Load positioning method based on screen printing structure integrated sensing network

Abstract

The invention discloses a load positioning method based on a screen printing structure integrated sensing network, and belongs to the field of aircraft intelligent skin. Comprising the following steps: (1) setting a sensing network: setting a sensor array and a signal lead array on the same printing screen; (2) sensor network integration: integrating the sensor network set in the step (1) onto a skin structure; and (3) load positioning monitoring: acquiring signals of the sensing network, and performing mode recognition on the acquired signals to realize a load positioning function. According to the invention, the sensor array and the large-area signal lead are directly integrated on the surface of the structure through the silk-screen printing technology, so that the method has the advantages of simple process, high adaptability and the like, and is beneficial to realizing the integration of a large-area and light-weight sensor network and an aircraft structure.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent skins and intelligent materials for aircraft, and particularly to a load positioning method based on a screen-printed structure integrated sensing network. Background Art

[0002] The intelligent skin of an aircraft is one of the important applications of intelligent materials and structure technologies in the aerospace field, and is a revolutionary new technology that changes the design of future advanced aircraft. By integrating sensor devices, drive devices, and microprocessors into the aircraft skin structure, electrical signals are collected by the sensor devices and combined with signal processing, enabling the aircraft structure to have capabilities such as self-diagnosis, self-adaptation, self-learning, and self-repair, thus meeting the requirements for aircraft structure state assessment.

[0003] During the service of an aircraft, the structure surface is vulnerable to the influence of strong overloads generated by the surrounding harsh environment, resulting in varying degrees of damage inside the structure. For example, snow accumulation, lightning strikes, or impact loads such as bird strikes and collisions with aircraft debris can cause damage such as fiber fractures, delaminations, metal structure fractures, and local yield deformations in the composite material structure of the structure. This may lead to a significant decline in the mechanical properties of aircraft components such as the fuselage and wings, directly affecting the service safety of the aircraft. In addition, the aircraft usually uses large-sized composite material structures in parts such as the wings, fuselage, vertical tail, and horizontal tail. With a large load-bearing range, it is necessary to monitor the load over a large area. To accurately evaluate the structural health status and perform repairs as appropriate, it is crucial to obtain the load conditions of the structure in a timely manner.

[0004] In actual structural health monitoring, the load positioning of the structure is often achieved by integrating sensors with the structure, collecting relevant conductive or dielectric properties such as the strain and pressure of the structure, and combining with a positioning algorithm. Currently, common sensors include strain gauges, fiber gratings, piezoelectric wafers, etc., which are integrated with the structure through surface paste type or embedded structure type. This method of integrating one by one and point by point has problems such as low efficiency and high cost, which is not conducive to the realization of a large-area and lightweight sensing network. Summary of the Invention

[0005] The present invention aims to provide a load positioning method based on a screen-printed structure integrated sensing network to achieve the layout of a large-area and lightweight sensing network.

[0006] To achieve the above object, the technical solution of the present invention includes the steps:

[0007] (1) Sensing network setting: Set a sensor array and a signal lead array on the same printing screen.

[0008] (2) Sensing network integration: Integrate the sensing network set in (1) onto the skin structure.

[0009] (3) Load positioning monitoring, collecting signals of the sensing network, and performing pattern recognition on the collected signals to achieve the load positioning function.

[0010] In one embodiment, the printing stencil is a nylon polyester fiber screen printing stencil with 200 meshes and oil-based adaptability.

[0011] In one embodiment, the sensor array includes 9 sensors of the same size, with a lateral spacing of 165 mm and a longitudinal spacing of 125 mm between adjacent two sensors.

[0012] In one embodiment, the gate length of the sensor is 20 mm, the gate pitch is 1.4 mm, the line width is 0.3 mm, and the number of gates is 12.

[0013] In one embodiment, the printing material of the sensor array is a high-impedance conductive carbon paste.

[0014] In one embodiment, the signal lead array includes 9 signal leads, the line width of the signal leads is 0.6 mm, and the minimum spacing is 2 mm.

[0015] In one embodiment, the printing material of the signal lead array is a low-impedance conductive silver paste.

[0016] In one embodiment, the sensors and the signal leads are connected in one-to-one correspondence.

[0017] In one embodiment, in step (3), a neural network is used for the pattern recognition method to achieve the positioning function.

[0018] In one embodiment, the neural network pattern recognition method is one or several of BP neural network, convolutional neural network, long short-term memory neural network, etc.

[0019] Advantageous effects: The load positioning method of the present invention based on the screen printing structure integrated sensing network directly integrates the sensor array and the large-area signal leads on the structure surface through screen printing technology. This method has the advantages of large-area manufacturing, batch manufacturing, simple equipment, low cost, good adaptability of the substrate and ink, etc., which is beneficial to the preparation of the flexible sensor network and the integration with the structure. In addition, the load positioning method using pattern recognition can learn the mutual relationship between the input data and the target category without having to reveal and describe the mutual relationship between the input and output in advance before learning, which is beneficial to solving the problem that it is difficult to determine the input-output relationship mapping due to unstable resistance.

[0020] To make the above features and advantages of the invention more obvious and understandable, the following specific embodiments are given and detailed descriptions are made in conjunction with the accompanying drawings as follows. Description of the Drawings

[0021] Figure 1 This is the flowchart of the structure-integrated sensor network based on screen printing and its load positioning method in the present invention.

[0022] Figure 2 This is the schematic diagram of the application of the structure-integrated sensing network based on screen printing to the skin structure in the present invention.

[0023] Figure 3 This is the schematic diagram of the structure of the structure-integrated sensing network based on screen printing in the present invention.

[0024] Figure 4 This is the schematic diagram of the sensor array in the present invention.

[0025] Figure 5 This is the schematic diagram of the signal lead array in the present invention.

[0026] Figure 6 This is the schematic diagram of the integrated state of the sensor network in the present invention.

[0027] Figure 7 This is the schematic diagram of the orientation of a single sensor in the present invention.

[0028] Figure 8 This is the voltage response diagram when loading the head and tail orientations of the sensor in the present invention.

[0029] Figure 9 This is the voltage response diagram when loading the left and right side orientations of the sensor in the present invention.

[0030] Figure 10 This is the loading position diagram of the 16 equal division regions of the skin structure in the present invention.

[0031] Figure 11 This is the schematic diagram of the neural network architecture for load positioning monitoring in the present invention.

[0032] Figure 12(a) is the confusion matrix diagram of the prediction results of the large-area load positioning by the sensing network.

[0033] Figure 12(b) is the experimental result diagram of the large-area load positioning by the sensing network. Detailed implementation manners

[0034] To make the objectives and technical solutions of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions of the embodiments of the present invention with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0035] In the following embodiments, many details are described to enable a better understanding of the present invention. However, those skilled in the art can easily recognize that some of these features can be omitted in different situations, or can be replaced by other components, materials, and methods. In some cases, some operations related to the present invention are not shown or described in the specification to avoid overwhelming the core part of the present invention with excessive descriptions. For those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations based on the descriptions in the specification and the general technical knowledge in the art.

[0036] Taking a typical epoxy material aircraft skin panel as an example, the load positioning method of the present invention based on the screen-printed structure integrated sensing network is specifically introduced. However, the material of the aircraft skin panel is only used as an example in the present invention and does not limit the present invention.

[0037] As Figure 1 shown, the present invention provides a load positioning method based on the screen-printed structure integrated sensing network, including the steps of:

[0038] (1) Sensing network setting;

[0039] Please combine Figures 2 to 5 , the sensing network 1 includes a sensor array 11 and a signal lead array 12. According to the actual size, material, and other parameters of the skin structure 2, the sensor array 11 and the signal lead array 12 are set on the same printing screen 13.

[0040] The printing screen 13 is a 200-mesh nylon polyester fiber screen printing screen with an oil-based adaptability.

[0041] As Figure 4 shown, in this embodiment, the sensor array 11 includes 9 sensors of the same size, namely sensor 111, sensor 112, sensor 113, sensor 114, sensor 115, sensor 116, sensor 117, sensor 118, and sensor 119. The above 9 sensors form a 3×3 array, and the lateral spacing between adjacent two sensors is 165 mm, and the longitudinal spacing is 125 mm.

[0042] In this embodiment, the gate length of the sensor is 20 mm, the gate pitch is 1.4 mm, the line width is 0.3 mm, and the number of gates is 12.

[0043] In this embodiment, the printing material of the sensor array 11 is a high-impedance conductive carbon paste. Further, it can be the CH-8 conductive carbon paste of Shijiao in Japan, and its resistivity is 1.00×10 -2 Ω·cm, and the viscosity is 30000 mPa·s.

[0044] As Figure 5 shown, corresponding to the sensor array 11, the signal lead array 12 includes nine signal leads, namely signal lead 21, signal lead 22, signal lead 23, signal lead 24, signal lead 25, signal lead 26, signal lead 27, signal lead 28, and signal lead 29.

[0045] In this embodiment, the line width of the signal lead is 0.6 mm, and the minimum pitch is 2 mm. Both the line width and the pitch are within the process preparation range.

[0046] When aligning the sensor array 11 and the signal lead array 12 to prepare the sensing network 1, each sensor is connected to the signal lead in a one-to-one correspondence. Specifically, the sensor 111 is connected to the signal lead 121, the sensor 112 is connected to the signal lead 122, the sensor 113 is connected to the signal lead 123, the sensor 114 is connected to the signal lead 124, the sensor 115 is connected to the signal lead 125, the sensor 116 is connected to the signal lead 126, the sensor 117 is connected to the signal lead 127, the sensor 118 is connected to the signal lead 128, and the sensor 119 is connected to the signal lead 129.

[0047] In this embodiment, the printing material of the signal lead array 12 is a low-impedance conductive silver paste. Further, it can be the low-impedance Shenzhen Sheng Tian Feng 823SS model conductive silver paste, whose resistivity is 6.35×10 -5 Ω·cm and the viscosity is 30000 mPa·s.

[0048] And the high-impedance conductive carbon paste and the low-impedance conductive silver paste can be adjusted to the appropriate viscosity by adding a diluent. The mass ratio of the high-impedance conductive carbon paste to the diluent is 4:1, and the mass ratio of the low-impedance conductive silver paste to the diluent is 10:1.

[0049] (2) Sensing network integration;

[0050] As Figure 6 shown, integrate the sensing network 1 set in step (1) onto the skin structure 2.

[0051] Integrate the sensor array 11 and the skin structure 2.

[0052] Adjust the position of the printing stencil 13, align the sensor 15 at the center of the sensor array 11 with the exact center of the skin structure 2 for positioning, set the squeegee speed to 200 mm / s, the squeegee angle to 60°, the off-contact distance to 0.5 mm, the number of printing passes to 2 times, and cure the high-impedance conductive carbon paste by heating at 120 °C for 15 min, thereby realizing the integrated preparation of the sensor array 11 on the skin structure 2, as Figure 6 shown in states P1 to P2 in

[0053] Integrate the signal lead array 12 and the skin structure 2.

[0054] Adjust the position of the printing stencil 13, make the signal leads correspond to the sensors one by one, set the squeegee speed to 200 mm / s, the squeegee angle to 60°, the off-contact distance to 0.5 mm, the number of printing passes to 2 times, and cure the low-impedance conductive silver paste by heating at 120 °C for 40 min, thereby realizing the integrated preparation of the signal lead array 12 on the skin 2, as Figure 6 shown in states P2 to P3 in

[0055] Among them, the signal leads are connected to the sensors one by one. The printing materials of the sensor network 1, that is, the printing materials of the sensor array 11 and the signal lead array 12, are directly deposited on the surface of the skin structure 2 through the screen printing process to realize the integrated preparation of the sensor network 1 and the skin structure 2. And, the cable is connected to the electrodes of the signal leads in the sensor network 1 through silver paste to realize the final signal output.

[0056] (3) Load positioning monitoring;

[0057] Collect the signals of the sensing network 1 and perform pattern recognition on the collected signals to realize the load positioning function.

[0058] First, take the sensor 111 and the sensor 115 as examples for test characterization. As Figure 7 shown, it is a schematic diagram of the azimuth of a single sensing unit. In the figure, a is the head of the sensor, b is the tail of the sensor, c is the left side of the sensor, and d is the right side of the sensor. Figure 8 is the voltage response diagram when loading the head and tail azimuths of the sensing unit 111. Figure 9 is the voltage response diagram when loading the left and right side azimuths of the sensing unit 115. It can be found from the figure that the response of the sensor to the head loading is more sensitive than that to the tail, while there is little difference between the left and right loadings. According to the characteristic that the sensor has different responses to loading signals in different azimuths, this lays a foundation for the sensing network 1 to realize the function of load positioning.

[0059] The following introduces the large-area load positioning method implemented by the integrated position sensor network based on screen printing provided by the present invention.

[0060] As Figure 10 shown, the skin structure 2 can equally divide the load recognition area into 16 areas of 4×4, and each area is cyclically loaded with weights, and a signal acquisition system is used for real-time acquisition. A total of two sets of data need to be collected, namely training set data and prediction data. Among them, the training data and the prediction data are respectively the voltage signals of the sensing network corresponding to 15 consecutive and 5 separate loadings.

[0061] In this embodiment, a neural network is used for pattern recognition to achieve the positioning function. The schematic diagram of the neural network architecture is as Figure 11 shown. The neural network contains two hidden layers. The number of neurons in the previous layer is 15, and the number of neurons in the latter layer is 15. First, the training data under 30 separate loadings is input into the BP neural network for training. Subsequently, 20 prediction sets are input into this model. The calculated accuracy rate of the prediction set reaches 97.184%. The confusion matrix and the results are shown in Figure 12. Figure 12(a) is the confusion matrix diagram of the prediction results of the large-area load positioning of the sensing network. Figure 12(b) is the experimental result diagram of the large-area load positioning of the sensing network. It can be seen from Figure 12 that the sensing network is feasible for load positioning.

[0062] The neural network pattern recognition methods adopted include one or several of BP neural network, convolutional neural network, long short-term memory neural network, etc. Preferably, the number of hidden layers of the neural network is 1-2, and the number of neurons in the neural network is 6-15. The number of neurons in the neural network is L, and L takes 6-15. Where n is the number of input units, m is the number of output units, and a is a constant between the intervals (1, 10), satisfying:

[0063]

[0064] The load positioning method of the integrated sensing network based on the screen printing structure of the present invention directly integrates the sensor array and the large-area signal lead on the surface of the skin structure through screen printing technology. This method has the advantages of large-area manufacturing, batch manufacturing, simple equipment, low cost, good adaptability to the substrate and ink, etc., which is beneficial to the preparation of the flexible sensor network and the integrated integration with the structure. In addition, the load positioning method using pattern recognition can learn the mutual relationship between the input data and the target category without having to reveal and describe the mutual relationship between the input and output in advance before learning, which is beneficial to solving the problem that it is difficult to determine the mapping of the input-output relationship due to the instability of the resistance.

[0065] Although the present invention has been disclosed above by way of embodiments, it is not intended to limit the present invention. Any person having ordinary knowledge in the relevant technical field may make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to that defined by the appended claims for patent application.

Claims

1. A load positioning method based on a screen-printed structure integrated sensor network, characterized in that: include: (1) Sensor network setting: setting the sensor array and signal lead array on the same printing screen; (2) sensor network integration, integrating the sensor network provided in (1) into the skin structure; (3) Load positioning monitoring, collecting signals from the sensor network and performing pattern recognition on the collected signals to realize the load positioning function.

2. The load positioning method based on the screen printing structure integrated sensor network according to claim 1 is characterized in that: The printing screen is a nylon polyester fiber screen printing screen with 200 meshes and oil adaptability.

3. The load positioning method based on the screen-printed structure integrated sensor network according to claim 1 is characterized in that: The sensor array includes 9 sensors of the same size, and the lateral spacing between two adjacent sensors is 165 mm, and the longitudinal spacing is 125 mm.

4. The load positioning method based on the screen printing structure integrated sensor network according to claim 3 is characterized in that: The sensor has a gate length of 20 mm, a gate pitch of 1.4 mm, a line width of 0.3 mm, and 12 gates.

5. The load positioning method based on the screen printing structure integrated sensor network according to claim 4 is characterized in that: The printing material of the sensor array is high-impedance conductive carbon paste.

6. The load positioning method based on the screen printing structure integrated sensor network according to claim 5 is characterized in that: The signal lead array includes 9 signal leads, the line width of the signal leads is 0.6 mm, and the minimum spacing is 2 mm.

7. The load positioning method based on the screen-printed structure integrated sensor network according to claim 6 is characterized in that: The printing material of the signal lead array is low-impedance conductive silver paste.

8. The load positioning method based on the screen printing structure integrated sensor network according to claim 7 is characterized in that: The sensors are connected to the signal leads in a one-to-one correspondence.

9. The load positioning method based on the screen-printed structure integrated sensor network according to claim 1, characterized in that: In step (3), a neural network is used to perform pattern recognition to achieve the positioning function.

10. The load positioning method based on the screen printing structure integrated sensor network according to claim 9, characterized in that: The neural network pattern recognition method is one or more of BP neural network, convolutional neural network, long short-term memory neural network and the like.

Images