Error separation optimization method for large high-speed rotary equipment based on genetic algorithm
Through the optimization method based on genetic algorithm, the problem of harmonic suppression in error separation of large high-speed slewing equipment is solved, the error separation accuracy is improved, and the accurate measurement of leaf tip gap data is ensured.
Patent Information
- Application Number
- CN201911367201.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-12-26
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2039-12-26
AI Technical Summary
The error separation method of existing large-scale high-speed rotary equipment can easily lead to harmonic suppression of weight function ω(n), affecting the error separation accuracy.
Using the optimization method based on genetic algorithm, by establishing an objective function and a fitness function that optimizes the installation angle of the sensor S2 and S3 relative to the sensor S1, the installation angles α and β are optimized, so as to avoid harmonic suppression and improve error separation accuracy.
It effectively avoids harmonic suppression during error separation, improves error separation accuracy, and ensures the accuracy of blade tip gap data measurement.
Smart Images

Figure CN111125844B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a large-scale high-speed rotating equipment error separation optimization method based on a genetic algorithm, and belongs to the technical field of large-scale high-speed rotating equipment error separation optimization. Background Art
[0002] The assembly uniformity of the rotor-stator clearance of large-scale rotating equipment is an important indicator for evaluating its assembly quality. Taking aircraft engines as an example, uneven distribution of blade tip clearance will lead to rotor vibration, reduced thrust, poor reliability, low work efficiency, high fuel consumption, and poor aerodynamic performance. In severe cases, it will also cause compressor surge, rotor friction, blade friction fracture and serious engine failure. The current measurement technology of the rotor-stator clearance of large-scale high-speed rotating equipment is relatively mature, but there is a lack of control methods for the clearance. In fact, the size of the clearance is affected by many factors, including the roundness error of the inner wall of the stator of large-scale high-speed rotating equipment, the eccentricity of the rotor-stator installation, and the rotation error of the rotor of large-scale high-speed rotating equipment. Separating these errors and guiding the assembly process of large-scale high-speed rotating equipment based on this is conducive to improving the uniformity of its rotor-stator clearance. Previously, there has been a multi-error separation algorithm based on the three-sensor single-position method. The algorithm is based on three sets of clearance data measured by three sensors distributed at a certain angle on the rotor of large-scale high-speed rotating equipment. The rotor roundness error, the rotor-stator installation eccentricity error and the rotor rotation error can be separated in turn. The angle setting between the three sensors will affect the accuracy of the stator roundness error separation of large high-speed rotating equipment. If the angle setting is unreasonable, it will cause harmonic suppression of the weight function ω(n), resulting in the loss of harmonics that affect error separation, which in turn affects the accuracy of the entire error separation algorithm. At present, there is a lack of research on the distribution angle optimization method of the three sensors on the rotor of large high-speed rotating equipment. Summary of the invention
[0003] The present invention proposes an error separation optimization method for large-scale high-speed rotating equipment based on genetic algorithm, the purpose of which is to solve the problem that the existing method easily causes harmonic suppression of the weight function ω(n), resulting in the loss of harmonics affecting error separation, thereby affecting the accuracy of the entire error separation algorithm.
[0004] The error separation optimization method of large-scale high-speed rotary equipment based on genetic algorithm comprises the following steps:
[0005] Step 1: Establish an optimization objective function for optimizing the sensor installation angle;
[0006] Step 2: Establish the fitness function of the genetic algorithm according to the optimization objective function. Specifically, use the maximum entropy function method to perform an equivalent transformation on the optimization objective function to obtain:
[0007]
[0008] Further equivalent transformations are performed to obtain the fitness formula required by the genetic algorithm:
[0009]
[0010] Step 3: Set the constraints of each parameter;
[0011] Step 4: Use genetic algorithm to optimize the installation angles α and β of sensor S2 and sensor S3 relative to sensor S1;
[0012] Step 5: Determine the blade number corresponding to the optimal installation angle based on the optimization result.
[0013] Further, in step 1, specifically, in the three-sensor single-inversion method, ω(n) is defined as a harmonic weighted function, referred to as a weight function:
[0014]
[0015] The value of ω(n) depends on the installation angle of sensors S2 and S3 relative to sensor S1. and And the harmonic order n, the optimization objective function can be expressed as:
[0016]
[0017] Furthermore, in step 3, for equation (3), select g = 10 9 , set the constraints: α+15°≤β, 15°≤α and β≤345°,
[0018] Furthermore, in step 4, specifically, the optimization process includes the following steps:
[0019] Step 41: Randomly generate an initial population;
[0020] Step 42: Set the constraints α+15°≤β, 15°≤α and β≤345°, and remove individuals that do not meet the constraints from the population;
[0021] Step 43: Set the fitness function:
[0022]
[0023] Step 44: Calculate the fitness of each individual in the population;
[0024] Step 4 and 5: Perform selection operation based on the calculated fitness, followed by crossover and mutation operations to generate a new population;
[0025] Step 46: Determine whether the maximum number of iterations has been reached. If so, the optimization process terminates.
[0026] Step 47: If the maximum number of iterations has not been reached, repeat steps 42 to 46 until the maximum number of iterations is reached, and the optimization process is terminated.
[0027] Step 48: Use a genetic algorithm to optimize the values of α and β to obtain the best installation angles of sensor S2 and sensor S3 relative to sensor S1.
[0028] Further, in step five, specifically, according to and Determine the rotor plate numbers B' and B" corresponding to the optimal installation angle.
[0029] The present invention has the following beneficial effects: the present invention installs three sensors according to the blade numbers 0, B' and B" obtained by genetic algorithm optimization, measures the tip clearance data and performs error separation, which can effectively avoid the harmonic suppression phenomenon in the error separation process and improve the error separation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 The present invention is a flow chart of the error separation optimization method for large-scale high-speed rotating equipment based on genetic algorithm. DETAILED DESCRIPTION
[0031] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings 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.
[0032] Reference Figure 1 As shown, the error separation optimization method of large-scale high-speed rotary equipment based on genetic algorithm includes the following steps:
[0033] Step 1: Establish an optimization objective function for optimizing the sensor installation angle;
[0034] Step 2: Establish the fitness function of the genetic algorithm according to the optimization objective function;
[0035] Step 3: Set the constraints of each parameter;
[0036] Step 4: Use genetic algorithm to optimize the installation angles α and β of sensor S2 and sensor S3 relative to sensor S1;
[0037] Step 5: Determine the blade number corresponding to the optimal installation angle based on the optimization result.
[0038] In the preferred embodiment of this part, in step 1, specifically, in the three-sensor single-inversion method, ω(n) is defined as a harmonic weighting function, referred to as a weight function:
[0039]
[0040] The value of ω(n) depends on the installation angle of sensors S2 and S3 relative to sensor S1. and As well as the harmonic order n, in fact, the roundness error signal is mainly composed of low-frequency signals of order 2 to 50, so the selection of α and β needs to make the weight function modulus of the 2nd to 50th harmonics far away from 0 at each order of frequency, and the minimum value searched should be the largest. The optimization objective function can be expressed as:
[0041]
[0042] In the preferred embodiment of this part, in step 2 and step 3, specifically, the maximum entropy function method is used to perform an equivalent transformation on formula (2) to obtain:
[0043]
[0044] The choice of g value is closely related to the equivalent transformation accuracy of the function. In practical applications, it can be selected according to the use requirements. Usually, the g value is between 10 7 10 9 To ensure a higher conversion accuracy, here we select g = 10 9 , set the constraints: α+15°≤β, 15°≤α and β≤345°,
[0045] Further equivalent transformations are performed to obtain the fitness formula required by the genetic algorithm:
[0046]
[0047] In the preferred embodiment of this part, in step 4, specifically, the optimization process includes the following steps:
[0048] Step 41: Randomly generate an initial population;
[0049] Step 42: Set the constraints α+15°≤β, 15°≤α and β≤345°, and remove individuals that do not meet the constraints from the population;
[0050] Step 43: Set the fitness function:
[0051]
[0052] Step 44: Calculate the fitness of each individual in the population;
[0053] Step 4 and 5: Perform selection operation based on the calculated fitness, followed by crossover and mutation operations to generate a new population;
[0054] Step 46: Determine whether the maximum number of iterations has been reached. If so, the optimization process terminates.
[0055] Step 47: If the maximum number of iterations has not been reached, repeat steps 42 to 46 until the maximum number of iterations is reached, and the optimization process is terminated.
[0056] Step 48: Use a genetic algorithm to optimize the values of α and β to obtain the best installation angles of sensor S2 and sensor S3 relative to sensor S1.
[0057] In the preferred embodiment of this part, in step five, specifically, according to and Determine the rotor plate numbers B' and B" corresponding to the optimal installation angle.
Claims
1. A genetic algorithm-based error separation optimization method for large-scale high-speed rotary equipment, characterized in that: The optimization method comprises the following steps: Step 1: Establish the objective function for optimizing the sensor installation angle. In step 1, specifically, in the three-sensor single-inversion method, ω(n) is defined as a harmonic weighting function, referred to as weight function: The value of ω(n) depends on the installation angle of sensors S2 and S3 relative to sensor S1. and and harmonic order n, where B' and B" are the rotor plate numbers corresponding to the optimal installation angle. The optimization objective function can be expressed as: Step 2: Establish the fitness function of the genetic algorithm according to the optimization objective function. Specifically, use the maximum entropy function method to perform an equivalent transformation on the optimization objective function to obtain: Where g = 10 9 , α and β are the installation angles of sensor S2 and sensor S3 relative to sensor S1, n is the harmonic order, Further equivalent transformations are performed to obtain the fitness formula required by the genetic algorithm: Step 3: Set the constraints of each parameter; Step 4: Use genetic algorithm to optimize the installation angles α and β of sensor S2 and sensor S3 relative to sensor S1; In step 4, specifically, the optimization process includes the following steps: Step 41: Randomly generate an initial population; Step 42: Set the constraints α+15°≤β, 15°≤α and β≤345°, and remove individuals that do not meet the constraints from the population; Step 43: Set the fitness function: Step 44: Calculate the fitness of each individual in the population; Step 4 and 5: Perform selection operation based on the calculated fitness, followed by crossover and mutation operations to generate a new population; Step 46: Determine whether the maximum number of iterations has been reached. If so, the optimization process terminates. Step 47: If the maximum number of iterations has not been reached, repeat steps 42 to 46 until the maximum number of iterations is reached, and the optimization process is terminated. Step 48: Use a genetic algorithm to optimize the values of α and β to obtain the best installation angles of sensor S2 and sensor S3 relative to sensor S1; Step 5: Determine the blade number corresponding to the optimal installation angle based on the optimization result.
2. The error separation optimization method for large-scale high-speed rotary equipment based on genetic algorithm according to claim 1 is characterized in that: In step 3, for equation (3), select g = 10 9 , set the constraints: α+15°≤β, 15°≤α and β≤345°.
3. The error separation optimization method for large-scale high-speed rotary equipment based on genetic algorithm according to claim 1 is characterized in that: In step five, specifically, according to and Determine the rotor plate numbers B' and B" corresponding to the optimal installation angle.
