Six-dimensional force multi-parameter decoupling method

Abstract

The invention relates to a six-dimensional force multi-parameter decoupling method, which is based on a six-dimensional force sensor measurement principle, and comprises the following steps of: analyzing the stress condition of each part through Ansys simulation software, and pasting a strain gauge at a stress sensitive position; acquiring strain data of each position through a process of applying or unloading the six-dimensional force; a decoupling model is established based on an SVM algorithm, the strain data is used as input, the borne force is used as output for training, and the borne Fx, Fy, Fz, Mx, My and Mz can be decoupled. According to the method, the six-dimensional force detection efficiency and precision of the rocket cabin can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of six-dimensional force measurement, and in particular to a six-dimensional force multi-parameter decoupling method in force monitoring scenarios such as a robotic arm, a space station, and an engine casing. Background Art

[0002] The cross-sectional load (tension, compression, bending, shearing, and torsion) of the rocket structure during actual flight is an important basis for structural design, and its accuracy is directly related to the reliability and design of the rocket structure. At present, the main way to obtain flight loads is to use simulation calculations, using external forces obtained from engine tests, wind tests, or simulation results from upstream professionals combined with the rocket body dynamics model to calculate the loads of each cross section as the basis for structural design and strength verification.

[0003] However, there is a large gap between the theoretical value and the actual value. During actual flight, there is no high-precision sensor to meet the structural pressure measurement under extreme conditions. Especially in complex climate environments, the real-time measurement accuracy is difficult to guarantee and the adaptability is low. Summary of the invention

[0004] In view of the problems of low precision, slow measurement speed and large volume of the existing rocket compartment six-dimensional force measurement methods, this paper provides a high-precision rocket compartment six-dimensional force multi-parameter decoupling method based on the measurement principle of six-dimensional force sensors. This method can realize the real-time monitoring of the six-dimensional force of the compartment shell, and the algorithm is highly efficient and adaptable.

[0005] The six-dimensional force multi-parameter decoupling algorithm provided in the present disclosure mainly includes the following steps:

[0006] S1, using Ansys simulation software to analyze the six-dimensional forces and moments acting on the rocket compartment, and attaching strain gauges to the force-sensitive locations;

[0007] S2, build the six-dimensional force calibration system of the rocket cabin and place the cabin on the calibration platform;

[0008] S3, loading and unloading the rocket compartment with six-dimensional forces and moments on the calibration platform to obtain strain data at each position;

[0009] S4, repeat S3 operation several times;

[0010] S5, a decoupling model is established based on the SVM algorithm. The obtained strain data is used as input and the force is used as output. The data is input into the SVM model for training to minimize the model output error. This model can be used to decouple the six-dimensional forces and moments acting on the rocket compartment.

[0011] Furthermore, in step S1, the number of strain gauge sticking points is not less than 12.

[0012] Furthermore, in step S3, when loading force and moment, the magnitude and graduation value of the selected force and moment are determined according to the manufacturing materials of the rocket compartment and the simulation results.

[0013] Furthermore, in step S3, for a rocket compartment made of aviation aluminum alloy material with a maximum force range of 800N in the simulation results, the specific method of loading force and moment includes:

[0014] S31, when Fx is loaded, the forces and moments in other directions are controlled to be 0, and Fx is applied in sequence of 0N, 100N, 200N, 300N, 400N, and 500N;

[0015] S32, when Fx is unloaded, the forces and moments in other directions are controlled to be 0, and Fx is applied in sequence of 500N, 400N, 300N, 200N, 100N, and 0N;

[0016] S33, the force and moment loading methods in other directions are the same as S31 and S32.

[0017] Furthermore, in step S5, when training the SVM model, the RBF Gaussian kernel function is selected and the SVM parameters are adjusted to minimize the model output error, wherein the parameters are adjusted according to the maximum force range error of the rocket compartment.

[0018] Compared with the prior art, the beneficial effects of the present invention are: (1) in order to solve the problem of six-dimensional force detection of rocket compartment, a high-efficiency, high-real-time, and high-cost-effective detection algorithm is proposed, thereby improving the efficiency and accuracy of six-dimensional force detection of rocket compartment; (2) the algorithm has good adaptability in harsh environments; (3) it has good versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other objects, features and advantages of the present disclosure will become more apparent through a more detailed description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present disclosure.

[0020] Figure 1 is a flowchart of the algorithm disclosed in the present invention;

[0021] Figure 2 A structural diagram of an exemplary six-dimensional force sensor. DETAILED DESCRIPTION

[0022] The preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0023] The present invention is based on the measurement principle of six-dimensional force sensors and provides a solution for the six-dimensional force detection of rocket compartment shells (the six-dimensional force on the rocket compartment shell is equivalent to the force and torque on the six-dimensional force sensor). After determining the force conditions of each part during force loading through Ansys simulation software, the position of strain gauge pasting is determined; then, through the loading and unloading process of six-dimensional force (Fx, Fy, Fz, Mx, My, Mz), the strain data of each position transmitted by the strain gauge is obtained, and a decoupling model is established based on the SVM algorithm. The strain data is used as input and the force received is used as output for training, so that the received Fx, Fy, Fz, Mx, My, and Mz can be decoupled.

[0024] In an exemplary embodiment, a six-dimensional force sensor is provided. Figure 2 The decoupling process is shown in the attached Figure 1 As shown, it mainly includes the following steps:

[0025] S1. Perform force analysis on the six-dimensional force sensor in Ansys simulation software, paste the strain gauge, and obtain 12 strain paste points through simulation;

[0026] S2. Build a six-dimensional force sensor calibration system and place the six-dimensional force sensor on the calibration platform;

[0027] S3, loading and unloading the six-dimensional force sensor with force / torque (Fx, Fy, Fz, Mx, My, Mz) on the calibration platform to collect strain data;

[0028] S4, repeat S3 operation three times;

[0029] S5. Input the data obtained in S3 and S4 into the SVM model for training, select the RBF Gaussian kernel function, and adjust the SVM parameters (C, lambda, epsilon, kerneloption) to minimize the model output error.

[0030] When the force / torque is loaded in step S3, the magnitude and graduation value of the force / torque selected shall be determined according to the manufacturing material of the six-dimensional force sensor and the simulation result.

[0031] For example, for a six-dimensional force sensor made of aviation aluminum alloy, simulation data shows that the maximum range of the sensor is 800N. For safety reasons, 500N is taken as the maximum range of the actual object, and the graduation value is 100N.

[0032] Then step S3 specifically includes:

[0033] S 301 When Fx is loaded, the forces / torques in other directions are controlled to be 0, and Fx is applied in sequence of 0N, 100N, 200N, 300N, 400N, and 500N, and the outputs of the 12 strain gauges under each loading force are collected;

[0034] S 302 , when Fx is unloaded, the forces / torques in other directions are controlled to be 0, and Fx is applied in sequence of 500N, 400N, 300N, 200N, 100N, and 0N, and the outputs of the 12 strain gauges under each loading force are collected;

[0035] S 303 , force / torque loading methods in other directions and S 301 , S 302 Same, collect data.

[0036] The above technical scheme is only an exemplary embodiment of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the application methods and principles disclosed in the present invention, and it is not limited to the method described in the above specific embodiment of the present invention. Therefore, the method described above is only preferred and does not have a restrictive meaning.

Claims

1. A six-dimensional force multi-parameter decoupling method, comprising the following steps: S1, using Ansys simulation software to analyze the six-dimensional forces and moments acting on the rocket compartment, and attaching strain gauges to the force-sensitive locations; S2, build the six-dimensional force calibration system of the rocket cabin and place the cabin on the calibration platform; S3, loading and unloading the rocket compartment with six-dimensional forces and moments on the calibration platform to obtain strain data at each position; S4, repeat S3 operation several times; S5, a decoupling model is established based on the SVM algorithm. The obtained strain data is used as input and the force is used as output. The data is input into the SVM model for training to minimize the model output error. This model can be used to decouple the six-dimensional forces and moments acting on the rocket compartment.

2. The method according to claim 1, characterized in that In the step S1, the number of strain gauge sticking points is not less than 12.

3. The method according to claim 1 or 2, characterized in that: In step S3, when loading force and moment, the magnitude and division value of the selected force and moment are determined according to the manufacturing material of the rocket compartment and the simulation results.

4. The method according to claim 3, characterized in that In step S3, for a rocket compartment made of aviation aluminum alloy material and having a maximum force range of 800N in the simulation results, the specific method of loading force and moment includes: S31, when Fx is loaded, the forces and moments in other directions are controlled to be 0, and Fx is applied in sequence of 0N, 100N, 200N, 300N, 400N, and 500N; S32, when Fx is unloaded, the forces and moments in other directions are controlled to be 0, and Fx is applied in sequence of 500N, 400N, 300N, 200N, 100N, and 0N; S33, the force and moment loading methods in other directions are the same as S31 and S32.

5. The method according to claim 1, characterized in that: In step S5, for SVM When training the model, the RBF Gaussian kernel function is selected and the SVM parameters are adjusted to minimize the model output error. Among them, when adjusting the parameters, they are set according to the maximum force range error of the rocket compartment.

Images