Shafting dynamic centering method

Abstract

The invention relates to a shafting dynamic centering method, which comprises the following steps of: (1) collecting vibration displacement signals at the cross section of each bearing through an eddy current sensor in a shafting operation state; (2) calculating the real-time relative axis position of each bearing section according to the vibration displacement signal; (3) applying an initial offset to the bearing to generate an initial influence coefficient matrix; (4) determining a theoretical dynamic axis position in combination with a shafting elevation curve, and calculating a bearing adjustment amount based on the influence coefficient matrix; and (5) iteratively correcting the influence coefficient matrix and adjusting the position of the bearing until the actual axis position is matched with the theoretical axis position. According to the method, the problem that dynamic centering cannot be achieved under shafting static centering adjustment is solved. By continuously correcting the influence coefficient and adjusting the offset of the bearing, the accurate dynamic centering of the shaft system is realized, and meanwhile, the centering influence coefficient matrix of the shaft system is obtained, so that the working process of next centering is simplified to a great extent.

Description

Technical Field

[0001] The present invention belongs to the technical field of mechanical equipment condition monitoring and maintenance, and specifically relates to a shaft alignment method based on dynamic vibration signals, which is suitable for real-time alignment adjustment of rotating machinery such as steam turbine generator sets and ship propulsion shaft systems. Background Art

[0002] Shaft misalignment can lead to uneven bearing load distribution, deviating from the designed value. Overloaded bearings can deform and overheat the lubricating oil over time, resulting in wear, metal breakage, and bearing burnout. Underloaded bearings can cause oil film turbulence.

[0003] The current method of shaft alignment is static alignment, that is, after the coupling of the shaft system (steam turbine generator set, ship propulsion shaft system, etc.) is disconnected, a dial indicator is used to measure the lateral and vertical deviations of the shaft ends at both ends of the coupling; then, based on the deviation and the shaft system lift curve, the thickness and offset required for each bearing seat position are calculated to achieve static alignment of the unit.

[0004] However, after the unit was put into operation, the center position of the bearing shaft was lifted and offset to varying degrees. Under dynamic operation, the unit returned to the misaligned state. Static alignment methods cannot achieve dynamic shaft alignment.

[0005] Shaft misalignment is a common source of failure in rotating machinery. Static alignment methods (such as dial indicator and laser alignment) can only be adjusted when the machine is stopped and cannot compensate for dynamic offsets caused by thermal deformation and changes in the bearing oil film during operation. Existing technologies such as CN117451354A measure displacement and infer load by lifting the bearing, but this requires frequent machine shutdowns; CN105912873B relies on finite element modeling, which is computationally complex and cannot be corrected in real time. The present invention proposes a dynamic alignment method that overcomes these drawbacks by iteratively adjusting the vibration signal in real time. Summary of the Invention

[0006] To address the problem that static alignment methods cannot achieve dynamic alignment of a shafting system during operation, this paper proposes a novel dynamic shafting alignment method. Initial position adjustments are first performed on each bearing. After the shafting system is operational, the influence coefficient matrix of the adjustment amount on the relative shaft center position change is calculated using the vibration displacement signal. The bearing position adjustment amount is then calculated based on the theoretical relative shaft center position and the influence coefficient matrix. By repeating this process, dynamic shafting alignment can be achieved, and an accurate influence coefficient matrix can be obtained, simplifying the next alignment process.

[0007] To achieve the above objectives, the technical solutions of the present invention are as follows:

[0008] A method for dynamic alignment of a shaft system comprises the following steps:

[0009] (1) When the shaft system is in operation, the vibration displacement signal at each bearing section is collected by eddy current sensors;

[0010] (2) calculating the real-time relative axis position of each bearing cross section based on the vibration displacement signal;

[0011] (3) Applying an initial offset to the bearing to generate an initial influence coefficient matrix;

[0012] (4) determining the theoretical dynamic axis position in combination with the shaft lift curve, and calculating the bearing adjustment amount based on the influence coefficient matrix;

[0013] (5) Iteratively modify the influence coefficient matrix and adjust the bearing position until the actual axis position matches the theoretical axis position.

[0014] Furthermore, the calculation formula of the real-time relative axis position is:

[0015]

[0016] Among them, U and V are the vibration displacement values of two eddy current sensors arranged at an angle of 45 degrees.

[0017] Furthermore, the calculation of the initial influence coefficient matrix includes:

[0018] (1) Apply lateral and vertical offsets to the bearing and measure the change in the axis position;

[0019] (2) The linear relationship between the offset and the axis position change is fitted by the least squares method to generate matrix A1.

[0020] Furthermore, the modification of the influence coefficient matrix is achieved by the following steps:

[0021] (1) Recollect axis position data after adjustment;

[0022] (2) Update matrix A1 according to the actual axis position deviation to generate correction matrix A2.

[0023] Furthermore, the theoretical dynamic axis position is determined based on the shafting uplift curve, which reflects the deflection distribution caused by the deadweight of the shafting.

[0024] Furthermore, the iterative correction process includes:

[0025] If the actual axis position after adjustment does not reach the theoretical value, recalculate the influence coefficient matrix and make adjustments;

[0026] Record the final influence coefficient matrix for subsequent alignment work.

[0027] A shafting dynamic alignment system, comprising:

[0028] (1) Eddy current sensor group, arranged at a 45° angle on each bearing section, used to collect vibration displacement signals;

[0029] (2) a data processing unit configured to execute the method according to any one of claims 1 to 6;

[0030] (3) Bearing adjustment device, which realizes lateral and vertical position adjustment according to the calculation results.

[0031] Furthermore, the data processing unit is further configured to:

[0032] (1) Real-time display of axis position deviation and adjustment suggestions;

[0033] (2) Store the historical influence coefficient matrix to optimize subsequent alignment.

[0034] Furthermore, the arrangement spacing of the eddy current sensors is 1 / 2-2 / 3 of the bearing span to capture high-order vibration modes.

[0035] Furthermore, the iteration termination condition is that the axis position deviation is less than 5 microns and the bearing load distribution error is less than 3%.

[0036] The beneficial effects brought about by the technical solution of the present invention are:

[0037] This paper proposes a novel method for dynamic shaft alignment, resolving the issue of static shaft alignment, where dynamic alignment is impossible. By continuously modifying the influence coefficients, precise dynamic shaft alignment is achieved. A matrix of influence coefficients for the shaft alignment is also generated, significantly simplifying the next alignment process.

[0038] The present invention achieves dynamic centering and technical effects through the following technical solutions:

[0039] 1. Dynamic signal acquisition: dual eddy current sensors are deployed to obtain axis trajectory in real time;

[0040] 2. Influence coefficient modeling: Establish a quantitative relationship between the adjustment amount and the axis position through the least squares method;

[0041] 3. Iterative optimization: Dynamically correct the bearing position based on the lift curve until it matches the theoretical axis.

[0042] 4. Technical effects:

[0043] (1) The centering accuracy is improved to 5μm level, reducing the vibration amplitude by more than 30%;

[0044] (2) By reusing historical matrices, subsequent adjustment efficiency is increased by 60%;

[0045] (3) Avoid frequent downtime and reduce annual maintenance costs by 25%. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a schematic diagram of the shaft vibration displacement measurement;

[0047] Figure 2 This is a schematic diagram of the series dual-rotor shafting structure;

[0048] Figure 3 is the shaft lift curve;

[0049] Figure 4 Process diagram for dynamic alignment. DETAILED DESCRIPTION

[0050] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0051] A shaft system dynamic centering method of the present invention is to Figure 1 The shafting support structure of the tandem twin rotors shown in the figure is used as an example. The bearing numbers from left to right are 1, 2, 3, and 4.

[0052] (1) Calculation of the axis center position of the cross-section based on the vibration displacement

[0053] The general arrangement of the shaft vibration displacement sensor is as follows Figure 1 As shown in Figure 2, the two eddy current sensors are arranged at an angle of 45°. The vibration displacement of the measuring point close to the rotation direction is U, and the vibration displacement of the measuring point along the rotation direction is V. The relative axis position of the measuring point section in the operating state is:

[0054]

[0055] The shaft bearing seat is usually equipped with the above-mentioned eddy current vibration displacement sensor. After the unit is running, according to the formula, the relative axis positions of bearings 1 to 4 are:

[0056]

[0057] (2) Apply the offset to calculate the centering influence coefficient matrix

[0058] In the static state, the offsets applied to the bearing section positions 1 to 4 are:

[0059]

[0060] After the shaft system is running, under the influence of the initial offset, the relative axis positions of bearings 1 to 4 are:

[0061]

[0062] Then the coefficient matrix A1 of the offset's influence on the relative axis position of each bearing is:

[0063]

[0064] The approximate influence coefficient matrix A1 can be obtained using the least squares method.

[0065] (3) Calculate the adjustment amount based on the influence coefficient

[0066] After obtaining the influence coefficient of the offset on the relative axis position, the bearing can be adjusted according to the theoretical dynamic relative axis position.

[0067] Generally, due to its own weight, the shaft system will have a certain degree of deflection in the middle section, such as Figure 3 The entire shaft will bend slightly, and the amount of bending is called deflection.

[0068] Taking into account the influence of deflection, the elevation of each bearing is different during dynamic alignment. The curve formed by the curved shaft system is called the lift curve. Combined with the lift curve, the theoretical transverse axis position and theoretical vertical axis position of each bearing are calculated as follows:

[0069]

[0070] Then the transverse axis position of each bearing δ′ xi and vertical axis position δ′ yi The adjustment amount is:

[0071]

[0072] Adjust the bearing position according to the calculated horizontal and vertical axis position adjustments. If the horizontal and vertical axis positions reach the theoretical values after startup, the alignment adjustment is complete; otherwise, proceed to the next step.

[0073] (4) Correct the influence coefficient and recalculate the adjustment amount

[0074] After restarting the machine, if the relative axis position change does not meet the requirements and there is still deviation, the influence coefficient should be corrected.

[0075] Assume that the actual relative axis positions of bearings 1 to 4 after restarting are:

[0076]

[0077] Then the modified influence coefficient matrix A2 is:

[0078]

[0079] Then the transverse axis position of each bearing is δ″ xiand vertical axis position δ″ yi The adjustment amount is:

[0080]

[0081] The modified influence coefficient matrix can be obtained using the least squares method.

[0082] (5) Iterate the above steps

[0083] If the actual axis center position after adjustment reaches the theoretical axis center position, the alignment is completed, and the precise alignment influence coefficient matrix of the shaft system is obtained, which simplifies the process of the next alignment work.

[0084] If the actual axis position after adjustment still does not meet the theoretical axis position, continue to iterate the above steps until the actual axis position reaches the theoretical axis position to achieve centering.

[0085] (6) Achieving dynamic centering

[0086] After the above work, the relative axial center positions of each bearing after the unit is started can meet the theoretical axial center positions of the bearings under the lift curve, realizing dynamic alignment of the shaft system under working conditions.

[0087] 6. Embodiment

[0088] The present invention is a new type of dynamic alignment method for shafting. Figure 4 As shown, the steps are as follows:

[0089] Step 1: Make the shaft system work at rated speed and rated load, and calculate the relative axis position of each bearing shaft surface.

[0090] Step 2: Stop the machine and change the position of the bearing, such as adjusting the lateral offset and elevation.

[0091] Step 3: Restart the machine and calculate the coefficient matrix of the influence of the adjustment amount on the change of the axis center position based on the change of the axis center position of the bearing shaft surface after starting up.

[0092] Step 4: Calculate the offset adjustment of the bearing position based on the lift curve, the theoretical shaft center position under dynamic alignment conditions, and the influence coefficient matrix, and adjust the bearing.

[0093] Step 5: Restart the machine and calculate whether the axis position meets the requirements. If so, dynamic alignment is completed. If not, proceed to the next step.

[0094] Step 6: Based on the axis center position from Step 5 and Step 3, as well as the bearing offset adjustment from Step 4, modify the influence coefficient matrix, calculate the bearing offset adjustment, and adjust the bearings. If the axis center position after power-on meets the theoretical axis center position, dynamic alignment is complete. If not, repeat this process until dynamic alignment is achieved.

[0095] Step 7: Record the last influence coefficient matrix calculated before the dynamic alignment work is completed to provide support for the next alignment work.

[0096] Example 1:

[0097] Take a marine propulsion shaft system (4 bearing support) as an example:

[0098] 1. Collect U and V vibration signals at rated speed and calculate the initial axis position X1 = [12, 15, 18, 20] μm, Y1 = [8, 10, 12, 14] μm;

[0099] 2. Apply δx = [10, 0, 0, 0] μm offset and measure the new axis X2 = [18, 16, 19, 21] μm. The k in matrix A1 is obtained by the least squares method. 11 =0.6;

[0100] 3. According to the theoretical value of the lift curve D_x=[15,16,17,18]μm, the calculation needs to adjust δx'=A1 -1 (D_x-X1), the vibration amplitude decreased by 28% after adjustment.

[0101] Example 2:

[0102] For gas turbine generator sets:

[0103] 1. Residual deviation after initial adjustment ΔX = [2, 1, 3, 2] μm, correction matrix A2;

[0104] 2. After the second adjustment, the deviation ΔX' = [0.5, 0.3, 0.7, 0.4] μm, meeting the termination condition; the A2 matrix is stored and directly called upon during the next alignment, shortening the adjustment time from 8 hours to 3 hours.

[0105] Example 3:

[0106] Development of dedicated alignment systems:

[0107] 1. The sensor spacing was set to 60% of the bearing span, successfully identifying the second-order vibration mode;

[0108] 2. Equipped with an adaptive filtering algorithm, it maintains 5μm-level accuracy even under oil film oscillation conditions;

[0109] 3. After 10 iterations, the bearing life was extended to 18,000 hours, a 40% increase compared to traditional methods.

Claims

1. A method for dynamic centering of a shaft system, characterized in that: The following steps are involved: (1) When the shaft system is in operation, the vibration displacement signal at each bearing section is collected by eddy current sensors; (2) calculating the real-time relative axis position of each bearing cross section based on the vibration displacement signal; (3) Applying an initial offset to the bearing to generate an initial influence coefficient matrix; (4) determining the theoretical dynamic axis position in combination with the shaft lift curve, and calculating the bearing adjustment amount based on the influence coefficient matrix; (5) Iteratively modify the influence coefficient matrix and adjust the bearing position until the actual axis position matches the theoretical axis position.

2. The method according to claim 1, characterized in that The calculation formula of the real-time relative axis position is: Among them, U and V are the vibration displacement values of two eddy current sensors arranged at an angle of 45 degrees.

3. The method according to claim 1, characterized in that The calculation of the initial influence coefficient matrix includes: (1) Apply lateral and vertical offsets to the bearing and measure the change in the axis position; (2) The linear relationship between the offset and the axis position change is fitted by the least squares method to generate matrix A1.

4. The method according to claim 3, characterized in that The modification of the influence coefficient matrix is achieved by the following steps: (1) Recollect axis position data after adjustment; (2) Update matrix A1 according to the actual axis position deviation to generate correction matrix A2.

5. The method according to claim 1, wherein The theoretical dynamic axis position is determined based on the shafting uplift curve, which reflects the deflection distribution caused by the deadweight of the shafting.

6. The method according to claim 1, characterized in that The iterative correction process includes: If the actual axis position after adjustment does not reach the theoretical value, recalculate the influence coefficient matrix and make adjustments; Record the final influence coefficient matrix for subsequent alignment work.

7. A shafting dynamic alignment system, characterized in that: include: (1) Eddy current sensor group, arranged at a 45° angle on each bearing section, used to collect vibration displacement signals; (2) a data processing unit configured to execute the method according to any one of claims 1 to 6; (3) Bearing adjustment device, which realizes lateral and vertical position adjustment according to the calculation results.

8. The system according to claim 7, characterized in that The data processing unit is further configured to: (1) Real-time display of axis position deviation and adjustment suggestions; (2) Store the historical influence coefficient matrix to optimize subsequent alignment.

9. The method according to claim 1, characterized in that The eddy current sensors are arranged at a spacing of 1 / 2-2 / 3 of the bearing span to capture high-order vibration modes.

10. The method according to claim 1, characterized in that The iteration termination condition is that the axis position deviation is less than 5 microns and the bearing load distribution error is less than 3%.

Images