Method and device for sequencing multi-stage mistuned bladed disc rotor blades of large high-speed rotating equipment

Abstract

The application discloses a large-scale high-speed rotating equipment multi-stage mistuned bladed disc rotor blade sequencing optimization method and device, belongs to the technical field of rotor balancing, and solves the problem of low precision of multi-stage rotor unbalance control. The method comprises the following steps: according to the reference axis transformation principle, an unbalance amount transmission model of multiple blades of a multi-stage rotor is established to obtain a target function; a cloud adaptive genetic algorithm is used to obtain an optimized blade sequencing order, specifically comprising the following steps: a group of blades is regarded as a chromosome, and the blade arrangement is coded; the size of an initial population and a parent population, and the number of iterations are set; the target function is used to obtain an adaptability function, an adaptability relative value, a crossover probability and a mutation probability; iterative calculation of the cloud adaptive genetic algorithm is carried out to obtain the optimized blade sequencing order. The application is suitable for sequencing of mistuned bladed discs of large-scale high-speed rotating equipment.

Description

Technical Field

[0001] This application relates to the field of rotor balancing technology, and in particular to rotor balancing technology for multi-stage detuned bladed disks in large high-speed rotating equipment. Background Technology

[0002] As the most important component of an aircraft, the engine's operating condition directly affects flight safety. The majority of the engine's power is provided by the bladed disk. While aero-engine bladed disks have a cyclically symmetrical structure, during actual manufacturing, the mass and shape of each blade can vary, resulting in mistuning. This disrupts the original cyclic symmetry, causing vibrational energy to concentrate on certain blades during operation. These blades experience vibration amplitudes several times greater than normal blades, leading to localized high-cycle fatigue failure and posing a serious threat to the overall structure. Therefore, utilizing rotor balancing technology to reduce vibration localization caused by mistuning is of great practical significance.

[0003] To address the problem of optimizing the sorting of blades in detuned bladed disks, traditional blade sorting algorithms introduce significant randomness due to their inherent characteristics. Therefore, it is urgent to propose an optimization model for the sorting of blades in detuned bladed disks of large high-speed rotating equipment to solve the problem of unclear transmission of imbalance in multi-stage rotors. By optimizing and analyzing the blade sorting order, the magnitude of imbalance in multi-stage rotors can be precisely controlled. Summary of the Invention

[0004] The purpose of this invention is to solve the problem of low precision in controlling the imbalance of multi-stage rotors due to the unclear transmission process of multi-stage rotor imbalance in the existing technology, and to provide a method and device for optimizing the sorting of multi-stage detuned blade disk rotor blades in large high-speed rotary equipment.

[0005] This invention is achieved through the following technical solution: In one aspect, this invention provides a method for optimizing the arrangement of blades on a multi-stage detuned bladed disk rotor in a large high-speed rotating equipment. The method includes:

[0006] Based on the principle of reference axis transformation, an unbalance transmission model for multi-stage rotors is established to obtain the objective function;

[0007] The optimal leaf sorting order is obtained using a cloud-adaptive genetic algorithm, specifically including:

[0008] A group of leaves is treated as a chromosome, and the leaf arrangement is encoded.

[0009] Set the initial population and parent population sizes, and the number of iterations;

[0010] Using the objective function, obtain the fitness function, relative fitness value, crossover probability, and mutation probability;

[0011] Based on the size of the initial population and the parent population, the number of iterations, the fitness function, the relative fitness value, the crossover probability, and the mutation probability, the cloud adaptive genetic algorithm is used for iterative calculation to obtain the optimized leaf sorting order.

[0012] Furthermore, the objective function is specifically:

[0013]

[0014] in,

[0015] In the formula, zi is the height of the rotor's center of mass at stage i, zM is the height of the correction surface M, zN is the height of the correction surface N, and Q is the overall imbalance.

[0016]

[0017] Where S is the total number of blades, Mi is the mass of the i-th stage rotor disk, and mi is the mass of the i-th stage rotor blade. Let be the coordinates of the centroid of the i-th stage rotor blade, with the actual rotation axis as the reference. The coordinates of the centroid of the i-th stage rotor blade disk, with the actual rotation axis as the reference, are given. The subscript x represents the x-direction component, and the subscript y represents the y-direction component. Q i Let be the initial imbalance of the i-th stage rotor.

[0018] Furthermore, the encoding of the blade arrangement specifically includes:

[0019] Sequential coding is used as the coding method to correspond the blades and their installation positions one by one.

[0020] Furthermore, the specific method for setting the installation location is as follows:

[0021] Establish a coordinate system with the center of the impeller as the origin, with the positive direction of the x-axis as the installation position of the first blade and the counterclockwise rotation direction as the positive installation direction.

[0022] Furthermore, the method for obtaining the fitness function specifically includes:

[0023] Using the objective function, obtain the fitness function:

[0024]

[0025] Furthermore, the method for obtaining the relative fitness value specifically includes:

[0026] Using the fitness function, obtain the relative fitness value:

[0027]

[0028] Where Pop is the population size, p i Then f is the probability that this group of blades is selected. i Let be the fitness of the i-th blade group.

[0029] Furthermore, the aforementioned variations specifically include:

[0030] A leaf group is selected according to the mutation probability. Two leaves in the selected leaf group are randomly swapped to form a new leaf group. Mutation operations are performed on all parent nodes according to the mutation probability until the next generation population is generated. The mutation probability is obtained using the following formula:

[0031]

[0032] Where, p m The mutation probability, Let f be the average fitness, f' be the fitness of the mutated individual, and f' be the larger fitness value of the two crossover individuals. max The maximum fitness of the variant individual is represented by k1~k4 and c1~c4, which are control parameters. k1~k4∈[0~1]. RANDN(En,He) generates a normal random number with an expected value of En and a standard deviation of He.

[0033] Furthermore, the intersection specifically includes:

[0034] A leaf group is selected according to the crossover probability. Two leaves in the selected leaf group are randomly swapped to form a new leaf group. Crossover operations are performed on all parent nodes according to the crossover probability until the next generation of the population is generated. The formula for obtaining the crossover probability is:

[0035]

[0036] Where, p c This represents the crossover probability.

[0037] Secondly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, it executes the above-described method for optimizing the sorting of blades of a multi-stage detuned impeller rotor for a large high-speed rotating equipment.

[0038] Thirdly, the present invention provides a computer-readable storage medium storing a plurality of computer instructions, the plurality of computer instructions being used to cause a computer to execute the above-described method for optimizing the sorting of blades on a multi-stage detuned impeller rotor of a large high-speed rotating equipment.

[0039] The beneficial effects of this invention are:

[0040] Due to processing errors and material properties, detuning of blades in bladed disks can occur, disrupting the original structural symmetry and causing high-cycle fatigue failure of some blades, threatening the normal operation of the overall structure. This invention proposes to use a benchmark transformation as the objective function and a genetic algorithm as the optimization method to optimize the arrangement sequence of detuned blades in large high-speed rotating equipment.

[0041] First, this invention proposes to consider using a reference transformation as the objective function. The reference transformation makes the various transmission processes of the imbalance more accurate. Therefore, the objective function derived based on the reference transformation is more accurate, thereby improving the accuracy of the control of multi-stage rotor imbalance.

[0042] Secondly, this invention combines the objective function obtained by benchmark transformation with the cloud adaptive genetic algorithm. By utilizing the characteristics of the cloud adaptive genetic algorithm, such as its ability to combine qualitative and quantitative information, the optimization of the blade sorting of the detuned blade disk of a large high-speed rotating equipment is achieved. Ultimately, the blade sorting with the smallest imbalance among all populations after the multi-stage blade disk is assembled is obtained, thereby achieving precise control of the multi-stage rotor imbalance.

[0043] This invention is applicable to the sorting of blades in multi-stage detuned bladed disks of large high-speed rotating equipment. Attached Figure Description

[0044] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a schematic diagram of the genetic algorithm flow of the method of the present invention;

[0046] Figure 2 This refers to the rotor system imbalance under the genetic algorithm search of the method of this invention. Detailed Implementation

[0047] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0048] The cloud adaptive genetic algorithm is based on the cloud theory model. It selects individuals with higher fitness by assessing the fitness of different individuals in the population. The X-cloud conditional generator model is a stable set of random numbers following a normal distribution, characterized by expected value Ex, entropy En, and hyperentropy He. The cloud model, based on its linguistic qualitative and quantitative uncertainty transformation model, can simultaneously guarantee randomness and fuzziness, facilitating the combination of qualitative and quantitative information. The model determines the individuals to be retained by assessing their fitness and modifying the mutation and crossover operators using model generation theory. This iterative process is repeated until a convergence criterion is met. The process is as follows: Figure 1 As shown.

[0049] Implementation Method 1: A method for optimizing the arrangement of blades on a multi-stage detuned bladed disk rotor in a large high-speed rotating equipment, the method comprising:

[0050] Based on the principle of reference axis transformation, an unbalance transmission model for multi-stage rotors is established to obtain the objective function;

[0051] It should be noted that the traditional assembly method uses the axis of the circle passing through the lower end face of the first stage rotor as the reference plane and perpendicular to it as the reference axis. However, in the actual assembly of aero engines, the line connecting the centers of the bearings at both ends of the multi-stage rotor is used as the reference axis. Therefore, the equation for the transfer of unbalance of the multi-stage rotor is first obtained under the former as the reference axis, and then a rotation transformation is performed to obtain the equation for the transfer of unbalance of the multi-stage rotor under the condition that the line connecting the centers of the bearings at both ends of the multi-stage rotor is used as the reference axis.

[0052] The optimal leaf sorting order is obtained using a cloud-adaptive genetic algorithm, specifically including:

[0053] A group of leaves is treated as a chromosome, and the leaf arrangement is encoded.

[0054] Set the initial population and parent population sizes, and the number of iterations;

[0055] Using the objective function, obtain the fitness function, relative fitness value, crossover probability, and mutation probability;

[0056] Based on the size of the initial population and the parent population, the number of iterations, the fitness function, the relative fitness value, the crossover probability, and the mutation probability, the cloud adaptive genetic algorithm is used for iterative calculation to obtain the optimized leaf sorting order.

[0057] In this embodiment, firstly, it is proposed to consider using a reference transformation as the objective function. The reference transformation makes the various transmission processes of the unbalance more accurate. Therefore, the objective function derived based on the reference transformation is more accurate, thereby improving the accuracy of the control of the multi-stage rotor unbalance.

[0058] Secondly, this implementation method combines the objective function obtained by the benchmark transformation with the cloud adaptive genetic algorithm. By utilizing the characteristics of the cloud adaptive genetic algorithm, such as its ability to combine qualitative and quantitative information, the optimization of the blade sorting of the detuned blade disk of the large high-speed rotating equipment is finally achieved. The final result is the blade sorting with the smallest imbalance among all populations after the multi-stage blade disk is assembled, thereby achieving precise control of the multi-stage rotor imbalance.

[0059] It should be noted that this implementation method can follow the general design process of genetic algorithms, performing selection, crossover, mutation, and fitness design on the leaf groups. By setting the initial population and parent population sizes, crossover and mutation probabilities, and the number of iterations, iterative calculations are performed on a computer to obtain the optimal leaf sorting order.

[0060] It should be noted that the number of iterations in this embodiment is the termination condition of the method. That is, when the cloud adaptive genetic algorithm reaches the set number of iterations, the leaf sorting order obtained is the final result, and the calculation stops.

[0061] Implementation Method Two is a further definition of the method for optimizing the arrangement of blades on a multi-stage detuned bladed disk rotor in a large high-speed rotating equipment as described in Implementation Method One. In this implementation method, the objective function is further defined, specifically including:

[0062] The objective function is specifically:

[0063]

[0064] in,

[0065] In the formula, zi is the height of the centroid of the i-th stage rotor, zM is the height of the correction surface M, zN is the height of the correction surface N, Q is the overall imbalance, and the correction surface M can be set as a plane 40mm away from the lower end face of the multi-stage rotor, and the correction surface N can be set as a plane 80mm away from the lower end face of the multi-stage rotor.

[0066]

[0067] Where S is the total number of blades, Mi is the mass of the i-th stage rotor disk, and mi is the mass of the i-th stage rotor blade. Let be the coordinates of the centroid of the i-th stage rotor blade, with the actual rotation axis as the reference. The coordinates of the centroid of the i-th stage rotor blade disk, with the actual rotation axis as the reference, are given. The subscript x represents the x-direction component, and the subscript y represents the y-direction component. Q i Let be the initial imbalance of the i-th stage rotor.

[0068] In this embodiment, the optimization objective is to reduce the forced vibration assignment of the bladed disk system and alleviate the localization of system vibration. The cause of system vibration is that the unbalance error of the rotor is too large during the assembly of multi-stage rotors. The current unbalance transmission principle is very complex and is affected by various factors such as the bladed disk and blades. It is necessary to take into account the actual assembly situation of the engine. Based on the common axis, an unbalance transmission model after the assembly of multi-stage rotors is established. Therefore, this embodiment uses the principle of reference axis transformation to establish a transmission model of multiple blades of multi-stage rotors and establish an objective function.

[0069] The objective function of this implementation method has high accuracy, thereby improving the accuracy of multi-stage rotor imbalance control.

[0070] It should be noted that after obtaining the unbalance of each stage of the rotor according to formula (3), they are projected onto the two correction planes M and N (the planes located 40mm and 80mm away from the lower end face of the multi-stage rotor, respectively). The unbalance expressions located on these two correction planes are the two formulas in formula (2). Then, according to formula (2), the final overall multi-stage rotor unbalance expression (3) can be obtained.

[0071] Implementation method three is a further limitation on the method for optimizing the arrangement of blades in a multi-stage detuned bladed disk rotor of a large high-speed rotating equipment as described in implementation method one. In this implementation method, the encoding of the blade arrangement is further limited, specifically including:

[0072] Sequential coding is used as the coding method to correspond the blades and their installation positions one by one.

[0073] It should be noted that in this embodiment, a group of leaves is considered as a chromosome, and each leaf can be regarded as a gene of the chromosome. For example, 200 groups of leaves are randomly generated as the initial population, and the population size is set to 100 groups. Before sorting the leaves of the leaf disc, the leaf arrangement needs to be encoded. This embodiment uses sequential encoding as the encoding method, that is, there is a one-to-one correspondence between the leaves and the installation position.

[0074] This embodiment uses sequential coding based on the installation position of the blades, which facilitates the sorting of blades on detuned bladed disks in large high-speed rotating equipment.

[0075] Implementation Method Four: This implementation method further defines the method for optimizing the arrangement of blades on a multi-stage detuned bladed disk rotor in a large high-speed rotating equipment as described in Implementation Method One. In this implementation method, the specific method for setting the installation position is further defined, specifically including:

[0076] Establish a coordinate system with the center of the impeller as the origin, with the positive direction of the x-axis as the installation position of the first blade and the counterclockwise rotation direction as the positive installation direction.

[0077] This embodiment takes into account the actual situation of the detuned blades of large high-speed rotating equipment. For example, if the blade is circular, the installation position can be set to effectively realize the sorting of the detuned blades of large high-speed rotating equipment.

[0078] Implementation method five is a further limitation on the method for optimizing the sorting of blades on a multi-stage detuned bladed disk rotor of a large high-speed rotating equipment described in implementation method one. In this implementation method, the method for obtaining the fitness function is further limited, specifically including:

[0079] Using the objective function, obtain the fitness function:

[0080]

[0081] The fitness function in this embodiment enables the genetic algorithm to search in the optimal direction and also achieves sufficiently high computational accuracy.

[0082] The fitness function is the foundation of genetic algorithm evolution. Choosing a suitable fitness function can guide the genetic algorithm's search towards the optimal direction and also achieve sufficiently high computational accuracy. Considering that the bladed disk has no initial imbalance and the blade only has imbalance caused by the mass moment, the fitness function given in this implementation method is adopted.

[0083] Implementation method six is ​​a further limitation on the method for optimizing the sorting of blades on a multi-stage detuned bladed disk rotor in a large high-speed rotating equipment as described in implementation method one. In this implementation method, the method for obtaining the relative fitness value is further defined, specifically including:

[0084] Using the fitness function, obtain the relative fitness value:

[0085]

[0086] Where Pop is the population size, p i Then f is the probability that this group of blades is selected. i Let be the fitness of the i-th blade group.

[0087] In this embodiment, the roulette wheel selection algorithm is used as the selection operator. The blade group with higher fitness has a higher probability of being selected. The fitness of the i-th blade group is set to f. i .

[0088] Using the above relative fitness values ​​as the probability of the population being selected can improve computational accuracy.

[0089] It should be noted that the fitness f of the i-th blade group i It can be obtained from formula (4).

[0090] Implementation method seven is a further limitation on the method for optimizing the arrangement of blades in a multi-stage detuned bladed disk rotor of a large high-speed rotating equipment as described in implementation method one. In this implementation method, the variations are further limited, specifically including:

[0091] A leaf group is selected according to the mutation probability. Two leaves in the selected leaf group are randomly swapped to form a new leaf group. Mutation operations are performed on all parent nodes according to the mutation probability until the next generation population is generated. The mutation probability is obtained using the following formula:

[0092]

[0093] Where, p m The mutation probability, Let f be the average fitness, f' be the fitness of the mutated individual, and f' be the larger fitness value of the two crossover individuals. max The maximum fitness of the variant individual is represented by k1~k4 and c1~c4, which are control parameters. k1~k4∈[0~1]. RANDN(En,He) generates a normal random number with an expected value of En and a standard deviation of He.

[0094] The mutation operation in this embodiment uses a two-element optimized mutation operator, which can restore the undeveloped or lost genetic material of the leaf group, so as to prevent the population from converging too early in the process of finding the optimal solution.

[0095] Implementation method eight is a further definition of the method for optimizing the arrangement of blades on a multi-stage detuned bladed disk rotor in a large high-speed rotating equipment as described in implementation method one. In this implementation method, the "crossing" is further defined, specifically including:

[0096] A leaf group is selected according to the crossover probability. Two leaves in the selected leaf group are randomly swapped to form a new leaf group. Crossover operations are performed on all parent nodes according to the crossover probability until the next generation of the population is generated. The formula for obtaining the crossover probability is:

[0097]

[0098] Where, p c This represents the crossover probability.

[0099] The crossover operation in this embodiment maintains the renewal of the offspring of the leaf group, ensuring the possibility of more suitable leaf groups emerging. To reduce computation and save time, this invention uses mutation instead of crossover. Based on the basic principles of biological evolution, populations with high fitness are selected and retained, while those with low fitness are eliminated.

[0100] It should be noted that the mutation probability and crossover probability mentioned above are both given by the cloud condition generator.

[0101] Implementation method nine, this implementation method is a specific embodiment of the multi-stage mistunted blade disk rotor blade sorting optimization method for large high-speed rotating equipment as described above, specifically including:

[0102] The optimization objective of this embodiment is to reduce the forced vibration assignment of the bladed disk system and alleviate the localization of system vibration. The cause of system vibration is the excessive unbalance error of the rotor during multi-stage rotor assembly. Currently, the principle of unbalance transfer is very complex and is affected by various factors such as the bladed disk and blades. It is necessary to consider the actual assembly situation of the engine. Taking the common axis as the reference, an unbalance transfer model after multi-stage rotor assembly is established. Therefore, this embodiment uses the principle of reference axis transformation to establish an unbalance transfer model of multi-stage rotor and establish an objective function, specifically:

[0103]

[0104] In the formula M i Let m be the mass of the i-th stage rotor blade disk; i Let be the mass of the i-th stage rotor blade;

[0105] The coordinates of the centroid of the i-th stage rotor blade are based on the actual rotation axis.

[0106] The coordinates of the centroid of the i-th stage rotor blade disk, with the actual rotation axis as the reference.

[0107] Q i The initial imbalance of the i-th stage rotor; S is the total number of blades;

[0108] After obtaining the unbalance of each stage of the rotor, project them onto two correction planes, M and N (planes located 40mm and 80mm away from the lower end face of the multi-stage rotor, respectively). The expressions for the unbalance on these two correction planes are as follows:

[0109]

[0110]

[0111] Finally, the overall expression for the multi-stage rotor imbalance is:

[0112]

[0113] In the formula, z i Let z be the height of the centroid of the i-th stage rotor. M For the height of the correction surface M, z N Let N be the height of the correction surface, Q be the overall imbalance, M be the plane 40mm from the lower end face of the multi-stage rotor, and N be the plane 80mm from the lower end face of the multi-stage rotor.

[0114] If we consider a group of leaves as a chromosome, then each leaf can be regarded as a gene on that chromosome. 200 leaf groups are randomly generated as the initial population, with a population size of 100 groups. Before sorting the leaves in the leaf disc, the leaf arrangement needs to be encoded. This embodiment uses sequential encoding, meaning there is a one-to-one correspondence between leaves and their installation positions. Since the leaf disc is circular, a coordinate system is established with the center of the circle as the origin. The positive x-axis represents the installation position of the first leaf, and counterclockwise rotation is the positive installation direction.

[0115] The fitness function is the foundation of genetic algorithm evolution. Choosing a suitable fitness function can guide the genetic algorithm's search towards the optimal direction and also achieve sufficiently high computational accuracy. Considering that the bladed disk has no initial imbalance and the blade only has imbalance caused by the mass moment, the fitness function given in this embodiment is as follows:

[0116]

[0117] This embodiment follows the general design process of genetic algorithms, performing selection, crossover, mutation, and fitness design on the leaf groups.

[0118] Using roulette wheel selection as the algorithm's selection operator, blade groups with higher fitness are more likely to be selected. The fitness of the i-th blade group is set to f. i .

[0119] Its relative fitness value is as follows:

[0120]

[0121] Pop is the population size, p i Then f is the probability that this group of blades is selected. i Let be the fitness of the i-th blade group.

[0122] Mutation operations are used to restore undeveloped or lost genetic material in leaf groups to prevent the population from converging too early in the process of finding the optimal solution. A two-element optimization mutation operator is used, that is, a leaf group is selected according to the mutation probability, and the positions of two leaves in the leaf group are randomly swapped to form a new leaf group. Mutation operations are performed on all parent nodes according to the mutation probability until the next generation of population is generated.

[0123] The purpose of crossover is to maintain the renewal of offspring in the blade group, ensuring the possibility of a more suitable blade group emerging. To reduce computation and save time, this embodiment uses mutation instead of crossover. Based on the basic principles of biological evolution, populations with high fitness are selected and those with low fitness are eliminated. The mutation probability and crossover probability are both given by the cloud condition generator, and their expressions are as follows:

[0124]

[0125]

[0126] In the formula is the average fitness; f is the fitness of the mutated individual; f' is the larger value of the fitness of the two crossover individuals; k1~k4, c1~c4 are control parameters, k1~k4∈[0~1], RANDN(En,He) generates normal random numbers with expected value En and standard deviation He.

[0127] By setting the initial and parent population sizes, crossover and mutation probabilities, and the number of iterations, iterative calculations are performed on a computer to obtain the optimal leaf sorting order.

[0128] Using the same parameters, a cloud-based adaptive genetic algorithm was employed to optimize the imbalance of a multi-stage disk-separated rotor system. The cloud-based adaptive genetic algorithm parameters were set as k1 = k3 = 0.9, k2 = k4 = 0.4, c1 = 2.8, c3 = 3.1, c2 = c4 = 9. The relationship between the imbalance and the number of iterations was obtained as follows: Figure 2 As shown in the figure, the cloud adaptive genetic algorithm has a fast fitness update speed. When the 100th generation of offspring is reached, the optimal imbalance value in the population has been reduced to 345.62 g·mm, and the search time is 355.27 s. In this embodiment, the threshold for selecting and retaining the fitness of offspring is continuously changed by the population fitness, which improves the search effect of population genetics and finally finds the blade order with the smallest imbalance among all populations after the multi-stage bladed disk blade assembly.

Claims

1. A method for optimizing the arrangement of blades in a multi-stage mistunted bladed disk rotor of a large high-speed rotating equipment, characterized in that... The method includes: Based on the principle of reference axis transformation, an unbalance transmission model for multi-stage rotors is established to obtain the objective function; The optimal leaf sorting order is obtained using a cloud-adaptive genetic algorithm, specifically including: A group of leaves is treated as a chromosome, and the leaf arrangement is encoded. Set the initial population and parent population sizes, and the number of iterations; Using the objective function, obtain the fitness function, relative fitness value, crossover probability, and mutation probability; Based on the size of the initial population and the parent population, the number of iterations, the fitness function, the relative fitness value, the crossover probability, and the mutation probability, the cloud adaptive genetic algorithm is used for iterative calculation to obtain the optimized leaf sorting order. The objective function is specifically: in, , In the formula, z i For the first i Stage rotor center of mass height, z M For the correction surface M height, z N For the correction surface N height, Q This represents the overall imbalance. Where S is the total number of blades. M i Let i be the mass of the i-th stage rotor blade disk. m i Let be the mass of the i-th stage rotor blade. Let be the coordinates of the centroid of the i-th stage rotor blade, with the actual rotation axis as the reference. The coordinates of the centroid of the i-th stage rotor blade disk, with the actual rotation axis as the reference, are given. The subscript x represents the x-direction component, and the subscript y represents the y-direction component. Let be the initial imbalance of the i-th stage rotor.

2. The method for optimizing the arrangement of blades in a multi-stage mistunted bladed disk rotor of a large high-speed rotating equipment according to claim 1, characterized in that, The encoding of the blade arrangement specifically includes: Sequential coding is used as the coding method to correspond the blades and their installation positions one by one.

3. The method for optimizing the blade arrangement of a multi-stage mistuned bladed disk rotor in a large high-speed rotating equipment according to claim 2, characterized in that, The specific method for setting the installation location is as follows: Establish a coordinate system with the center of the impeller as the origin, with the positive direction of the x-axis as the installation position of the first blade and the counterclockwise rotation direction as the positive installation direction.

4. The method for optimizing the blade arrangement of a multi-stage mistuned bladed disk rotor in a large high-speed rotating equipment according to claim 1, characterized in that, The method for obtaining the fitness function specifically includes: Using the objective function, obtain the fitness function: 。 5. The method for optimizing the blade arrangement of a multi-stage mistuned bladed disk rotor in a large high-speed rotating equipment according to claim 1, characterized in that, The method for obtaining the relative fitness value specifically includes: Using the fitness function, obtain the relative fitness value: in, Pop For the size of the population, p i This represents the probability that the group of blades will be selected. Let be the fitness of the i-th blade group.

6. The method for optimizing the blade arrangement of a multi-stage detuned bladed disk rotor in a large high-speed rotating equipment according to claim 1, characterized in that, The variations specifically include: A leaf group is selected according to the mutation probability. Two leaves in the selected leaf group are randomly swapped to form a new leaf group. Mutation operations are performed on all parent nodes according to the mutation probability until the next generation population is generated. The mutation probability is obtained using the following formula: in, The mutation probability, For average fitness, For the fitness of the variant individuals, The larger of the fitness values ​​of the two crossover individuals is used. The maximum fitness of the mutated individual is represented by k1~k4, and c1~c4 are control parameters, where k1~k4∈[0~1]. RANDN(En, He) The expected value generated is En The standard deviation is He Normal random numbers.

7. The method for optimizing the blade arrangement of a multi-stage mistuned bladed disk rotor in a large high-speed rotating equipment according to claim 6, characterized in that, The intersection specifically includes: A leaf group is selected according to the crossover probability. Two leaves in the selected leaf group are randomly swapped to form a new leaf group. Crossover operations are performed on all parent nodes according to the crossover probability until the next generation of the population is generated. The formula for obtaining the crossover probability is: in, This represents the crossover probability.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, The steps of the method according to any one of claims 1 to 7 are performed when the processor runs the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of computer instructions, which are used to cause a computer to perform the method of any one of claims 1 to 7.

Images