Single-path multimode flexible sensor array based on impedance modulation signal and decoupling method

Abstract

The invention relates to the technical field of flexible electronic sensors, in particular to a single-path multi-mode flexible sensor array based on impedance modulation signals and a decoupling method, and the single-path multi-mode flexible sensor array comprises N multi-mode sensor units which are connected in parallel, each multimode sensor unit is a passive band-pass filter and comprises M modal measurement modules which are connected in series, the M modal measurement modules are used for responding to changes of M different physical quantities, and N and M are positive integers larger than or equal to 2; the number of the modal measurement modules is three, and the modal measurement modules are respectively a pressure-sensitive measurement module, a temperature-sensitive measurement module and a frequency calibration measurement module. The pressure-sensitive measurement module is a capacitor and is used for responding to the stress change, and the capacitance value of the pressure-sensitive measurement module changes along with the stress; the capacitor comprises a voltage-sensitive capacitor and a bias capacitor which are connected in parallel; the problems of low integration degree caused by high harness redundancy and high system complexity caused by multiple measurement circuits of the existing multimode sensor array are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of flexible electronic sensor, in particular to a single-channel multi-mode flexible sensor array based on impedance modulation signal and a decoupling method. BACKGROUND

[0002] The flexible multi-mode sensor array refers to an integrated sensing system capable of synchronously measuring multiple physical signals (such as stress, temperature), which can realize high-resolution spatial mapping of multi-modal stimulation and is crucial for the next generation of wearable electronics and soft robots. Existing multi-mode sensors can be divided into multi-material discrete type and single-material multi-parameter type according to the sensitive materials. The multi-material discrete type is composed of multiple sensitive materials, and each sensitive material only responds to a single physical quantity, such as material A only being sensitive to stress and material B only being sensitive to temperature. However, each sensor unit needs to be connected by independent wires, resulting in 4N signal lines for an array containing N units, and the serious redundancy of wire bundles severely limits the integration of the array. The single-material multi-parameter type is composed of a single sensitive material, and different physical properties of the material respond to different physical quantities, such as resistance changing with temperature and capacitance changing with stress. Although the number of materials is reduced, two sets of independent measurement circuits, such as resistance detection and capacitance detection circuits, need to be configured for each unit, resulting in an exponential increase in system complexity with the number of units.

[0003] Therefore, it is necessary to invent a single-channel multi-mode flexible sensor array based on impedance modulation signal and a decoupling method to solve the above problems. SUMMARY

[0004] The present application provides a single-channel multi-mode flexible sensor array based on impedance modulation signal and a decoupling method to solve the problems of low integration caused by high wire redundancy and high system complexity caused by multiple measurement circuits in existing multi-mode sensor arrays.

[0005] The present application is implemented by using the following technical solutions:

[0006] The single-channel multi-mode flexible sensor array based on impedance modulation signal comprises N multi-mode sensor units connected in parallel; each multi-mode sensor unit is a passive band-pass filter and comprises M modal measurement modules connected in series, the M modal measurement modules are used to respond to changes in M different physical quantities, and N and M are positive integers greater than or equal to 2.

[0007] Further, the number of modal measurement modules is 3, and they are a pressure-sensitive measurement module, a temperature-sensitive measurement module, and a frequency calibration measurement module.

[0008] Further, the pressure-sensitive measurement module is a capacitor, which is used to respond to stress changes, the capacitance value of which changes with the stress, and the initial capacitance values of the capacitors of the N multi-mode sensor units are different; the capacitor includes a pressure-sensitive capacitor and a bias capacitor connected in parallel; the pressure-sensitive capacitor is a capacitor based on a liquid metal elastomer foam composite dielectric material; and the bias capacitor is a patch capacitor with a fixed capacitance value.

[0009] Further, the temperature-sensitive measurement module is a resistor, which is used to respond to temperature changes, the impedance value of which changes with the temperature, and includes two negative temperature coefficient thermistors connected in parallel.

[0010] Further, the frequency calibration measurement module is a fixed inductor, and the inductance value range is 33 μH-56 μH.

[0011] Further, the sensor array is integrally formed by an elastomer package, and the base and the package layer are connected to the signal acquisition system through two oxidized liquid metal signal lines; the N multi-mode sensor units are connected in parallel through oxidized liquid metal wires.

[0012] Further, the signal acquisition system includes:

[0013] a signal generation module, which is used to apply an AC voltage signal with adjustable frequency to the sensor array;

[0014] a signal conversion module, which is used to convert the current signal flowing through the sensor array into a voltage signal;

[0015] a processing module, which is used to collect the voltage signal and perform frequency domain analysis thereon to obtain the impedance-frequency spectrum of the sensor array.

[0016] The decoupling method of the single-channel multi-mode flexible sensor array based on impedance modulation signals, which is used to decouple the single-channel multi-mode flexible sensor array based on impedance modulation signals as described in any one of the present application, includes the following steps:

[0017] obtaining the impedance-frequency spectrum data of the sensor array under different modal combination conditions as a training data set;

[0018] constructing a deep learning model, inputting the impedance-frequency spectrum data into the deep learning model, and outputting the modal values of each multi-mode sensor unit;

[0019] training the deep learning model using the training data set;

[0020] inputting the real-time collected impedance-frequency spectrum data into the trained deep learning model, and outputting the modal values of each multi-mode sensor unit.

[0021] Further, the deep learning model is a hybrid model comprising a convolutional neural network layer and a long short-term memory network layer.

[0022] Further, the training of the deep learning model adopts a transfer learning strategy, which comprises the following steps:

[0023] First, the model is pre-trained using a simulation data set generated by circuit simulation;

[0024] Then, the pre-trained model is fine-tuned using a measured data set obtained by actual measurement.

[0025] Compared with the existing multi-mode sensor array, the single-channel multi-mode flexible sensor array based on impedance modulation signals provided by the application has the following beneficial effects:

[0026] 1. High integration: the entire sensor array realizes multi-modal signal acquisition of all multi-mode sensor units through only two signal lines, greatly reducing the number of wire bundles and significantly improving the integration and flexible adaptability of the system.

[0027] 2. Simplified system structure: a single measurement circuit based on impedance-frequency scanning is used to replace the traditional multi-channel independent detection circuit, reducing the system complexity and hardware cost.

[0028] 3. Precise decoupling capability: by configuring different bias capacitors for each multi-mode sensor unit, the resonant frequencies are separated in the frequency domain, and the impedance spectrum is analyzed by a deep learning model, realizing high-precision decoupling and reconstruction of multi-modal signals.

[0029] 4. Strong anti-interference and reliability: the frequency domain response characteristics of the passive band-pass filter are used to effectively suppress the crosstalk between units, improving the stability and reliability of signal measurement. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 is a structural schematic diagram of the sensor array of the application.

[0031] Figure 2 is an equivalent circuit diagram of the sensor array of the application.

[0032] Figure 3 is a schematic diagram of the sensor array of embodiment 1 in the application.

[0033] Figure 4 is a structural schematic diagram of the pressure-sensitive capacitor of embodiment 1 in the application.

[0034] Figure 5 is a schematic diagram of the pressure and temperature sensitive principle of the multi-mode sensor unit of embodiment 2 in the application.

[0035] Figure 6 is a specific architecture diagram of the mixed neural network model of embodiment 2 in the present application.

[0036] Figure 7 is an impedance-frequency response curve diagram example of the sensor array of embodiment 2 in the present application under multi-modal stress and temperature conditions.

[0037] In the figure: 1, multi-modal sensor unit; 2, capacitor; 3, pressure-sensitive capacitor; 4, bias capacitor; 5, resistor; 6, fixed inductor; 7, liquid metal elastic foam; 8, liquid metal electrode; 9, substrate and packaging layer; 10, signal acquisition system. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0039] Embodiment 1

[0040] Single-channel multi-modal flexible sensor array based on impedance modulation signal, as shown in FIGS. 1-3 of the accompanying drawings, includes four multi-modal sensor units 1 connected in parallel by oxidized liquid metal wires; each multi-modal sensor unit 1 is a passive band-pass filter, and each includes three modal measurement modules connected in series by oxidized liquid metal wires, which are a pressure-sensitive measurement module, a temperature-sensitive measurement module, and a frequency calibration measurement module, forming an RLC series resonance circuit. Figure 1 Figure 3 Figure 4 The pressure-sensitive measurement module is a capacitor 2 for responding to stress changes, and the capacitance value changes with the stress it bears. The initial capacitance values of the capacitors 2 of the four multi-modal sensor units 1 are different, so that the four multi-modal sensor units 1 have mutually non-overlapping initial resonance frequencies. The capacitor 2 includes a pressure-sensitive capacitor 3 and a bias capacitor 4 connected in parallel.

[0041] The temperature-sensitive measurement module is a resistor 5 for responding to temperature changes, and its resistance value changes with the temperature it bears. The initial resistance values of the resistors 5 of the four multi-modal sensor units 1 are different, so that the four multi-modal sensor units 1 have mutually non-overlapping initial resonance frequencies.

[0042] ​​The pressure-sensitive capacitor 3 is a parallel-plate capacitor composed of a liquid metal elastic foam 7 and two liquid metal electrodes 8, and the capacitance varies from 0 to 18 pF, specifically, the liquid metal elastic foam 7 is sandwiched between the two liquid metal electrodes 8 and encapsulated to form; the liquid metal elastic foam 7 is prepared by a stencil method, which is to mix eutectic gallium-indium liquid metal, Ecoflex 00-30 silicone of Smooth-On company and water-soluble template, remove the template after curing to obtain, the porosity of the porous structure is 60%, and the pore size is 1 mm; the water-soluble template in the embodiment is selected as sugar particles.

[0043] The bias capacitor 4 is a patch capacitor with a fixed capacitance value, and the four bias capacitors 4 in the four multi-mode sensor units 1 are respectively selected as 470 pF, 200 pF, 91 pF and 32 pF patch capacitors, so as to separate the resonant frequencies of the four passive band-pass filters and avoid crosstalk.

[0044] The temperature-sensitive measurement module is a resistor 5 for responding to the change of temperature, and the impedance value changes with temperature, and includes two negative temperature coefficient thermistors connected in parallel, and the resistance change of the negative temperature coefficient thermistor is 800Ω at 0-120℃.

[0045] The frequency calibration measurement module is a fixed inductor 6, and the inductance value is 56μH.

[0046] The four multi-mode sensor units 1 constitute a multi-stage passive band-pass filter, and convert the capacitance and resistance output by the multi-mode sensor unit 1 into an impedance modulation signal.

[0047] The whole sensor array is encapsulated by a polydimethylsiloxane (PDMS) to form a substrate and an encapsulation layer 9, and is connected to a signal acquisition system 10 through two oxidized liquid metal signal lines, so as to realize the measurement of the multi-mode signal of the sensor array.

[0048] In the embodiment, the oxidized liquid metal is selected as oxidized eutectic gallium-indium, which is obtained by mechanical stirring oxidation of eutectic gallium-indium.

[0049] The preparation method of the sensor array in the embodiment specifically includes the following steps:

[0050] S1: preparing a sensor mold by a 3D printer;

[0051] S2: mixing and stirring the main agent and the curing agent of PDMS at a mass ratio of 10:1 by a centrifugal electric mixer for 3 minutes, the PDMS is produced by Dowcorning company, to obtain elastomer mixed liquid I, and part of the elastomer mixed liquid I is injected into the prepared sensor mold to form the substrate part of the substrate and the encapsulation layer 9 after curing.

[0052] S3: Using a centrifugal electric mixer, mix the A and B components of Ecoflex 00-30 silicone gel from Smooth-On Company at a mass ratio of 1:1 for 3 minutes to obtain elastomer mixture II; add 20wt% of eutectic gallium-indium liquid metal and 60wt% of sugar particles to the elastomer mixture II, and further mix for 3 minutes, then inject the mixture into the liquid metal elastomer foam mold prepared in advance, and after curing at a temperature of 60°C for 2 hours, remove it and place it in deionized water, and ultrasonic for 30 minutes to remove the sugar particles, to obtain liquid metal elastomer foam 7.

[0053] S4: Use a mechanical stirrer to stir 300g of eutectic gallium-indium for 30 minutes to obtain oxidized eutectic gallium-indium.

[0054] S5: Laser cutting paperboard to obtain a template for the lower liquid metal electrode 8 of the pressure-sensitive capacitor 3 and the oxidized liquid metal wire, leaving space for the bias capacitor 4 and the fixed inductor 6, and placing the template on the upper surface of the substrate and the base portion of the packaging layer 9 prepared in step S2. The oxidized eutectic gallium-indium is evenly applied to the template, and after the oxide layer of the oxidized eutectic gallium-indium is stable, the template is removed to form the lower liquid metal electrode 8 of the pressure-sensitive capacitor 3 and the oxidized liquid metal wire.

[0055] S6: Place the corresponding bias capacitor 4 and fixed inductor 6 in the reserved position of the oxidized liquid metal wire.

[0056] S7: Place the liquid metal elastomer foam 7 prepared in step S3 in the corresponding position of the lower liquid metal electrode 8 of the pressure-sensitive capacitor 3 in the sensor mold, and inject part of the elastomer mixture I. After curing at a temperature of 60°C for 4 hours, the dielectric layer of the pressure-sensitive capacitor 3 is formed.

[0057] S8: Laser cutting paperboard to obtain a template for the upper liquid metal electrode 8 of the pressure-sensitive capacitor 3 and the oxidized liquid metal wire, leaving space for the resistor 5, and placing the template on the upper surface of the substrate and the base portion of the packaging layer 9 prepared in step S2. The oxidized eutectic gallium-indium is evenly applied to the template, and after the oxide layer of the oxidized eutectic gallium-indium is stable, the template is removed to form the upper liquid metal electrode 8 of the pressure-sensitive capacitor 3 and the oxidized liquid metal wire, and the corresponding resistor 5 is placed in the reserved position.

[0058] S9: Inject the remaining elastomer mixture I into the sensor mold, and after curing at a temperature of 60°C for 4 hours, the packaging of the sensor array is completed, and a complete single-channel multi-mode flexible sensor array based on impedance modulation signal is obtained.

[0059] In the implementation process, the selection of the oxidized liquid metal includes but is not limited to oxidized eutectic gallium-indium; the material selection of the substrate and the encapsulation layer 9 includes but is not limited to polydimethylsiloxane (PDMS) of Dow Corning Company; the material selection of the liquid metal elastic foam 7 includes but is not limited to Ecoflex 00-30 silicone of Smooth-On Company; the selection of the water-soluble template includes but is not limited to sugar particles; the 3D printer is selected from a 3D printer of A1 type of Shenzhen Tuozhu Technology Co., Ltd.; the centrifugal electric mixer is selected from a centrifugal electric mixer of SC-S300 type of Shenzhen Zhanhua Technology Co., Ltd.; the mechanical stirrer is selected from a mechanical stirrer of JB-80SH type of Shanghai Xinniu Co., Ltd.; and the eutectic gallium-indium is purchased from Shenyang Baijujie Scientific Instrument Co., Ltd.

[0060] In operation, the sensor array is attached to the surface of the measured object and connected to the signal acquisition system 10, which is electrically connected to the host computer.

[0061] The signal acquisition system 10 comprises:

[0062] a signal generation module for applying an AC voltage signal with adjustable frequency to the sensor array;

[0063] a signal conversion module connected in series with the sensor array for converting the current signal flowing through the sensor array into a voltage signal;

[0064] a processing module for collecting the voltage signal and extracting the amplitude of the voltage signal at different frequencies through fast Fourier transform (FFT), thereby obtaining the impedance-frequency spectrum of the sensor array.

[0065] The hardware of the signal acquisition system 10 is built with a field programmable gate array (FPGA) as the control core. The FPGA controls the digital-to-analog converter (DAC) to generate a sinusoidal voltage signal with a frequency linearly scanned from 1 MHz to 5 MHz, which is applied to one end of the sensor array after being amplified by a low-noise amplifier. The other end of the array is connected to a current-voltage (I / V) conversion circuit, which converts the current signal into a voltage signal. The voltage signal is sampled by an analog-to-digital converter (ADC) and sent back to the FPGA. Custom transmission and reception codes for AC voltage control are developed through the FPGA programming interface. The processed data is transmitted from the FPGA to the host computer, and the time-domain voltage signal collected is subjected to fast Fourier transform (FFT) to obtain its frequency-domain amplitude spectrum, i.e., the impedance-frequency spectrum.

[0066] Example 2

[0067] The decoupling method of the single-channel multi-mode flexible sensor array based on impedance modulation signals, which is used for decoupling the single-channel multi-mode flexible sensor array based on impedance modulation signals as described in Embodiment 1, comprises the following steps:

[0068] S1: Obtain training data: As shown in the attached Figure 7 Figure, obtain the impedance-frequency spectrum data of the sensor array under different modal combination conditions as the training data set through circuit simulation.

[0069] S2: Build a deep learning model: build a hybrid neural network model, i.e. a deep learning model, whose input is one-dimensional impedance-frequency spectrum data and whose output is the multi-mode measurement value of each multi-mode sensor unit 1; the hybrid neural network model includes a convolutional neural network (CNN) layer for extracting local frequency domain features and a long short-term memory (LSTM) network layer for capturing frequency domain sequence dependencies, as shown in the attached Figure 6 Figure, the specific architecture consists of the following layer order: input layer (1000 frequency points) → 1D convolutional layer (64 filters) → pooling layer → 1D convolutional layer (32 filters) → LSTM layer (100 units) → Dropout layer → fully connected layer (8 output nodes, corresponding to 4 pressures and 4 temperatures).

[0070] S3: Model training and transfer learning: first, pre-train the model using the simulation data set generated by circuit simulation; then fine-tune the pre-trained model using the measured data set obtained by actual measurement to make up for the difference between simulation and reality.

[0071] S4: Real-time decoupling: input the real-time collected impedance-frequency spectrum data into the trained deep learning model to directly output the multi-modal results of each multi-mode sensor unit 1, i.e. pressure values and temperature values.

[0072] The theoretical basis of this decoupling method is that the multi-mode sensor unit 1 is essentially an RLC series resonant circuit, whose impedance characteristics are determined by the physical quantity modulation elements. The complex impedance Z total of each multi-mode sensor unit 1 can be represented as:

[0073]

[0074] where R is the resistance value of the resistor 5, C is the capacitance value of the capacitor 2, L is the inductance value of the inductor 6, ω is the angular frequency, and j is the imaginary unit. The resonance frequency f0, the quality factor Q, the bandwidth BW, and the resonance point impedance Z(f0) of the resonant circuit can be represented as:

[0075]

[0076] |Z(f0)|=R; (5)

[0077] By combining equations (2) to (5), a model relating the characteristic parameters of the multimode sensor unit 1 signal and the resonant signal can be constructed:

[0078] R = |Z(f0)|; (6)

[0079]

[0080] It can be seen that by extracting the characteristic data of the resonant frequency f0, quality factor Q, bandwidth BW and resonant impedance Z(f0) of the output resonant signal, the response characteristics of each mode in the multimode sensor unit 1 can be obtained, and the effective decoupling measurement of the multimode signal can be realized.

[0081] Furthermore, multiple multimode sensor units 1 are connected in parallel to construct the sensor array described in this invention. (See attached diagram) Figure 2 This is an equivalent circuit diagram of the sensor array described in this invention. The array can be equivalent to a multi-order passive bandpass filter, with a total admittance Y. total (jω) can be represented as:

[0082]

[0083] As can be seen, the overall response of this sensor array is the superposition of the output signals of multiple passive bandpass filters. To achieve accurate measurement, it is necessary to avoid overlap of the resonant frequencies of each multimode sensor unit 1 to suppress crosstalk between channels. By tuning the resonant frequency of each multimode sensor unit 1 within a fixed, non-overlapping frequency band, multiple independent resonant peaks can be distinguished in the frequency domain. By extracting the characteristic parameters of each resonant peak separately, the measurement of all physical quantities in the entire sensor array can be completed simultaneously, ultimately achieving efficient parallel acquisition and decoupling of the multimode signals of the sensor array from a single signal line.

[0084] As attached Figure 5 As shown, attached Figure 5 The left figure shows the relationship between the impedance modulation signal and the varistor capacitor 3. When mechanical stress is applied to capacitor 2, its capacitance increases, leading to a decrease in the resonant frequency of the passive bandpass filter. This trend is consistent with the result of theoretical formula (2), verifying that capacitor 2 has a modulation effect on the frequency characteristics of the passive bandpass filter under stress. By extracting the resonant frequency, stress measurement of the sensor can be achieved. (Appendix) Figure 5The right graph in FIG. 1 shows the response relationship between the impedance modulation signal and the negative temperature coefficient thermistor. When the ambient temperature gradually rises from 0 degrees Celsius to 120 degrees Celsius, the current signal at the resonance frequency rises significantly. This phenomenon is consistent with the calculation results of the theoretical formula (5), indicating that the negative temperature coefficient thermistor has good temperature response characteristics and can realize accurate temperature measurement through the amplitude value of the impedance modulation signal resonance frequency. The above results prove the effectiveness and reliability of the passive band-pass filter system in stress and temperature sensing. The piezoresistor 3 and the negative temperature coefficient thermistor realize high sensitivity response to mechanical stress and temperature, respectively, and the experimental results are consistent with the theoretical model, indicating that the system is suitable for high-precision sensing and detection of various physical quantities, and has good practical application prospects.

[0085] In summary, the single-channel multi-mode flexible sensor array based on the impedance modulation signal provided by the present application has the following advantages:

[0086] First, high integration greatly reduces the complexity of wiring:

[0087] The sensor array described in the present application adopts a parallel structure, and all multi-mode sensor units 1 are connected to the signal acquisition system 10 through only two oxidized liquid metal signal lines. Regardless of how many multi-mode sensor units 1 (N≥2) are included in the array, only one pair of signal lines is needed to complete the collection and transmission of all data. This design significantly reduces the high wire redundancy problem caused by the need for independent wiring for each unit in traditional multi-mode sensor arrays, greatly improving the integration and flexible wearable adaptability of the system, especially for applications such as electronic skin and soft robots that have extremely high requirements for space and flexibility.

[0088] Second, high system structure simplification, reducing hardware complexity and cost:

[0089] Traditional multi-mode sensor arrays often need to configure multiple independent measurement circuits for each unit, resulting in a sharp increase in system complexity as the number of units increases. The present application introduces a single measurement mechanism based on alternating current impedance frequency scanning, which only requires one set of signal generation, conversion and processing module to simultaneously obtain the multi-modal information of all units. This design not only simplifies the system structure and reduces hardware costs, but also improves the scalability and stability of the system.

[0090] Third, high-frequency domain separation combined with deep learning to realize high-precision multi-modal decoupling:

[0091] Each multi-mode sensor unit 1 is designed as a passive band-pass filter, and by configuring different bias capacitors for each unit, the resonant frequencies of the units are mutually non-overlapping, thereby achieving natural separation of signals in the frequency domain. Further, by obtaining the impedance-frequency spectrum of the entire array and using a deep learning model to extract features and perform regression analysis on the spectrum data, the multi-modal physical quantities of each unit can be directly decoupled. This method overcomes the limitations of traditional decoupling algorithms under complex multi-modal coupling conditions, achieving high-precision and high-efficiency signal reconstruction.

[0092] Fourth, high anti-interference and high reliability:

[0093] Since the resonant frequencies of each multi-mode sensor unit 1 are separated from each other in the frequency domain, the crosstalk problem between units is effectively suppressed. At the same time, using liquid metal oxide as the wire material, it has high conductivity, flexibility and oxidation resistance, improving the long-term stability and reliability of the sensor in complex environments. In addition, by optimizing the deep learning model through transfer learning strategy, the generalization ability and robustness of the system in practical application are further enhanced.

[0094] Fifth, high multi-modal synchronous measurement, expanding application prospects:

[0095] The present application can realize the synchronous measurement and spatial mapping of stress, temperature and other physical quantities, has good multi-modal sensing ability, and has broad application prospects.

[0096] In the description of the present application, it should be understood that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0097] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A single channel multi-mode flexible sensor array based on impedance modulated signals, characterized by: The N multi-mode sensor units are connected in parallel; each multi-mode sensor unit is a passive band-pass filter, and each multi-mode sensor unit includes M modal measurement modules connected in series, the M modal measurement modules are used to respond to changes in M different physical quantities, and N and M are positive integers greater than or equal to 2.

2. The single-path multi-mode flexible sensor array based on impedance modulated signals of claim 1, wherein: The number of modal measurement modules is 3, and the modal measurement modules are a pressure-sensitive measurement module, a temperature-sensitive measurement module, and a frequency calibration measurement module.

3. The single-path multi-mode flexible sensor array based on impedance modulated signals of claim 2, wherein: The pressure-sensitive measurement module is a capacitor, which is used to respond to stress changes, and the capacitance value changes with the stress; the initial capacitance values of the capacitors of the N multi-mode sensor units are different; the capacitor includes a pressure-sensitive capacitor and a bias capacitor connected in parallel. The pressure-sensitive capacitor is a capacitor based on a liquid metal elastomer foam composite dielectric material; the bias capacitor is a patch capacitor with a fixed capacitance value.

4. The single-path multi-mode flexible sensor array based on impedance modulated signals of claim 2, wherein: The temperature-sensitive measurement module is a resistor, which is used to respond to temperature changes, and the impedance value changes with temperature; and the temperature-sensitive measurement module includes two negative temperature coefficient thermistors connected in parallel.

5. The single-path multi-mode flexible sensor array based on impedance modulated signals of claim 2, wherein: The frequency calibration measurement module is a fixed inductor, and the inductance value range is 33 μH-56 μH.

6. The single-path multi-mode flexible sensor array based on impedance modulated signals of claim 1, wherein: The sensor array is formed by an elastomer package, and the base and the package layer are connected to the signal acquisition system through two oxidized liquid metal signal lines; the N multi-mode sensor units are connected in parallel through oxidized liquid metal wires.

7. The single-path multi-mode flexible sensor array based on impedance modulated signals of claim 6, wherein: The signal acquisition system includes: a signal generation module for applying an AC voltage signal with adjustable frequency to the sensor array; a signal conversion module for converting the current signal flowing through the sensor array into a voltage signal; a processing module for collecting the voltage signal and performing frequency domain analysis to obtain the impedance-frequency spectrum of the sensor array.

8. A decoupling method for an impedance-modulated signal-based single-path multi-mode flexible sensor array, the decoupling method being used for decoupling the impedance-modulated signal-based single-path multi-mode flexible sensor array according to any one of claims 2-7, characterized in that: The method includes the following steps: obtaining impedance-frequency spectrum data of the sensor array under different modal combination conditions as a training data set; building a deep learning model, inputting the impedance-frequency spectrum data into the deep learning model, and outputting the modal values of each multi-mode sensor unit; training the deep learning model using the training data set; inputting the real-time collected impedance-frequency spectrum data into the trained deep learning model to output the modal values of each multi-mode sensor unit.

9. The method of decoupling of a single channel multi-mode flexible sensor array based on impedance modulated signals according to claim 8, characterized in that: The deep learning model is a hybrid model including a convolutional neural network layer and a long short-term memory network layer.

10. The method of decoupling of a single channel multi-mode flexible sensor array based on impedance modulated signals according to claim 8, wherein: The training of the deep learning model adopts a transfer learning strategy, and the transfer learning strategy includes the following steps: first, pre-training the model using a simulation data set generated by circuit simulation; then, fine-tuning the pre-trained model using a measured data set obtained by actual measurement.

Images