A device and method for measuring the torque output characteristics of a control moment gyroscope

Abstract

A kind of control moment gyro torque output characteristic measuring device and test method, belong to the field of aircraft control and ground simulation technique.To solve the problem of realizing high-precision torque measurement.The present application is composed of super static suspension system, precision torque sensitive system, high-precision torque calibration system and data monitoring and control system;Super static suspension system includes air floating ball bearing, instrument installation platform and support unloading device, air floating ball bearing is arranged in air floating ball bowl, air floating ball bearing is fastened and connected with instrument installation platform by bolt, measured equipment is placed on air floating ball bearing, pressure sensor is installed on support unloading device;Precision torque sensitive system is composed of horizontal torque measurement system and vertical torque measurement system;High-precision torque calibration system is composed of inertial navigation unit and single-shaft mass adjustment robot, data monitoring and control system is connected with super static suspension system, precision torque sensitive system, high-precision torque calibration system through wireless communication device.

Description

Technical Field

[0001] This invention belongs to the field of aircraft control and ground simulation technology, specifically relating to a control torque gyroscope torque output characteristic measurement device and testing method. Background Technology

[0002] Control moment gyroscopes are important attitude actuators in spacecraft and are widely used in the aerospace field. To address the problem of fuzzy output characteristics, it is necessary to study a high-precision method for measuring the output torque of control moment gyroscopes for testing their output characteristics.

[0003] The invention patent "Torque Testing Air-Floating Turntable for Controlled Torque Gyroscope" (CN200610011829.7) measures the output torque of a controlled torque gyroscope using a single-axis air-floating turntable. It operates in two modes: free rotation and torque rebalancing. In free rotation, an encoder calculates the turntable spindle speed to obtain the output torque of the torque gyroscope along the air-floating turntable's rotation axis. In torque rebalancing, a torque loop is constructed based on a high-precision position closed loop, enabling high-precision measurement of the torque along the air-floating turntable's rotation axis. However, it can only measure the output torque along the air-floating turntable's rotation axis and cannot obtain the complete output torque characteristics of the controlled torque gyroscope. Furthermore, the measuring device is not calibrated before measurement; the controlled torque calculated by the program is a nominal value, not the true output torque. Additionally, the measurement system may contain unmodeled characteristics, all of which lead to significant differences between the accuracy and the actual output torque.

[0004] The invention patent "Measuring Device and Method for High-Precision Controlled Torque Gyroscope Torque Output" (CN201610191170.1) uses a dual-axis air-bearing platform as the measuring platform. During torque measurement, the device uses force sensor data collected by three torque measuring mechanisms, combined with the distance between the centers of the air-bearing ball bearings of the force sensors, to calculate the output torque and torque fluctuation of the controlled torque gyroscope. However, it does not consider the influence of errors such as the tilt of the air-bearing platform and the measurement of the lever arm during the measurement process, resulting in unreliable output torque accuracy. Furthermore, it lacks calibration of the measuring device, making the accuracy of the measurement results questionable. Additionally, due to the sensor layout, this device can only measure horizontal output torque and cannot measure vertical torque, making it unsuitable for testing the output characteristics of all controlled torque gyroscopes. Summary of the Invention

[0005] The problem to be solved by this invention is to achieve high-precision torque measurement, and a device and test method for measuring the torque output characteristics of a control torque gyroscope are proposed.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A control torque gyroscope torque output characteristic measurement device includes an ultra-quiet suspension system, a precision torque sensing system, a high-precision torque calibration system, and a data monitoring and control system;

[0008] The ultra-quiet suspension system includes an air-floating ball bearing, an instrument mounting platform, and a support and unloading device. The air-floating ball bearing is set in an air-floating ball cup and is fastened to the instrument mounting platform. The instrument mounting platform is placed on the instrument mounting platform and a pressure sensor is installed on the support and unloading device.

[0009] The precision torque sensing system consists of a horizontal torque measurement system and a vertical torque measurement system;

[0010] Four sets of horizontal torque measurement systems are evenly arranged at 90° intervals on the instrument mounting platform to convert horizontal torque measurements into force measurements. The vertical torque measurement system includes a high-precision force sensor set, an air bearing, and an auxiliary extension mechanism. The auxiliary extension mechanism is rigidly connected to the instrument mounting platform. The high-precision force sensor set consists of two force sensors, which are installed on the outer side of the two air bearing surfaces formed between the air bearing and the auxiliary extension mechanism. This is used to realize z-axis torque measurement and decouple z-axis torque from xy-axis torque measurement.

[0011] The high-precision torque calibration system is placed on the instrument mounting platform and includes an inertial navigation unit and two single-axis mass adjustment robots, orthogonally distributed between the device under test and the horizontal torque measurement system. These robots are used to output high-precision standard torque while leveling the instrument mounting platform. The inertial navigation unit is placed between the device under test and the horizontal torque measurement system, facing the single-axis mass adjustment robots, and is used to monitor the attitude information of the ultra-static suspension system in real time.

[0012] The data monitoring and control system is connected to the ultra-quiet suspension system, the precision torque sensing system, and the high-precision torque calibration system via a wireless communication device.

[0013] Furthermore, the horizontal torque measurement system includes a micro-touch unit, a high-precision force sensor, a high-precision electric linear actuator, an air cushion, a pneumatic actuator, a pneumatic preload unit, a composite beam, an air-bearing guide rail, and an air-bearing slider;

[0014] The composite beam is fixedly mounted on the air-bearing guide rail. A high-precision electric linear actuator is rigidly connected to the air-bearing slider through the composite beam. A pneumatic actuator passes through the airtight cavity of the composite beam and acts on the upper surface of the air-bearing slider. A pneumatic pre-tightening unit is integrated inside the pneumatic actuator and is independently controlled by a pneumatic pipeline. The output end of the pneumatic pre-tightening unit is coupled to the air-bearing slider, which is suspended and installed in the raceway of the air-bearing guide rail. The lower surface of the air-bearing slider is fixedly connected to a high-precision force sensor. A micro-touch unit is connected to the measuring end of the high-precision force sensor. The high-precision electric linear actuator pre-presses the micro-touch unit onto the air cushion placed on the instrument mounting platform through the high-precision force sensor. With the locking action of the pneumatic pre-tightening unit, the high-precision force sensor and the instrument mounting platform are pre-pressed together.

[0015] Furthermore, there are three sets of support unloading devices, distributed at 120°, used to support and roughly balance the instrument installation platform, and two single-axis mass adjustment robots, namely Robot A and Robot B.

[0016] Furthermore, the data monitoring and control system is responsible for receiving the output information of sensitive and control elements in each system, sending quality adjustment and loading commands to the high-precision torque compensation system and the precision torque sensing system, and simultaneously calculating the compensation information and measured torque.

[0017] Furthermore, the measured force results of the four sets of horizontal torque measurement systems are F1, F2, F3, and F4, respectively, with theoretical lever arms between each force sensor and the center of rotation being L1, L2, L3, and L4. The measured force results of the vertical torque measurement system are F5 and F6, with theoretical lever arms between each force sensor and the center of rotation being L5 and L6, respectively. Based on the equilibrium method for torque measurement, the formula for the ideal output torque is:

[0018]

[0019] in, , and The measured torque results are shown in the x-direction, y-direction, and z-direction, respectively.

[0020] A method for measuring the torque output characteristics of a control torque gyroscope, based on the aforementioned control torque gyroscope torque output characteristic measuring device, includes the following steps:

[0021] S1. Confirm whether the air-bearing ball bearing is ventilated and whether the power supply of a control torque gyroscope torque output characteristic measuring device is turned on. The air-bearing ball bearing enters the working state after being ventilated and the power supply is turned on.

[0022] S2. Manually add load counterweights to the instrument mounting platform, read the pressure sensor values ​​on the support unloading device, and when the pressure sensor values ​​are basically equal, complete the coarse balancing of a control torque gyroscope torque output characteristic measurement device.

[0023] S3. Remove the support unloading device, and then use the inertial navigation unit as feedback to control the movement of the programmable mass adjustment component in the single-axis mass adjustment robot. When the inertial navigation unit displays that the attitude of the instrument mounting platform has reached the predetermined level requirement, the fine balancing of a control torque gyroscope torque output characteristic measurement device is completed.

[0024] S4. Based on the high-precision standard torque self-calibration method, complete the self-calibration of the high-precision torque calibration system to obtain the high-precision standard torque;

[0025] S5. Based on the self-calibration method of the measurement system of high-precision standard torque, a precision torque sensing system is loaded to complete the self-calibration of a control torque gyroscope torque output characteristic measurement device, and a predictive neural network model and a z-axis error compensation relationship model are obtained;

[0026] S6. Adjust a control torque gyroscope torque output characteristic measurement device to the self-calibration initial state, turn on the device under test, perform torque measurement, calculate the output torque of the device under test based on the predictive neural network model, the z-axis error compensation relationship model, and the output torque formula, save the experimental data, and complete the experiment.

[0027] Furthermore, the specific implementation method of the high-precision standard torque self-calibration method in step S4 includes the following steps:

[0028] S4.1. Under the condition of fine balancing based on a control torque gyroscope torque output characteristic measurement device, a position code is set on the instrument mounting platform with a mass load of known weight. , ;

[0029] S4.2. Without loading the precision torque-sensitive system, the data monitoring and control system reads the data from the inertial navigation unit and controls the movement of the programmable mass adjustment components in robot A and robot B to achieve horizontal leveling of the ultra-quiet suspension system;

[0030] S4.3. Add a known mass load to the instrument mounting platform, installing the mass load at position P1. Level the ultrastatic suspension system again, and record the state of the measurement system at this point as the initial zero-position state of self-calibration. Record the mass load mass at this point. And the positions of the programmed quality adjustment components in Robot A and Robot B;

[0031] S4.4. Sequentially encode according to position. The mass load was moved in ascending order. After each movement, the horizontal leveling of the ultra-quiet levitation system was repeated, and the relative displacement of the mass load was recorded for each movement. And the relative displacement after each horizontal leveling movement of the programmable quality adjustment components in Robot A and Robot B. and , where i represents the number of times the mass load is moved, and N sets of data are obtained;

[0032] S4.5. Define the unknown quantities as including the total mass of the programmable mass adjustment components in robots A and B, respectively. and The included angle of the installation error between robot A and the x-axis The installation error angle between the robot's B and Y axes is divided. The high-precision standard torque self-calibration calculation method is constructed as follows:

[0033] S4.5.1. Construct a least-squares system, based on N sets of measurements, and build the following system of linear equations:

[0034]

[0035] Define the variable to be solved as ;

[0036] S4.5.2. Solving for intermediate variables based on the improved total least squares method: The coefficient matrix is ​​denoted as... The observation vector is denoted as Constructing an augmented matrix And perform SVD singular value decomposition as follows:

[0037]

[0038] Wherein, the coefficient matrix Observation vector , It is a singular value matrix. It is a left singular value vector matrix;

[0039] make The last column is And it was revised to ,in If is the regularization parameter, then the above variables to be solved can be written as:

[0040] ;

[0041] S4.5.3. Calculate the mass of the programmable quality adjustment component, and obtain:

[0042] ;

[0043] S4.5.4. Separate the angle parameters and solve for the final error installation angle based on N sets of data, to obtain:

[0044]

[0045] ;

[0046] S4.5.5. After the above self-calibration, the high-precision standard torque self-calibration output formula is as follows:

[0047]

[0048] in, For high-precision standard torque on the x-axis, For high-precision standard torque on the y-axis, and These represent the relative positions of the programmable mass adjustment components of the two robots under program control.

[0049] Furthermore, the specific implementation method of the high-precision standard torque measurement system self-calibration method in step S5 includes the following steps:

[0050] S5.1. A neural network prediction model is constructed for horizontal torque self-calibration. The neural network prediction model adopts an input layer-two hidden layers-output layer structure. The input layer consists of the output information from the force sensor and the inertial navigation unit, including a 7-dimensional input vector. and , The output is the three-axis attitude data of the inertial navigation unit. The output layer is the horizontal torque of the x-axis and y-axis. The two hidden layers consist of 10 neurons and 5 neurons respectively.

[0051] S5.2. The vertical moment self-correction method is constructed by using polynomial fitting to obtain the z-axis error compensation relationship model; based on and The z-axis disturbance torque is calculated using the following formula:

[0052]

[0053] in, The z-axis disturbance torque. and They are respectively and The corresponding nominal lever arm;

[0054] Establish the mapping relationship between attitude angle and disturbance torque. The expression is:

[0055]

[0056] Where k is the index of the data sample, and N is the total number of data samples. For the x-axis attitude data, y-axis attitude data, and z-axis attitude data of the k-th sample, The disturbance torque around the z-axis corresponding to the k-th sample;

[0057] A multinomial regression was performed on the mapping relationship between attitude angle and disturbance torque to generate the following z-axis error compensation relationship model:

[0058]

[0059] in, For z-axis compensation torque, For polynomial basis functions, , Here, is the fitting coefficient, j is the term index of the polynomial basis function, and m is the highest order of the polynomial regression. The construction of this z-axis error compensation relationship model realizes the self-correction of the vertical moment.

[0060] Furthermore, in step S5, the method for loading a precision torque sensing system to complete the self-calibration of a control torque gyroscope torque output characteristic measurement device is as follows:

[0061] Before formally measuring the device under test, the ultra-quiet suspension system is first ensured to be in a leveled state. Then, a precision torque sensing system is loaded onto the instrument mounting platform, and the output of each force sensor is set to 50% of the full scale to obtain the self-calibration initial state of a control torque gyroscope torque output characteristic measurement device.

[0062] Then, the data monitoring and control system sends the programmed torque sequence to the single-axis mass adjustment robot in the high-precision torque calibration system. Based on the self-calibration results, the single-axis mass adjustment robot outputs high-precision programmed torque by changing the relative position of the programmed mass adjustment components.

[0063] Then, the output information of the high-precision programmable torque, force sensor and inertial navigation unit is stored in real time and synchronized. The self-calibration method of the high-precision standard torque measurement system is used to perform self-calibration operations on the horizontal torque and vertical torque respectively.

[0064] After self-calibration is completed, a measurement system and model are used to achieve high-precision torque measurement.

[0065] The beneficial effects of this invention are:

[0066] The control torque gyroscope torque output characteristic measurement device of the present invention does not require the introduction of additional high-precision torque output equipment. It can use the measurement system itself to obtain high-precision output torque, and the cost of obtaining high-precision output torque is low and the calibration method is simple. It does not require additional output torque correction equipment and can realize self-calibration of the measurement system based on high-precision output torque.

[0067] The control torque gyroscope torque output characteristic measurement device described in this invention uses data-driven measurement system after calibration and correction. It eliminates the need for complex torque measurement model establishment, avoids the handling of unknown modeling characteristics and complex nonlinear factors, reduces algorithm complexity, and improves the measurement accuracy of output torque.

[0068] The control torque gyroscope torque output characteristic measuring device described in this invention can realize three-axis torque measurement and can realize high-precision measurement of the output torque of control torque gyroscopes of all structures. Attached Figure Description

[0069] Figure 1 This is a schematic diagram of the structure of a control torque gyroscope torque output characteristic measuring device according to the present invention;

[0070] Figure 2 This is a top view of a control torque gyroscope torque output characteristic measuring device according to the present invention;

[0071] Figure 3 This is a schematic diagram of the vertical torque measurement system of the present invention;

[0072] Figure 4 This is a schematic diagram of the horizontal torque measurement system of the present invention;

[0073] Figure 5 This is a schematic diagram of the single-axis mass adjustment machine structure of the present invention;

[0074] Figure 6 This is a diagram showing the installation location of the mass load of the present invention;

[0075] Figure 7 This is a flowchart of the self-calibration method for the high-precision standard torque measurement system of the present invention;

[0076] Figure 8 This is a flowchart of a method for measuring the torque output characteristics of a control torque gyroscope according to the present invention. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described specific embodiments are merely a part of the embodiments of the invention, and not all of them. The components of the specific embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations, and the invention may also have other embodiments.

[0078] Therefore, the following detailed description of specific embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected specific embodiments of the invention. All other specific embodiments obtained by those skilled in the art based on these specific embodiments without inventive effort are within the scope of protection of this invention.

[0079] To further understand the invention's content, features, and effects, the following specific embodiments are provided, along with accompanying drawings. Figure 1 -Appendix Figure 8 Detailed explanation is as follows:

[0080] Example 1:

[0081] A control torque gyroscope torque output characteristic measurement device includes an ultra-quiet suspension system, a precision torque sensing system, a high-precision torque calibration system, and a data monitoring and control system;

[0082] The ultra-quiet suspension system includes an air-floating ball bearing 1, an instrument mounting platform 2, and a support and unloading device 17. The air-floating ball bearing 1 is set in the air-floating ball bowl 3 and is fastened to the instrument mounting platform 2. The instrument mounting platform 2 is on which the device under test 16 is placed. The support and unloading device 17 is equipped with a pressure sensor 18.

[0083] The precision torque sensing system consists of a horizontal torque measurement system 20 and a vertical torque measurement system 21.

[0084] Four sets of horizontal torque measurement systems 20 are evenly arranged at 90° on the instrument mounting platform 2 to convert horizontal torque measurement into force measurement; the vertical torque measurement system 21 includes a high-precision force sensor group 22, an air bearing 23, and an auxiliary extension mechanism 25, with the auxiliary extension mechanism 25 rigidly connected to the instrument mounting platform 2; the high-precision force sensor group 22 consists of two force sensors, which are installed on the outside of the two air surfaces 24 formed between the air bearing 23 and the auxiliary extension mechanism 25, to realize z-axis torque measurement and decouple z-axis torque from xy-axis torque measurement;

[0085] The high-precision torque calibration system is placed on the instrument mounting platform 2, including an inertial navigation unit 13 and a single-axis mass adjustment robot 14. There are two single-axis mass adjustment robots 14, which are orthogonally distributed between the device under test 16 and the horizontal torque measurement system 20, that is, on the negative side of the torque axis x-axis. The attitude output axis of the inertial navigation unit (13) coincides with the torque measurement axis of the measurement system, which is used to output a high-precision standard torque while the instrument mounting platform 2 is leveling. The inertial navigation unit 13 is placed between the device under test 16 and the horizontal torque measurement system 20 facing the single-axis mass adjustment robot 14, which is used to monitor the attitude information of the ultra-static suspension system in real time.

[0086] The data monitoring and control system 15 is connected to the ultra-quiet suspension system, the precision torque sensing system, and the high-precision torque calibration system via a wireless communication device.

[0087] Furthermore, the horizontal torque measurement system 20 includes a micro-touch unit 4, a high-precision force sensor 5, a high-precision electric linear actuator 6, an air cushion 7, a pneumatic actuator 8, a pneumatic pre-tensioning unit 9, a composite beam 10, an air-bearing guide rail 11, and an air-bearing slider 12.

[0088] The composite beam 10 is fixedly mounted on the air-bearing guide rail 11. The high-precision electric linear actuator 6 is rigidly connected to the air-bearing slider 12 through the composite beam 10. The pneumatic actuator 8 passes through the airtight cavity of the composite beam 10 and acts on the upper surface of the air-bearing slider 12. The pneumatic pre-tightening unit 9 is integrated inside the pneumatic actuator 8 and is independently controlled by the air pressure pipeline. The output end of the pneumatic pre-tightening unit 9 is coupled to the air-bearing slider 12. The air-bearing slider 12 is suspended and installed in the raceway of the air-bearing guide rail 11. The lower surface of the air-bearing slider 12 is fixedly connected to the high-precision force sensor 5. The micro-touch unit 4 is connected to the measuring end of the high-precision force sensor 5. The high-precision electric linear actuator 6 pre-presses the micro-touch unit 4 onto the air cushion 7 placed on the instrument mounting platform 2 through the high-precision force sensor 5. With the locking action of the pneumatic pre-tightening unit 9, the high-precision force sensor 5 and the instrument mounting platform 2 are pre-pressed together.

[0089] Furthermore, there are three sets of support unloading devices 17, distributed at 120°, used to support and roughly balance the instrument mounting platform 2, and two single-axis mass adjustment robots 14, namely robot A and robot B.

[0090] Furthermore, the data monitoring and control system 15 is responsible for receiving the output information of the sensitive elements and control elements in each system, sending quality adjustment and loading commands to the high-precision torque compensation system and the precision torque sensing system, and calculating the compensation information and measured torque.

[0091] Furthermore, the measured force results of the four sets of horizontal torque measurement systems 20 are F1, F2, F3, and F4, respectively, with the theoretical lever arms between each force sensor and the center of rotation being L1, L2, L3, and L4. The measured force results of the vertical torque measurement system 21 are F5 and F6, respectively, with the theoretical lever arms between each force sensor and the center of rotation being L5 and L6. Based on the equilibrium method for measuring torque, the formula for the ideal output torque is:

[0092]

[0093] in, , and The measured torque results are shown in the x-direction, y-direction, and z-direction, respectively.

[0094] Furthermore, a high-pressure air film is formed between the air-float ball cup and the air-float ball bearing. When no platform constraints are applied, the air-float ball bearing can create a micro-friction, micro-interference environment, enabling the instrument mounting platform to rotate freely along its three axes. The instrument mounting platform is made of high-strength structural steel and features standard mounting holes and positioning holes machined by high-precision CNC machine tools, facilitating the installation of the device under test 16 and system equipment. The support and unloading device can support the instrument mounting platform to prevent it from tipping over when not in operation, and it also has a pressure sensor 18 for guiding rough balancing.

[0095] Furthermore, in the vertical torque measurement system, a high-precision force sensor assembly is mounted on an air bearing, with the air bearing surface serving as the force transmission structure for both parts. This design not only accommodates the minute movements of the air bearing platform's xy-axis attitude but also ensures decoupling between the z-axis torque measurement and the xy-axis torque measurement. The force sensors are installed in pairs, with two force sensors forming a sensor group, denoted as force sensor A and force sensor B. To reduce force transmission errors caused by poor contact, the force sensors are also pre-loaded, ensuring close contact between the force sensors and the auxiliary extension mechanism via the air bearing surface. When the air bearing platform exhibits a z-axis rotational tendency due to the z-axis torque output by the Control Moment Gyro (CMG), both z-axis force sensors simultaneously detect the applied force. By combining this with the stress arm information, the z-axis torque can be differentially calculated.

[0096] Furthermore, the high-precision torque calibration system mounted on the x-axis is called Robot A, and the one mounted on the y-axis is called Robot B. It can perform leveling work on the instrument mounting platform while also outputting a high-precision standard torque. This system enables self-calibration of the output torque measurement system. The inertial navigation unit is an HG-120 type inertial assembly, and the single-axis mass adjustment robot is a HIWIN single-axis robot. The standard torque can be achieved by the single-axis mass adjustment robot 14 in the high-precision torque calibration system. The programmable mass adjustment component on the single-axis mass adjustment robot consists of a moving load 14-1 and a load-bearing slider 14-2.

[0097] Furthermore, when the CMG being measured is installed on the leveled instrument mounting platform and outputs torque, the ultra-quiet suspension system will generate a corresponding motion trend under the pre-tightening effect of the precision torque sensing system. At this time, the force sensor in the precision torque sensing system will output the force measurement data corresponding to the current torque.

[0098] Example 2:

[0099] A method for measuring the torque output characteristics of a control torque gyroscope, based on the control torque gyroscope torque output characteristic measuring device described in Example 1, includes the following steps:

[0100] S1. Confirm whether the air-bearing ball bearing is ventilated and whether the power supply of a control torque gyroscope torque output characteristic measuring device is turned on. The air-bearing ball bearing enters the working state after being ventilated and the power supply is turned on.

[0101] S2. Manually add load counterweights to the instrument mounting platform, read the pressure sensor values ​​on the support unloading device, and when the pressure sensor values ​​are basically equal, complete the coarse balancing of a control torque gyroscope torque output characteristic measurement device.

[0102] S3. Remove the support unloading device, and then use the inertial navigation unit as feedback to control the movement of the programmable mass adjustment component in the single-axis mass adjustment robot. When the inertial navigation unit displays that the attitude of the instrument mounting platform has reached the predetermined level requirement, the fine balancing of a control torque gyroscope torque output characteristic measurement device is completed.

[0103] S4. Based on the high-precision standard torque self-calibration method, complete the self-calibration of the high-precision torque calibration system to obtain the high-precision standard torque;

[0104] When using this system to measure output torque in practice, two issues arise. First, because the force sensor is not perfectly rigid, the measuring platform will deflect due to the torque applied, introducing the influence of gravitational torque and thus reducing the accuracy of the output torque. Second, the actual lever arm information may differ from the design value, further reducing the accuracy of the output torque measurement. To address these accuracy-influencing factors, a standard torque calibration measurement system is needed. When the data monitoring and control system sends a mass adjustment command to the single-axis mass adjustment robot, the programmable mass adjustment component moves to change its distance from the rotation center of the ultra-quiet suspension system, thereby achieving programmable torque output. However, the programmable torque output by this method has an error compared to the ideal standard torque. This error stems from two sources: firstly, the unknown error angle between the moving direction of the programmable mass adjustment component on the single-axis mass adjustment robot and the actual installation direction; and secondly, the unknown mass characteristics of the supporting slider.

[0105] Furthermore, the specific implementation method of the high-precision standard torque self-calibration method in step S4 includes the following steps:

[0106] S4.1. Under the condition of fine balancing based on a control torque gyroscope torque output characteristic measurement device, a position code is set on the instrument mounting platform with a mass load of known weight. , ;

[0107] S4.2. Without loading the precision torque-sensitive system, the data monitoring and control system reads the data from the inertial navigation unit and controls the movement of the programmable mass adjustment components in robot A and robot B to achieve horizontal leveling of the ultra-quiet suspension system;

[0108] S4.3. Add a known mass load to the instrument mounting platform, installing the mass load at position P1. Level the ultrastatic suspension system again, and record the state of the measurement system at this point as the initial zero-position state of self-calibration. Record the mass load mass at this point. And the positions of the programmed quality adjustment components in Robot A and Robot B;

[0109] S4.4. Sequentially encode according to position. The mass load was moved in ascending order. After each movement, the horizontal leveling of the ultra-quiet levitation system was repeated, and the relative displacement of the mass load was recorded for each movement. And the relative displacement after each horizontal leveling movement of the programmable quality adjustment components in Robot A and Robot B. and , where i represents the number of times the mass load is moved, and N sets of data are obtained;

[0110] S4.5. Define the unknown quantities as including the total mass of the programmable mass adjustment components in robots A and B, respectively. and The included angle of the installation error between robot A and the x-axis The installation error angle between the robot's B and Y axes is divided. The high-precision standard torque self-calibration calculation method is constructed as follows:

[0111] S4.5.1. Construct a least-squares system, based on N sets of measurements, and build the following system of linear equations:

[0112]

[0113] Define the variable to be solved as ;

[0114] S4.5.2. Solving for intermediate variables based on the improved total least squares method: The coefficient matrix is ​​denoted as... The observation vector is denoted as Constructing an augmented matrix And perform SVD singular value decomposition as follows:

[0115]

[0116] Wherein, the coefficient matrix Observation vector , It is a singular value matrix. It is a left singular value vector matrix;

[0117] make The last column is And it was revised to ,in If is the regularization parameter, then the above variables to be solved can be written as:

[0118] ;

[0119] S4.5.3. Calculate the mass of the programmable quality adjustment component, and obtain:

[0120] ;

[0121] S4.5.4. Separate the angle parameters and solve for the final error installation angle based on N sets of data, to obtain:

[0122]

[0123] ;

[0124] S4.5.5. After the above self-calibration, the high-precision standard torque self-calibration output formula is as follows:

[0125]

[0126] in, For high-precision standard torque on the x-axis, For high-precision standard torque on the y-axis, and These represent the relative positions of the programmable mass adjustment components of the two robots under program control.

[0127] S5. Based on the self-calibration method of the measurement system of high-precision standard torque, a precision torque sensing system is loaded to complete the self-calibration of a control torque gyroscope torque output characteristic measurement device, and a predictive neural network model and a z-axis error compensation relationship model are obtained;

[0128] After obtaining the aforementioned high-precision standard torque, and before formally measuring the output torque of the device under test, a self-calibration method for the measurement system based on the high-precision standard torque is invented to achieve self-calibration of the measurement system. This addresses the impact of gravitational torque introduced by the tilt of the platform, lever arm variations, and other unknown errors on the horizontal torque measurement, thereby truly achieving high-precision measurement of the output torque. The setup method is as follows:

[0129] Before formally measuring the device under test, the ultra-quiet suspension system is first ensured to be in a leveled state. Then, a precision torque sensing system is loaded onto the instrument mounting platform, and the output of each force sensor is set to 50% of its full scale using a pre-tightening device. This is considered the initial state of the measurement system's self-calibration. Next, the data monitoring and control system sends a programmed torque sequence to the single-axis mass adjustment robot in the high-precision torque calibration system. Based on the self-calibration results, the single-axis mass adjustment robot outputs high-precision programmed torque by changing the relative positions of the programmed mass adjustment components. Subsequently, considering the accuracy of the calibration process, the programmed torque sequence includes multiple levels of static calibration points and dynamic composite calibration points. Based on these two torque sequences, the process of outputting high-precision programmed torque is divided into two stages: multi-level static torque output and dynamic composite torque output.

[0130] When outputting multi-stage static torque, step torques are applied independently along the x-axis and y-axis in sequence (sequence: 0→10%FS→…→100%FS→…→0 symmetrical increasing and decreasing sequence, with a predetermined duration for each step).

[0131] During dynamic composite torque output, adjustable sinusoidal torques are applied independently along the x-axis and y-axis sequentially. Where A is the amplitude, f is the frequency, φ is the phase, and B is the bias. This torque is applied using a frequency sweep followed by an amplitude sweep. During the frequency sweep phase, the amplitude A is fixed, and the frequency f is gradually increased from a low frequency to the measurement system bandwidth. After the frequency sweep is complete, the amplitude sweep phase begins, where the frequency f is fixed, and the amplitude A is changed in a stepwise manner. The parameters in the above process... It can be configured as needed.

[0132] During the process of outputting high-precision programmable torque, the output information of high-precision programmable torque, force sensor and inertial navigation unit is stored in real time and automatically corrected for horizontal torque and vertical torque respectively.

[0133] Furthermore, the specific implementation method of the high-precision standard torque measurement system self-calibration method in step S5 includes the following steps:

[0134] S5.1. A neural network prediction model is constructed for horizontal torque self-calibration. The neural network prediction model adopts an input layer-two hidden layers-output layer structure. The input layer consists of the output information from the force sensor and the inertial navigation unit, including a 7-dimensional input vector. and , The output is the three-axis attitude data of the inertial navigation unit. The output layer is the horizontal torque of the x-axis and y-axis. The two hidden layers consist of 10 neurons and 5 neurons respectively.

[0135] Furthermore, for horizontal self-calibration, based on the above data, the output information of the force sensor in the horizontal torque measurement system and the inertial navigation unit in the high-precision torque calibration system are used as the model input information, and the high-precision programmable torque is used as the output information. A reliable neural network prediction model is trained to complete the horizontal self-calibration. The neural network model adopts a 7-10-5-2 network structure. The input layer is the output information of the force sensor and the inertial navigation unit, and the output layer is the horizontal torque along the x and y axes. It contains two hidden layers consisting of 10 and 5 neurons respectively. The mathematical expressions for each layer are shown below:

[0136] Input layer is

[0137]

[0138] in, Given a 7-dimensional input vector, For force sensor output data, This is the three-axis attitude data output by the inertial navigation unit.

[0139] The first hidden layer has 10 neurons, and the input-output expression is:

[0140]

[0141] in, Represents the output vector of the first hidden layer, and the weight matrix. Bias vector .

[0142] The second hidden layer has 5 neurons, and the input-output expression is:

[0143]

[0144] in, Represents the output vector of the first hidden layer, and the weight matrix. Bias vector .

[0145] The output layer is

[0146]

[0147] in, Represents the output vector of the output layer. The horizontal moments along the x and y axes are represented by weight matrices. Bias vector .

[0148] S5.2. The vertical moment self-correction method is constructed by using polynomial fitting to obtain the z-axis error compensation relationship model; based on and The z-axis disturbance torque is calculated using the following formula:

[0149]

[0150] in, The z-axis disturbance torque. and They are respectively and The corresponding nominal lever arm;

[0151] Establish the mapping relationship between attitude angle and disturbance torque. The expression is:

[0152]

[0153] Where k is the index of the data sample, and N is the total number of data samples. For the x-axis attitude data, y-axis attitude data, and z-axis attitude data of the k-th sample, The disturbance torque around the z-axis corresponding to the k-th sample;

[0154] A multinomial regression was performed on the mapping relationship between attitude angle and disturbance torque to generate the following z-axis error compensation relationship model:

[0155]

[0156] in, For z-axis compensation torque, For polynomial basis functions, , Here, is the fitting coefficient, j is the term index of the polynomial basis function, and m is the highest order of the polynomial regression. The construction of this z-axis error compensation relationship model realizes the self-correction of the vertical moment.

[0157] Furthermore, in step S5, the method for loading a precision torque sensing system to complete the self-calibration of a control torque gyroscope torque output characteristic measurement device is as follows:

[0158] Before formally measuring the device under test, the ultra-quiet suspension system is first ensured to be in a leveled state. Then, a precision torque sensing system is loaded onto the instrument mounting platform, and the output of each force sensor is set to 50% of the full scale to obtain the self-calibration initial state of a control torque gyroscope torque output characteristic measurement device.

[0159] Then, the data monitoring and control system sends the programmed torque sequence to the single-axis mass adjustment robot in the high-precision torque calibration system. Based on the self-calibration results, the single-axis mass adjustment robot outputs high-precision programmed torque by changing the relative position of the programmed mass adjustment components.

[0160] Then, the output information of the high-precision programmable torque, force sensor and inertial navigation unit is stored in real time and synchronized. The self-calibration method of the high-precision standard torque measurement system is used to perform self-calibration operations on the horizontal torque and vertical torque respectively.

[0161] After self-calibration is completed, a measurement system and model are used to achieve high-precision torque measurement.

[0162] Furthermore, when the device under test outputs torque, the output information of the force sensor and the inertial navigation unit is read. On the one hand, it is input into the trained neural network prediction model. The output of the neural network prediction model is the high-precision horizontal output torque obtained by measurement. On the other hand, it is input into the z-axis error compensation relationship model. The z-axis error torque calculated by the z-axis error compensation relationship model is subtracted from the real-time z-axis calculated torque to obtain the high-precision vertical output torque.

[0163] S6. Adjust a control torque gyroscope torque output characteristic measurement device to the self-calibration initial state, turn on the device under test, perform torque measurement, calculate the output torque of the device under test based on the predictive neural network model, the z-axis error compensation relationship model, and the output torque formula, save the experimental data, and complete the experiment.

[0164] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0165] Although this application has been described above with reference to specific embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of this application. In particular, as long as there is no structural conflict, the features in the specific embodiments disclosed in this application can be combined with each other in any way. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, this application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A device for measuring the torque output characteristics of a control torque gyroscope, characterized in that, This includes an ultra-quiet suspension system, a precision torque sensing system, a high-precision torque calibration system, and a data monitoring and control system; The ultra-quiet suspension system includes an air-floating ball bearing (1), an instrument mounting platform (2), and a support unloading device (17). The air-floating ball bearing (1) is set in the air-floating ball bowl (3). The air-floating ball bearing (1) is fastened to the instrument mounting platform (2). The device under test (16) is placed on the instrument mounting platform (2). A pressure sensor (18) is installed on the support unloading device (17). The precision torque sensing system consists of a horizontal torque measurement system (20) and a vertical torque measurement system (21); Four sets of horizontal torque measurement systems (20) are evenly arranged at 90° intervals on the instrument mounting platform (2) to convert horizontal torque measurement into force measurement; the vertical torque measurement system (21) includes a high-precision force sensor group (22), an air bearing (23), and an auxiliary extension mechanism (25), and the auxiliary extension mechanism (25) is rigidly connected to the instrument mounting platform (2); the high-precision force sensor group (22) consists of two force sensors, and the high-precision force sensor group (22) is installed on the outside of the two air-floating surfaces (24) formed between the air bearing (23) and the auxiliary extension mechanism (25) to realize z-axis torque measurement and decouple z-axis torque from xy-axis torque measurement; The high-precision torque calibration system is placed on the instrument mounting platform (2), including an inertial navigation unit (13) and a single-axis mass adjustment robot (14). There are two single-axis mass adjustment robots (14), which are orthogonally distributed between the device under test (16) and the horizontal torque measurement system (20) to output high-precision standard torque while the instrument mounting platform (2) is leveling. The inertial navigation unit (13) is placed between the device under test (16) and the horizontal torque measurement system (20) facing the single-axis mass adjustment robot (14) to monitor the attitude information of the ultra-static suspension system in real time. The data monitoring and control system (15) is connected to the ultra-quiet suspension system, the precision torque sensing system, and the high-precision torque calibration system via a wireless communication device.

2. The control torque gyroscope torque output characteristic measuring device according to claim 1, characterized in that, The horizontal torque measurement system (20) includes a micro-touch unit (4), a high-precision force sensor (5), a high-precision electric linear actuator (6), an air cushion (7), a pneumatic actuator (8), a pneumatic preload unit (9), a composite beam (10), an air-bearing guide rail (11), and an air-bearing slider (12). The composite beam (10) is fixedly mounted on the air-bearing guide rail (11). The high-precision electric linear actuator (6) is rigidly connected to the air-bearing slider (12) through the composite beam (10). The pneumatic actuator (8) passes through the airtight cavity of the composite beam (10) and acts on the upper surface of the air-bearing slider (12). The pneumatic pre-tensioning unit (9) is integrated inside the pneumatic actuator (8) and is independently controlled through the air pressure pipeline. The output end of the pneumatic pre-tensioning unit (9) is coupled to the air-bearing slider (12). The air-bearing slider (12) is suspended. The float is installed in the raceway of the air-bearing guide rail (11). The lower surface of the air-bearing slider (12) is fixed to the high-precision force sensor (5). The micro-touch unit (4) is connected to the measuring end of the high-precision force sensor (5). The high-precision electric linear push rod (6) presses the micro-touch unit (4) onto the air cushion (7) placed on the instrument mounting platform (2) through the high-precision force sensor (5). With the locking action of the pneumatic pre-tightening unit (9), the high-precision force sensor (5) and the instrument mounting platform (2) are pre-tightened.

3. The control torque gyroscope torque output characteristic measuring device according to claim 1, characterized in that, There are three sets of support unloading devices (17), which are distributed at 120° intervals and are used to support and roughly balance the instrument installation platform (2). The two single-axis mass adjustment robots (14) are robot A and robot B, respectively.

4. The control torque gyroscope torque output characteristic measuring device according to claim 1, characterized in that, The data monitoring and control system (15) is responsible for receiving the output information of sensitive elements and control elements in each system, sending quality adjustment and loading instructions to the high-precision torque calibration system and the precision torque sensitive system, and calculating the compensation information and measured torque.

5. The control torque gyroscope torque output characteristic measuring device according to claim 1, characterized in that, The measured force results of the four sets of horizontal torque measurement systems (20) are F1, F2, F3, and F4, respectively, and the theoretical lever arms between each force sensor and the rotation center are L1, L2, L3, and L4, respectively. The measured force results of the vertical torque measurement system (21) are F5 and F6, respectively, and the theoretical lever arms between each force sensor and the rotation center are L5 and L6, respectively. The formula for the output torque under ideal conditions obtained by measuring torque using the equilibrium method is as follows: in, , and The measured torque results are shown in the x-direction, y-direction, and z-direction, respectively.

6. A method for measuring the torque output characteristics of a control torque gyroscope, implemented using the control torque gyroscope torque output characteristic measuring device as described in any one of claims 1-4, characterized in that... Includes the following steps: S1. Confirm whether the air-bearing ball bearing is ventilated and whether the power supply of the control torque gyroscope torque output characteristic measuring device is turned on. The air-bearing ball bearing enters the working state after being ventilated and the power supply is turned on. S2. Manually add load counterweights to the instrument mounting platform, read the pressure sensor values ​​on the support unloading device, and when the pressure sensor values ​​are basically equal, complete the coarse balancing of the control torque gyroscope torque output characteristic measuring device. S3. Remove the support unloading device, and then use the inertial navigation unit as feedback to control the movement of the programmable mass adjustment component in the single-axis mass adjustment robot. When the inertial navigation unit displays that the attitude of the instrument mounting platform has reached the predetermined level requirement, complete the fine balancing of the control torque gyroscope torque output characteristic measurement device. S4. Based on the high-precision standard torque self-calibration method, complete the self-calibration of the high-precision torque calibration system to obtain the high-precision standard torque; S5. Based on the self-calibration method of the measurement system of high-precision standard torque, a precision torque sensing system is loaded to complete the self-calibration of the control torque gyroscope torque output characteristic measurement device, and the predictive neural network model and z-axis error compensation relationship model are obtained; S6. Adjust the control torque gyroscope torque output characteristic measuring device to the self-calibration initial state, turn on the device under test, perform torque measurement, calculate the output torque of the device under test based on the predictive neural network model, the z-axis error compensation relationship model, and the output torque formula, save the experimental data, and complete the experiment.

7. The method for measuring the torque output characteristics of a control torque gyroscope according to claim 6, characterized in that: The specific implementation method of the high-precision standard torque self-calibration method in step S4 includes the following steps: S4.

1. Under the condition of fine balancing based on the control torque gyroscope torque output characteristic measurement device, set the position code of the mass load with known weight added on the instrument mounting platform. , ; S4.

2. Without loading the precision torque-sensitive system, the data monitoring and control system reads the data from the inertial navigation unit and controls the movement of the programmable mass adjustment components in robot A and robot B to achieve horizontal leveling of the ultra-quiet suspension system; S4.

3. Add a known mass load to the instrument mounting platform, installing the mass load at position P1. Level the ultrastatic suspension system again, and record the state of the measurement system at this point as the initial zero-position state of self-calibration. Record the mass load mass at this point. And the positions of the programmed quality adjustment components in Robot A and Robot B; S4.

4. Sequentially encode according to position. The mass load was moved in ascending order. After each movement, the horizontal leveling of the ultra-quiet levitation system was repeated, and the relative displacement of the mass load was recorded for each movement. And the relative displacement after each horizontal leveling movement of the programmable quality adjustment components in Robot A and Robot B. and , where i represents the number of times the mass load is moved, and N sets of data are obtained; S4.

5. Define the unknown quantities as including the total mass of the programmable mass adjustment components in robots A and B, respectively. and The included angle of the installation error between robot A and the x-axis The included angle of the installation error between the robot's B and Y axes The high-precision standard torque self-calibration calculation method is constructed as follows: S4.5.

1. Construct a least-squares system, based on N sets of data, and build the following system of linear equations: Define the variable to be solved as ; S4.5.

2. Solving for intermediate variables based on the improved total least squares method: The coefficient matrix is ​​denoted as... The observation vector is denoted as Constructing an augmented matrix And perform SVD singular value decomposition as follows: Wherein, the coefficient matrix Observation vector , It is a singular value matrix. It is a left singular value vector matrix; make The last column is And it was revised to ,in If is the regularization parameter, then the above variables to be solved can be written as: ； S4.5.

3. Calculate the mass of the programmable quality adjustment component, and obtain: ； S4.5.

4. Separate the angle parameters and calculate the final installation error angle based on N sets of data, obtaining: ； S4.5.

5. After the above self-calibration, the high-precision standard torque self-calibration output formula is as follows: in, For high-precision standard torque on the x-axis, For high-precision standard torque on the y-axis, and These represent the relative positions of the programmable mass adjustment components of the two robots under program control.

8. The method for measuring the torque output characteristics of a control torque gyroscope according to claim 7, characterized in that: The specific implementation method of the high-precision standard torque measurement system self-calibration method in step S5 includes the following steps: S5.

1. Construct a predictive neural network model for horizontal torque self-calibration. The predictive neural network model adopts an input layer-two hidden layers-output layer structure. The input layer consists of the output information from the force sensor and the inertial navigation unit, including a 7-dimensional input vector. and , The output is the three-axis attitude data of the inertial navigation unit. The output layer is the horizontal torque of the x-axis and y-axis. The two hidden layers consist of 10 neurons and 5 neurons respectively. S5.

2. The vertical moment self-correction method is constructed by using polynomial fitting to obtain the z-axis error compensation relationship model; based on and The z-axis disturbance torque is calculated using the following formula: in, The z-axis disturbance torque. and They are respectively and The corresponding nominal lever arm; Establish the mapping relationship between attitude angle and disturbance torque. The expression is: Where k is the index of the data sample, and N is the total number of data samples. For the x-axis attitude data, y-axis attitude data, and z-axis attitude data of the k-th sample, The disturbance torque around the z-axis corresponding to the k-th sample; A multinomial regression was performed on the mapping relationship between attitude angle and disturbance torque to generate the following z-axis error compensation relationship model: in, For z-axis compensation torque, For polynomial basis functions, , Here, is the fitting coefficient, j is the term index of the polynomial basis function, and m is the highest order of the polynomial regression. The construction of this z-axis error compensation relationship model realizes the self-correction of the vertical moment.

9. A method for measuring the torque output characteristics of a control torque gyroscope according to claim 8, characterized in that: The method for loading the precision torque sensing system in step S5 to complete the self-calibration of the control torque gyroscope torque output characteristic measurement device is as follows: Before formally measuring the device under test, first ensure that the ultra-quiet suspension system is in a leveled state, then load the precision torque sensing system onto the instrument mounting platform, and set the output of each force sensor to 50% of the full scale to obtain the self-calibration initial state of the control torque gyroscope torque output characteristic measuring device. Then, the data monitoring and control system sends the programmed torque sequence to the single-axis mass adjustment robot in the high-precision torque calibration system. Based on the self-calibration results, the single-axis mass adjustment robot outputs high-precision programmed torque by changing the relative position of the programmed mass adjustment components. Then, the output information of the high-precision programmable torque, force sensor and inertial navigation unit is stored in real time and synchronized. The self-calibration method of the high-precision standard torque measurement system is used to perform self-calibration operations on the horizontal torque and vertical torque respectively. After self-calibration is completed, a measurement system and model are used to achieve high-precision torque measurement.

Images