Self-reconfigurable superstructure, single-channel identification system and vibration sensing method

Abstract

The application discloses a self-reconfigurable superstructure, which comprises a plurality of chiral unit cells, the plurality of chiral unit cells are rotatably connected with each other, the chiral unit cell comprises a hexagonal frame and an internal chiral structure arranged in the hexagonal frame, the internal chiral structure comprises a chiral elastic beam and a mass block arranged at the center of the hexagonal frame, one end of the chiral elastic beam is connected with the inner wall surface of the hexagonal frame, and the other end is connected with the mass block. Correspondingly, the application also discloses a single-channel identification system, which comprises an excitation module, a signal acquisition module, a signal processing module and a plurality of self-reconfigurable superstructures according to the application; wherein the self-reconfigurable superstructure is provided with a piezoelectric sensor, a plurality of piezoelectric sensors are connected in series and send response signals to the signal acquisition module, and the signal acquisition module is respectively connected with the excitation module and the signal processing module in communication; and the application also discloses a vibration sensing method, which is implemented based on the single-channel identification system according to the application.

Description

Technical Field

[0001] This invention relates to the field of vibration sensing technology, specifically to a self-reconfigurable superstructure, a single-channel identification system, and a vibration sensing method. Background Technology

[0002] As is well known, vibration sensing technology has wide applications in equipment fault diagnosis, sound source identification, smart devices, and the Internet of Things. Currently, to reduce the complexity of the testing system, vibration sensing technology can be implemented by setting up a superstructure. The spatially identifiable transmission characteristics of the superstructure can be used to achieve physical encoding, which can then be combined with compressed sensing, machine learning algorithms, and other methods to decode vibration information from a single channel signal.

[0003] However, research has found that existing conventional vibration sensing technology systems often employ superstructures with weak structural reconfigurability and insufficient functional integration in order to improve their vibration response recognition. Such conventional superstructures cannot simultaneously achieve external stimulus recognition and internal state monitoring, making them unsuitable for the comprehensive sensing needs of miniaturized intelligent devices and complex scenarios.

[0004] Therefore, there is an urgent need to design a new superstructure that can be applied to single-channel vibration sensing and identification, and to apply it to single-channel identification systems and vibration sensing processes, so as to obtain stronger reconfigurability and functional integration to meet the comprehensive sensing needs in complex scenarios. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that the present invention discloses a self-reconfigurable superstructure, a single-channel identification system and a vibration sensing method, so as to solve the problems of weak structural reconfigurability and insufficient functional integration of existing conventional superstructures.

[0006] To address the aforementioned technical problems, one objective of this invention is to disclose a self-reconfigurable superstructure comprising: a plurality of chiral unit cells rotatably connected to each other; each chiral unit cell includes a hexagonal frame and an internal chiral structure disposed within the hexagonal frame; the internal chiral structure includes a chiral elastic beam and a mass block disposed at the center of the hexagonal frame; one end of the chiral elastic beam is connected to the inner wall of the hexagonal frame, and the other end of the chiral elastic beam is connected to the mass block.

[0007] Furthermore, in the self-reconfigurable superstructure described in this invention, the internal chiral structure includes a plurality of chiral elastic beams, which are uniformly distributed between the outer wall of the mass block and the inner wall of the hexagonal frame.

[0008] Furthermore, in the self-reconfigurable superstructure described in this invention, the internal chiral structure includes three chiral elastic beams, with adjacent chiral elastic beams spaced 120° apart from each other relative to the center point of the mass block.

[0009] Furthermore, in the self-reconfigurable superstructure described in this invention, one end of the chiral elastic beam is connected to the corner of the hexagonal frame.

[0010] Furthermore, in the self-reconfigurable superstructure described in this invention, the six border corners of the hexagonal frame are classified into a first group and a second group, each comprising three border corners. In the first group or the second group, adjacent border corners are spaced apart by a border corner from the other group. In a first connection mode, one end of each of the three chiral elastic beams is connected to the three border corners of the first group of the hexagonal frame. In a second connection mode, one end of each of the three chiral elastic beams is connected to the three border corners of the second group of the hexagonal frame.

[0011] Furthermore, in the self-reconfigurable superstructure described in this invention, the outer contour of the mass block is circular or polygonal; wherein, when the outer contour of the mass block is polygonal, the other end of the chiral elastic beam is connected to the corner of the mass block.

[0012] Furthermore, in the self-reconfigurable superstructure described in this invention, two adjacent chiral unit cells are rotatably connected to each other by means of hinges or magnets, so that multiple chiral unit cells can be arbitrarily rotated and reconfigured to each other.

[0013] Accordingly, another objective of the present invention is to disclose a single-channel recognition system for recognizing external stimuli and monitoring internal states, thereby meeting the vibration recognition requirements in complex scenarios. The single-channel recognition system includes an excitation module, a signal acquisition module, a signal processing module, and multiple self-reconfigurable superstructures as described above. Each self-reconfigurable superstructure is equipped with a piezoelectric sensor, and the multiple piezoelectric sensors are connected in series and send response signals to the signal acquisition module. The signal acquisition module is communicatively connected to both the excitation module and the signal processing module.

[0014] Furthermore, another object of the present invention is to disclose a vibration sensing method, which is implemented based on the single-channel identification system described above, and includes the following steps: S1: The excitation module of the single-channel recognition system is used to simulate external stimuli and act on multiple self-reconfigurable superstructures; S2: The signal acquisition module of the single-channel identification system is used to acquire the response signals output by the piezoelectric sensors corresponding to multiple self-reconfigurable superstructures and the input signals of the excitation module, and send them to the signal processing module; S3: The signal processing module of the single-channel recognition system uses corresponding algorithms to process signal data according to different application scenarios and calculates the vibration transfer function to realize the recognition of external excitation and internal state.

[0015] Furthermore, in the vibration sensing method of the present invention, in step S3, when identifying external stimuli of the self-reconstructing superstructure, the signal processing module performs a fast Fourier transform on the response signal output by the piezoelectric sensor to obtain the vibration transfer function and construct the observation matrix for subsequent compressed sensing algorithm. When used for identifying the internal state of a self-reconfigurable superstructure, the vibration transfer functions of multiple sets of self-reconfigurable superstructures with different configurations are obtained to construct a training dataset. The training dataset is used to train a neural network model to obtain an optimized neural network model. The superstructure output response is used as the input to the optimized neural network model to output the corresponding information decoding.

[0016] The beneficial effects of this invention are as follows: This invention proposes a novel self-reconfigurable superstructure for single-channel dynamic recognition, which eliminates the need for complex multi-channel signal processing and achieves single-channel vibration sensing and recognition while maintaining a small size, making it effective for application in micro-intelligent devices. This self-reconfigurable superstructure can be effectively applied to single-channel recognition systems and, in conjunction with the signal acquisition and signal processing modules within the single-channel recognition system, achieves integrated functions of external stimulus recognition and internal state monitoring, thereby meeting the comprehensive sensing needs in complex scenarios. It has good prospects for promotion and application value.

[0017] Accordingly, the vibration sensing method designed in this invention is specifically implemented based on the self-reconfigurable superstructure and the single-channel identification system described above, and therefore also has the aforementioned advantages and beneficial effects. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the chiral unit cell of the self-reconfigurable superstructure described in this invention in one embodiment; Figure 2 This is a schematic diagram illustrating the working principle of the self-reconfigurable superstructure described in this invention under one embodiment. Figure 3 This is a schematic diagram of the hexagonal frame structure of the self-reconfigurable superstructure described in this invention in one embodiment; Figure 4 This is a schematic diagram of the chiral structure of the chiral unit cell in another embodiment of the self-reconfigurable superstructure described in this invention, under a first connection mode. Figure 5This is a schematic diagram of the chiral structure of the chiral unit cell in another embodiment of the self-reconfigurable superstructure described in this invention, under a second connection mode. Figure 6 The diagram illustrates the structure of the self-reconfigurable superstructure described in this invention, which is based on the combination of multiple chiral unit cells. Figure 7 This is a schematic diagram illustrating the working principle of the single-channel identification system described in this invention under one embodiment. Figure 8 This is a reconstruction result diagram of the single-channel recognition system of the present invention, which excites six different positions in one embodiment. Detailed Implementation

[0019] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0020] To address the shortcomings of existing conventional vibration sensing technology systems, such as weak structural reconfigurability and insufficient functional integration of superstructures, this invention designs a novel self-reconfigurable superstructure. By optimizing the construction of this self-reconfigurable superstructure and changing the connection mode between adjacent chiral unit cells, the superstructure can switch between multiple different configurations, achieving reconfigurable control of the overall stiffness distribution. This results in distinguishable dynamic responses under different configurations and generates a single-channel output signal when subjected to external excitation. The single-channel output signal is then processed using compressed sensing and machine learning methods to accurately identify the location of external impact and the internal structural state.

[0021] Please refer to Figure 1 As shown, and in conjunction with references Figure 2 , Figure 3 , Figure 4 and Figure 5 In this invention, a novel self-reconfigurable superstructure is designed. Specifically, the self-reconfigurable superstructure includes multiple chiral unit cells, which are rotatably connected to each other. Each chiral unit cell includes a hexagonal frame and an internal chiral structure disposed within the hexagonal frame. The internal chiral structure includes a chiral elastic beam and a mass block disposed at the center of the hexagonal frame. One end of the chiral elastic beam is connected to the inner wall of the hexagonal frame, and the other end of the chiral elastic beam is connected to the mass block.

[0022] It should be noted that in practical applications, such as Figure 3 As shown, the framework of this chiral unit cell is specifically selected as a hexagonal framework, and the internal chiral structure set in the hexagonal framework can be selected according to the actual application requirements; such as Figure 1 , Figure 4 and Figure 5As shown, the mass blocks in the internal chiral structure can be selected as circular or polygonal mass blocks according to specific requirements. In the self-reconfigurable superstructure described in this invention, the aforementioned internal chiral structure may include multiple chiral elastic beams, which are uniformly distributed between the outer wall of the mass block and the inner wall of the hexagonal frame. Wherein, Figure 1 or Figure 4 In the embodiment shown, the internal chiral structure may specifically include three chiral elastic beams, with adjacent chiral elastic beams spaced 120° apart from each other relative to the center point of the mass block.

[0023] It should be noted that when setting the chiral elastic beam, the connection method of the chiral elastic beam can be specifically controlled. One end of the chiral elastic beam is connected to the corner of the hexagonal frame, and the other end can be connected to the outer surface of the circular mass block or to the corner of the polygonal mass block.

[0024] Different connection patterns can be obtained by varying the corners of the hexagonal frames being connected. For example, see [link to example]. Figure 4 and Figure 5 As shown, in actual setup, the six border corners of the hexagonal frame of the chiral unit cell can be classified into a first group and a second group, each consisting of three border corners. In the first group or the second group, adjacent border corners are spaced apart by a border corner from the other group. In the first connection mode, one end of each of the three chiral elastic beams is connected to the three border corners of the first group of the hexagonal frame. In the second connection mode, one end of each of the three chiral elastic beams is connected to the three border corners of the second group of the hexagonal frame.

[0025] In other words, in practical applications, see Figure 1 As shown, combined with Figure 4 and Figure 5 The designed chiral unit cell exhibits two connection modes due to its geometric chirality: a first connection mode and a second connection mode, denoted as "0" and "1" respectively. These modes can be independently rotated and switched. N unit cells can form 2... N Different configurations (such as) Figure 6 As shown, different configurations exhibit unique global stiffness distributions due to differences in local connection modes, resulting in low-correlation dynamic responses. Specifically, the mass blocks and chiral elastic beams in the hexagonal frame can be manufactured using 3D printing additive manufacturing.

[0026] It should be noted that in practical applications, multiple chiral unit cells can be connected to each other by magnets or hinges to form a superstructure, which can be arbitrarily recombined through the connection structure to achieve global stiffness distribution control; the hexagonal frame in the chiral unit cell applied to the self-reconfigurable superstructure can be divided into flexible frame and rigid frame according to stiffness, and the discernibility of the vibration response of the superstructure is negatively correlated with the frame stiffness.

[0027] To this end, users can preset a stiffness value and compare the actual stiffness of the hexagonal frame of the chiral unit cell in the self-reconfigurable superstructure with the preset stiffness value. When the actual stiffness is greater than the preset stiffness value, the hexagonal frame can be classified as a rigid frame. When the actual stiffness is less than the preset stiffness value, the hexagonal frame can be classified as a flexible frame.

[0028] In practical applications, depending on specific usage requirements, multiple chiral unit cells can be hinged together. Specifically, this can be achieved by extending connecting beams from the corners of the hexagonal frame of the chiral unit cells. Alternatively, in some specific embodiments, multiple chiral unit cells can also be magnetically connected to each other using magnets. In this case, magnets of different polarities can be attached to each side of the hexagonal frame of the chiral unit cells to achieve magnetic connection, allowing multiple chiral unit cells to be arbitrarily rotated and reconstructed to obtain different configurations.

[0029] For ease of understanding, the present invention also addresses... Figure 1 The specific structure of the chiral unit cell in this self-reconfigurable superstructure is explained and described in detail below: like Figure 1 and Figure 2 As shown, in actual setup, the outer contour of the mass block of the chiral unit cell is hexagonal, and each side of the mass block is parallel to each side of the hexagonal frame; at this time, each corner of the mass block corresponds to each corner of the hexagonal frame.

[0030] In this embodiment, the internal chiral structure may specifically include multiple chiral elastic beams, namely three chiral elastic beams, which are evenly distributed between the outer wall of the mass block and the inner wall of the hexagonal frame. One end of each chiral elastic beam is used to connect with a corner of the hexagonal frame, and the other end of each chiral elastic beam is specifically connected to an adjacent corner of the mass block near the corner of the frame, which corresponds to the corner of the frame.

[0031] like Figure 1In the embodiment shown, the three chiral elastic beams can be arranged apart from each other by a corner of the mass block and the hexagonal frame, so as to be evenly distributed between the hexagonal frame and the mass block; wherein, the chiral elastic beams may specifically include a first side, a first V-shaped side, a second side, a second V-shaped side and a third side connected in sequence, the first side is specifically connected to the corner of the frame of the hexagonal frame, and the third side is connected to the corner of the frame of the hexagonal frame adjacent to the mass block.

[0032] It should be noted that, in order to improve the application effect of the self-reconfigurable superstructure designed in this invention, in practical applications, the first side can be specifically set to connect with the first V-shaped side to form a 120° angle, the first V-shaped side to connect with the second side to form a 60° angle, the second side to connect with the second V-shaped side to form a 120° angle, and the second V-shaped side to connect with the third side to form a 60° angle; wherein the first V-shaped side and the second V-shaped side are arranged parallel to each other, and both the first V-shaped side and the second V-shaped side have a V-shaped angle, which is specifically set to 120°. Based on this setting, the combined application of the three chiral elastic beams can effectively fill the gap between the outer wall of the mass block and the inner wall of the hexagonal frame, thereby improving the subsequent use effect.

[0033] Accordingly, such as Figure 2 As shown, and in conjunction with references Figure 7 The present invention also designs a single-channel identification system, which specifically includes an excitation module, a signal acquisition module, a signal processing module, and multiple self-reconfigurable superstructures mentioned above. Among them, the self-reconfigurable superstructures are provided with piezoelectric sensors, which are connected in series with each other and send response signals to the signal acquisition module. Furthermore, the signal acquisition module is also communicatively connected to the excitation module and the signal processing module respectively.

[0034] It should be noted that in the single-channel recognition system designed in this invention, each chiral unit cell in the self-reconfigurable superstructure contains a hexagonal frame and an internal chiral structure, thus forming a "mass-spring-matrix" resonant system. Specifically, the mass is represented as a mass block, the spring as a chiral elastic beam, and the matrix as a hexagonal frame.

[0035] In practical applications, such as Figure 7 As shown, the excitation module applied to this single-channel recognition system can specifically consist of an impact hammer or signal generator, a power amplifier, and a vibration excitation table, which is used to simulate external stimuli in actual applications.

[0036] In practical applications, the single-channel identification system described above can be specifically applied based on the following vibration sensing method, which specifically includes the following steps: S1: The excitation module of the single-channel recognition system is used to simulate external stimuli and act on multiple self-reconfigurable superstructures; S2: The signal acquisition module of the single-channel identification system is used to acquire the response signals output by the piezoelectric sensors corresponding to multiple self-reconfigurable superstructures and the input signals of the excitation module, and send them to the signal processing module; S3: The signal processing module uses corresponding algorithms to process signal data and calculate vibration transfer functions according to different application scenarios, so as to realize the identification of external excitation and internal state.

[0037] S3: The signal processing module of the single-channel recognition system uses corresponding algorithms to process signal data according to different application scenarios and calculates the vibration transfer function to realize the recognition of external excitation and internal state.

[0038] Furthermore, in the vibration sensing method of the present invention, in step S3, when identifying external stimuli of the self-reconstructing superstructure, the signal processing module performs a fast Fourier transform on the response signal output by the piezoelectric sensor to obtain the vibration transfer function and construct the observation matrix for subsequent compressed sensing algorithm. When used for identifying the internal state of a self-reconfigurable superstructure, vibration transfer functions of multiple sets of self-reconfigurable superstructures with different configurations are obtained to construct a training dataset. This training dataset is used to train a neural network model to obtain an optimized neural network model. The superstructure output response is then used as the input to the optimized neural network model to decode the corresponding information. It should be noted that the superstructure's data acquisition module relies on a single-channel piezoelectric sensor. The piezoelectric element is embedded in the beam of the end unit cell of the superstructure to extract the dynamic response signal of the superstructure under external excitation, such as the electrical signal corresponding to vibration displacement and stress.

[0039] Accordingly, in the vibration sensing method designed in this invention, in step S3, the signal data will be processed by a corresponding algorithm according to the specific content being identified. For example, when used for identifying external stimuli of self-reconstructing superstructures, the above signal processing module performs a fast Fourier transform on the response signal output by the piezoelectric sensor to obtain the vibration transfer function, and then uses principal component analysis to reduce the dimensionality and construct the observation matrix for subsequent compressed sensing algorithm. When used for identifying the internal state of self-reconfigurable superstructures, vibration transfer functions of multiple sets of self-reconfigurable superstructures with different configurations can be obtained to construct a training dataset. This training dataset is used to train a neural network model to obtain an optimized neural network model. The superstructure output response is then used as the input to the optimized neural network model to decode the corresponding information. In other words, in practical applications, a fully connected neural network can be used, combined with a cross-entropy loss function and the ADAM algorithm (Adaptive Momentum Estimation Algorithm) to optimize configuration identification. The superstructure output response is then used as the input to the optimized neural network model and mapped to Base32 characters to complete the information decoding.

[0040] For ease of understanding, this invention is aimed at... Figure 7 The single-channel identification system of the illustrated embodiment will be further explained and described, such as... Figure 7 As shown, in this embodiment, the hexagonal frame of the chiral unit cell of the self-reconfigurable superstructure in the single-channel recognition system is specifically a flexible frame.

[0041] Figure 7 The system shown includes an excitation module, a signal acquisition module, a signal processing module, and multiple self-reconfigurable superstructures. The excitation module is specifically a force hammer. The self-reconfigurable superstructures are flexible frame structures composed of multiple hexagonal chiral unit cells connected by hinges. Rotating the chiral unit cells changes the configuration of the self-reconfigurable superstructure, thereby achieving global stiffness distribution control. The superstructure consists of six groups of self-reconfigurable superstructures with different configurations arranged radially at equal intervals, fixed inside a grid-like substrate. The substrate is filled with Blu-Tack material to suppress impact transmission. A pressure sensor is attached to the end beam of each group of self-reconfigurable superstructures. All pressure sensors are connected in series to the signal acquisition module to achieve single-channel signal acquisition.

[0042] When an external impact is applied to a certain location on the outside of the substrate, the self-reconfigurable superstructure in the corresponding region receives vibration excitation. Due to its unique chiral structure and connection method, self-reconfigurable superstructures of different configurations produce highly uncorrelated dynamic responses to vibration signals.

[0043] The signal acquisition module is connected to the force hammer and pressure sensor, and accordingly acquires the impact signal of the input superstructure and the response signal output by the pressure sensor; the signal analysis module uses a compressed sensing algorithm to process the acquired signals, specifically including constructing a measurement matrix and using the TWIST algorithm (Two-Step Iterative Shrinkage / Thresholding) to reconstruct the impact location and occurrence time. Its vibration sensing method specifically includes a calibration part and a testing part: The calibration process includes the following steps: selecting the location to be calibrated and applying excitation at that location; acquiring and processing the output signal of the pressure sensor set on the superstructure; obtaining the vibration transfer function using the fast Fourier transform algorithm; performing principal component analysis to reduce the dimensionality of the obtained vibration transfer function and constructing the observation matrix for subsequent compressed sensing algorithms.

[0044] The testing section includes the following steps: exciting the superstructure at a randomly calibrated location; acquiring signals from pressure sensors set on the superstructure; and reconstructing and identifying the vibration source information using a compressed sensing algorithm, which may include an observation matrix algorithm and a reconstruction algorithm.

[0045] like Figure 8 As shown, see Figure 8 The reconstruction results of six different excitation positions are shown. Experimental results show that the single-channel recognition system can identify six impact positions by using only one channel output signal under laboratory conditions and applying compressed sensing reconstruction algorithm, which significantly reduces the system complexity and is suitable for application scenarios with high integration and limited space.

[0046] Therefore, to address the shortcomings of existing superstructures, this invention proposes a novel self-reconfigurable superstructure for single-channel dynamic identification. It eliminates the need for complex multi-channel signal processing, achieving single-channel vibration sensing and identification while maintaining a small size. Its structure possesses instantaneous reconfigurability, generating multiple identifiable vibration responses through simple geometric changes, simplifying the encoding process and reducing costs and manufacturing difficulty. This self-reconfigurable superstructure can be effectively applied to single-channel identification systems, combining with signal acquisition and processing modules for acquisition and identification. This achieves functional integration of external stimulus identification and internal state monitoring, overcoming the insufficient functional integration of existing technologies and meeting the comprehensive sensing needs in complex scenarios. It has promising prospects and application value.

[0047] As can be seen from the above description, the self-reconfigurable superstructure designed in this invention has the following beneficial effects: 1. This self-reconfigurable superstructure can achieve single-channel vibration sensing, breaking through the dependence of existing technologies on multiple sensors or multiple channels. It adopts an integrated structural and functional design, which is convenient for the integration and miniaturization of intelligent devices. 2. This self-reconfigurable superstructure combines external stimulus recognition and internal state monitoring functions, which can solve the functional integration deficiency of existing superstructures. 3. This self-reconfigurable superstructure has instant reconfigurability, generating a variety of different configurations through simple geometric changes. It can identify a variety of vibration responses, simplify the coding process, and reduce costs and manufacturing difficulty.

[0048] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A self-reconfigurable superstructure, characterized in that, include: Multiple chiral unit cells are rotatably connected to each other. Each chiral unit cell includes a hexagonal frame and an internal chiral structure disposed within the hexagonal frame. The internal chiral structure includes a chiral elastic beam and a mass block disposed at the center of the hexagonal frame. One end of the chiral elastic beam is connected to the inner wall surface of the hexagonal frame, and the other end of the chiral elastic beam is connected to the mass block. The internal chiral structure includes a plurality of chiral elastic beams, which are evenly distributed between the outer wall of the mass block and the inner wall of the hexagonal frame. The internal chiral structure includes three chiral elastic beams, with adjacent chiral elastic beams spaced 120° apart from each other relative to the center point of the mass block; One end of the chiral elastic beam is connected to the corner of the frame of the hexagonal frame; The six border corners of the hexagonal frame are classified into a first group and a second group, each comprising three border corners. In the first group or the second group, adjacent border corners are spaced apart by a border corner from the other group. In a first connection mode, one end of each of the three chiral elastic beams is connected to one of the three border corners of the first group of the hexagonal frame. In a second connection mode, one end of each of the three chiral elastic beams is connected to one of the three border corners of the second group of the hexagonal frame.

2. The self-reconfigurable superstructure according to claim 1, characterized in that, The outer contour of the mass block is circular or polygonal; wherein, when the outer contour of the mass block is polygonal, the other end of the chiral elastic beam is connected to the corner of the mass block.

3. The self-reconfigurable superstructure according to claim 1, characterized in that, Adjacent chiral cells are rotatably connected to each other by means of hinges or magnets, so that multiple chiral cells can be arbitrarily rotated and reconfigured.

4. A single-channel identification system, characterized in that, It includes an excitation module, a signal acquisition module, a signal processing module, and multiple self-reconfigurable superstructures as described in any one of claims 1-3; wherein, the self-reconfigurable superstructure is provided with a piezoelectric sensor, the multiple piezoelectric sensors are connected in series and send response signals to the signal acquisition module, and the signal acquisition module is communicatively connected to the excitation module and the signal processing module respectively.

5. A vibration sensing method, implemented based on the single-channel identification system as described in claim 4, characterized in that, Including the following steps: S1: The excitation module of the single-channel recognition system is used to simulate external stimuli and act on multiple self-reconfigurable superstructures; S2: The signal acquisition module of the single-channel identification system is used to acquire the response signals output by the piezoelectric sensors corresponding to multiple self-reconfigurable superstructures and the input signals of the excitation module, and send them to the signal processing module; S3: The signal processing module of the single-channel recognition system uses corresponding algorithms to process signal data according to different application scenarios and calculates the vibration transfer function to realize the recognition of external excitation and internal state.

6. The vibration sensing method according to claim 5, characterized in that, In step S3, when identifying external stimuli of the self-reconstructing superstructure, the signal processing module performs a fast Fourier transform on the response signal output by the piezoelectric sensor to obtain the vibration transfer function and construct the observation matrix for subsequent compressed sensing algorithm. When used for identifying the internal state of a self-reconfigurable superstructure, the vibration transfer functions of multiple sets of self-reconfigurable superstructures with different configurations are obtained to construct a training dataset. The training dataset is used to train a neural network model to obtain an optimized neural network model. The superstructure output response is used as the input to the optimized neural network model to output the corresponding information decoding.

Images