A seal whisker sensor array device and measurement system for flow velocity vector measurement

Abstract

The application discloses a kind of for flow velocity vector measurement imitates seal mustache sensor array device and measurement system.Sensor array device includes support base and multiple sensing elements arranged thereon, each sensing element has different installation reference direction, to obtain the multichannel response information under fluid action.Sensing element includes piezoresistive strain gauge and imitates seal mustache, imitates seal mustache has the structure of axial periodic change of elliptical section, it helps to weaken vortex-induced vibration interference.When fluid acts on imitates seal mustache, imitates seal mustache converts fluid load into root moment and is transmitted to piezoresistive strain gauge, makes it occur flexural deformation, in turn causes strain resistance wire strain and resistance value change, output electric signal.Multichannel electric signal is handled after signal conditioning, data acquisition and flow velocity vector identification module, realizes the joint identification of flow velocity and flow direction information.The application is compact, easy to integrate, suitable for underwater vehicle, marine environment monitoring and underwater flow field perception etc.Situations.

Description

Technical Field

[0001] This invention belongs to the field of underwater fluid parameter measurement technology, and particularly relates to a seal whisker-inspired sensor array device and measurement system for flow velocity vector measurement. Background Technology

[0002] Flow velocity and direction are important parameters describing the state of underwater flow fields and are key fluid characteristics for attitude control, path planning, and environmental perception of underwater vehicles. Existing flow velocity measurement methods mainly include mechanical current meters, electromagnetic current meters, acoustic Doppler current meters, and optical current measurement devices. Among them, mechanical current meters have a relatively simple structure, but they usually suffer from problems such as large size, easy wear of moving parts, and limited dynamic response; electromagnetic current meters rely on a conductive medium environment and are susceptible to interference from complex underwater electromagnetic signals; acoustic Doppler current meters have strong measurement capabilities, but the equipment structure is complex, and the size and power consumption are large, which is not conducive to the miniaturization and integration of compact underwater devices; optical measurement methods have high requirements for water transparency and illumination conditions, and are easily affected in turbid or low-light environments.

[0003] To address the aforementioned issues, biomimetic fluid sensing technology offers a novel research approach. Seal whiskers in nature possess a unique geometric structure with a non-circular cross-section and periodic undulations along the axial direction. This structure exhibits directional response characteristics under the influence of water flow and helps to mitigate eddy-induced vibration interference caused by background flow. Therefore, seal whisker-inspired structures have certain application potential in the field of fluid sensing.

[0004] However, most existing biomimetic flow velocity sensing structures use a single sensing unit for measurement. The output signal of a single sensing unit is affected by both the magnitude and direction of the flow velocity, making it difficult to obtain complete flow velocity vector information directly from a single channel signal. Furthermore, existing multi-sensor devices still have limitations in fusing and utilizing multi-directional fluid response information, and their ability to characterize the nonlinear relationship between multi-channel output signals and flow velocity vectors is limited. Therefore, it is still necessary to further optimize the sensor array structure and flow velocity vector recognition method to achieve joint recognition of flow velocity magnitude and direction. Summary of the Invention

[0005] To address the problems in existing technologies, such as the coupling of flow velocity magnitude and flow direction information in a single biomimetic sensing unit, the difficulty in directly obtaining the complete flow velocity vector, and the insufficient utilization of multi-directional fluid response information fusion by multiple sensing devices, this invention provides a seal whisker-inspired sensor array device and measurement system for flow velocity vector measurement.

[0006] This invention provides a seal whisker-inspired sensor array device for flow velocity vector measurement, comprising a support base and multiple sensing elements disposed on the support base. The number of sensing elements is no less than three, and each sensing element has a different mounting reference orientation to acquire multi-channel response information under fluid action. Each sensing element includes a piezoresistive strain gauge and a seal whisker-inspired structure, with the bottom end of the seal whisker-inspired structure connected to the piezoresistive strain gauge. The seal whisker-inspired structure has an elliptical cross-section structure that varies periodically along the axial direction. Each sensing element is configured to independently output an electrical signal to identify the magnitude and direction of the fluid flow through the differences in the multi-channel electrical signals.

[0007] Preferably, the support base is an annular base, and the plurality of sensing elements are evenly distributed along the circumference of the annular base.

[0008] Preferably, the number of sensing elements is 6, and the central angle between two adjacent sensing elements is 60°.

[0009] Preferably, the major axis of the cross-section of the imitation seal whiskers in each of the sensing elements is aligned with the radial direction of the annular base, so that the reference directions of each imitation seal whisker are spatially staggered.

[0010] Preferably, each of the sensing elements further includes a sub-base, the piezoresistive strain gauge is disposed on the sub-base, and the bottom end of the seal whisker-like element is fixedly connected to the central region of the piezoresistive strain gauge.

[0011] Preferably, the piezoresistive strain gauge includes a capping layer, a substrate, and four strain resistance wires disposed between the capping layer and the substrate; the capping layer and the substrate are made of flexible insulating film material to form a sandwich composite structure, and the four strain resistance wires are symmetrically arranged around the center of the seal whiskers and electrically connected by a Wheatstone full-bridge circuit.

[0012] This invention also provides a flow velocity vector measurement system, including the aforementioned seal whisker-like sensor array device for flow velocity vector measurement, and further including a signal conditioning module, a data acquisition module, and a flow velocity vector recognition module. The signal conditioning module is configured to amplify and filter the multi-channel electrical signals output by the seal whisker-like sensor array; the data acquisition module is configured to acquire the multi-channel data processed by the signal conditioning module; and the flow velocity vector recognition module is configured to identify the magnitude and direction of the fluid flow based on the multi-channel data acquired by the data acquisition module.

[0013] Preferably, the signal conditioning module includes an instrumentation amplifier circuit, a low-pass filter circuit, and a non-inverting amplifier circuit that are connected in sequence.

[0014] Preferably, the flow velocity vector recognition module employs a backpropagation neural network model optimized by a genetic algorithm, wherein the genetic algorithm is used to optimize the initial weights and thresholds of the backpropagation neural network model.

[0015] Preferably, the flow velocity vector identification module takes the time-domain characteristics of the multi-channel electrical signal as input, and takes the flow velocity magnitude, the sinusoidal component of the flow direction, and the cosine component of the flow direction as output, and determines the flow direction angle based on the sinusoidal component of the flow direction and the cosine component of the flow direction.

[0016] Compared with the prior art, the present invention has the following beneficial effects: First, this invention arranges multiple seal whisker-like sensing elements with axially periodic elliptical cross-sections in a circumferential array along a ring base, allowing different sensing elements to have different reference angles relative to the incoming flow direction, thereby obtaining multi-channel differentiated response information. This structure helps reduce the impact of coupling flow velocity and flow direction information in a single sensing unit on the recognition results, providing multi-dimensional input features for the joint recognition of flow velocity and flow direction.

[0017] Second, based on the basic spatial configuration required for flow velocity vector recognition, this invention introduces multiple sensing elements to form a circumferentially redundant layout. This array configuration can reduce the impact of local flow disturbances, single-channel fluctuations, or individual element response deviations on the recognition results, thereby improving the stability and reliability of flow velocity vector recognition.

[0018] Third, this invention employs a sensing element structure integrating a seal whisker-like structure, a piezoresistive strain gauge, and a sub-base. The seal whisker-like structure, under the influence of fluid, transmits the torque at its root to the piezoresistive strain gauge, causing it to flex and deform. The resulting change in the resistance of the strain gauge wire outputs an electrical signal. This sensing element structure is relatively simple, with no external rotating parts, facilitating multi-element array integration within a limited installation space.

[0019] Fourth, the present invention amplifies and filters the multi-channel weak differential voltage signal output by the sensor array through a signal conditioning module, which can improve the signal-to-noise ratio and acquisition quality, reduce the impact of power frequency electromagnetic interference and high-frequency noise on the output signal, and provide stable multi-channel input data for subsequent flow velocity vector identification.

[0020] Fifth, this invention utilizes a flow velocity vector recognition module to establish a nonlinear mapping relationship between multi-channel electrical signals and flow velocity vectors, and decomposes the flow direction angle into a sine component and a cosine component for recognition. This reduces the impact of 0° / 360° boundary discontinuities on flow direction recognition, enables the joint output of flow velocity magnitude and flow direction information, and improves the applicability of the sensor array in underwater flow field perception scenarios. Attached Figure Description

[0021] Figure 1This is a schematic diagram of the overall structure of the seal whisker-like sensor array device provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the sensing element structure provided in an embodiment of the present invention; Figure 3 A schematic diagram of the axial periodic undulations and elliptical cross-section structure of the imitation seal whiskers provided in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the flow direction definition provided in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the principle of velocity vector sensing and the change of feature length of the sensing element provided in an embodiment of the present invention. Figure 6 This is a schematic diagram of the strain gauge wire arrangement provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the Wheatstone full-bridge circuit connection provided in an embodiment of the present invention; Figure 8 This is a block diagram of the flow velocity vector measurement system provided in an embodiment of the present invention; Figure 9 The figure shows the simulation results of the average strain of the piezoresistive strain gauge as a function of flow velocity under different flow direction conditions, as provided in the embodiments of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. This invention provides a seal whisker-inspired sensor array device and measurement system for flow velocity vector measurement. To further illustrate the structure of this invention, detailed description is provided below with reference to the accompanying drawings: Example 1 This embodiment provides a seal whisker-inspired sensor array device for flow velocity vector measurement. Please refer to [link to relevant documentation]. Figure 1The array includes a support base 1 and multiple sensing elements 2 disposed on the support base. The number of sensing elements is no less than three, and each sensing element has a different mounting reference direction. The support base 1 is an annular base, and the multiple sensing elements 2 are evenly distributed along the circumference of the annular base. Preferably, the number of sensing elements 2 is six, and the central angle between two adjacent sensing elements 2 is 60°. In other embodiments, the number of sensing elements 2 can be adjusted according to measurement requirements, installation space, and the number of signal channels, preferably no less than three. To ensure the rigidity and geometric stability of the array structure under underwater high pressure or strong current impact environments, and to prevent common-mode interference coupling caused by base deformation in each channel, the annular base is integrally formed from a hard, corrosion-resistant insulating material or a metal material with an insulating layer on its surface. The annular base has internal cable trays to accommodate the independently led-out electrical signal transmission lines of each sensing element. In this embodiment, the overall outer diameter of the sensor array is preferably 135 mm, the height is preferably 76.5 mm, and the overall mass is approximately 75 g.

[0023] Each of the aforementioned sensing elements includes a base 2-1, a piezoresistive strain gauge, and a seal whisker-like component 2-2. (See also...) Figure 2 The piezoresistive strain gauge is disposed on the sub-base, and the bottom end of the seal whisker-like structure is fixedly connected to the central region of the piezoresistive strain gauge. The seal whisker-like structure has an elliptical cross-section structure that varies periodically along the axial direction. The major axis direction of the cross-section of the seal whisker-like structure in each of the sensing elements coincides with the radial direction of the annular base, so that the reference directions of each seal whisker-like structure are spatially staggered. Each of the sensing elements is configured to independently output an electrical signal, so as to collaboratively identify the flow velocity and direction of the fluid through the differences in multi-channel electrical signals.

[0024] In terms of spatial arrangement, by uniformly arranging six sensing elements around the circumference of the annular base, and aligning the major axis of each seal whisker-like cross-section with the radial direction of the annular base, different sensing elements can form different reference orientations in space. When an external flow field acts on this array, the local angles between each sensing element and the incoming flow direction differ, resulting in variations in the fluid action and root bending moment experienced by each element, thus generating multi-channel differentiated electrical signal outputs. Comprehensive analysis of the multi-channel output signals provides input features for the subsequent joint identification of flow velocity and direction information.

[0025] To improve the force transmission efficiency between the seal whiskers and the piezoresistive strain gauge, the sub-base is used to support and constrain the piezoresistive strain gauge; the bottom end of the seal whiskers is provided with a connecting end, which is fixedly connected to the capping surface of the central region of the piezoresistive strain gauge. Thus, the bending moment generated by the seal whiskers under the action of fluid can be transmitted to the sensitive area of ​​the piezoresistive strain gauge, causing the piezoresistive strain gauge to flex and deform, and resulting in a change in the resistance of the strain gauge wire.

[0026] Please see Figure 2 The piezoresistive strain gauge includes a capping layer 2-3, a substrate 2-4, and four strain resistance wires 2-5 disposed between the capping layer and the substrate. The capping layer and the substrate are constructed as a sandwich composite structure using a flexible insulating thin film material. The four strain resistance wires are symmetrically arranged around the center of the seal whiskers and electrically connected by a Wheatstone full-bridge circuit. In terms of microstructure and process adaptation, the capping layer and the substrate are made of polyimide (PI) film or polydimethylsiloxane (PDMS) flexible substrate with excellent mechanical toughness, insulating resistance, and low Young's modulus. The strain resistance wires are deposited and patterned on the substrate using magnetron sputtering or photolithography processes of microelectromechanical systems (MEMS), and their material is constantan, platinum-tungsten alloy, or semiconductor-doped silicon material with a high resistance strain sensitivity coefficient. In this embodiment, the radius of the sensitive region of the piezoresistive strain gauge is preferably 10 mm, and the thickness of the PI film is preferably 100 μm.

[0027] The four strain gauge wires are spatially distributed within the inner and outer sensitive areas of the piezoresistive strain gauge. Specifically, the first and second strain gauge wires are located in the inner sensitive area, respectively on either side of the center of the seal whisker-like mounting; the third and fourth strain gauge wires are located in the outer sensitive area, respectively on either side of the center of the seal whisker-like mounting (see...). Figure 6 Four strain gauge wires are symmetrically arranged around the mounting center of the simulated seal whiskers and connected using a Wheatstone bridge. When the simulated seal whiskers transmit the root torque to the piezoresistive strain gauge under fluid load, the strain gauge undergoes flexural deformation. The inner and outer sensitive areas produce different strain responses, and the strain gauge wires on opposite sides of the same sensitive ring can generate tensile or compressive strain respectively, thus causing a change in resistance value. The deformation of the piezoresistive strain gauge is converted into a bridge output voltage signal through a Wheatstone bridge circuit.

[0028] Please see Figure 3The simulated seal whiskers have an elliptical cross-section structure that varies periodically along the axial direction. Specifically, the cross-section of the simulated seal whiskers is elliptical, and the major and minor axes of the elliptical cross-section vary periodically along the axial direction of the simulated seal whiskers, thus forming an axially periodically undulating structure that alternates between a flat, elongated elliptical cross-section and a short, wide elliptical cross-section. This structure can simulate the non-circular cross-section and axially periodically undulating characteristics of seal whiskers, helping to reduce eddy-induced vibration interference under the action of water flow and improve the stability of the output signal of the sensing element. Simultaneously, since the equivalent characteristic length of the simulated seal whiskers differs under different incoming flow directions, the fluid action and root bending moment it experiences also vary with the incoming flow direction, thus enabling the sensing element to produce differentiated responses to changes in flow velocity and direction. In this embodiment, the simulated seal whiskers can be integrally formed using 3D printing technology, with photosensitive resin being the preferred printing material. The height of the simulated seal whiskers... Preferably, the axial distance between adjacent characteristic elliptical cross sections is 10.92 mm. Preferably 1.82 mm; wherein the semi-major axis of the flattened elongated elliptical cross-section and short half shaft The preferred semi-axis sizes are 1.19 mm and 0.48 mm, respectively, for the short, stout elliptical cross-section. and short half shaft The preferred sizes are 0.95 mm and 0.58 mm, respectively. These dimensional parameters can be adjusted according to the actual flow rate measurement range, sensor installation space, and processing conditions.

[0029] Example 2 This embodiment also provides a flow velocity vector measurement system, including the seal whisker-like sensor array device for flow velocity vector measurement as described above, and further including a signal conditioning module, a data acquisition module, and a flow velocity vector recognition module. The signal conditioning module amplifies and filters the multi-channel electrical signals output by the sensor array; the data acquisition module acquires the multi-channel data processed by the signal conditioning module; and the flow velocity vector recognition module identifies the magnitude and direction of the fluid flow based on the acquired multi-channel data. In practical applications, this system can be integrated into the bow or flow field sensing surface of an underwater vehicle. When the vehicle is moving underwater or in a steady flow field, the fluid acts on multiple seal whisker-like structures with axial periodic undulations. The Wheatstone full-bridge circuit inside each sensing element converts the deformation of the piezoresistive strain gauge into a weak differential voltage signal, which is then input to the subsequent signal conditioning module, data acquisition module, and flow velocity vector recognition module via a signal transmission line, thereby achieving joint recognition of flow velocity magnitude and direction information.

[0030] The signal conditioning module includes an instrumentation amplifier circuit, a low-pass filter circuit, and a non-inverting amplifier circuit connected in sequence. Due to potential power frequency noise, electromagnetic interference, and mechanical vibration interference caused by the vehicle's propulsion system in the underwater environment, the raw electrical signal directly output by the sensing element is typically weak, with amplitudes in the microvolt (μV) range, making direct high-resolution analog-to-digital conversion unsuitable. Therefore, in this embodiment, the instrumentation amplifier circuit uses an instrumentation amplifier with a high common-mode rejection ratio to form a preamplifier stage, used for primary amplification of the weak differential signal output from the closed-loop bridge circuit and suppression of common-mode interference; the low-pass filter circuit is connected to the output of the instrumentation amplifier circuit to filter out high-frequency noise outside the effective signal frequency band; the non-inverting amplifier circuit serves as a subsequent conditioning circuit, used to further amplify and condition the filtered signal, adapting its amplitude range to the input range of the subsequent data acquisition module, and also providing impedance matching and enhanced output drive capability.

[0031] The data acquisition module is connected to the output of the in-phase amplifier circuit and may include a multi-channel sampling circuit and an analog-to-digital converter (ADC) for converting the conditioned multi-channel analog voltage signal into a digital signal and transmitting it to the flow velocity vector recognition module.

[0032] The flow velocity vector recognition module employs a backpropagation (BP) neural network model optimized by a genetic algorithm (GA). The genetic algorithm is used to optimize the initial weights and thresholds of the backpropagation neural network model. For the nonlinear mapping relationship between multi-channel electrical signals and flow velocity vectors, the training results of traditional BP neural networks are easily affected by the selection of initial weights and thresholds. To improve the stability of model training, this embodiment introduces a genetic algorithm to optimize the initial parameters of the BP neural network. Specifically, the connection weights and thresholds of the BP neural network are encoded as chromosomes, and iterative optimization is performed through operations such as selection, crossover, and mutation to obtain a better combination of initial weights and thresholds. Subsequently, this set of parameters is assigned to the BP neural network, and training is performed using an error backpropagation algorithm. Through this method, a nonlinear mapping relationship between multi-channel electrical signals and flow velocity vectors can be established, improving the training stability and recognition capability of the flow velocity vector recognition module.

[0033] Specifically, the flow velocity vector recognition module takes the time-domain characteristics of the multi-channel electrical signal as input and outputs the flow velocity magnitude, the sinusoidal component of the flow direction, and the cosine component of the flow direction, and determines the flow direction angle based on the sinusoidal and cosine components of the flow direction. In constructing the algorithm output, directly using the flow direction angle as a single output variable is easily affected by the discontinuity at the 0° / 360° boundary. To reduce the impact of this boundary discontinuity on model training and recognition results, this embodiment uses the flow direction angle to be measured... Decomposed into sinusoidal components of the flow direction. and the cosine component of the flow The network model predicts the two components mentioned above, and finally calculates the flow direction angle using the four-quadrant arctangent function. By using the above-mentioned sine and cosine component output methods, the continuity and stability of the flow direction angle recognition process can be improved.

[0034] The response process of the seal whisker-like sensor array under fluid action can be described by the mapping relationship between fluid load, structural bending deformation, and bridge output, as follows: like Figure 5 As shown, due to the unique axially periodically varying elliptical cross-sectional structure of the seal whiskers, its projected geometry changes in a specific flow direction. The equivalent feature length of a single sensing element... satisfy:

[0035] In the formula, and These are the major and minor semi-axises of the elongated elliptical cross-section of the simulated seal whiskers, respectively. and These are the major and minor semi-axises of the short, wide elliptical cross-section, respectively. The angle of the incoming flow direction relative to the reference direction of the simulated seal whiskers.

[0036] When density is The fluid flows at a constant velocity When fluid flows around a single sensing element, the fluid resistance experienced by the seal whisker-like surface under fluid load is... satisfy:

[0037] In the formula, The drag coefficient, Let be the total axial vertical height of the simulated seal whiskers. Since the fluid resistance acts as a distributed load along the whisker height, its equivalent point of application can be approximated as being located at half the vertical height. Therefore, the combined bending moment generated by this fluid resistance on the piezoresistive strain gauge at the base of the simulated seal whiskers is... satisfy:

[0038] The bending moment at the root is transmitted to the sensitive area of ​​the piezoresistive strain gauge below through the connecting end at the bottom of the seal whisker-like structure, causing the piezoresistive strain gauge to flex and deform. The maximum deflection generated at the edge of the contact area at the center of the moment is... satisfy:

[0039] In the formula, Let the radius of the contact area be the equivalent circle radius. The ratio of the outer radius of the sensitive area of ​​the piezoresistive strain gauge to the equivalent circle radius of the contact area is given. The bending stiffness of the piezoresistive strain gauge is given.

[0040] As the piezoresistive strain gauge flexes and deforms, the four strain resistance wires positioned between the substrate and the capping layer undergo corresponding tensile or compressive strain, resulting in a change in resistance. The initial resistance of each of the four strain resistance wires is... The changes in their resistance are respectively , , and .like Figure 6 As shown, the first strain resistance wire R 1 and second strain resistance wire R 2 are set in the inner ring sensitive area, and are located on both sides of the imitation seal whisker mounting center; the third strain resistance wire R 3 and the fourth strain resistance wire R 4 are positioned in the outer ring sensitive area, respectively on both sides of the center of the seal whisker-like mounting. Based on the piezoresistive effect of semiconductor or metallic materials, the change in resistance of each resistance wire is expressed as... (in (This refers to the sensitivity coefficient of the resistance wire). For example... Figure 7 As shown, these four strain gauge wires are electrically connected to form a Wheatstone full-bridge circuit, under a constant bias supply voltage. Below, the output voltage of the sensing element satisfy: ; Based on the relationship between the resistance change and the output voltage of the Wheatstone full-bridge circuit, under the above-mentioned symmetrical arrangement and minor deformation conditions, the bridge output voltage can be approximately expressed as: ; According to the thin plate bending theory, under fixed geometry, boundary conditions, and a small deflection range, the equivalent strain of the sensitive region where the strain resistance wire is located is approximately linearly related to the maximum deflection of the piezoresistive strain gauge, which can be expressed as: ; In the formula, , These are the equivalent proportionality coefficients for the sensitive regions of the outer and inner rings, respectively (mainly determined by the geometric parameters of the piezoresistive strain gauge, the layout of the resistance wire, and the flow direction, and are independent of the flow velocity).

[0041] Finally, the analytical output voltage of a single sensing element can be obtained. for: ; From the above equation, it can be seen that, under the conditions that the structural parameters of the piezoresistive strain gauge, the bridge connection method, and the supply voltage are determined, the output voltage of the sensing element is... With flow rate and feature length The characteristic length varies with the direction of the incoming flow. Therefore, the output voltage of a single sensing element simultaneously contains information about both the flow velocity and the flow direction.

[0042] For a ring array consisting of six sensing elements, the six sensing elements can be labeled clockwise from a specific reference direction as the first sensing element to the sixth sensing element (S1-S6). Taking the radial direction of the ring base where the third sensing element (S3) is located as the 0° spatial reference direction, since each sensing element has a fixed angle of 60° on the circumference, the corresponding installation angle of each sensing element is... These are specific fixed spatial phase values. At this time, the first... Local feature length corresponding to each sensing element This can be uniformly represented by the following expression that includes the installation angle offset: ; In the formula, , for the first The mounting angle of each sensing element relative to the reference direction. In this embodiment, the six sensing elements are evenly distributed around the circumference of the annular base, and the central angle between two adjacent sensing elements is 60°. Therefore, the output signal of each channel (i.e., the voltage amplitude difference of each sensing element independently output) exhibits different and regular nonlinear spatial misalignment changes with the direction of incoming flow.

[0043] For an array consisting of six sensing elements, each sensing element outputs an independent voltage signal, forming a multi-channel response feature that includes information on flow velocity magnitude and inflow direction. Given fixed sensor array structural and electrical parameters, the flow velocity vector recognition module inputs the six-channel time-domain features into a neural network model optimized by a genetic algorithm, establishing a nonlinear mapping relationship between the multi-channel electrical signals and the flow velocity vector, thereby achieving joint recognition of flow velocity magnitude and flow direction angle.

[0044] like Figure 8 The flow velocity vector measurement system provided in this embodiment uses a seal whisker-like sensor array to output multi-channel electrical signals. The signal conditioning module amplifies and filters the multi-channel electrical signals, the data acquisition module acquires and converts the conditioned multi-channel electrical signals, and the flow velocity vector recognition module identifies the flow velocity and direction information of the external fluid based on the multi-channel electrical signals.

[0045] The signal conditioning module may include an instrumentation amplifier circuit, a low-pass filter circuit, and a non-inverting amplifier circuit. The instrumentation amplifier circuit is used to perform primary amplification of the weak differential signal output from the sensor array and suppress common-mode interference. The low-pass filter circuit is used to filter out high-frequency noise outside the effective signal frequency band. The non-inverting amplifier circuit is used to further amplify and condition the filtered signal. The data acquisition module is used to complete multi-channel signal acquisition, A / D conversion, and data transmission.

[0046] The flow velocity vector recognition module can employ a backpropagation neural network model optimized by a genetic algorithm. This model takes the multi-channel electrical signal characteristics output by the sensor array as input and outputs flow velocity magnitude and flow direction information. Preferably, the flow velocity vector recognition module takes the time-domain characteristics of the multi-channel electrical signal as input and outputs flow velocity magnitude, sinusoidal flow direction components, and cosine flow direction components, and calculates the flow direction angle based on the sinusoidal and cosine flow direction components. Through this method, a nonlinear mapping relationship can be established between the multi-channel response of the sensor array and the flow velocity vector, achieving joint recognition of flow velocity magnitude and flow direction angle.

[0047] like Figure 9 As shown, under different flow direction conditions, the average strain of the piezoresistive strain gauge increases with increasing flow velocity, indicating that the sensing element has the ability to respond to changes in flow velocity. Under the same flow velocity conditions, the average strain differs for different flow directions, indicating that the seal whiskers analog exhibits a differentiated response to the incoming flow direction. This result shows that the output signal of the sensing element simultaneously contains information on both flow velocity magnitude and flow direction, which can provide a basis for subsequent multi-channel array identification of flow velocity vectors.

[0048] Example 3 This embodiment provides a flow velocity vector measurement method based on a seal whisker-inspired sensor array device. The method, implemented using the aforementioned seal whisker-inspired sensor array device and measurement system, includes the following steps: S1. Install the seal whisker-like sensor array device in the flow field to be measured, so that multiple sensing elements correspond to different installation reference directions. S2. The fluid to be measured acts on the seal whiskers in each sensing element. The seal whiskers convert the fluid load into root torque and transmit it to the piezoresistive strain gauge. S3. The piezoresistive strain gauge undergoes flexural deformation under the action of the root torque, causing the strain resistance wire to produce strain and resistance changes, and outputs the corresponding electrical signal through the Wheatstone full bridge circuit. S4, the signal conditioning module amplifies and filters the multi-channel electrical signals output by multiple sensing elements; S5. The data acquisition module acquires multi-channel data processed by the signal conditioning module; S6. The flow velocity vector recognition module extracts the time-domain features of the multi-channel data and inputs the time-domain features into the flow velocity vector recognition model; S7. The flow velocity vector recognition model outputs the flow velocity magnitude, the sinusoidal component of the flow direction, and the cosine component of the flow direction based on the multi-channel time domain characteristics, and calculates the flow direction angle based on the sinusoidal component of the flow direction and the cosine component of the flow direction.

[0049] Through the above steps, multi-channel differentiated response information can be obtained using sensing elements with different installation reference directions, and combined with the flow velocity vector recognition model to achieve joint measurement of flow velocity magnitude and flow direction information.

[0050] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A seal whisker-inspired sensor array device for flow velocity vector measurement, characterized in that, include: Support base; And a plurality of sensing elements disposed on the support base, wherein the number of sensing elements is not less than three, and each of the sensing elements has a different mounting reference direction, so as to obtain multi-channel response information under the action of fluid. Each of the sensing elements includes a piezoresistive strain gauge and a seal whisker-like structure, the bottom end of which is connected to the piezoresistive strain gauge; the seal whisker-like structure has an elliptical cross-section structure that varies periodically along the axial direction. Each of the aforementioned sensing elements is configured to output an independent electrical signal, so as to collaboratively identify the flow velocity and direction of the fluid through the differences in the multi-channel electrical signals.

2. The seal whisker-inspired sensor array device for flow velocity vector measurement according to claim 1, characterized in that, The support base is an annular base, and the plurality of sensing elements are evenly distributed along the circumference of the annular base.

3. The seal whisker-inspired sensor array device for flow velocity vector measurement according to claim 2, characterized in that, The number of sensing elements is 6, and the central angle between two adjacent sensing elements is 60°.

4. The seal whisker-inspired sensor array device for flow velocity vector measurement according to claim 2, characterized in that, The major axis of the cross-section of the imitation seal whiskers in each of the aforementioned sensing elements is aligned with the radial direction of the annular base, so that the reference directions of each imitation seal whisker are spatially staggered.

5. The seal whisker-inspired sensor array device for flow velocity vector measurement according to claim 1, characterized in that, Each of the sensing elements also includes a sub-base, on which the piezoresistive strain gauge is disposed, and the bottom end of the seal whisker-like element is fixedly connected to the central region of the piezoresistive strain gauge.

6. The seal whisker-inspired sensor array device for flow velocity vector measurement according to claim 1, characterized in that, The piezoresistive strain gauge includes a capping layer, a substrate, and four strain resistance wires disposed between the capping layer and the substrate; the capping layer and the substrate are made of flexible insulating film material to form a sandwich composite structure, and the four strain resistance wires are symmetrically arranged around the center of the seal whiskers and electrically connected by a Wheatstone full-bridge circuit.

7. A flow velocity vector measurement system, characterized in that, The seal whisker-like sensor array device for flow velocity vector measurement as described in any one of claims 1 to 6 further includes: The signal conditioning module is configured to amplify and filter the multi-channel electrical signals output by the simulated seal whiskers sensor array device. The data acquisition module is configured to acquire multi-channel data processed by the signal conditioning module; The velocity vector recognition module is configured to identify the velocity magnitude and direction information of the fluid based on the multi-channel data acquired by the data acquisition module.

8. The velocity vector measurement system according to claim 7, characterized in that, The signal conditioning module includes an instrumentation amplifier circuit, a low-pass filter circuit, and a non-inverting amplifier circuit that are connected in sequence.

9. The velocity vector measurement system according to claim 7, characterized in that, The flow velocity vector recognition module employs a backpropagation neural network model optimized by a genetic algorithm, which is used to optimize the initial weights and thresholds of the backpropagation neural network model.

10. The velocity vector measurement system according to claim 7, characterized in that, The flow velocity vector recognition module takes the time-domain characteristics of the multi-channel electrical signal as input, and outputs the flow velocity magnitude, the sinusoidal component of the flow direction, and the cosine component of the flow direction, and determines the flow direction angle based on the sinusoidal component of the flow direction and the cosine component of the flow direction.

Images