensated, while deviations with a negative effect are compensated, thereby calculating the load compensation amount of each bearing. S7. Calculate the bearing height compensation amount and adjust the actual shaft system: After obtaining the bearing load that needs to be compensated, solve the bearing height compensation amount by designing the shaft system load influence matrix. The bearing height adjustment amount calculated by S3 plus the bearing height compensation amount is the actual shaft system bearing height adjustment amount. Adjust the bearings according to the actual shaft system bearing height adjustment amount. S8. Determine whether the load of each bearing meets the target requirements. If the target requirements are not met, repeat S3-S8 until the load of each bearing meets the target requirements.

2. The multi-support shaft system load compensation method based on the banded symmetric deviation model according to claim 1, characterized in that, In step S1, during the calibration and adjustment process, the total adjustment height of each bearing after each adjustment and the bearing load sample set are recorded. Used to indicate the current state of the shaft system. This represents the total number of adjustments, and , For the first This adjustment ; For the first i After the adjustment Total adjustment height of each bearing For the number of bearings, For the first i After the adjustment, the first The total adjustment height of each bearing; for For the first i After the adjustment The load on each bearing, For the first i After the adjustment, the first The load on each bearing, upper right corner This represents the transpose of a vector.

3. The multi-support shaft system load compensation method based on the banded symmetric deviation model according to claim 1, characterized in that, Step S2 specifically includes: S2.1 Determine the number of adjacent bearings that significantly affect any bearing based on the shaft system structure, and use this to determine the bandwidth; S2.2 Construct a banded symmetric deviation matrix, determine the non-zero elements in the banded symmetric deviation matrix based on the bandwidth, and obtain a banded symmetric deviation matrix containing only the necessary unknowns; S2.

3. Determine the sliding window size based on the total number of intermediate bearings, ensuring that the sliding window size multiplied by the total number of intermediate bearings is not less than the number of unknowns in the banded symmetric deviation matrix.

4. The multi-support shaft system load compensation method based on the banded symmetric deviation model according to claim 3, characterized in that, The banded symmetric deviation matrix constructed in step S2.2 The structure is as follows: （1） in, The Middle Line 1 Column elements are , , , n For the number of bearings, Indicates adjustment Bearing unit height relative to Deviation caused by bearing load, due to If it is a symmetric matrix, then we have ; The specific method for determining the non-zero elements in the band-symmetric deviation matrix based on bandwidth is as follows: when the bandwidth is... At that time, In the case of .

5. The multi-support shaft system load compensation method based on the banded symmetrical deviation model according to claim 1, characterized in that, Step S4 specifically includes: S4.1, in the After this adjustment, there is The state of the shaft system, through calculation before Total adjustment height of the shaft system With the Total adjustment height of the shaft system The Euclidean distance between them, for the former The states of the group axes are sorted from farthest to nearest to obtain a set of sample pairs. ,in, To be ranked in The total adjustment height of the shaft system under the specified condition, numbered From 1 to , This refers to the bearing load under the corresponding conditions; S4.2 Calculate the first row after permutation The state of the shaft system relative to the first The bearing adjustment height between different shaft system states and the corresponding actual bearing load changes are used to obtain a set of sample pairs. : （2） S4.3, Design the shaft system load influence matrix Adjusting the height in conjunction with the bearing The calculation shows that the bearing adjustment height is... The design bearing load variation under certain conditions; S4.4 Subtract the designed bearing load change from the actual bearing load change to obtain the final data set. ,in For each bearing at the adjusted height The deviation between the actual bearing load change and the designed bearing load change is given by: （3） S4.5, with As a collection of historical measured data samples.

6. The multi-support shaft system load compensation method based on a banded symmetrical deviation model according to claim 1 or 5, characterized in that, The shaft system load influence matrix The results are obtained from the shaft alignment calculation book, or calculated using the three bending moment method based on the parameters in the alignment calculation book.

7. The multi-support shaft system load compensation method based on the banded symmetric deviation model according to claim 5, characterized in that, The specific method for step S5 is as follows: in the first... Before this adjustment, based on the size of the sliding window Choose the European style closest to you. Shaft system status ; When the existing measured data volume is less than When using the data, all measured data were used; combined with the strip-shaped symmetrical deviation matrix. The internal element definition and measured data are used to solve the banded symmetric deviation matrix using the least squares method. The least-squares identification form of the banded symmetric deviation matrix is ​​as follows: （4）。 8. The multi-support shaft system load compensation method based on the banded symmetric deviation model according to claim 1, characterized in that, Step S6 includes: S6.1, Obtaining the first through school-based strategies Adjustment amount of each bearing height Then, through the banded symmetric deviation matrix Calculate the load deviation under this set of height adjustment amounts. : （5） ,in For the first time under this set of height adjustment amounts Deviation of bearing load; S6.2 For each bearing, first determine whether the result of the deviation will make the bearing load more closely approach the target value. If so, the bearing load deviation is set to 0 during compensation; otherwise, the bearing deviation value is multiplied by a confidence factor to obtain the amount of compensation required. The calculation formula for the load compensation amount of each bearing is as follows: （6） In the formula, For bearing height adjustment amount The bearing load compensation amount below, of which For the first time under this set of height adjustment amounts The load compensation amount of each bearing; As a credibility factor; Represents a diagonal matrix, where the deviation has a positive effect. ,otherwise .

9. The multi-support shaft system load compensation method based on a banded symmetrical deviation model according to claim 1, characterized in that, The specific method for step S7 is as follows: Obtain the result by inverting the load influence matrix of the designed shaft system. Multiply by the load compensation amount The bearing height that needs compensation is obtained, plus the bearing height adjustment amount calculated by the calibration strategy. Get the final Second adjustment of height As shown in the following formula: （7）。 10. A computer storage medium, characterized in that, It contains a computer program that can be executed by a processor, which performs the multi-support shaft load compensation method based on the banded symmetric deviation model as described in any one of claims 1-9.

Images