Intelligent assembly process for aero-engine or gas turbine blade

Abstract

The invention discloses an intelligent assembly technology for blades of an aero-engine or a gas turbine. The intelligent assembly technology comprises the steps that S1, aerodynamic performance testing is conducted on the blades through an experiment system; s2, blade mass moment measurement, blade inherent frequency measurement and blade appearance scanning are carried out through a measurement system; s3, blade pneumatic and vibration simulation and simulation data checking under the experiment working condition are carried out through a simulation system; s4, judgment is carried out, and if the simulation data check in the above steps is not passed, pneumatic and vibration full-cycle optimal installation position optimization is carried out through an optimization system; and if the simulation data passes the verification, pneumatic and vibration simulation and pneumatic and vibration full-cycle optimal installation position optimization of the optimization system under the real working condition are carried out. According to the method, the aerodynamic performance and the vibration performance of the blades can be comprehensively considered, and the optimal phase angle for mounting each blade is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft engines or gas turbines, and in particular to an intelligent assembly process for aircraft engine or gas turbine blades. Background Art

[0002] The compressor and turbine are the two key components of aircraft engines and gas turbines. Whether it is the compressor blades or the turbine blades, their assembly process has a great impact on the performance of the unit. Research on key technologies for the intelligent assembly process of aircraft engine or gas turbine blades is of great significance to improving the performance of aircraft engines and gas turbines.

[0003] In the compressor and turbine of a gas turbine, the rotor blades work under high aerodynamic load and high speed. The circumferential imbalance of the blade aerodynamic load and the imbalance of the rotor mass seriously affect the performance and working reliability of the impeller machinery. According to statistics, about 60% of the vibration problems of turbine machinery are caused by rotor imbalance. The uneven circumferential aerodynamic load of the compressor blades will reduce the surge margin.

[0004] Although the manufacturing of gas turbine blades is strictly in accordance with the process flow of the drawings, each blade produced by the manufacturer has its own characteristics. Whether it is aerodynamic characteristics or vibration characteristics, each blade will be different from other blades of the same model produced in the same batch. Therefore, for blades that meet the processing accuracy requirements, optimizing their circumferential installation position can obtain an annular blade row with balanced aerodynamic performance, better vibration and dynamic balance effects throughout the whole cycle. Especially for compressor blade rows, optimizing the circumferential installation position of the blades can also delay the occurrence of rotating stall and effectively improve the surge margin of the compressor.

[0005] Aerodynamic imbalance and mass imbalance of the rotor are one of the important causes of gas turbine failure. The impact of rotor imbalance can be effectively reduced by rationally planning the blade assembly sequence. Some foreign companies have developed blade arrangement software that can perform mathematical balancing based on the actual rotation centerline of the rotor, but there is little relevant public information. A typical example is the assembly method for rotating mechanical blades proposed by General Electric Company of the United States. This assembly method requires measuring the weight moment of each blade, and continuously iterates and replaces the circumferential sequence of the rotor blades based on the calculation of a computer system composed of software code segments, so that the final total weight moment of the rotor is less than the specified threshold, thereby controlling the rotor imbalance. Rolls-Royce Limited has proposed a method to achieve dynamic balancing of rotating parts by predicting the mass and moment weight of the blade according to the relative position of the blade assembly and filling the cavity in the blade. Nowadays, with the development of computer technology, international researchers have innovatively handled the assembly sequence optimization problem by establishing new algorithms or improving existing algorithms, which not only improves the accuracy and robustness of problem solving, but also improves the efficiency of problem solving.

[0006] In domestic engineering practice, the assembly of blades generally adopts the traditional method. First, the blades are weighed and numbered, and then the disk is divided into several quadrants. The total mass difference of blades in adjacent quadrants is minimized to meet the static balancing requirements, and the oscillation frequency between adjacent blades is maximized, that is, the "staggered frequency method" is used to reduce the probability of engine vibration failure, and finally the optimization algorithm is used to solve the blade sorting. In addition, the application of optimization algorithms in rotor balancing has also achieved the expected results, such as annealing single-parent algorithm, genetic algorithm, ant colony algorithm and particle swarm algorithm.

[0007] At present, domestic researchers have made remarkable achievements in solving the optimization problem of blade assembly sequence, but the efficiency of calculating the relatively optimal assembly sequence is still not high. In addition, for large domestic steam turbine manufacturers or gas turbine manufacturers, the blade's own weight, resistance characteristics and vibration characteristics are not fully considered in the blade installation process, resulting in obvious rotor imbalance after the blade assembly is completed, which seriously hinders the improvement of the aerodynamic and vibration performance of the gas turbine or steam turbine. Summary of the invention

[0008] In view of the shortcomings of the prior art, the purpose of the present invention is to provide an intelligent assembly process for aircraft engine or gas turbine blades. Based on the measurement results of local or partial aerodynamic, geometric and vibration parameters, the aerodynamic and vibration characteristics of the aircraft engine or gas turbine blades are extracted by means of numerical analysis methods. With the goal of optimizing the aerodynamic and vibration performance of the actual blade cascade, the optimal installation position of the blade in the turbine / compressor blade cascade is recommended. In order to achieve the above-mentioned purpose and other advantages according to the present invention, an intelligent assembly process for aircraft engine or gas turbine blades is provided, including:

[0009] S1. Test the aerodynamic performance of blades through the experimental system;

[0010] S2. Measure blade mass moment, blade natural frequency and blade shape through the measuring system;

[0011] S3. Perform blade aerodynamic and vibration simulation and simulation data verification under experimental conditions through the simulation system;

[0012] S4. Make a judgment. If the simulation data verification in the above steps fails, the optimization system is used to optimize the best installation position of pneumatic and vibration in the whole circumference; if the simulation data verification passes, the pneumatic and vibration simulation under the actual working conditions and the optimization system are used to optimize the best installation position of pneumatic and vibration in the whole circumference.

[0013] Preferably, in step S4, when the simulation data is verified, the data other than the mass moment in the fitness function used by the optimization system are the pneumatic and simulation data under the actual working conditions;

[0014] When the simulation data verification fails, the data other than the mass moment in the fitness function used by the optimization system are the data measured by the blade natural frequency measurement in step S1 and step S2.

[0015] Preferably, the measurement system includes a measurement component, and the measurement component is used to measure the mass moment, natural frequency and geometric shape of the blade;

[0016] The measurement assembly includes a mass moment module and a natural frequency and geometry module.

[0017] Preferably, the mass moment module comprises a horizontal guide rail, a support frame 1 and a support frame 2 movably arranged on the horizontal guide rail, a mass sensor 1 and a mass sensor 2 respectively arranged on the support frame 1 and the support frame 2, and a blade to be measured mounted on the mass sensor 1 and the mass sensor;

[0018] The two ends of the blade to be measured are respectively provided with a laser sensor 1 and a laser sensor 2;

[0019] A laser transmitter is fixedly connected to the horizontal guide rail.

[0020] Preferably, the natural frequency and geometry module includes a workbench, a data acquisition device fixedly mounted on the workbench, a robotic arm, a vise and a blade to be measured fixed on the vise, an end effector fixed to the end of the robotic arm, a vibration applying device clamped on the end effector, and an exciter fixed on the vibration applying device.

[0021] Preferably, the experimental system comprises a temporary wind tunnel device, which is used to make the blades operate at a predetermined temperature, speed and angle of attack to obtain the force, torque and temperature field exerted on the blades.

[0022] Preferably, the simulation system is used to evaluate the aerodynamic performance and vibration performance of the blade under experimental conditions and real conditions, and comprises a high-performance computer and simulation software;

[0023] Specifically, the simulation system is used to calculate the blade indicators including the blade axial force, tangential force, torque, bending moment, maximum temperature and natural frequency under experimental conditions and real conditions;

[0024] The boundary conditions of the simulation loading under the test conditions are the incoming flow velocity and temperature of the measurement system, and the boundary adjustments of the simulation loading under the actual conditions are the inlet total temperature, inlet total pressure, outlet static pressure and impeller speed in the design parameters of the aircraft engine or gas turbine.

[0025] Compared with the prior art, the present invention has the following beneficial effects: by using the measurement results of local or partial aerodynamic, geometric and vibration parameters and with the help of numerical analysis methods, the aerodynamic and vibration characteristics of aircraft engine or gas turbine blades are extracted, and the optimal installation position of the blades in the turbine / compressor blade grid is recommended with the goal of optimizing the aerodynamic and vibration performance of the actual blade grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of the process flow of the intelligent assembly process of aero-engine or gas turbine blades according to the present invention;

[0027] Figure 2 It is a schematic diagram of a mass moment module of a measurement system for an intelligent assembly process of an aircraft engine or a gas turbine blade according to the present invention;

[0028] Figure 3 It is a schematic diagram of a mass moment module of a measurement system for an intelligent assembly process of an aircraft engine or a gas turbine blade according to the present invention. DETAILED DESCRIPTION

[0029] 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.

[0030] Reference Figure 1 , an intelligent assembly process for aero-engine or gas turbine blades, comprising the following steps:

[0031] First, the aerodynamic performance test of the blade is performed in step S1 using the experimental system.

[0032] The measuring components of the measuring system include a fan, an air storage tank, a regulating valve, a heating section, a stable section, a transition section, a rotating platform, a blade to be measured, a six-component force balance and a thermal imager.

[0033] The specific working process of the experimental system is as follows:

[0034] Fix the blades on the rotating disc and adjust the angle control motor until the blade installation angle is consistent with the incoming flow direction. During the adjustment process, observe the signal feedback of the angle sensor. When the angle reaches within 0.5° of the set angle, set the angle controller to stop automatic rotation and change to manual control, and adjust until the angle is consistent;

[0035] Turn on the fan switch and adjust the frequency converter to control the fan speed, thereby adjusting the fan flow. When the flow meter indicates that the test flow has reached the specified flow, the fan frequency converter will no longer be adjusted;

[0036] Observe the pressure index gauge in the gas tank. If the pressure change is less than 0.5% within 1 minute, it means that the pressure has stabilized and you can proceed to the next step.

[0037] Turn on the heater and adjust the heater power to control the airflow temperature. When the thermometer indicates that the airflow temperature reaches the specified temperature, the heater power is no longer adjusted;

[0038] Every 30 seconds, use a thermal imager to record the two-dimensional temperature distribution on the blade surface through the infrared observation window until the temperature stops changing;

[0039] Record the torque and axial force in the X-axis direction, and the bending moment and tangential force in the Y-axis direction displayed by the six-component force balance;

[0040] Use a thermal imager to record the two-dimensional temperature distribution image on the blade surface through the infrared observation window;

[0041] Next, the blade mass moment measurement of step S2 is performed using the mass moment module of the measurement system:

[0042] Adjust the position of the horizontal guide rail slider so that it can place the complete blade. Place the blade on the support platform, fix the blade tenon on support frame 1, and fix the blade tip on support frame 2. Place a laser sensor at the blade tip and blade root respectively;

[0043] Adjust the height of support frame 1 and 2, and observe the display at the same time to make the height of the leaf tip and root consistent with the height of the laser transmitter;

[0044] Observe the display screen and record the readings of mass sensor 1 and mass sensor 2;

[0045] Observe the scale of the horizontal guide rail and record the distance between the support frame 1 and the blade root hub surface;

[0046] Observe the scale of the horizontal guide rail and record the distance between support frame 2 and support frame 1;

[0047] Using Newton's third law and the static equilibrium condition, we know that the blade mass moment is:

[0048] M=M1*(R+L1)+M2*(R+L1+L2)

[0049] Where, M1 is the reading of mass sensor 1, M2 is the reading of mass sensor 2, R is the distance between the blade root and the rotation center, L1 is the distance between support frame 1 and the blade root hub surface, and L2 is the distance between support frame 2 and support frame 1;

[0050] Next, the natural frequency of the S3 blade is measured using the natural frequency and geometry module of the measurement system:

[0051] Use a clamping vise device to fix the root tenon of the blade in the fixture, push the clamp groove to fix the blade in the jaws. Ensure the stability of the fixture to fix the blade to avoid vibration or displacement during the test. Check that the contact between the fixture and the root of the blade is uniform and the clamping force is moderate to prevent slippage or unstable fixation;

[0052] Fix the excitation application device to the end effector interface of the robot arm through the slot. First, adjust the position of the end of the robot arm so that it is in the middle of the blade. Then, adjust the length of the exciter guide rail so that the exciter head can directly contact the blade, and use an interface or gasket between the exciter and the blade to improve the contact quality and stability. Next, adjust the angle of the ring arm so that it can cover all the blades, and finally tighten the screws on the surface of the ring arm so that the blade can be completely and stably fixed in the excitation platform;

[0053] Use adhesive to fix the accelerometer to the top of the blade;

[0054] Use the signal generator control system to drive the exciter, apply vibration, and gradually scan the frequency range from low frequency 10HZ to high frequency 5000HZ, with the step size gradually increasing from 0.1HZ to 2HZ. The frequency step size selection should be adjusted according to the required resolution, usually with a smaller step size at a lower frequency and a larger step size at a higher frequency;

[0055] Record the sensor vibration signal through the data acquisition system, configure the data acquisition system, including sampling rate and sampling time, to ensure the accuracy and integrity of the data;

[0056] Preprocess the collected signals to remove noise, smooth and process windows, and use the window functions of Hening window and Hamming window to reduce signal leakage and spectrum artifacts;

[0057] Perform Fourier transform on the acquired time domain signal, convert it to the frequency domain, and generate a spectrum. Fourier transform uses the fast Fourier transform (FFT) algorithm to convert the time domain signal into a frequency domain signal and process possible spectral artifacts or noise;

[0058] In the spectrum graph, look for obvious peaks. Each peak represents a natural frequency or the resonant frequency of a vibration mode. Determine the natural frequency by analyzing the peaks on the spectrum graph, and determine the vibration mode based on the frequency peaks;

[0059] Next, the blade profile of the S4 is scanned using the natural frequency and geometry module of the measurement system:

[0060] Remove the vibration applying device at the end of the robot arm in the previous step;

[0061] Fix the 3D scanner to the end effector interface of the robot arm to ensure a stable connection between the 3D scanner and the robot arm;

[0062] Before scanning, use the calibration target to calibrate the 3D scanner and complete the detection of each image corner point;

[0063] Adjust the robot arm's travel range and the end-point alignment angle so that the scanner's field of view can cover the entire blade. Confirm that the robot arm can rotate 360 ​​degrees around the blade to ensure that the blade is fully scanned from all angles. Check the scanner's installation angle and position to ensure that the scanner can capture all geometric details of the blade;

[0064] In the data acquisition system, the 3D scanner is controlled to project stripe light onto the blade;

[0065] Keep moving the end of the robotic arm. During the movement, make sure that the horizontal and vertical rotation angles do not exceed 30 degrees, and the translation distance does not exceed 3 / 4 of the scanning range to ensure that there are enough markers for the two acquisitions to stitch the images.

[0066] In the data acquisition system, the collected point cloud is screened and the noise points are deleted;

[0067] Export the required project files;

[0068] Next, use the simulation system to conduct aerodynamic and vibration simulation of S5 under experimental conditions:

[0069] For the measured blade model, the two-dimensional blade profile coordinates at three typical cross sections of 10%, 50% and 90% are extracted. Based on the two-dimensional CFD simulation method and the boundary conditions of the experimental system (incoming wind speed and incoming flow temperature), the blade surface pressure and temperature fields are solved.

[0070] For the discrete surface data of the blade, the axial force, tangential force, torque and bending moment are solved by integration. The formula is:

[0071] Mx≈∑ i-1 n (y i ·P z,i ·△S i -Z i ·P y,i ·△S i )

[0072] Among them, △S i is the infinitesimal area at discrete point i, P z,i , P y,i is the corresponding pressure component;

[0073] The measured blade model is divided into solid domain grids. Based on the finite element analysis method, the boundary conditions of the experimental system (incoming wind speed and incoming flow temperature) are loaded to solve the natural frequency of the blade;

[0074] Next, use the simulation system to verify the simulation data of S6:

[0075] Compare the blade surface axial force, tangential force, bending moment and torque obtained in the previous step with the experimental values. If the error is less than 2%, the aerodynamic check is considered to have passed.

[0076] Compare the first-order natural frequency of the blade obtained in the previous step with the experimental value. If the error is less than 1%, it is considered that the natural frequency verification has passed;

[0077] If both the aerodynamic check and the natural frequency check are passed, the simulation check is considered passed, otherwise it is considered failed;

[0078] If the simulation check in the previous step is passed, the simulation system is used to perform aerodynamic and vibration simulation of S7 under real working conditions:

[0079] Based on the measured two-dimensional blade profile coordinates at three typical cross-sections of 10%, 50% and 90% of the blade model, the pressure and temperature fields on the blade surface are solved based on the two-dimensional CFD simulation method and the boundary conditions of the real gas turbine or aircraft engine design parameters (inlet total temperature, inlet total pressure, outlet static pressure, speed);

[0080] For the discrete surface data of the blade, the axial force, tangential force, torque and bending moment are solved by integration;

[0081] The measured blade model is divided into solid domain meshes. Based on the finite element analysis method, the boundary conditions of the actual gas turbine or aircraft engine design parameters (inlet total temperature, inlet total pressure, outlet static pressure, speed) are loaded to solve the natural frequency of the blade;

[0082] Next, use the optimization system to optimize the best installation position for the S8's pneumatics and vibrations:

[0083] Set the optimization variable to the blade arrangement number;

[0084] The performance parameters of each blade are set, including data selection adjustment. If the simulation data of S6 passes the verification, the data in the blade performance data except the mass moment are the aerodynamic and simulation data under the actual working conditions of S7; if the simulation data of S6 fails the verification, the data in the blade performance data except the mass moment are the data measured by S1 and S3.

[0085] Set optimization parameters, including crossover rate, mutation rate, elite rate, and initial population size;

[0086] The fitness function is set to be the weighted sum of the performance unevenness of the leaves after rearrangement:

[0087] OBF=C1·Mg+C2·Fx+C3·Fy+C4·Mx+C5·My+C6·T+C7·ω

[0088] Among them, Ci is the weight coefficient, Mg is the mass moment imbalance, Fx and Fy are the unevenness of the axial and tangential forces, Mx and My are the unevenness of the torque and bending moment, T is the unevenness of the maximum temperature, and ω is the unevenness of the natural frequency;

[0089] The sorting was optimized by genetic algorithm 100 times relatively independently, and the one with the highest fitness was selected as the optimal sorting of leaves in the final optimization result.

[0090] In summary, the present application uses a temporary wind tunnel device to make the blade work at a certain temperature, speed and angle of attack to obtain the force, torque and temperature field of the blade; the measurement system uses the measurement equipment to measure the mass moment, natural frequency and geometric shape of the blade, and the simulation system uses high-performance computers and simulation software to evaluate the aerodynamic performance and vibration performance of the blade under experimental conditions and real conditions, obtain the natural frequency of the blade based on the finite element method, obtain the flow field and temperature field of the typical cross-section of the blade based on the two-dimensional computational fluid dynamics method, and integrate to obtain the circumferential force, tangential force, bending moment and torque of the blade; the optimization system uses high-performance computers and optimization software, based on the multi-objective optimization platform of the genetic algorithm, and uses the multiple performance data of each blade as the fitness function of the genetic algorithm to optimize the best sorting order of the blade. Taking into account the aerodynamic performance and vibration performance of the blade, the optimal phase angle for the installation of each blade is proposed.

[0091] The number of devices and processing scales described here are used to simplify the description of the present invention, and the application, modification and variation of the present invention will be obvious to those skilled in the art.

[0092] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and implementation modes. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.

Claims

1. An intelligent assembly process for aircraft engine or gas turbine blades, characterized in that: The following steps are involved: S1. Test the aerodynamic performance of blades through the experimental system; S2. Measure blade mass moment, blade natural frequency and blade shape through the measuring system; S3. Perform blade aerodynamic and vibration simulation and simulation data verification under experimental conditions through the simulation system; S4. Make a judgment. If the simulation data verification in the above steps fails, the optimization system is used to optimize the best installation position of pneumatic and vibration in the whole circumference; if the simulation data verification passes, the pneumatic and vibration simulation under the actual working conditions and the optimization system are used to optimize the best installation position of pneumatic and vibration in the whole circumference.

2. The intelligent assembly process for aero-engine or gas turbine blades according to claim 1, characterized in that: In step S4, when the simulation data is verified, the data other than the mass moment in the fitness function used by the optimization system are the aerodynamic and simulation data under the actual working conditions; When the simulation data verification fails, the data other than the mass moment in the fitness function used by the optimization system are the data measured by the blade natural frequency measurement in step S1 and step S2.

3. The intelligent assembly process for aero-engine or gas turbine blades according to claim 1, characterized in that: The measurement system includes a measurement component, which is used to measure the mass moment, natural frequency and geometric shape of the blade; The measurement assembly includes a mass moment module and a natural frequency and geometry module.

4. The intelligent assembly process for aero-engine or gas turbine blades according to claim 3, characterized in that: The mass moment module comprises a horizontal guide rail, a support frame 1 and a support frame 2 movably arranged on the horizontal guide rail, a mass sensor 1 and a mass sensor 2 respectively arranged on the support frame 1 and the support frame 2, and a blade to be measured mounted on the mass sensor 1 and the mass sensor; The two ends of the blade to be measured are respectively provided with a laser sensor 1 and a laser sensor 2; A laser transmitter is fixedly connected to the horizontal guide rail.

5. The intelligent assembly process for aircraft engine or gas turbine blades according to claim 3, characterized in that: The natural frequency and geometry module includes a workbench, a data collector fixedly arranged on the workbench, a mechanical arm, a vise and a blade to be measured fixed on the vise, an end effector fixedly connected to the end of the mechanical arm, a vibration applying device clamped on the end effector, and an exciter fixedly connected to the vibration applying device.

6. The intelligent assembly process for aero-engine or gas turbine blades according to claim 1, characterized in that: The experimental system comprises a temporary wind tunnel device, which is used to make the blade work at a predetermined temperature, speed and angle of attack to obtain the force, moment and temperature field received by the blade.

7. The intelligent assembly process for aircraft engine or gas turbine blades according to claim 1, characterized in that: The simulation system is used to evaluate the aerodynamic performance and vibration performance of the blade under experimental conditions and real conditions, and includes a high-performance computer and simulation software; Specifically, the simulation system is used to calculate the blade indicators including the blade axial force, tangential force, torque, bending moment, maximum temperature and natural frequency under experimental conditions and real conditions; The boundary conditions of the simulation loading under the test conditions are the incoming flow velocity and temperature of the measurement system, and the boundary adjustments of the simulation loading under the actual conditions are the inlet total temperature, inlet total pressure, outlet static pressure and impeller speed in the design parameters of the aircraft engine or gas turbine.

8. The intelligent assembly process for aero-engine or gas turbine blades according to claim 1, characterized in that: The optimization system is used to optimize the best sorting order of blades, and includes a high-performance computer and optimization software; Specifically, the optimization goal of the optimization system through the multi-objective genetic algorithm is global aerodynamic and vibration optimization, where the fitness function is: OBF=C1·M g +C2·F x +C3·F y +C4·M x +C5·M y +C6·T+C7·ω Among them, C i is the weight coefficient, M g is the mass moment unbalance, F x and F y is the unevenness of the axial and tangential forces, M x and M y is the non-uniformity of torque and bending moment, T is the non-uniformity of maximum temperature, and ω is the non-uniformity of natural frequency.

Images