A model load self-sensing support device based on distributed strain and its usage method

Abstract

This invention discloses a model load self-sensing support device and its usage method based on distributed strain, relating to the field of wind tunnel testing technology. It includes a measuring element segment with an octagonal cross-section, front and rear connecting segments, and a labyrinth-style overlapping protective cover. The measuring element segment has an axial through hole at its center, and different types of strain gauges (full-bridge or half-bridge) are attached to each surface according to their function, enabling simultaneous measurement of five aerodynamic components: normal force, pitching moment, lateral force, yaw moment, and roll moment. The protective cover uses a labyrinth overlap and brush sealing to effectively isolate interference from high-speed airflow. This invention solves the problem of being unable to obtain the full-scale aerodynamic force in hinge moment tests and pressure tests due to the inability to arrange a main balance. Static calibration linearity error is <0.7%, coupled loading relative error is <1.5%, cross-interference is <2%, and the test cycle is shortened by more than 50%. This invention has a compact structure, accurate measurement, and compatibility with existing tail support interfaces, making it suitable for various high-speed wind tunnel model tests.

Description

Technical Field

[0001] This invention relates to the field of wind tunnel testing technology, and more specifically to the field of a model load self-sensing support device based on distributed strain and its usage method. Background Technology

[0002] The balance is the most important measuring device in wind tunnel force measurement tests, used to measure the magnitude, direction, and point of application of aerodynamic loads (forces and moments) acting on the model. In hinge moment tests and pressure tests, due to the extremely limited internal space of the model and the need to lay out a large number of measuring cables, it is usually impossible to install the main balance, making it impossible to directly obtain the aerodynamic forces experienced by the model in high-speed airflow. This leads to two serious shortcomings: First, in order to obtain complete aerodynamic characteristics, the same shape often requires the design of separate force measurement models, pressure measurement models, and hinge moment models, and different tests are carried out separately, which greatly prolongs the development cycle. At the same time, multiple tests involve the design and processing of multiple sets of models and the repeated use of wind tunnel equipment, resulting in huge resource consumption. Second, in hinge moment, pressure measurement and other tests, due to the lack of a main balance to measure the aerodynamic forces of the whole aircraft in real time, active damping and other technologies cannot be effectively used. Many aircraft often experience severe flutter due to airflow and structural coupling at high angles of attack and large sideslip angles, making it impossible to accurately determine the model position and attitude. This directly leads to attitude errors in the hinge moment and pressure measurement data, and in severe cases, it can even affect the construction and use of the aerodynamic database.

[0003] In existing technologies, there have been attempts to perform simple measurements by directly attaching strain gauges to the support rods. However, these methods have the following inherent drawbacks: the strain gauges are directly exposed to high-speed airflow, and are affected by airflow impact and drastic temperature changes, resulting in severe output signal drift and unusable data; the strain sensitivity of circular or square support rods is low, and the coupling between different load components is large, making accurate decoupling difficult; the support rod surface has no protective structure, making the strain gauges and wiring extremely easy to damage, which cannot meet the reliability and repeatability requirements of engineering applications.

[0004] Therefore, there is an urgent need for an integrated tail support device that meets both the requirements for support strength and stiffness and can reliably measure aerodynamic loads. Summary of the Invention

[0005] The purpose of this invention is to address the aforementioned technical problems by providing a model load self-sensing support device and its usage method based on distributed strain. This shortens the testing cycle, reduces resource consumption, and provides real-time load input for active vibration damping.

[0006] To achieve the above objectives, the present invention specifically adopts the following technical solution: One aspect of the present invention provides a model load self-sensing support device based on distributed strain, comprising: The measuring element segment has a regular octagonal cross-section and an axial through hole at its center; The front connection section, located at the front end of the measuring element section, is used to connect to the wind tunnel model; The rear connecting section, located at the rear end of the measuring element section, is used to connect with the wind tunnel support mechanism; A protective cover is fitted over the outside of the measuring element section and fixedly connected to the front end of the measuring element section to isolate high-speed airflow. Multiple sets of strain gauges are attached to the outer surfaces of a regular octagonal cross section according to a predetermined layout to measure the strain signal experienced by the wind tunnel model on the corresponding outer surface. And the wiring structure, which is located inside the measuring element section and connected to the axial through hole, is used to guide the strain gauge leads out from inside the measuring element section; Multiple sets of strain gauges collect strain signals of corresponding components, which are then transmitted to external acquisition devices via a wiring structure. The strain signals are decoupled and calculated using a pre-generated calibration matrix to obtain the normal force, pitching moment, lateral force, yaw moment, and rolling moment acting on the model.

[0007] In one embodiment, the circumcircle diameter of the octagonal cross-section of the measuring element segment is 12 mm, the axial length is 88 mm, and each edge is provided with a rounded corner with a radius of 2 mm; the diameter of the axial through hole is 8 mm.

[0008] Specifically, the octagonal cross-section of the measuring element segment exhibits good symmetry, resulting in uniform strain distribution across all edges under normal and lateral loads. Furthermore, the planar structure facilitates strain gauge bonding and pressure application. Compared to a circular cross-section, the octagonal plane provides a stable strain bonding reference surface; compared to a rectangular cross-section, the octagon exhibits higher bending stiffness for the same circumcircle diameter and can accommodate strain gauges with six components simultaneously.

[0009] In one embodiment, the protective cover adopts a clasping structure, consisting of two halves of the shell with a labyrinthine overlapping structure at their joint surface; a gap is left between the free end of the protective cover and the tail support body, and a brush is pasted in the gap. The combination of labyrinthine overlapping and brush can effectively prevent high-speed airflow from entering the measuring element, while not transferring additional structural load.

[0010] Specifically, the joint surface has a labyrinthine overlapping structure (the cross-section is sawtooth-shaped, forming a tortuous air passage), and is pressed and fixed to the front end of the measuring element section with a circular boss (Φ12mm, thickness 3mm) using 4 M3 screws; a 0.5mm gap is left between the free end of the protective cover and the tail support body, and a brush (bristles 1mm long, density 20~30 bristles / mm) is pasted in this gap. 2 ).

[0011] In one implementation, the arrangement of multiple sets of measuring strain gauges is as follows: Two strain gauges are attached to the upper and lower surfaces of a regular octagon to form a full bridge, which is used to measure the normal force and pitching moment. Two strain gauges are attached to the left and right surfaces of a regular octagon to form a full bridge, which is used to measure lateral force and yaw moment. Two strain gauges are attached to each of the first and second sets of symmetrical inclined planes of a regular octagon to form a full bridge for measuring rolling torque.

[0012] Specifically, after each strain gauge is attached, a layer of silicone rubber protective adhesive is applied, and then a polyimide film is covered to prevent moisture and dust from affecting it.

[0013] In one embodiment, the wiring structure includes two radial wiring holes disposed on the measuring element segment. The radial wiring holes communicate with the axial through hole, and all measuring strain gauge leads converge into the axial through hole through the radial wiring holes and are led out to the rear.

[0014] Specifically, all strain gauge leads converge from the radial routing holes into the axial through-hole and are then led out backwards. They are then led out to the strain acquisition device via the sealed wiring fitting at the rear connection section. The junction of the radial routing holes and the axial through-hole is rounded to prevent damage to the cables.

[0015] In one embodiment, the front connecting section is provided with an interface that mates with the rear cone of the wind tunnel model or the rear cone of the force balance, and the interface is a 1:5 taper fit with key positioning; the rear connecting section is provided with a balance interface.

[0016] The balance interface size conforms to industry standards (such as Φ20mm×1mm threaded or flanged connection), and can directly replace the conventional tail support without modifying the existing wind tunnel equipment.

[0017] In one embodiment, the tail support device is made of 30CrMnSiA material, 7075-T6 aluminum alloy, or QBe2 beryllium bronze.

[0018] In one implementation, the size of the measuring element segment can be scaled proportionally to accommodate wind tunnel models of different sizes.

[0019] Another aspect of the present invention provides a method for measuring aerodynamic loads on a wind tunnel model, using the aforementioned model load self-sensing support device based on distributed strain, comprising the following steps: S1. Connect the tail support device to the model via the front connecting section and to the wind tunnel support mechanism via the rear connecting section. S2. Connect multiple sets of strain gauges to the strain acquisition device according to the predetermined bridge circuit; S3. During the wind tunnel test, the aerodynamic load on the wind tunnel model is transferred to the measuring element section, causing strain in the regular octagonal cross section. S4. Multiple sets of strain gauges collect strain signals of corresponding components, which are then transmitted to external acquisition devices via wiring structures. S6. The strain signal is decoupled and calculated using a pre-generated calibration matrix to obtain the normal force, pitching moment, lateral force, yaw moment and roll moment acting on the model.

[0020] The working principle is as follows: Before the wind tunnel test begins, the tail support device is rigidly connected to the rear cone of the wind tunnel model via the front connecting section, and then connected to the wind tunnel support mechanism via the rear connecting section. After multiple sets of measuring strain gauges are connected in a full-bridge configuration, their leads are led out to the external strain acquisition equipment through radial wiring holes, axial through holes, and sealed wiring fittings.

[0021] When the wind tunnel model is subjected to aerodynamic loads in a high-speed airflow, these loads are transferred to the measuring element section via the front connecting section. Under the action of normal force, pitching moment, lateral force, yaw moment, and roll moment, the measuring element with a regular octagonal cross-section experiences different tensile or compressive strains on its various surfaces. Specifically: The normal force and pitching moment mainly cause large axial tensile and compressive strains on the upper and lower surfaces. The full-bridge strain gauge attached here outputs a voltage signal that is linearly related to the normal force and pitching moment. Lateral forces and yaw moments primarily cause strain on the left and right surfaces, corresponding to the relevant signals output by the full-bridge. The rolling torque causes torsional shear strain in the measuring element. The shear-sensitive strain gauge (ZF650-4HA) attached to the ±45° inclined plane is most sensitive to this shear strain and outputs a signal corresponding to the rolling torque.

[0022] The strain signals are led out through the wiring structure and sent to the acquisition system. Using a 6×6 calibration matrix generated in advance through a six-component calibration platform, the original strain voltage signals are decoupled and calculated to eliminate cross-interference between components, and finally the accurate six-component aerodynamic load on the model is obtained.

[0023] Throughout the air-blowing process, the protective cover effectively isolates the external high-speed airflow through labyrinthine overlaps and brushes, preventing the strain gauge from being directly affected by airflow impact and temperature fluctuations. Simultaneously, it does not transmit structural loads, thus ensuring the purity and stability of the measurement signal. The use of two parallel strain gauges mounted at the same location and bridged on opposite sides effectively offsets the heat output caused by ambient temperature changes, further improving measurement accuracy.

[0024] The beneficial effects of this invention are as follows: 1. Functional Integration and Reduced Cycle Time: For the first time, direct measurement of five-component aerodynamic loads on the tail support is achieved, solving the long-standing problem of not being able to simultaneously obtain the aerodynamic forces of the entire aircraft during hinge moment and pressure tests. The same wind tunnel model no longer needs to undergo separate force, pressure, and hinge moment tests, reducing the test cycle by more than 50%, reducing the number of wind tunnel models required by at least two, and decreasing wind tunnel model machine time by approximately 40%, significantly lowering R&D costs.

[0025] 2. Measurement accuracy meets engineering requirements: After static calibration verification, the linear correlation coefficient between the output of each component and the load is greater than 0.999; under coupled loading, the relative error of each component is ≤1.5%, which is better than the 2% index of conventional small balances; repeatability error is ≤0.5% FS, and hysteresis error is ≤0.3% FS.

[0026] 3. High anti-interference capability: The regular octagonal cross section and differentiated strain gauge layout result in cross-interference between components of <2%, which can be reduced to less than 0.5% after decoupling matrix correction; the protective cover labyrinth overlap + brush has been verified by wind tunnel blowing at Mach numbers of 0.3~1.2. The zero drift change of the measuring element is <0.05% FS under conditions with / without protective cover, the dynamic signal signal-to-noise ratio is improved by more than 15dB, and the airflow interference zero drift can be reduced by more than 90%.

[0027] 4. Structural compatibility and redundancy: The rear connection section adopts a balance interface, which can directly replace the conventional tail support without modifying the wind tunnel bevel or bracket; when the main balance is damaged, this device can be put into use as a backup force measuring unit immediately, avoiding losses of hundreds of thousands of yuan caused by test interruption.

[0028] 5. Provide load input for active damping: In hinge torque or pressure measurement tests where the main balance cannot be installed, this device can output the aerodynamic force of the entire machine in real time for use by the active damping control system, effectively suppressing model chattering at large angles of attack and improving the accuracy of attitude angle measurement.

[0029] 6. Excellent dynamic response performance: The device has a first-order bending frequency of 285Hz and a second-order torsional frequency of 620Hz, which is much higher than the typical buffeting frequency (<50Hz) of wind tunnel models, fully meeting the real-time requirements of active vibration reduction control. Attached Figure Description

[0030] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0031] Figure 1This is an external contour diagram of a model load self-sensing support device based on distributed strain.

[0032] Figure 2 yes Figure 1 A sectional view.

[0033] Figure 3 It is a view of the measuring element segment.

[0034] Figure 4 yes Figure 3 The left view.

[0035] Figure 5 yes Figure 3 The right view.

[0036] Figure 6 This is a schematic diagram of the structure when the protective cover and measuring element section are assembled.

[0037] Figure 7 This is a view of the protective shield.

[0038] Figure 8 This is a schematic diagram of the brush mechanism.

[0039] Figure 9 This is a comparison chart of the predicted and actual values ​​of the normal force.

[0040] Figure 10 This is a comparison chart of the predicted and actual lateral force values.

[0041] Figure 11 This is a comparison chart of the predicted and actual rolling torque values.

[0042] Figure 12 This is a comparison chart of the predicted and actual yaw moment values.

[0043] Figure 13 This is a comparison chart of the predicted pitch moment and the actual value.

[0044] Figure 14 This is a graph evaluating the predictive performance of the correlation coefficients between the predicted values ​​and the actual values ​​of the linear and nonlinear modified models.

[0045] Figure 15 This is a graph showing the variation of zero-drift strain with Mach number under two conditions: with and without a protective shield.

[0046] Figure 16 This is a comparison chart of the accuracy of the prediction model in this paper and that of a traditional force balance.

[0047] Reference numerals: 1. Front connecting section; 2. Labyrinth overlap structure; 3. Brush; 4. Protective cover; 5. Axial through hole; 6. Rear connecting section; 7. Balance interface; 8. Measuring element section; 9. Radial wiring hole. Detailed Implementation

[0048] To make the technical problems, technical solutions, and technical effects of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0049] Therefore, the following detailed description of the 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 embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0050] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0051] In the description of the embodiments of the present invention, it should be noted that the terms "inner", "outer", "upper", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the present invention.

[0052] This invention provides a model load self-sensing support device based on distributed strain, comprising: The measuring element segment 8 has a regular octagonal cross-section and an axial through hole 5 at its center; Front connection section 1, located at the front end of measuring element section 8, is used to connect with the wind tunnel model; Rear connection section 6, located at the rear end of measuring element section 8, is used to connect with the wind tunnel support mechanism; The protective cover 4 is fitted over the outside of the measuring element section 8 and is fixedly connected to the front end of the measuring element section 8 to isolate high-speed airflow; Multiple sets of strain gauges are attached to the outer surfaces of a regular octagonal cross section according to a predetermined layout to measure the strain signal experienced by the wind tunnel model on the corresponding outer surface. And the wiring structure is located inside the measuring element section 8 and is connected to the axial through hole 5 to guide the strain gauge lead wire out from inside the measuring element section 8; Multiple sets of strain gauges collect strain signals of corresponding components, which are then transmitted to external acquisition devices via a wiring structure. The strain signals are decoupled and calculated using a pre-generated calibration matrix to obtain the normal force, pitching moment, lateral force, yaw moment, rolling moment, and axial force acting on the model.

[0053] In one embodiment, the circumcircle diameter of the regular octagonal cross-section of the measuring element segment 8 is 12mm, the axial length is 88mm, and each edge is provided with a rounded corner with a radius of 2mm; the diameter of the axial through hole 5 is 8mm.

[0054] Specifically, the octagonal cross-section of measuring element segment 8 has good symmetry, and the strain distribution on each edge is uniform under normal and lateral loads. Furthermore, the planar structure facilitates strain gauge bonding and pressure application. Compared to a circular cross-section, the octagonal plane provides a stable strain bonding reference surface; compared to a rectangular cross-section, the octagon has higher bending stiffness for the same circumscribed circle diameter and can simultaneously accommodate strain gauges of six components.

[0055] In one embodiment, the protective cover 4 adopts a hugging structure, consisting of two halves of the shell, with the joint surface being a labyrinth overlap structure 2; a gap is left between the free end of the protective cover 4 and the tail support body, and a high-temperature resistant brush 3 is pasted in the gap to seal it. The combination of the labyrinth overlap and the brush 3 can effectively prevent high-speed airflow from entering the measuring element, while not transferring additional structural load.

[0056] Specifically, the joint surface is a labyrinthine overlapping structure 2 (the cross-section is sawtooth-shaped, forming a tortuous air passage), which is pressed and fixed to the circular boss (diameter Φ12mm, thickness 3mm) set at the front end of the measuring element section 8 by 4 M3 screws; a 0.5mm gap is left between the free end of the protective cover 4 and the tail support body, and a high-temperature resistant brush 3 (brush bristle length 1mm, density 20~30 bristles / mm2) is pasted in this gap to seal it.

[0057] In one embodiment, the arrangement of multiple sets of measuring strain gauges is as follows: Two strain gauges are attached to the upper and lower surfaces of a regular octagon to form a full bridge, which is used to measure the normal force and pitching moment. Two strain gauges are attached to the left and right surfaces of a regular octagon to form a full bridge, which is used to measure lateral force and yaw moment.

[0058] Two strain gauges are attached to each of the first and second sets of symmetrical inclined planes of a regular octagon to form a full bridge for measuring rolling torque.

[0059] Specifically, after each strain gauge is attached, a layer of silicone rubber protective adhesive is applied, and then a polyimide film is covered to prevent moisture and dust from affecting it.

[0060] In one embodiment, the wiring structure includes two radial wiring holes 9 disposed on the measuring element segment 8. The radial wiring holes 9 communicate with the axial through hole 5. All measuring strain gauge leads converge into the axial through hole 5 through the radial wiring holes 9 and are led out to the rear.

[0061] Specifically, all strain gauge leads converge into the axial through-hole 5 via the radial routing hole 9 and are then led out to the strain acquisition device via the sealed cable threading part of the rear connecting section 6. The junction of the radial routing hole 9 and the axial through-hole 5 is rounded to prevent damage to the cables.

[0062] In one embodiment, the front connecting section 1 is provided with an interface that mates with the rear cone of the wind tunnel model, and the interface is a 1:5 taper fit with key positioning; the rear connecting section 6 is provided with a balance interface 7.

[0063] The balance interface 7 dimensions conform to industry standards (such as Φ20mm×1mm threaded or flanged connections), and can directly replace conventional tail supports without requiring modifications to existing wind tunnel equipment.

[0064] In one embodiment, the tail support device is made of 30CrMnSiA material, 7075-T6 aluminum alloy, or QBe2 beryllium bronze.

[0065] In one embodiment, the size of the measuring element segment 8 can be scaled proportionally to accommodate wind tunnel models of different tonnages.

[0066] Another aspect of the present invention provides a method for measuring aerodynamic loads on a wind tunnel model, using the aforementioned model load self-sensing support device based on distributed strain, comprising the following steps: S1. Connect the tail support device to the model through the front connecting section 1 and to the wind tunnel support mechanism through the rear connecting section 6. S2. Connect multiple sets of strain gauges to the strain acquisition device according to the predetermined bridge circuit; S3. During the wind tunnel test, the aerodynamic load on the wind tunnel model is transmitted to the measuring element section 8, causing strain in the regular octagonal cross section. S4. Multiple sets of strain gauges collect strain signals of corresponding components. The strain signals are eliminated by temperature-compensated strain gauges to eliminate heat output caused by changes in ambient temperature, and then transmitted to external acquisition devices through the wiring structure. S6. The strain signal is decoupled and calculated using a pre-generated calibration matrix to obtain the normal force, pitching moment, lateral force, yaw moment and roll moment acting on the model.

[0067] Example 1 like Figures 1 to 8 As shown, this embodiment provides a model load self-sensing support device based on distributed strain. The specific assembly, testing, and calibration steps are as follows: A1. Processing and Assembly: In this embodiment, the tail support device is made of 30CrMnSiA material, which undergoes solution treatment and aging to achieve a hardness of HRC38~42. The tail support device is precision machined according to the drawings, with the following key dimensional tolerances controlled: octagonal side-to-side distance tolerance ±0.02mm, axial through-hole 5 coaxiality Φ0.03mm, and surface roughness Ra0.8 for all bonding surfaces. After machining, each bonding surface is cleaned sequentially with acetone and anhydrous ethanol, and then dried for later use. A2. Strain gauge bonding and wiring: Surface preparation: Apply CC-33A primer to the cleaned bonding surface, ensuring a uniform thickness.

[0068] Application: Using a special fixture, fix the strain gauges onto the designated surfaces of the regular octagon, apply pressure of 0.3 MPa, and cure at 150°C for 2 hours. After curing, check the resistance of each strain gauge; the deviation should not exceed ±0.5 Ω.

[0069] Lead wire soldering: Silver-plated copper wire with a diameter of 0.1mm was used to spot weld the strain gauge lead-out end to the high-temperature shielding wire (outer diameter Φ0.8mm). All solder joints were full and free of defects.

[0070] Bridge and road connections: according to Figures 3 to 5 The layout shown connects the strain gauges of each component into a full bridge. For example, the normal force full bridge consists of two sets of strain gauges on each of the upper and lower surfaces, with the following output relationship: .

[0071] Protection: Apply a layer of 703 silicone rubber protective adhesive, 0.3mm thick, cure at room temperature for 24 hours, then cover with a layer of polyimide film and seal the edges with high-temperature resistant tape.

[0072] A3. Routing and Packaging: All strain gauge leads are routed into the radial routing hole 9 at the rear end of the measuring element section 8, converging into the Φ8mm axial through hole 5, and exiting from the tail support. A layer of PTFE sheath is wrapped around the rounded corner at the junction of the routing hole and the axial through hole 5 to prevent cable scratches. Finally, the protective cover 4 is installed, and the left and right halves of the housing are fitted onto the measuring element section 8, ensuring a tight fit between the labyrinthine joint surfaces. High-temperature resistant brushes 3 are applied to the gaps at the free ends to seal the gaps, and four M3 screws are tightened.

[0073] A4. Static calibration: The assembled tail support device is fixed to the dedicated six-component calibration platform via the rear connecting section 6. A standard load is applied to the front end of the device using a weight loading method via a dedicated code disk.

[0074] Table 1. Loading range and loading performance of five-component loads

[0075] In the table, Y represents the measured normal force, Mz represents the pitching moment, Z represents the lateral force, My represents the yaw moment, and Mx represents the roll moment. The loading range specifies the magnitude of the load for each component, which directly affects the applicability of subsequent calibration formulas. If a load point deviates significantly from the loading range, the performance of the prediction model trained using that loading data may be affected. The number of calibration points represents the number of loading steps. For example, if the Y range is 0~780 and the number of calibration points is 6, it means dividing 0~780 into 6 equal parts for step loading. Linearity describes the correlation between strain measurements and load, while repeatability is the degree of dispersion of multiple loading results.

[0076] The calibration data for each component under independent and coupled loading are as follows: Establish the calibration matrix equation: ; Where [C] is the sensitivity matrix, obtained by least squares fitting. Y is the normal force. Z is the pitching moment, and Z is the lateral force. For yaw moment, This refers to the rolling torque. This paper primarily focuses on strain measurement, forming five characteristic channels based on strain bridging and characteristic normalization, and using... , , , , This indicates that it can be considered This represents the characteristic voltage difference corresponding to the Y component.

[0077] A5. Coupled Loading Verification: A composite load was applied: Y = 1200 N, Mz = 60 Nm, X = 500 N, while a small number of other components were applied to simulate the actual working condition. The outputs of each channel were collected, and the results were calculated using the calibration matrix described above. The results are as follows: Table 2 Comparison of Actual Load Values ​​and Predicted Results

[0078] Table 2 presents a comparison between the actual load state and the predicted values. It can be seen that the actual load state is Y=1200N, Mz=60N.m, while the predicted values ​​using the method presented in this paper are Y=1190.4N and Mz=60.6, with relative errors all less than 1%, fully meeting engineering requirements. The actual loads of the other three components Z, My, and Mx are 0, and the predicted results are also very small. Simultaneously, the outputs of the cross-interference components are all less than 1% of their corresponding ranges, proving that the decoupling effect is excellent (if the decoupling effect is poor, a larger Y will induce a larger Z, My, and Mx).

[0079] A6. Comparison and verification with a conventional balance: A metrologically certified six-component standard balance (model: 2N6-30I, accuracy: 0.3%) was connected in series at the front end of the tail support device. Eight different composite loads were applied simultaneously. The measurement results of this device and the standard balance are compared below (three typical loads are selected): Table 3 Comparison results between this device and the standard balance

[0080] Table 3 presents three sets of measurements, taken using both a conventional balance and the method developed in this paper to measure the same load condition. By comparing the conventional balance measurements with the standard, it can be seen that for different load conditions (the first and second groups are dominated by longitudinal loads of the Y\Mz combination, and the third group is dominated by transverse loads of the Z\My\Mx combination), the difference between the measurements of this device and the conventional balance measurements is less than 1.2%, and the maximum deviation of the five components is 1.16%, which meets the requirements of wind tunnel force measurement test engineering (generally ≤2%).

[0081] A7. Airflow disturbance test of protective cover 4: In a 0.6m × 0.6m transient wind tunnel, with wind speeds Ma = 0.3~1.2 and angles of attack α = 0°~20°, zero drift of the measuring elements was tested in two states: with and without protective cover 4 (maze + brush 3). (No model, no aerodynamic load, only airflow impact). The results are as follows: Table 4 Comparison of zero drift at measuring points with and without protective shield at different incoming Mach numbers

[0082] In actual use, the temperature distribution sensed by the testing equipment varies depending on the incoming Mach number. Table 4 compares the zero drift at different Mach numbers with and without a protective shield. It can be seen that with the protective shield, the zero drift value for each Mach number is lower than that without the shield. Compared to the unprotected design, the protective shield can reduce airflow interference zero drift by more than 90%. The zero drift voltage difference with the protective shield, converted into an equivalent load, has a maximum of 0.81, which is far less than the conventional aerodynamic load and can be ignored, demonstrating the effectiveness of the protection scheme proposed in this paper.

[0083] A8. Dynamic Response Test: The model was excited using the hammer impact method, and the natural frequencies of the tail support device were measured. The first-order bending frequency was 285Hz, and the second-order torsional frequency was 620Hz, which are much higher than the typical buffeting frequency (<50Hz) of wind tunnel models, indicating that the device has sufficient dynamic response bandwidth and can be used for active vibration damping control.

[0084] Example 2 In this embodiment, the size of the measuring element segment 8 can be scaled proportionally according to the different wind tunnel and model size requirements. In this embodiment, for large wind tunnels (model weight ≥ 200kg), the diameter of the circumscribed circle of the regular octagon is increased to Φ20mm, the length of the measuring element segment 8 is increased to 120mm, the diameter of the axial through hole 5 is increased to Φ12mm, and the design values ​​of each load are increased proportionally. The strain gauge is correspondingly adjusted to a model with a larger grid length (such as ZF2000-4AA) to match the larger bonding area. The size of the protective cover 4 is simultaneously enlarged, and the labyrinth overlap gap is adjusted to 0.8mm. The balance interface 7 of the rear connecting segment 6 is changed to a flange connection (Φ40mm) conforming to the standards for large wind tunnels. The remaining structure, bonding layout, and signal processing flow are the same as in Embodiment 1. This embodiment demonstrates that the present invention has good scalability and can cover a wide range of applications from small to large wind tunnels.

[0085] Example 3 This embodiment replaces the materials of the device to address different load levels and stiffness requirements. In addition to 30CrMnSiA stainless steel, high-strength aluminum alloy 7075-T6 is selected. This material has a lower elastic modulus (approximately 71 GPa), resulting in a larger strain output from the regular octagonal cross-section under the same load, improving sensitivity by approximately 30%, but at the cost of reduced absolute stiffness. To maintain support stiffness, the diameter of the circumscribed circle of the regular octagon is increased from Φ12mm to Φ14mm. The strain gauge selection, bonding process, and protective cover structure remain unchanged. After calibration, the sensitivity of the aluminum alloy version of the device is significantly improved, but the full-scale load is reduced by approximately 20%, making it suitable for pressure testing scenarios where the wind tunnel model is lightweight and higher measurement sensitivity is required.

[0086] Another alternative material is beryllium bronze QBe2, which has an elastic modulus between that of steel and aluminum, and possesses excellent corrosion resistance and elastic stability. All strain gauges employing a regular octagonal measurement cross-section and the aforementioned strain gauge layout are equivalent substitutions of this invention and do not fall outside the scope of protection of this invention due to material variations.

[0087] in, Figure 3 The colors in the diagram represent the corresponding bridge configurations. Strain gauges of the same color form a full-bridge circuit; for example, the two green strain gauges in the upper right and the two green strain gauges in the lower left together form a Wheatstone bridge. Meanwhile, Figure 3 The colors in Figure 4 , Figure 5 The colors in the images correspond one-to-one, and the combination of the two images shows the position and orientation of the strain gauges.

[0088] Figures 9 to 13 This is a comparison chart of the predicted and actual values ​​of each load component. The horizontal axis represents the actual load value, and the vertical axis represents the predicted load value. Theoretically, an accurate prediction should fall on a 45° angled line; the closer the scatter points are to the 45° line, the more accurate the prediction. Figure 9This is a comparison chart of the predicted and actual values ​​of the normal force. Figure 10 This is a comparison chart of the predicted and actual lateral force values. Figure 11 This is a comparison chart of the predicted and actual rolling torque values. Figure 12 This is a comparison chart of the predicted and actual yaw moment values. Figure 13 This is a comparison chart of the predicted pitch moment and the actual value.

[0089] Figure 14 This is a graph evaluating the prediction performance of the correlation coefficient between the predicted values ​​and the actual values ​​of the linear model and the nonlinear modified model. The closer the correlation coefficient is to 1, the better the prediction performance. It can be clearly seen from the graph that the correlation coefficient of the nonlinear model is generally greater than that of the linear model.

[0090] Figure 15 This graph shows the variation of zero drift of strain with Mach number under two conditions: with and without a protective cover. Theoretically, it is desirable for the zero drift value to be 0, that is, no zero drift. However, in practice, because different incoming Mach numbers correspond to different flow field temperatures, the temperature field near the measuring point will be non-uniform, causing zero drift of strain. Furthermore, as the incoming Mach number increases, the temperature non-uniformity will increase, and the zero drift value will also increase. In order to reduce zero drift, a protective cover is added. It can be seen that the zero drift value is significantly smaller with a protective cover than without one.

[0091] Figure 16 This is a comparison chart of the accuracy of the prediction model presented in this paper and that of a traditional force balance. The comparison focuses on relative error, expressed as a percentage, representing the percentage of the measurement error relative to the true value. A relative error of 0 indicates that the prediction is completely accurate. In practice, the accuracy achievable by a traditional strain balance is represented by the yellow line in the chart. The method developed in this paper corresponds to the blue line for the linear model and the gray line for the modified model. As can be seen, the prediction accuracy of both the linear and nonlinear models is basically at the same level as that of the traditional balance. The linear model's error is slightly worse than that of the traditional balance, but it is acceptable. The modified model's prediction effect is comparable to that of the traditional balance, demonstrating the effectiveness of the method presented in this paper.

Claims

1. A model load self-sensing support device based on distributed strain, characterized in that, include: The measuring element segment has a regular octagonal cross-section and an axial through hole at its center; The front connection section, located at the front end of the measuring element section, is used to connect to the wind tunnel model; The rear connecting section, located at the rear end of the measuring element section, is used to connect with the wind tunnel support mechanism; A protective cover is fitted over the outside of the measuring element segment and fixedly connected to the front end of the measuring element segment to isolate high-speed airflow; Multiple sets of strain gauges are attached to the outer surfaces of the regular octagonal cross-section according to a predetermined layout, and are used to measure the strain signals of the wind tunnel model on the corresponding outer surfaces. And a wiring structure, which is set inside the measuring element segment and communicates with the axial through hole, is used to guide the strain gauge lead wire out from inside the measuring element segment; Multiple sets of strain gauges respectively collect strain signals from the corresponding outer surface. The strain signals are eliminated by a full-bridge bridging method to eliminate heat output caused by changes in ambient temperature, and then transmitted to the external acquisition device through the wiring structure. The strain signals are decoupled and calculated by a pre-generated calibration matrix to obtain the normal force, pitching moment, lateral force, yaw moment and rolling moment on the model.

2. The model load self-sensing support device based on distributed strain according to claim 1, characterized in that, The circumcircle diameter of the regular octagonal cross-section of the measuring element segment is 12mm, the axial length is 88mm, and each edge is provided with a rounded corner with a radius of 2mm; the diameter of the axial through hole is 8mm.

3. The model load self-sensing support device based on distributed strain according to claim 1, characterized in that, The protective cover adopts a clasp-like structure, consisting of two halves of the shell, with a labyrinthine overlapping structure at the joint surface. A gap is left between the free end of the protective cover and the tail support body, and a brush is pasted in this gap. The combination of labyrinthine overlapping and brush can effectively prevent high-speed airflow from entering the measuring element, while not transferring additional structural load.

4. The model load self-sensing support device based on distributed strain according to claim 1, characterized in that, The arrangement of the multiple sets of measuring strain gauges is as follows: Two strain gauges are attached to the upper and lower surfaces of a regular octagon to form a full bridge, which is used to measure the normal force and pitching moment. Two strain gauges are attached to the left and right surfaces of a regular octagon to form a full bridge, which is used to measure lateral force and yaw moment. Two strain gauges are attached to each of the first and second sets of symmetrical inclined planes of a regular octagon to form a full bridge for measuring rolling torque.

5. The model load self-sensing support device based on distributed strain according to claim 1, characterized in that, The wiring structure includes two radial wiring holes disposed on the measuring element segment. The radial wiring holes are connected to the axial through hole. All the measuring strain gauge leads are drawn into the axial through hole through the radial wiring holes and led out to the rear.

6. The model load self-sensing support device based on distributed strain according to claim 1, characterized in that, The front connecting section is equipped with an interface that mates with the rear cone of the wind tunnel model or a force balance. The interface is a 1:5 taper fit with key positioning.

7. The model load self-sensing support device based on distributed strain according to claim 1, characterized in that, The rear connecting section is equipped with a wind tunnel support mechanism interface.

8. The model load self-sensing support device based on distributed strain according to claim 1, characterized in that, The tail support device is made of 30CrMnSiA material, 7075-T6 aluminum alloy or QBe2 beryllium bronze.

9. The model load self-sensing support device based on distributed strain according to any one of claims 1 to 8, characterized in that, The size of the measuring element segment can be scaled proportionally to accommodate wind tunnel models of different sizes.

10. A method of using a model load self-sensing support device based on distributed strain, comprising using the model load self-sensing support device based on distributed strain as described in any one of claims 1 to 9, characterized in that, Includes the following steps: S1. Connect the tail support device to the model via the front connecting section and to the wind tunnel support mechanism via the rear connecting section. S2. Connect multiple sets of strain gauges to the strain acquisition device according to the predetermined bridge circuit; S3. During the wind tunnel test, the aerodynamic load on the wind tunnel model is transferred to the measuring element section, causing strain in the regular octagonal cross section. S4. Multiple sets of strain gauges collect strain signals of corresponding components. The strain signals are eliminated by the full-bridge group to eliminate the heat output caused by changes in ambient temperature, and then transmitted to the external acquisition device through the wiring structure. S6. The strain signal is decoupled and calculated using a pre-generated calibration matrix to obtain the normal force, pitching moment, lateral force, yaw moment and roll moment acting on the model.

Images