Tolerance dimension chain optimization method for marine steam turbine rotor

Abstract

The invention discloses a tolerance dimension chain optimization method for a marine steam turbine rotor, and belongs to the technical field of marine steam turbine manufacturing. In order to solve the problems of large tolerance dimension chain deviation and low dynamic balance precision in steam turbine rotor part machining, the method comprises the steps of obtaining probability distribution of mass centers of all stages of cylindrical discs in the spatial position, generating geometric offset deviation, obtaining mass center offset deviation, randomly selecting one end face and obtaining the position of an end face center point. A normal deviation angle is obtained according to the mass center deviation and the end face center point position; taking the geometric deviation of the center of the cylindrical disc, the end face center point position and the normal deviation angle as geometric features, constructing a dimension chain centroid deviation prediction model, and outputting a concentric and centroid collaborative deviation prediction value; and carrying out feature sensitivity evaluation on the deviation prediction value, and if the feature sensitivity meets geometric and centroid precision indexes, taking the current concentric and centroid collaborative deviation prediction value as a rotor tolerance dimension chain. The method is used for dynamic balance of the steam turbine rotor.

Description

Technical Field

[0001] The invention relates to a tolerance dimension chain optimization method for a marine steam turbine rotor, and belongs to the technical field of marine steam turbine manufacturing. Background Art

[0002] The working environment and use conditions of marine steam turbines are different from those of power station steam turbines. They are installed on a deformable hull base and are often affected by the swaying and impact of the hull. Their normal operation is directly related to the safety of the entire ship, so they have higher requirements for reliability. The rotor structure is one of the most important components in a steam turbine.

[0003] During the manufacturing and assembly process of the turbine rotor, the rotor may be unbalanced due to the difference in tolerance size chains. Rotor imbalance can cause rotor vibration, accelerate the wear of bearings, shaft seals and other components, and reduce the service life and efficiency of the machine. Therefore, during the manufacturing, maintenance and operation process, the rotor needs to be dynamically balanced. Dynamic balancing is to change the mass distribution of the rotor by removing or adding weights to the rotor, so that the rotor vibration caused by the eccentric centrifugal force of the center of mass or the dynamic load acting on the bearing is reduced to within the allowable range, so as to achieve the purpose of stable operation of the rotor.

[0004] The geometric concentricity and mass center offset of each rotating component of the turbine rotor structure will be accumulated along the internal dimension chain of the parts, causing the geometric center and mass center of the rotor as a whole to deviate from the ideal central axis, reducing the vibration quality. The key factors affecting the geometric concentricity and mass center offset of the rotor are the geometric offset and mass center offset of each cylindrical disk. Although the influencing factors of concentricity and mass center offset are similar, there are significant differences in the sensitivity and contribution of the influence.

[0005] The existing steam turbine rotor processing has the following problems:

[0006] 1. The processing accuracy is mainly based on geometric indicators such as concentricity, and the changes of corresponding factors such as mass eccentricity are not considered;

[0007] 2. After the rotor parts are processed, their dynamic balancing quality is mainly achieved by adjusting the position and weight of the balancing block. The dynamic balancing adjustment process relies on the operator's experience and lacks theoretical guidance, resulting in low efficiency and long cycle of the adjustment process.

[0008] Therefore, it is necessary to start with the tolerance dimension chain of steam turbine rotor parts and design the coordinated tolerance of rotor geometry and center of mass offset, so as to adjust and optimize the dynamic balance and improve the dynamic balance accuracy of steam turbine rotors. This fundamentally reduces the imbalance of the rotor itself and provides a design method for practical guidance of rotor adjustment and performance improvement. Summary of the invention

[0009] The invention aims to solve the problems of large tolerance dimension chain deviation and low dynamic balancing accuracy in the processing of steam turbine rotor parts, and provides a method for optimizing the tolerance dimension chain of a marine steam turbine rotor.

[0010] The present invention provides a method for optimizing the tolerance dimension chain of a marine steam turbine rotor, comprising:

[0011] S1. According to the size of the rotor parts, the geometric tolerances of the rotor axis and the cylindrical disks at each level are analyzed and obtained, and then the probability distribution of the center of mass of the cylindrical disks at each level in the spatial position is obtained;

[0012] S2, based on the probability distribution of the center of mass in the space position obtained in S1, generate a geometric offset deviation of the center of the cylindrical disk that conforms to the probability distribution;

[0013] S3, according to the rotor size, calculate the weight of the cylindrical disk, and combine the probability distribution of the center of mass in the spatial position obtained in S1 to obtain the corresponding center of mass offset deviation;

[0014] S4, randomly select an end face, obtain the position of the end face center point, and obtain the normal offset angle according to the centroid offset deviation and the end face center point position;

[0015] S5. Using the center geometric offset deviation of the cylindrical disk, the end face center point position and the normal offset angle as geometric features, a prediction model for the centroid offset of the dimension chain is constructed, and the prediction values ​​of the concentricity and centroid coordination deviation are output;

[0016] S6. Evaluate the characteristic sensitivity of the predicted value of the coordinated deviation between concentricity and center of mass to determine whether the characteristic sensitivity meets the geometric and center of mass accuracy indicators. Otherwise, perform dynamic balancing optimization. If yes, use the current predicted value of the coordinated deviation between concentricity and center of mass as the rotor tolerance dimension chain.

[0017] Preferably, the dynamic balance optimization specifically includes:

[0018] S2-1. According to the geometric and centroid coordination deviations, the probability distribution interval of the concentric and centroid coordination offsets is obtained by using statistical deviation simulation;

[0019] S2-2, setting the mass and position of the balancing block within the probability distribution interval of the co-center offset obtained in S2-1;

[0020] S2-3, constructing a prediction model for rotor mass center offset combined with a balancing block, and outputting a predicted value of the rotor mass center offset including the balancing block;

[0021] S2-4, evaluate the dynamic balancing performance of the predicted value of the center of mass offset of the rotor containing the balancing block to determine whether the dynamic balancing performance meets the dynamic balancing performance index; otherwise, return to execute S2-1, and if yes, use the mass and position of the current balancing block as dynamic balancing optimization parameters.

[0022] Preferably, the method for evaluating the dynamic balancing performance of the predicted value of the center of mass offset of a rotor containing a balancing block described in S2-4 adopts a finite element analysis method.

[0023] Preferably, using the current predicted value of the coordinated deviation between concentricity and center of mass as the rotor tolerance dimension chain also includes:

[0024] The concentricity and centroid coordinated deviations are set as optimization targets, and the optimization algorithm is used to optimize the tolerance zones of key geometric features at all levels to find the best optimized tolerance zone.

[0025] Preferably, the optimization algorithms include: genetic algorithm, game theory, fuzzy algorithm and artificial intelligence algorithm.

[0026] Advantages of the present invention: The tolerance dimension chain optimization method of a marine steam turbine rotor described in the present invention obtains the probability distribution of mass eccentricity and optimal balancing block compensation under a given combination of random geometric deviations of a steam turbine cylindrical disk according to the optimized design tolerance dimension chain. The unbalance of the rotor can be controlled within a small range by the design tolerance method to meet the dynamic balancing requirements of the rotor and improve the dynamic balancing accuracy of the steam turbine rotor. This method fundamentally reduces the unbalance of the rotor itself and provides a design method for actually guiding rotor adjustment and performance improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flowchart of the method for optimizing the tolerance dimension chain of a marine steam turbine rotor according to the present invention;

[0028] Figure 2 is a cross-sectional view of the rotor;

[0029] Figure 3 is the position where the rotor balancing weight is applied;

[0030] Figure 4 is the probability distribution of the turbine rotor axial offset;

[0031] Figure 5 is the probability distribution of radial offset of the turbine rotor;

[0032] Figure 6 is the probability distribution of radial unbalance of the turbine rotor;

[0033] Figure 7 is the probability distribution of the radial unbalance compensation amount of the turbine rotor. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0036] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0037] Embodiment 1:

[0038] Combine the following Figure 1 This embodiment is described. The tolerance dimension chain optimization method of a marine steam turbine rotor described in this embodiment includes:

[0039] S1. According to the size of the rotor parts, the geometric tolerances of the rotor axis and the cylindrical disks at each level are analyzed and obtained, and then the probability distribution of the center of mass of the cylindrical disks at each level in the spatial position is obtained;

[0040] S2, based on the probability distribution of the center of mass in the space position obtained in S1, generate a geometric offset deviation of the center of the cylindrical disk that conforms to the probability distribution;

[0041] S3, according to the rotor size, calculate the weight of the cylindrical disk, and combine the probability distribution of the center of mass in the spatial position obtained in S1 to obtain the corresponding center of mass offset deviation;

[0042] S4, randomly select an end face, obtain the position of the end face center point, and obtain the normal offset angle according to the centroid offset deviation and the end face center point position;

[0043] S5. Using the center geometric offset deviation of the cylindrical disk, the end face center point position and the normal offset angle as geometric features, a prediction model for the centroid offset of the dimension chain is constructed, and the prediction values ​​of the concentricity and centroid coordination deviation are output;

[0044] S6. Evaluate the characteristic sensitivity of the predicted value of the coordinated deviation between concentricity and center of mass to determine whether the characteristic sensitivity meets the geometric and center of mass accuracy indicators. Otherwise, perform dynamic balancing optimization. If yes, use the current predicted value of the coordinated deviation between concentricity and center of mass as the rotor tolerance dimension chain.

[0045] Furthermore, the dynamic balance optimization specifically includes:

[0046] S2-1. According to the geometric and centroid coordination deviations, the probability distribution interval of the concentric and centroid coordination offsets is obtained by using statistical deviation simulation;

[0047] S2-2, setting the mass and position of the balancing block within the probability distribution interval of the co-center offset obtained in S2-1;

[0048] S2-3, constructing a prediction model for rotor mass center offset combined with a balancing block, and outputting a predicted value of the rotor mass center offset including the balancing block;

[0049] S2-4, evaluate the dynamic balancing performance of the predicted value of the center of mass offset of the rotor containing the balancing block to determine whether the dynamic balancing performance meets the dynamic balancing performance index; otherwise, return to execute S2-1, and if yes, use the mass and position of the current balancing block as dynamic balancing optimization parameters.

[0050] Furthermore, the method described in S2-4 for evaluating the dynamic balancing performance of the predicted value of the center of mass offset of the rotor containing the balancing block adopts the finite element analysis method.

[0051] Furthermore, the predicted value of the current concentricity and center of mass coordinated deviation is used as the rotor tolerance dimension chain, which also includes:

[0052] The concentricity and centroid coordinated deviations are set as optimization targets, and the optimization algorithm is used to optimize the tolerance zones of key geometric features at all levels to find the best optimized tolerance zone.

[0053] Furthermore, the optimization algorithms include: genetic algorithm, game theory, fuzzy algorithm and artificial intelligence algorithm.

[0054] The present invention provides a method for optimizing the tolerance dimension chain of a marine steam turbine rotor, which reduces the imbalance of the rotor itself and solves the problems of low efficiency and long cycle in the current dynamic balancing adjustment process of the rotor, which relies on experience and lacks theoretical guidance.

[0055] Specifically include:

[0056] Step 1: Analyze the tolerance of rotor parts. Analyze the geometric tolerance of the rotor axis and various levels of cylindrical disks to obtain the probability distribution of their respective center of mass in space;

[0057] Step 2: Calculate the random deviation of the parts. Based on the probability distribution of the center of mass of each cylindrical disk in space obtained in step 1, generate the geometric offset deviation of the center of mass of the cylindrical disk that conforms to the distribution; calculate the weight of the cylindrical disk in combination with the rotor size, and calculate the corresponding center of mass offset according to the spatial position relationship between the center of mass of each cylindrical disk;

[0058] Step 3: Dimension chain and centroid offset prediction modeling. Based on the cylindrical disk center offset deviation, end face center point position and normal offset angle obtained in step 2, the deviation value applied for geometric feature modeling is used;

[0059] Step 4: Comprehensive concentricity and mass eccentricity evaluation. Based on the model established in step 2, calculate the changes in the comprehensive concentricity and mass eccentricity of the entire rotor.

[0060] Step 5: Statistical deviation analysis. Repeat the above process from step 2 to step 4 to obtain the statistical deviation simulation results and identify the sensitivity and contribution of key geometric features;

[0061] Step 6: Rotor tolerance optimization. Set the overall comprehensive concentricity and mass eccentricity as the optimization target, use intelligent optimization algorithms (such as genetic algorithms, game theory, fuzzy algorithms, artificial intelligence algorithms) to optimize the tolerance bands of key geometric features at all levels, and combine the three-dimensional deviation model of the wheel to predict the new concentricity and center of mass offset to solve the optimal design tolerance band. If the optimization target is achieved, the final rotor tolerance design is obtained. If it does not meet the requirements, repeat steps 1 to 5;

[0062] Step 7: Optimized rotor tolerance design. Based on step 6, the optimized tolerance design of the rotor rotating parts is obtained, and the probability distribution interval of the overall concentricity and the center of mass offset is obtained based on the statistical deviation simulation;

[0063] Step 8: Additional balance weight and position. Set the mass and position of the initial balance weight within the feasible domain of the center of mass offset obtained in step 7;

[0064] Step 9: Modeling the prediction of center of mass shift. Construct a prediction model of the center of mass shift of the rotor structure as a whole after considering the balancing mass.

[0065] Step 10: Evaluation of rotor structure mass eccentricity. Predict the center of mass offset of the rotor structure with balancing mass established in step 9;

[0066] Step 11: Dynamic balancing performance evaluation of the rotor structure. Based on the predicted value of mass eccentricity in step 10, the dynamic balancing performance of the rotor structure is evaluated (based on finite element or experimental methods);

[0067] Step 12: Dynamic balancing parameter optimization. According to the dynamic balancing performance results, the optimization algorithm is used to optimize the mass and position of the balancing block, and the new mass center offset is predicted in combination with the overall mass center offset model of the rotor structure. If the optimization target is achieved, the final balancing block mass and installation position can be obtained. If the target requirements are not met, repeat steps 8 to 11.

[0068] In the above process, the evaluation of concentricity and mass center deviation is the comprehensive amount of geometric eccentricity and mass eccentricity of each level of cylindrical disk; the overall concentricity and mass eccentricity control function of the multi-level cylindrical disk projected on the axial vertical plane is as follows:

[0069]

[0070] Among them, γ i is the geometric eccentricity weight coefficient of the i-th level, ε geo and ε mass are the control functions of comprehensive geometric concentricity and mass eccentricity, ε min is a comprehensive evaluation index, and α and β are the corresponding weight coefficients.

[0071] In the present invention, the feasible domain of dimensional tolerance is set, and the runout tolerance of the rotor disk is optimized with the rotor eccentric mass as the optimization target, so as to obtain the probability distribution of mass eccentricity and optimal balancing block compensation under a given combination of random geometric deviations of the turbine cylindrical disk, and optimize the compensation by adjusting the balancing block parameters during the dynamic balancing process; so as to achieve a steam turbine rotor structure that fundamentally reduces the imbalance of the rotor itself, and provide theoretical support for the actual steam turbine rotor dynamic balancing adjustment process.

[0072] In the present invention, the dynamic balancing process is optimized by designing the coordinated tolerance of the turbine rotor part geometry and the mass center offset, thereby improving the dynamic balancing accuracy of the turbine rotor. This can make up for the reliance on experience and lack of theoretical guidance in the traditional dynamic balancing adjustment process of the turbine rotor, shorten the rotor dynamic balancing adjustment process cycle, and improve efficiency.

[0073] In the present invention, Figure 2 The figure shows the cross-section of the rotor. Figure 3 The position between the balancing block and the center of mass applied on the rotor is determined by using the dynamic balancing optimization method of the rotor rotating parts based on the dimension chain mentioned above to obtain the probability distribution of the mass eccentricity and the optimal balancing block compensation amount under a given combination of random geometric deviations of the turbine cylindrical disk. The runout tolerance of each cylindrical disk is the key tolerance that affects the mass eccentricity of the part. Therefore, the following is mainly optimized for the runout tolerance. Combined with the actual process level, according to Table 1, the optimization feasible domain is set to optimize the runout tolerance of each rotor disk, and the runout tolerance of the rotor disk is optimized with the rotor eccentric mass as the optimization target, and the probability distribution of the corresponding compensation amount is calculated.

[0074] Table 1

[0075]

[0076] Figure 4 Shown is the probability distribution diagram of the axial offset of the rotor center of mass, with a distribution range of [-0.1186, 0.1197] mm.

[0077] like Figure 5 As shown, it is expressed as the probability distribution diagram of the radial offset of the rotor mass center. Figure 5 It can be seen that the distribution range is [0,1.985]×10 -3 mm.

[0078] like Figure 6As shown, it is represented as the probability distribution diagram of the radial imbalance of the rotor mass center, and the distribution range is [0,3.397]×10 3 g·mm.

[0079] When the radial distance between the two compensation blocks is 300 mm, the probability distribution of the radial imbalance compensation amount of the rotor center of mass is as follows: Figure 7 As shown, the imbalance distribution range is [0,10.324]g.

[0080] By using the dimension chain method to optimize the dynamic balance of a marine steam turbine rotor, the radial imbalance of the rotor mass center before and after the optimization of the runout tolerance of the rotor disc is compared. The mass of the balancing block can be reduced from 31.025g before optimization to 10.324g after optimization, and the mass of the balancing block is significantly reduced. The above results prove the effectiveness of the rotor imbalance optimization method with runout as the target.

[0081] The present invention can efficiently and quickly optimize the probability distribution of mass eccentricity and optimal balancing block compensation amount, select the balancing block compensation amount within a given range, and obtain a rotor with higher dynamic balancing accuracy.

[0082] This optimization method minimizes the rotor imbalance from the perspective of dimension chain design and improves the dynamic balancing accuracy of the turbine rotor. It makes up for the problem that the traditional dynamic balancing adjustment process of the turbine rotor relies on experience and lacks theoretical guidance, shortens the rotor dynamic balancing adjustment process cycle, and improves efficiency.

[0083] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. A method for optimizing the tolerance dimension chain of a marine steam turbine rotor, characterized in that: It includes: S1. According to the size of the rotor parts, the geometric tolerances of the rotor axis and the cylindrical disks at each level are analyzed and obtained, and then the probability distribution of the center of mass of the cylindrical disks at each level in the spatial position is obtained; S2, based on the probability distribution of the center of mass in the space position obtained in S1, generate a geometric offset deviation of the center of the cylindrical disk that conforms to the probability distribution; S3, according to the rotor size, calculate the weight of the cylindrical disk, and combine the probability distribution of the center of mass in the spatial position obtained in S1 to obtain the corresponding center of mass offset deviation; S4, randomly select an end face, obtain the position of the end face center point, and obtain the normal offset angle according to the centroid offset deviation and the end face center point position; S5. Using the center geometric offset deviation of the cylindrical disk, the end face center point position and the normal offset angle as geometric features, a prediction model for the centroid offset of the dimension chain is constructed, and the prediction values ​​of the concentricity and centroid coordination deviation are output; S6. Evaluate the characteristic sensitivity of the predicted value of the coordinated deviation between concentricity and center of mass to determine whether the characteristic sensitivity meets the geometric and center of mass accuracy indicators. Otherwise, perform dynamic balancing optimization. If yes, use the current predicted value of the coordinated deviation between concentricity and center of mass as the rotor tolerance dimension chain.

2. The method for optimizing the tolerance dimension chain of a marine steam turbine rotor according to claim 1, characterized in that: The dynamic balance optimization specifically includes: S2-1. According to the geometric and centroid coordination deviations, the probability distribution interval of the concentric and centroid coordination offsets is obtained by using statistical deviation simulation; S2-2, setting the mass and position of the balancing block within the probability distribution interval of the co-center offset obtained in S2-1; S2-3, constructing a prediction model for rotor mass center offset combined with a balancing block, and outputting a predicted value of the rotor mass center offset including the balancing block; S2-4, evaluate the dynamic balancing performance of the predicted value of the center of mass offset of the rotor containing the balancing block to determine whether the dynamic balancing performance meets the dynamic balancing performance index; otherwise, return to execute S2-1, and if yes, use the mass and position of the current balancing block as dynamic balancing optimization parameters.

3. The method for optimizing the tolerance dimension chain of a marine steam turbine rotor according to claim 2, characterized in that: The method described in S2-4 for evaluating the dynamic balancing performance of the predicted value of the center of mass offset of a rotor containing a balancing mass adopts the finite element analysis method.

4. The method for optimizing the tolerance dimension chain of a marine steam turbine rotor according to claim 1, characterized in that: The current predicted value of the concentricity and mass center coordinated deviation as the rotor tolerance dimension chain also includes: The concentricity and centroid coordinated deviations are set as optimization targets, and the optimization algorithm is used to optimize the tolerance zones of key geometric features at all levels to find the best optimized tolerance zone.

5. The method for optimizing the tolerance dimension chain of a marine steam turbine rotor according to claim 4, characterized in that: The optimization algorithms include: genetic algorithm, game theory, fuzzy algorithm and artificial intelligence algorithm.

Images