Method for correcting paddle adjusting impact simulation model of ram air turbine
By adopting Latin hypercube experimental design, response surface methodology, and conjugate gradient method in a ram air turbine system to correct the contact stiffness and damping of the pitch control mechanism, the problems of high manpower and material consumption and low calculation accuracy in the existing technology for experimental load acquisition are solved, and efficient and accurate load prediction is achieved.
Patent Information
- Application Number
- CN202510603781.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-12
- Publication Date
- 2025-10-17
AI Technical Summary
In the existing technology, the interface loads of the ram air turbine system's attachment points are mainly obtained through experiments, which results in a large consumption of manpower and material resources and low calculation accuracy. Simulation calculations have many uncertain parameters and it is difficult to accurately predict the loads under different wind speed conditions.
The Latin hypercube experimental design method was used to design the virtual test conditions. Dynamic simulation was performed using Simcenter 3D. Combining the response surface methodology and the conjugate gradient method, a global approximate model was constructed. The contact stiffness and damping of the pitch mechanism were corrected to obtain the hanging point interface load that matched the experimental value.
The accurate prediction of the interface load of the ram air turbine system under different wind speed conditions is achieved, which reduces the test requirements and improves the accuracy and efficiency of the calculation.
Smart Images
Figure CN120805285A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of dynamic model correction, and particularly relates to a ram air turbine pitch impact simulation model correction method. BACKGROUND
[0002] The hydraulic system is the most important electromechanical system of a modern aircraft, and plays a vital role in realizing the performance of the aircraft and ensuring flight safety. In the normal flight process of the aircraft, the actuator locks the ram air turbine in the cabin, and in an emergency, the actuator receives a release instruction, pushes it out of the cabin, and locks it in the release position. The turbine is exposed to high-speed airflow, the turbine blades extract the ram air pressure energy to generate rotary mechanical energy, and the power is transmitted to the generator through the gear train. Under the control of the generator controller, the generator starts to generate electricity to provide emergency power supply for the aircraft. The ram air turbine system can provide unlimited emergency AC power supply for the aircraft when the main AC power supply of the aircraft fails, and its normal operation is extremely important to ensure the flight safety of the aircraft.
[0003] After the ram air turbine is released out of the cabin door, in the process of starting to generate electricity, the blade pitch angle changes due to the rising of the turbine speed, and when the pitch angle is over-adjusted, the axial aerodynamic force of the blade changes suddenly, and the hanging point load of the ram air turbine system also changes suddenly, thereby generating an impact load.
[0004] At present, the hanging point interface load of the ram air turbine system is mostly obtained through tests, which consumes a lot of manpower and material resources. The simulation calculation has low calculation accuracy due to many uncertain parameters. There is a lack of research on the method of correcting the dynamic simulation model by using model correction to obtain accurate interface load. SUMMARY
[0005] The application solves the technical problem that the hanging point interface load of the ram air turbine system is mostly obtained through tests, which consumes a lot of manpower and material resources, and the simulation calculation has low calculation accuracy due to many uncertain parameters. The purpose of the application is to provide a ram air turbine pitch impact simulation model correction method, which can effectively correct the stiffness and damping of the contact between the cam and the rear spring seat in the pitch mechanism, obtain the ram air turbine dynamic model in the pitch stage with the minimum error between the hanging point interface load and the test value, and replace the test to predict the hanging point interface load of the ram air turbine system under different wind speed conditions.
[0006] Technical scheme:
[0007] A ram air turbine pitch impact simulation model correction method, comprising the following steps:
[0008] 1) Virtual test condition design and calculation: dynamic simulation of the impact process of the ram air turbine system during startup and pitch adjustment, determination of the target stiffness and damping value range based on the stiffness and damping empirical values of the contact between the cam and the rear spring seat in the turbine pitch adjustment mechanism, design of virtual test conditions, and dynamic simulation to obtain the actuator pivot point and motor box pivot point loads under different contact parameters of the pitch adjustment mechanism during the impact process of the ram air turbine system during startup and pitch adjustment;
[0009] 2) Global approximation model construction and optimization: construct a global approximation model, write a program to fit the function relationship between the actuator pivot point impact load and the motor box pivot point load and the contact stiffness and damping of the pitch adjustment mechanism during the startup and pitch adjustment process of the ram air turbine system; based on the constructed global approximation model, minimize the sum of the squared errors of the actuator pivot point load and the motor box pivot point load and the test point load as the correction target, and iteratively correct the objective function to obtain the pitch adjustment mechanism contact parameters and damping that can reflect the true structure.
[0010] Further, in step 1), dynamic simulation of the impact process of the ram air turbine system during startup and pitch adjustment is performed in Simcenter 3D.
[0011] Further, in step 1), the Latin hypercube experimental design method is used to design virtual test conditions.
[0012] Further, step 1) of virtual test condition design and calculation specifically includes the following steps:
[0013] 11) In Simcenter 3D, a dynamic model containing the actuator, motor box, support arm, and turbine head is established for dynamic simulation, the turbine is subjected to aerodynamic force, aerodynamic moment, and blade torque, the turbine blades undergo pitch adjustment when the turbine speed increases to a certain range, the pitch angle increases, and over-pitching leads to impact load at the ram air turbine system point, the initial contact parameters are set between the cam ball bearing and the rear spring seat of the pitch adjustment mechanism, the contact stiffness is 100000 N·mm, and the contact damping is 10 N / (mm / s);
[0014] 12) The contact stiffness and damping between the cam ball bearing and the rear spring seat of the pitch adjustment mechanism are designed using the Latin hypercube sampling method, assuming that the contact stiffness value range is 100000-a N·mm and the contact damping value range is 10-b N / (mm / s), where a and b are real numbers, a and b are determined according to engineering experience, m contact stiffness and n contact damping are selected within the value range, m*n contact parameters are obtained by combination, and the ram air turbine system point interface load results under m*n different contact parameters are calculated through dynamic simulation.
[0015] Further, in step 12), a and b take values in the range of 100000≤a≤200000 and 10≤b≤200.
[0016] Further, in step 2), a global approximation model is constructed by a multiple regression equation using a response surface method, and a program is written in MATLAB to fit the function relationship between the impact load of the actuator pivot hanging point and the motor box pivot hanging point and the contact stiffness and damping of the pitch control mechanism during the starting and pitch adjusting process of the ram air turbine system.
[0017] Further, in step 2), a program is written in MATLAB to iteratively correct the objective function using a conjugate gradient method to obtain the contact parameters and damping of the pitch control mechanism that can reflect the actual structure.
[0018] Further, the global approximation model construction and optimization process in step 2) includes the following steps:
[0019] 21) A global approximation model is obtained by fitting the numerical values of the ram air turbine system hanging point interface load, turbine contact stiffness and damping obtained from simulation using a response surface method through a multiple regression equation in MATLAB;
[0020] The conjugate gradient method is used to optimize the global approximation model established in MATLAB with the minimum error sum of squares of the actuator pivot hanging point load and the motor box pivot hanging point load and the test hanging point load as the correction target.
[0021] 22) The conjugate gradient method is used to optimize the global approximation model established in MATLAB with the minimum error sum of squares of the actuator pivot hanging point load and the motor box pivot hanging point load and the test hanging point load as the correction target.
[0022] Further, in step 21), the quadratic polynomial form of the regression equation is as follows:
[0023]
[0024] where n represents the total number of test factors, which in this case are the contact stiffness and contact damping between the cam and the ball bearing of the pitch control mechanism and the rear spring seat, n is 2, x i represents the i-th test factor, x j represents the j-th test factor, β0, β i , β ii , β ij represents the undetermined coefficients of different terms in the quadratic polynomial, and f i (x) represents the hanging point interface load corresponding to a set of contact stiffness and contact damping values.
[0025] Further, in step 22), the iteration expression is as follows:
[0026] x k+1 =β k+1 x k -g k+1
[0027] Where g k+1 To optimize the target in x k+1 The gradient at x k 、x k+1 They correspond to the independent variables after the kth and k+1th iterations in the modified iterative process, respectively. The superscript T represents the matrix transpose, and Δg is the gradient change.
[0028] Beneficial effects
[0029] The present invention designs virtual test conditions by adopting the Latin hypercube design method, obtains the interface loads of the actuator pivot hanging point and the motor box pivot hanging point under different contact stiffness and damping of the pitch control mechanism through dynamic simulation, constructs a global approximate model of the interface loads of the actuator pivot hanging point and the motor box pivot hanging point and the contact stiffness and damping of the pitch control mechanism through the response surface method, and finally uses the conjugate gradient method to iterate to obtain the hanging point interface load and the corresponding pitch control mechanism contact stiffness k and contact damping C that match the test results. The method can be used to correct the dynamic model of the ram air turbine system, and the corrected model simulation can be used instead of the test to obtain the hanging point interface load results under different wind speeds. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 It is a flow chart of the method of the present invention.
[0031] Figure 2 This is a schematic diagram of the Latin hypercube experimental design.
[0032] Figure 3 Define schematic diagrams for contact stiffness and damping.
[0033] Figure 4 Schematic diagram of the constructed global approximate model. DETAILED DESCRIPTION
[0034] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. The described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.
[0035] The present invention will be further described in detail below with reference to the accompanying drawings and specific examples:
[0036] One embodiment of the present invention is:
[0037] A method for modifying a ram air turbine simulation model is provided, comprising the following steps:
[0038] Step 1: dynamic modeling and simulation analysis are carried out on the ram air turbine system during the process of adjusting the propeller, the contact stiffness and damping of the cam ball bearing and the rear spring seat are determined based on engineering experience, and the value range is determined;
[0039] Preferably, the step 1 is modeled and simulated in Simcenter 3D.
[0040] Step 2: Latin hypercube design method is used to design virtual test conditions, and dynamic simulation is used to obtain the interface load of the actuator pivot hanging point and the motor box pivot hanging point under different contact stiffness and damping conditions;
[0041] Step 3: using the value range of the contact stiffness and damping obtained in step 1, and the actuator pivot hanging point load and motor box pivot hanging point load obtained in step 2, a global approximation model is constructed by a multiple regression equation using the response surface method;
[0042] Step 4: taking the minimum sum of squared errors of the actuator pivot hanging point load and the motor box pivot hanging point load as the correction target, the global approximation model in step 3 is iteratively optimized using the conjugate gradient method to obtain the contact stiffness and damping that can reflect the actual situation of the structure.
[0043] In this embodiment, the step 1 specifically includes: the contact stiffness value range is 100000-a N·mm, and the contact damping value range is 10-b N / (mm / s), wherein a and b are real numbers, and a and b can be determined according to engineering experience, generally, 100000≤a≤200000, 10≤b≤200;
[0044] In this embodiment, in step 2, a dynamic model including the actuator, the motor box, the support arm and the turbine head is established for dynamic simulation, the turbine is subjected to aerodynamic force, aerodynamic moment and blade torque, the turbine blades are adjusted when the turbine speed rises to a certain range, the pitch angle becomes larger, and the over-adjustment phenomenon causes the ram air turbine system hanging point impact load; a1 contact stiffness and b1 contact damping are selected within the value range of the contact stiffness and the contact damping obtained in step 1, a1*b1 contact parameters are combined, and the hanging point interface load results under a1*b1 different contact parameters are obtained through dynamic simulation calculation.
[0045] In this embodiment, in step 3, the process of constructing the global approximation model specifically includes:
[0046] The second-order response surface method is used to obtain the global approximate model of the actuator hanging point load and the motor box hanging point load and the values of the contact stiffness and the contact damping through the simulation of the multiple regression equation. The quadratic polynomial form of the regression equation is as follows:
[0047]
[0048] In the formula, n represents the total number of test factors, n is 2 in this case, xi represents the i-th test factor, xj represents the j-th test factor, b represents the undetermined coefficient, β0, β1, β2 and β3 represent the undetermined coefficients of different terms in the quadratic polynomial, and f(x) represents the hanging point interface load corresponding to the selected contact stiffness and contact damping. i j i ii ij i In the formula, n represents the total number of test factors, n is 2 in this case, xi represents the i-th test factor, xj represents the j-th test factor, b represents the undetermined coefficient, β0, β1, β2 and β3 represent the undetermined coefficients of different terms in the quadratic polynomial, and f(x) represents the hanging point interface load corresponding to the selected contact stiffness and contact damping.
[0049] After the undetermined coefficients are obtained, the approximate quadratic polynomial function relationship between the hanging point interface load and the contact stiffness and the contact damping of the pitch adjusting mechanism is obtained.
[0050] In this embodiment, in the step 3, the process of constructing the global approximate model specifically includes:
[0051]
[0052] The third-order response surface method is used to obtain the global approximate model of the actuator hanging point load and the motor box hanging point load and the values of the contact stiffness and the contact damping through the simulation of the multiple regression equation. The cubic polynomial form of the regression equation is as follows:
[0053]
[0054] In the formula, n represents the total number of test factors, n is 2 in this case, xi represents the i-th test factor, xj represents the j-th test factor, b represents the undetermined coefficient, β0, β1, β2 and β3 represent the undetermined coefficients of different terms in the cubic polynomial, and f(x) represents the hanging point interface load corresponding to the selected contact stiffness and contact damping. i i ii ij i In the formula, n represents the total number of test factors, n is 2 in this case, xi represents the i-th test factor, xj represents the j-th test factor, b represents the undetermined coefficient, β0, β1, β2 and β3 represent the undetermined coefficients of different terms in the cubic polynomial, and f(x) represents the hanging point interface load corresponding to the selected contact stiffness and contact damping.
[0055] After the undetermined coefficients are obtained, the approximate cubic polynomial function relationship between the hanging point interface load and the contact stiffness and the contact damping of the pitch adjusting mechanism is obtained.
[0056] In this embodiment, in the step 4, the specific process of using the conjugate gradient method to iteratively optimize the global approximate model of the step 3 includes:
[0057] With the error square sum of actuator pivot hanging point load and motor box pivot hanging point load and test hanging point load as the correction target, a program is written in MATLAB to optimize the established global approximation model by using the conjugate gradient method, and the iteration process is as follows:
[0058] When the search direction is perpendicular to the gradient change direction, then the conjugate. For the conjugate gradient method, the first search direction p0 is arbitrary, and the direction of the steepest descent method is usually selected to start the search:
[0059] p0=-g0 (3)
[0060] g0 is the gradient of the initial point, and an orthogonal vector p to {△g0,△g1,…,△g k-1} is constructed every time iteration. k The iteration form can be simplified as:
[0061] p k =-g k +β k p k-1 (4)
[0062] Wherein The superscript T represents the matrix transpose, and △g is the gradient change.
[0063] After the correction, the hanging point interface load matched with the test result and the corresponding pitch mechanism contact stiffness k and contact damping C are obtained.
[0064] The second embodiment of the application is:
[0065] Referring to Figure 1 , a ram air turbine pitch impact simulation model correction method provided by the application includes the following steps:
[0066] 1) Virtual test working condition design and calculation: in Simcenter 3D, the dynamics finite element simulation of the hanging point interface load in the start-up pitch impact process of the ram air turbine system is carried out, the value range of the target stiffness and damping is determined based on the stiffness and damping empirical value of the contact between the cam and the rear spring seat in the turbine pitch mechanism, the Latin hypercube test design method is used to design the virtual test working condition, and the actuator pivot hanging point and motor box pivot hanging point load under different contact parameters of the turbine pitch mechanism in the start-up pitch impact process of the ram air turbine system are obtained by dynamics simulation.
[0067] 2) Global approximation model construction and optimization: Response surface method is used to construct global approximation model through multivariate regression equation. The function relationship between the actuator pivot hanging point impact load and motor box pivot hanging point load and the contact stiffness and damping of the pitch control mechanism during the starting pitch control process of the ram air turbine system is fitted in MATLAB. On the basis of the constructed global approximation model, the minimum error sum of squares of the actuator pivot hanging point load and the motor box pivot hanging point load and the test hanging point load is taken as the correction target. The conjugate gradient method is used to correct the objective function in MATLAB, and the contact parameters and damping of the pitch control mechanism that can reflect the real structure are obtained.
[0068] The optimization object of the embodiment is a certain type of ram air turbine system. As shown in Figure 3 The contact stiffness of the cam ball bearing and the rear spring seat of the pitch control mechanism before correction is 100000 N·mm, and the contact damping is 10 N / (mm / s). The dynamic model of the actuator, motor box, support arm and turbine head is established in Simcenter 3D to perform dynamic simulation. The turbine is subjected to aerodynamic force, aerodynamic moment and blade torque. When the turbine speed rises to a certain range, the turbine blade is pitch controlled, the pitch angle becomes larger, and the over-pitch phenomenon leads to the hanging point impact load of the ram air turbine system. The initial contact parameters are set between the contact of the cam ball bearing and the rear spring seat of the pitch control mechanism. The Latin hypercube design method is used to design the contact stiffness and contact damping. The inner diameter of the plunger is in the range of 50000-200000 N / mm, and the contact damping is in the range of 10-200 N / (mm / s). The dynamic simulation of all test design conditions is performed in Simcenter 3D, and the hanging point interface load under all conditions is calculated. The actuator pivot hanging point and motor box pivot hanging point interface load under each condition is input into MATLAB together with the corresponding contact stiffness and damping. The global approximation model of the second-order multivariate regression equation is constructed by response surface method. The order of the multivariate regression equation can be defined by oneself, and the function relationship between the hanging point interface load and the contact stiffness and contact damping in the parameter value range is obtained as shown in Figure 4The minimum error square sum of the actuator pivot point load and the motor box pivot point load in the specified value range with the test pivot point load is the correction target, and the minimum point of the error square sum of the actuator pivot point load and the motor box pivot point load in the specified value range with the test pivot point load is found by the conjugate gradient method. The minimum point corresponds to the inner diameter of the plunger and the neck diameter. The simulation result after correction is consistent with the prediction result of the global approximation model, and the correction is completed. The specific application of the patent is many, and the above is only the preferred embodiment of the patent, not the limitation of the implementation and protection range of the patent. For those skilled in the art, under the premise of the patent principle, the schemes obtained by equivalent replacement and obvious changes should be included in the protection range of the patent.
Claims
1. A method for correcting a ram air turbine pitch adjustment impact simulation model, characterized in that: The following steps are included: 1) Design and calculation of virtual test conditions: Dynamic simulation of the interface loads at the attachment points during the start-up and pitch adjustment impact of the ram air turbine system is conducted. The target stiffness and damping ranges are determined based on the empirical values of the contact stiffness and damping between the cam and the rear spring seat in the turbine pitch adjustment mechanism. Virtual test conditions are designed, and dynamic simulation is used to obtain the loads at the actuator pivot attachment point and the motor box pivot attachment point under different contact parameters of the pitch adjustment mechanism during the start-up and pitch adjustment impact of the ram air turbine system. 2) Construction and optimization of the global approximate model: A global approximate model is constructed, and a program is written to fit the functional relationship between the impact load of the actuator pivot hanging point and the motor box pivot hanging point load and the contact stiffness and damping of the pitch control mechanism during the start-up and pitch control process of the ram air turbine system; based on the constructed global approximate model, the objective function is iteratively corrected with the minimization of the sum of squares of the errors between the actuator pivot hanging point load and the motor box pivot hanging point load and the test hanging point load as the correction target, so as to obtain the pitch control mechanism contact parameters and damping that can reflect the actual structural conditions.
2. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 1, characterized in that: In step 1), a dynamic simulation of the interface loads at the attachment point during the start-up and pitch adjustment impact process of the ram air turbine system is performed in Simcenter 3D.
3. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 1, characterized in that: In step 1), the Latin hypercube experimental design method is used to design the virtual test conditions.
4. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 3, characterized in that: Step 1) The virtual test condition design and calculation process specifically includes the following steps: 11) A dynamic model including the actuator, motor housing, support arm, and turbine head was established in Simcenter 3D for dynamic simulation. Aerodynamic forces, aerodynamic torques, and blade torques were applied to the turbine. When the turbine speed increased to a certain range, the turbine blades were adjusted, and the pitch angle increased. This overshoot phenomenon resulted in an impact load on the ram air turbine system attachment point. Initial contact parameters were set for the contact between the cam ball bearing and the rear spring seat of the pitch adjustment mechanism. The contact stiffness was 100,000 N·mm, and the contact damping was 10 N / (mm / s). 12) The Latin hypercube sampling method is used to design the contact stiffness and damping between the cam ball bearing and the rear spring seat of the pitch control mechanism. Assuming that the contact stiffness ranges from 100,000 to a N·mm, and the contact damping ranges from 10 to b N / (mm / s), where a and b are real numbers, the ranges of a and b are determined based on engineering experience. Within the range, m contact stiffnesses and n contact dampings are selected to obtain m*n contact parameters. Dynamic simulation calculations are performed to obtain the interface load results of the ram air turbine system under m*n different contact parameters.
5. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 4, characterized in that: In step 12), the value ranges of a and b are: 100000≤a≤200000, 10≤b≤200.
6. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 4, characterized in that: In step 2), the response surface method is used to construct a global approximate model through the multivariate regression equation, and a program is written in MATLAB to fit the functional relationship between the actuator pivot hanging point impact load and the motor box pivot hanging point load and the contact stiffness and damping of the pitch control mechanism during the starting and pitch control process of the ram air turbine system.
7. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 6, characterized in that: In step 2), a program is written in MATLAB to iteratively modify the objective function using the conjugate gradient method to obtain the contact parameters and damping of the pitch mechanism that can reflect the actual structural conditions.
8. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 6, characterized in that: Step 2) The global approximate model construction and optimization process specifically includes the following steps: 21) Write a program in MATLAB to use the response surface methodology to fit the numerical values of the ram air turbine system attachment point interface load and the turbine contact stiffness and damping obtained from the simulation through multiple regression equations to obtain a global approximate model; With the goal of minimizing the sum of squares of the errors between the actuator pivot point load and the motor box pivot point load and the test point load, a program was written in MATLAB to optimize the established global approximate model using the conjugate gradient method. 22) Taking the minimization of the sum of squares of the errors between the actuator pivot point load and the motor box pivot point load and the test point load as the correction target, a program was written in MATLAB to use the conjugate gradient method to optimize the established global approximate model.
9. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 8, characterized in that: In step 21), the quadratic polynomial form of the regression equation is expressed as follows: Where n represents the total number of test factors. Here, there are two test factors: contact stiffness and contact damping between the cam ball bearing and the rear spring seat of the pitch control mechanism. n is 2, x i represents the i-th experimental factor, x j represents the jth experimental factor, β0, β i , β ii , β ij represents the unknown coefficients of different terms in the quadratic polynomial, f i (x) represents the interface load at the hanging point corresponding to a set of contact stiffness and contact damping values.
10. The method for correcting a ram air turbine pitch adjustment impact simulation model according to claim 9, characterized in that: In step 22), the iterative expression is as follows: x k+1 =b k+1 x k -g k+1 Where g k+1 To optimize the target in x k+1 The gradient at x k 、x k+1 They correspond to the independent variables after the kth and k+1th iterations in the modified iterative process, respectively. The superscript T represents the matrix transpose, and Δg is the gradient change.