MATLAB-based blade sorting method and system after consideration of pitches

Abstract

The invention discloses a pitch-considered blade sorting method and system based on MATLAB, and the method comprises the following steps: determining the number of blades: determining the number of large-pitch blades, small-pitch blades and special blades participating in sorting according to the actual conditions of product design and assembly; parameters are marked, wherein the pitches, the weights and the weight moments of the large-pitch blades and the small-pitch blades participating in sorting are determined, and the parameters and blade serial numbers are marked; meanwhile, the weight of the special piece is determined; angle conversion: converting the blade pitch into a corresponding angle according to the mounting pitch circle; fixing the position of the special sheet: fixing the position of the special sheet at 0 degree; preliminarily arranging the scheme; and carrying out MATLAB (matrix laboratory) calculation. According to the method, the efficient matrix calculation capability of MATLAB is utilized, the arrangement number and positions of the large and small pitch blades are rapidly and accurately calculated, the balance precision of the assembled wheel disc is improved, a balance foundation is laid for a follow-up overall rotor dynamic balance experiment, and the safety and stability of unit operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of steam turbine assembly, and particularly relates to a blade sorting method and system based on MATLAB considering pitch. Background Art

[0002] To ensure the balance of the steam turbine disk, the moving blades to be assembled need to be assembled in sequence. In the traditional blade sorting, generally under the condition of equal pitch of the blades, the phase angle occupied by each blade is equal. Therefore, only the weight or weight moment of the blade is considered, so that the theoretical moment of the final weight or weight moment distributed to the center of the disk is within the required range. When the pitch of the blades varies greatly, it will cause the phase angle occupied by the weight or weight moment to be no longer equal. The result of the traditional sorting method that only considers the weight or weight moment is often a counter-effect. Therefore, for the sorting of steam turbine blades with large pitch differences, the traditional sorting software cannot be used.

[0003] During the steam turbine design process, when a single last-stage blade slot and a single last-stage blade are used on the disk, a large unbalance will occur locally. At this time, blades with large pitch and small pitch are needed. By using the density difference of the cascades to generate a difference in mass distribution to balance the unbalance. At this time, the arrangement order of the large-pitch and small-pitch blades cannot be obtained by using the traditional sorting software, resulting in the inability to carry out the assembly work, or the final unbalance cannot be controlled, and the overall dynamic balance of the rotor cannot be successfully completed. Summary of the Invention

[0004] The purpose of the present invention is to provide a blade sorting method and system based on MATLAB considering pitch, so as to solve the foregoing problems existing in the prior art.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] A blade sorting method based on MATLAB considering pitch includes the following steps:

[0007] S1. Determine the number of blades: According to the product design and actual assembly situation, determine the number of large-pitch blades, small-pitch blades, and special blades participating in the sorting;

[0008] S2. Parameter marking: Determine the pitch, weight / weight moment of the large-pitch blades and small-pitch blades participating in the sorting, and mark these parameters and the blade numbers; at the same time, determine the weight of the special blades;

[0009] S3. Angle conversion: Convert the blade pitch into the corresponding angle according to the installation pitch circle;

[0010] S4. Fix the position of the special blade: Fix the position of the special blade at 0°;

[0011] S5. Preliminary arrangement plan: Set the difference in the number of large and small pitch blades to m. Assume that n large-pitch blades are fixed in the 180° area, and (n+m) / 2 small-pitch blades are arranged on the left and right of the 0° special blades. The remaining large and small blades are arranged alternately in the 90° and 270° areas.

[0012] S6. MATLAB calculation: Read the blade weight and angle data through MATLAB, and use loop calculation to minimize the blade vector sum, so as to determine the optimal solution for the number n of large-pitch blades used in the 180° area, and find the most appropriate arrangement according to the weight of different blades; output the blade serial number and the remaining theoretical imbalance.

[0013] Furthermore, the cyclic calculation in step S6 specifically includes:

[0014] S61. Randomly disrupt the order of the large-pitch blades and the small-pitch blades using the randperm function;

[0015] S62, fix the position of the special slices and combine them into a top matrix R, a bottom matrix S, and two side matrices D1 and D2 according to the preliminary arrangement plan;

[0016] S63. Calculate the vector sum f of the combined leaf matrix F. The formula is:

[0017] f=sqrt(sum(F[:,2].*cos(F[:,3]))^2+sum(F[:,2].*sin(F[:,3]))^2);

[0018] S64. Compare the current vector sum f with the recorded optimal value. If the current vector sum f is smaller, update the optimal value and the corresponding sorting scheme.

[0019] S65, repeat the above steps until a preset number of cycles is reached or the vector sum f converges to below a certain set value.

[0020] Furthermore, the special blade is a hypothetical blade that is formed by integrating the last blade and the missing weight of the last blade window.

[0021] Furthermore, the method is applicable to the situation where the local imbalance caused by the single-window end blade in the turbine wheel is too large, and the imbalance is balanced by arranging a combination of large-pitch blades and small-pitch blades.

[0022] Based on the same concept, a blade sorting system considering pitch is developed based on MATLAB.

[0023] It includes: data input module, parameter conversion module, preliminary arrangement module, cyclic calculation module and result output module;

[0024] The data input module is used to input the number, pitch, weight, and weight moment parameters of the blades;

[0025] The parameter conversion module is used to convert the pitch into an angle;

[0026] The preliminary arrangement module is used to preliminarily arrange the blades according to preset rules;

[0027] The loop calculation module is used to perform loop calculations through MATLAB to optimize the blade sorting and output the optimal solution, and the result output module is used to output the serial numbers of the blades and the remaining theoretical unbalance.

[0028] The beneficial effects of the present invention are as follows: The present invention discloses a blade sorting method considering pitch based on MATLAB, including the following steps: determining the number of blades, parameter marking, angle conversion, fixing the position of special blades, and MATLAB calculation;

[0029] 1. Solve the problem that the current mainstream blade sorting software only considers weight and cannot arrange the blade order in the case of uneven cascade distribution when the blade pitches are unequal.

[0030] 2. The present invention utilizes the efficient matrix calculation ability of MATLAB to quickly and accurately calculate the arrangement quantity and position of large and small pitch blades.

[0031] 3. By using computer calculation, the process of manual continuous disassembly and assembly attempts can be saved, damage during the blade disassembly and assembly process can be avoided, and the cumbersome workload of workers can be reduced.

[0032] 4. Avoid repeated use of the low-speed balancing machine and repeated static balancing work, saving a certain amount of equipment cost.

[0033] 5. Improve the balance accuracy of the wheel disc after assembly, lay a good balance foundation for the subsequent overall rotor dynamic balance experiment, and improve the safety and stability of the unit operation. Description of the Drawings

[0034] Figure 1 is a flowchart of a blade sorting method considering pitch based on MATLAB of the present invention;

[0035] Figure 2 is a flowchart of the MATLAB algorithm of the present invention;

[0036] Figure 3 is a simplified diagram of the blade distribution matrix of the present invention;

[0037] Figure 4 is a structural diagram of a blade sorting system considering pitch based on MATLAB of the present invention. Detailed Embodiment

[0038] To make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention.

[0039] Referring to Figure 1 , Figure 2 , Figure 3 and Figure 4 a blade sorting method based on MATLAB considering pitch shown, which includes the following steps:

[0040] S1. Determine the number of blades: According to the actual situation of product design and assembly, determine the number of large-pitch blades, small-pitch blades and special blades participating in the sorting;

[0041] S2. Parameter marking: Determine the pitch, weight / weight moment of the large-pitch blades and small-pitch blades participating in the sorting, and mark these parameters and the blade numbers; at the same time, determine the weight of the special blade;

[0042] S3. Angle conversion: Convert the blade pitch to the corresponding angle according to the installation pitch circle;

[0043] S4. Fix the position of the special blade: Fix the position of the special blade at 0°;

[0044] S5. Preliminary layout plan: Set the difference in the number of large and small pitch blades as m. Assume that n large-pitch blades are fixed in the 180° opposite region, and (n + m) / 2 small-pitch blades are arranged on each side of the special blade at 0°. The remaining large and small blades are arranged at intervals in the 90° and 270° regions;

[0045] S6. MATLAB calculation: Read the blade weight and angle data through MATLAB, use loop calculation to minimize the vector sum of the blades, so as to determine the optimal solution of the number n of large-pitch blades used in the 180° region, and find the most suitable arrangement according to the weights of different blades; Output the serial numbers of the blades and the remaining theoretical unbalance.

[0046] Further, the loop calculation in step S6 specifically includes:

[0047] S61. Randomly shuffle the sorting of large-pitch blades and small-pitch blades using the randperm function;

[0048] S62. Fix the position of the special blade, and combine it into the top matrix R, the bottom matrix S, and the side matrices D1 and D2 according to the preliminary layout plan;

[0049] S63. Calculate the vector sum f of the combined blade matrix F, and the formula is:

[0050] f=sqrt(sum(F[:,2].*cos(F[:,3]))^2+sum(F[:,2].*sin(F[:,3]))^2);

[0051] S64. Compare the current vector sum f with the recorded optimal value. If the current vector sum f is smaller, update the optimal value and the corresponding sorting scheme.

[0052] S65, repeat the above steps until a preset number of cycles is reached or the vector sum f converges to below a certain set value.

[0053] Furthermore, the special blade is a hypothetical blade that is formed by integrating the last blade and the missing weight of the last blade window.

[0054] Furthermore, the method is applicable to the situation where the local imbalance caused by the single-window end blade in the turbine wheel is too large, and the imbalance is balanced by arranging a combination of large-pitch blades and small-pitch blades.

[0055] Reference Figure 2 The specific process of the algorithm shown is as follows: (1) The large and small pitch blades are grouped separately to form the large pitch blade group matrix A (serial number, weight / weight moment, angle), the small pitch blade group matrix B (serial number, weight / weight moment, angle), and the fixed blade matrix C (serial number, weight / weight moment, angle).

[0056] (2) Initially, the number of large blades n in the bottom S area is set to 2, the optimal value of the vector sum is 100, the number of matrices A is max, and the difference between the number of matrices A and B is m.

[0057] (3) Use the randperm function to randomly disrupt the matrix A to obtain the random matrix AA of the large-pitch blades, and use the randperm function to randomly disrupt the matrix B to obtain the random matrix BB of the small-pitch blades.

[0058] (4) The center position of the fixed matrix C is 0°.

[0059] (5) Take BB[1:(n+m) / 2], C, BB[1:(n+m) / 2] to form the top matrix R composed of the small leaves and fixed leaves in the top R area.

[0060] (6) Take AA[1,n] to form the matrix S of the bottom S region.

[0061] (7) Use D(1:2:end,:)=BB function to mix the remaining small leaves BB[n+m+1,max] and D(2:2:end,:)=AA function to mix the large leaves AA[n+1,max] into matrices D1 and D2.

[0062] (8) R, D1, S, D2 can form the final blade matrix F. Using the movesum function, accumulate the blade angles to obtain the vector angle of the blades.

[0063] (9) Use the function sqrt(sum(F[:,2].*cos(F[:,3]))^2 + sum(F[:,2].*sin(F[:,2]))^2) to calculate the vector sum f of matrix F.

[0064] (10) When f is greater than or equal to the optimal value, recalculate. When f is less than the optimal value, calculate the optimal value for this sorting and replace the optimal value used in the previous loop. Repeat the calculation 10^8 times to obtain the optimal sorting scheme for this n value.

[0065] (11) Substitute n with n + 2, that is, the number of large blades at the bottom is 2 more than that in the previous loop, and repeat the loop calculation from (3) to (10) until n reaches the total number of large blades. Through the repeated calculation from (1) to (11), obtain the optimal n value, the optimal sorting sequence number, and output the final remaining unbalance under this optimal scheme.

[0066] As Figure 3 shown, the present invention provides an efficient and accurate blade sorting method for the entire ring of moving blades composed of special blades and blades with different pitch sizes. It mainly includes determining the number of large pitch blades in the S area, the number of small pitch blades on both sides of the special blades in the top R area, and after confirming the number of large and small pitch blades at the bottom and top, according to the weight of each blade itself, output the optimal blade sorting sequence number and the remaining unbalance.

[0067] After actual assembly, after determining the large pitch blades and small pitch blades used, measure the weight / weight moment and blade pitch of all the blades used, convert the blade pitch to the corresponding angle, and group and number the large and small pitch blades separately.

[0068] Working principle and usage process:

[0069] The present invention equivalentizes the entire ring of blades to a circular ring, which is divided into four sections. The top and bottom are balanced, and both sides are balanced, so that the entire circular ring is balanced, that is, calculate the vector sum of the blade weight moments in four directions under different sortings to make it the smallest to achieve the best balance.

[0070] Utilize the excellent matrix calculation ability of MATLAB, write a script for the above algorithm using the MATLAB language and run it, and output the results. For subsequent assembly, assemble according to the output sequence number.

[0071] Example 1: Blade sorting of a certain type of steam turbine disk

[0072] Input parameters: 10 large-pitch blades, 20 small-pitch blades, and 1 special blade; Calculation process: Use MATLAB to optimize the value of n, and finally determine that the number of large-pitch blades in the 180° area is 4;

[0073] Output result: The optimal blade sorting order is [1, 3, 5, 7, 9, 2, 4, 6, 8, 10], and the remaining unbalance is 0.02 kg·m.

[0074] Implementation case 2: For the N5.6-3.5 / 378 type steam turbine, the first, fourth, fifth, and sixth stage impellers all adopt the form of large and small pitch blades. Taking the fifth stage as an example. When pre-installing the fifth stage, a total of 58 large-pitch blades, 60 small-pitch blades, and 2 special blades are used. The weights, angles, and serial numbers of the large-pitch blades are shown in Table 1, the weights, angles, and serial numbers of the small-pitch blades are shown in Table 2, and the special blade numbers are 1 and 2. Special blade serial number 1: weight 251.35 g, angle 3.533°; Special blade serial number 2: weight 645.9 g, angle 2.945°.

[0075]

[0076]

[0077] Using MATLAB to calculate, the optimal value of the number of large-pitch blades used is obtained as 24, and the output result of the blade serial numbers is as follows:

[0078]

[0079] Under this sorting, the final calculated vector sum is 0.2393. Combining with the pitch circle radius R of the blade being 300 mm, the unbalance can be obtained as 71.8 g·mm, which is much smaller than the allowable unbalance of this stage. That is, the blades installed in the order of the sequence numbers obtained by this algorithm have a high impeller balance accuracy level.

[0080] Refer to Figure 4 A blade sorting system based on MATLAB considering pitch as shown,

[0081] including: a data input module, a parameter conversion module, a preliminary arrangement module, a loop calculation module, and a result output module;

[0082] The data input module is used to input the number, pitch, weight, and weight moment parameters of the blades;

[0083] The parameter conversion module is used to convert the pitch into an angle;

[0084] The preliminary arrangement module is used to preliminarily arrange the blades according to preset rules;

[0085] The loop calculation module is used to perform loop calculations through MATLAB to optimize the blade sorting and output the optimal solution, and the result output module is used to output the serial numbers of the blades and the remaining theoretical unbalance.

[0086] The beneficial effects of the present invention are as follows: The present invention discloses a blade sorting method based on MATLAB considering pitch, including the following steps: determining the number of blades, parameter marking, angle conversion, fixing the positions of special blades, and MATLAB calculation;

[0087] 1. Solve the problem that the current mainstream blade sorting software only considers weight and cannot arrange the blade order in the case of uneven cascade distribution when the blade pitches are unequal.

[0088] 2. The present invention utilizes the high-efficiency matrix calculation ability of MATLAB to quickly and accurately calculate the arrangement quantity and positions of large and small pitch blades.

[0089] 3. By using computer calculation, the process of manual continuous disassembly and assembly attempts can be saved, damage during the blade disassembly and assembly process can be avoided, and the cumbersome workload of workers can be reduced.

[0090] 4. Avoid repeatedly going on the low-speed balancing machine and repeating the static balancing work, saving a certain amount of equipment costs.

[0091] 5. Improve the balance accuracy of the disk after assembly, lay a good balance foundation for the subsequent overall rotor dynamic balance experiment, and improve the safety and stability of the unit operation.

Claims

1. A blade sorting method based on MATLAB considering pitch, characterized in that, The following steps are involved: S1. Determine the number of blades: Based on the actual product design and assembly situation, determine the number of large-pitch blades, small-pitch blades, and special blades involved in the sorting process; S2. Parameter marking: Determine the pitch, weight, and weight moment of the large-pitch blades and small-pitch blades involved in the sorting, and mark these parameters and the blade serial number; at the same time, determine the weight of the special blade; S3. Angle conversion: convert the blade pitch to the corresponding angle according to the installation pitch circle; S4, special piece position fixation: fix the special piece position to 0°; S5. Preliminary arrangement plan: Set the difference in the number of large and small pitch blades to m. Assume that n large-pitch blades are fixed in the 180° area, and (n+m) / 2 small-pitch blades are arranged on the left and right of the 0° special blades. The remaining large and small blades are arranged alternately in the 90° and 270° areas. S6. MATLAB calculation: The blade weight and angle data are read through MATLAB, and the blade vector sum is minimized by loop calculation, thereby determining the optimal solution for the number n of large-pitch blades used in the 180° area and finding the most appropriate arrangement according to the weight of different blades; Outputs the blade serial number and the remaining theoretical unbalance.

2. The blade sorting method according to claim 1, wherein The cyclic calculation in step S6 specifically includes: S61. Randomly disrupt the order of the large-pitch blades and the small-pitch blades using the randperm function; S62, fix the position of the special slices and combine them into a top matrix R, a bottom matrix S, and two side matrices D1 and D2 according to the preliminary arrangement plan; S63. Calculate the vector sum f of the combined leaf matrix F. The formula is: f=sqrt(sum(F[:,2].*cos(F[:,3]))^2+sum(F[:,2].*sin(F[:,3]))^2); S64. Compare the current vector sum f with the recorded optimal value. If the current vector sum f is smaller, update the optimal value and the corresponding sorting scheme. S65, repeat the above steps until a preset number of cycles is reached or the vector sum f converges to below a certain set value.

3. The blade sorting method according to claim 1 or 2, characterized in that The special piece is a hypothetical blade formed by integrating the last blade and the missing weight of the last blade window.

4. The blade sorting method according to claim 1, wherein This method is applicable to the situation where the local imbalance caused by the single-window end blade in the turbine wheel is too large. The imbalance is balanced by arranging a combination of large-pitch blades and small-pitch blades.

5. A blade sorting system based on MATLAB considering pitch It is characterized by: It includes: data input module, parameter conversion module, preliminary arrangement module, cyclic calculation module and result output module; The data input module is used to input the number, pitch, weight and weight moment parameters of the blades; The parameter conversion module is used to convert the pitch into an angle; The preliminary arrangement module is used to preliminarily arrange the blades according to preset rules; The cyclic calculation module is used to perform cyclic calculations through MATLAB to optimize the blade sorting and output the optimal solution, and the result output module is used to output the serial number of the blade and the remaining theoretical unbalance amount.

Images