A flexible pressure sensor and its preparation method

Abstract

The present disclosure provides a method for preparing a piezoresistive sensor based on an electrospun hierarchical structure and the piezoresistive sensor, relating to the technical field of sensors. The preparation method includes using poly(vinylidene fluoride-trifluoroethylene) copolymer as a matrix material, performing electrospinning on the matrix material to obtain a poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer, uniformly coating a layer of PDA on the poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer, and growing a TiO2 nanorod structure on the PDA-coated poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer; then infiltrating it into a multi-walled carbon nanotube solution to obtain a conductive nanofiber membrane as a piezoresistive layer; repeating multiple times, and assembling the piezoresistive sensor after preparing and obtaining multiple layers of conductive nanofiber membranes; wherein, during assembly, an A4 paper cut is added between each layer of conductive nanofiber membranes, and the A4 paper cut is double-sidedly adhered to the conductive nanofiber membrane. The piezoresistive sensor prepared by the present disclosure can have ultra-high sensitivity under a wide range of pressures.

Description

Technical Field

[0001] The present disclosure relates to the technical field of sensors, and particularly to a flexible pressure sensor based on an electrospun hierarchical structure and a preparation method thereof. Background Art

[0002] The statements in this section merely provide background technical information related to the present disclosure and do not necessarily constitute prior art.

[0003] The Internet of Things consists of a series of devices interconnected by communication networks. It shows great potential in health prediction, human-computer interaction, intelligent sensing, and real-time response systems. In the practical application of intelligent systems combining pressure sensing technology and machine learning, data is mainly generated on irregular surfaces, such as human skin, biological tissues, and robot joints. Traditional rigid sensing technologies cannot conform to the surfaces of these objects, resulting in low-quality data acquisition. Machine learning is data-driven and requires a large number of high-fidelity data sets to complete a prediction task. In recent years, a new type of skin-like sensing technology called flexible sensors has been intensively studied. It can convert external pressure stimuli into electrical signals and has the characteristics of good softness, wearable, stretchable, and convenient to carry, facilitating attachment to the human body and enabling it to have a function similar to the tactile perception of human skin, reducing the influence of noise, thereby achieving high-fidelity data acquisition. According to different working mechanisms, flexible pressure sensors can be divided into piezoresistive, capacitive, piezoelectric, triboelectric, etc. Among them, piezoresistive pressure sensors have the advantages of simple structure and sensing mechanism, convenient preparation, and high sensitivity, and have become the research focus of flexible pressure sensors.

[0004] Currently, the following problems exist in flexible piezoresistive pressure sensors:

[0005] (1) The preparation of the sensing layer of the flexible piezoresistive pressure sensor is complex and costly, and it has the disadvantages of low sensitivity and small response range.

[0006] (2) Most of the studied sensors have poor stability, unable to collect high-quality data sets for machine learning model recognition, and are easily affected by humidity and not suitable for long-term use.

[0007] (3) Handwritten signature is one of the most important personal behavioral biometrics and occupies a very special position among a wide range of biometrics. At the same time, handwritten signatures are widely used in various civilian applications to verify personal identity, enhance security and privacy. The main challenge of handwritten signature recognition is how to obtain comprehensive handwritten information. General sensors are not sensitive to the external triggering force of handwritten signatures and cannot effectively improve the security strength of identity verification based on signature pressure, speed, and acceleration.

[0008] (4) The widespread use of smartphones has brought privacy and security issues. The current authentication method is usually password-based, that is, the user's identity is judged by whether the input password is correct. However, this knowledge-based authentication method is prone to password forgetting or password stealing. When typing the password, each user has a unique behavior pattern, such as the duration, intensity, and interval time of typing, which are unique. Generally, this typing behavior pattern is difficult to obtain through a general pressure sensor and cannot be extracted together with the touch events related to the input password, so machine learning technology cannot be used to identify legitimate users and enhance the security of unlocking. Summary of the Invention

[0009] To solve the above problems, the present disclosure proposes a flexible pressure sensor and its preparation method. The A4 paper-cutting and electrospinning processes are adopted, a structure of a multi-layer conductive nanofiber membrane is used, an air gap is introduced, and TiO2 is grown on the nanofiber thin film to increase the contact points of the conductive substance MWCNT.

[0010] According to some embodiments, the present disclosure adopts the following technical solutions:

[0011] A preparation method of a flexible pressure sensor, comprising:

[0012] Using poly(vinylidene fluoride-trifluoroethylene) copolymer (P(VDF-TrFE)) as a matrix material to prepare an electrospinning solution, performing electrospinning to obtain a P(VDF-TrFE) fiber layer, uniformly coating a layer of p-phenylenediamine (PDA) on the P(VDF-TrFE) fiber layer, and growing a TiO2 nanorod structure on the PDA-coated P(VDF-TrFE) fiber layer through two-step hydrothermal treatment to obtain a P(VDF-TrFE) / TiO2 fiber membrane; then infiltrating it into an MWCNT solution, taking it out, and drying. Obtaining a P(VDF-TrFE) / TiO2 / MWCNT conductive nanofiber membrane as a piezoresistive layer;

[0013] Repeat 3 times, and assemble a piezoresistive sensor after preparing and obtaining 3 layers of conductive nanofiber membranes; wherein, during assembly, A4 paper-cutting is added between each layer of conductive nanofiber membranes, and the A4 paper-cutting is double-sidedly adhered to the conductive nanofiber membranes.

[0014] Furthermore, when assembling the piezoresistive sensor, a conductive silver tape is used for the electrode layer.

[0015] Further, the preparation method of the fiber layer is as follows: Dissolve poly(vinylidene fluoride-trifluoroethylene) copolymer (P(VDF-TrFE)) powder in a mixed solution of N,N-dimethylformamide (DMF) and tetrahydrofuran (THF) to prepare a poly(vinylidene fluoride-trifluoroethylene) copolymer (P(VDF-TrFE)) solution. Then, ultrasonically treat the poly(vinylidene fluoride-trifluoroethylene) copolymer (P(VDF-TrFE)) solution until the poly(vinylidene fluoride-trifluoroethylene) copolymer (P(VDF-TrFE)) powder is completely dissolved. Then, electrospinning is carried out under certain conditions to obtain a poly(vinylidene fluoride-trifluoroethylene) copolymer (P(VDF-TrFE)) fiber layer.

[0016] Further, the method for growing TiO2 nanorod structures on the PDA-coated P(VDF-TrFE) fiber layer is as follows: First, dissolve dopamine hydrochloride in a tris(hydroxymethyl)aminomethane hydrochloride (Tris-HCl) buffer solution to prepare a dopamine solution. Then, immerse the P(VDF-TrFE) fiber layer in the dopamine solution for a period of time. After rinsing with deionized water, deposit TiO2 nanorods on the PDA-coated P(VDF-TrFE) fiber layer through two-step low-temperature hydrothermal treatment.

[0017] Further, the preparation method of the conductive nanofiber membrane is as follows:

[0018] Take a certain amount of multi-walled carbon nanotube solution MWCNT and put it into a test tube. Add multi-walled carbon nanotube MWCNT aqueous solution diluted to a certain extent with deionized water and pour it into a petri dish. Immerse a P(VDF-TrFE) fiber layer of a certain area that has been deposited with TiO2 nanorods and coated with PDA into the diluted multi-walled carbon nanotube MWCNT aqueous solution for a period of time, take it out, and put it in an oven to dry for a period of time;

[0019] Repeat several times to prepare a multi-layer conductive nanofiber membrane.

[0020] Further, the method for assembling the piezoresistive sensor is as follows:

[0021] Cut the prepared multi-layer conductive nanofiber membrane into the same size. Cut A4 paper into multiple pieces of the same size and hollow out the middle part according to a certain size. Put the paper pieces between the conductive nanofiber membranes and use a double-sided bonding method. Place the cut conductive silver tape as the electrode layer on the upper and lower surfaces of the piezoresistive layer, and finally encapsulate it face to face with a polyimide tape.

[0022] Further, the double-sided bonding method can be bonding with double-sided tape.

[0023] According to some embodiments, the present disclosure adopts the following technical solutions:

[0024] A flexible pressure sensor includes a plurality of acquisition circuits, and the plurality of acquisition circuits include single-channel acquisition circuits and multi-channel acquisition circuits.

[0025] Further, the single-channel acquisition circuit is used for signal acquisition of handwritten pressure recognition, and the multi-channel acquisition circuit is used for signal acquisition of a password lock system.

[0026] Further, both the single-channel acquisition circuit and the multi-channel acquisition circuit use a voltage of 5V.

[0027] Compared with the prior art, the beneficial effects of the present disclosure are as follows:

[0028] The solution of the present disclosure is convenient for fabricating a flexible piezoresistive pressure sensor device. Using A4 paper cutting and electrospinning technologies, the preparation process is simple. In terms of performance, the fabricated piezoresistive pressure sensor has the effects of ultra-stability, high sensitivity, wear resistance, and waterproofness. The high sensitivity depends on the structure of the multi-layer conductive nanofiber layer, introducing air gaps, and growing TiO2 on the nanofiber film to increase the contact points of the conductive material MWCNT.

[0029] The present disclosure improves the device packaging method. A layer of A4 paper is padded between the electrospinning layers, and the face-to-face adhesion packaging method enables the device to have high sensitivity while having super stability, which is beneficial for collecting high-quality data sets for subsequent machine learning training tasks, and has the advantages of wear resistance and waterproofness, and is applicable to application scenarios such as handwritten person recognition and identity authentication password unlocking systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The specification drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The schematic embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation to the present disclosure.

[0031] Figure 1 It is a flowchart of the preparation method of the piezoresistive pressure sensor in the embodiment of the present disclosure;

[0032] Figure 2 It is a schematic structural diagram of the piezoresistive pressure sensor in the embodiment of the present disclosure;

[0033] Figure 3 It is the fiber membrane structure formed at each stage in the preparation process of the piezoresistive pressure sensor in the embodiment of the present disclosure; (I) is the P(VDF-TrFE) fiber layer; (II) is the fiber membrane after growing TiO2 seeds on the P(VDF-TrFE) fiber membrane; (III) is the P(VDF-TrFE) / TiO2 fiber membrane; (IV) is the SEM of the P(VDF-TrFE) / TiO2 / MWCNT conductive nanofiber membrane;

[0034] Figure 4 The circuit diagrams for single-channel and multi-channel voltage signal acquisition in the embodiments of the present disclosure;

[0035] Figure 5 The handwritten voltage signals of four experimenters collected in the embodiments of the present disclosure;

[0036] Figure 6 The flowchart of the handwriting person recognition system and the recognition accuracy rate in the embodiments of the present disclosure;

[0037] Figure 6 In (a) is the flowchart of the handwriting person recognition system; Figure 6 In (b) is the recognition accuracy rate diagram;

[0038] Figure 7 The key voltage signals of three experimenters collected in the embodiments of the present disclosure;

[0039] Figure 8 The flowchart of the identity authentication password lock control system in the embodiments of the present disclosure; Specific embodiments

[0040] The present disclosure will be further described below in conjunction with the accompanying drawings and embodiments.

[0041] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further descriptions of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present disclosure belongs.

[0042] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless otherwise clearly specified in the context, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0043] Embodiment 1

[0044] In an embodiment of the present disclosure, a method for preparing a flexible pressure sensor is provided, as Figure 1 shown, including the following steps:

[0045] Preparation of P(VDF-TrFE) fibers: Dissolve P(VDF-TrFE) powder in 70:30 (N,N-dimethylformamide (DMF): tetrahydrofuran (THF)) to prepare an 18 wt% P(VDF-TrFE) solution, and then ultrasonically treat the solution until the P(VDF-TrFE) powder is completely dissolved. P(VDF-TrFE) nanofibers are prepared by electrospinning at a working voltage of 16 kV, a stock solution flow rate of 1 mL h -1 and a receiving distance of 10 cm.

[0046] Furthermore, the method for growing TiO2 nanorod structures on the PDA-coated P(VDF-TrFE) fiber layer is as follows: First, dissolve dopamine hydrochloride in Tris-HCl buffer solution to prepare a dopamine solution, then immerse the P(VDF-TrFE) fiber layer in the dopamine solution for a period of time, and after rinsing with deionized water, deposit TiO2 nanorods on the PDA-coated P(VDF-TrFE) fiber layer through two-step low-temperature hydrothermal treatment.

[0047] Specifically, prepare a dopamine (DA) solution by dissolving dopamine hydrochloride (100 mg, 98%, Aladdin, China) in Tris-HCl buffer solution. Immerse the fibers in the DA solution for 6 h and rinse with deionized water.

[0048] Preparation of P(VDF-TrFE) / TiO2 nanorod fibers: Deposit TiO2 nanorods on the PDA-coated P(VDF-TrFE) fibers through two-step low-temperature hydrothermal treatment.

[0049] The method for growing TiO2 nanorod structures on the PDA-coated P(VDF-TrFE) fiber layer is as follows: First, dissolve dopamine hydrochloride in Tris-HCl buffer solution to prepare a dopamine solution, then immerse the P(VDF-TrFE) fiber layer in the dopamine solution for a period of time, and after rinsing with deionized water, deposit TiO2 nanorods on the PDA-coated P(VDF-TrFE) fiber layer through two-step low-temperature hydrothermal treatment.

[0050] As an example, the preparation method of the conductive nanofiber membrane is as follows: Take a certain amount of MWCNT solution and put it into a test tube, add deionized water to dilute the MWCNT aqueous solution to a certain extent and pour it into a petri dish. Immerse a P(VDF-TrFE) fiber layer with TiO2 nanorods deposited and PDA-coated with a certain area in the diluted MWCNT aqueous solution for a period of time, take it out, and dry it in an oven for a period of time; repeat several times to prepare a multi-layer conductive nanofiber membrane.

[0051] Specifically, take 0.5 mL of MWCNT solution (0.15%, China) and put it into a test tube. Add deionized water to dilute it to a 0.015% MWCNT aqueous solution and pour it into a petri dish. Immerse the P(VDF-TrFE) / TiO2 nanofiber with an area of 1 cm * 1 cm into the diluted MWCNT solution for 10 s, take it out, and place it in an oven at 60 °C for drying for 15 min. Repeat this process 3 times.

[0052] As an example, after the piezoresistive layer of the sensor is prepared, each layer of the sensor is assembled. The assembly method is as follows:

[0053] Cut the prepared multi-layer conductive nanofiber membrane into the same size. Cut the A4 paper into multiple pieces of the same size and hollow out the middle part according to a certain size. Place the paper pieces between the conductive nanofiber membranes, and use a double-sided bonding method. Place the cut conductive silver tape as the electrode layer on the upper and lower surfaces of the piezoresistive layer, and finally encapsulate it face to face with polyimide tape.

[0054] The double-sided bonding method can be bonding with double-sided tape.

[0055] Specifically, prepare the three-layer conductive nanofiber membrane P(VDF-TrFE) / TiO2 / MWCNT, cut it into 3 pieces with a size of 1 cm * 1 cm. Cut the A4 paper into 2 pieces with a size of 1 cm * 1 cm and hollow out the middle part of 0.8 cm * 0.8 cm. Bond the P(VDF-TrFE) / TiO2 / MWCNT membrane and the trimmed paper pieces with double-sided tape. Place the cut silver tape electrodes on the upper and lower surfaces of the piezoresistive layer, and finally encapsulate it face to face with PI tape. In this way, a multi-layer highly sensitive, ultra-stable, wear-resistant and waterproof piezoresistive pressure sensor is successfully fabricated.

[0056] As Figure 2 shown, the assembly structure of the sensor, where: 1 is PI tape; 2 is the conductive silver tape as the electrode layer; 3 is the P(VDF-TrFE) / TiO2 / MWCNT thin film; 4 is the A4 paper cutout;

[0057] The piezoresistive pressure sensor prepared by the present disclosure has ultra-high sensitivity under a wide range of pressures;

[0058] Among them, the sensitivity of the piezoresistive pressure sensor is defined as:

[0059] S = δ(ΔI / I0) / ΔP

[0060] Where I0 and ΔI respectively represent the initial current when applying a constant voltage of 1 V and the relative change in current when applying a load, and P represents the applied load. The multi-layer structure adopted by the present disclosure improves the sensitivity of the piezoresistive pressure sensor.

[0061] The electrospinning process adopted in this disclosure has the advantages of low cost, simple process, and large-area preparation. Secondly, the device packaging method is improved. A layer of A4 paper is padded between the electrospun layers, and the device has high sensitivity and strong stability with the face-to-face adhesion packaging method using PI tape, which is beneficial to collecting high-quality data sets for subsequent machine learning training tasks, and has the advantages of wear resistance and waterproofness, and is suitable for application scenarios of handwritten person recognition and identity authentication password unlocking systems. This high sensitivity is attributed to the gaps caused by the paper-cutting between the electrospun layers. Its sensitivity roughly shows three regions. In the first stage, a very small pressure is applied to the sensor. At this stage, the gap between the fiber layers is significantly reduced. Because the gap between the fiber layers is easily squeezed, the contact area can increase rapidly with the applied pressure. Therefore, the sensor device has ultra-sensitive performance at this stage. In the second stage, the P(VDF-TrFE) / TiO2 fibers in the fiber membrane come into contact with each other and are compressed. This is the main reason for the increase in the MWCNT contact area at this stage. However, due to the low-speed deformation of the contact area, the sensitivity is lower than that in the first stage. In the third stage, all the structural components on the nanostructured membrane bear the applied pressure simultaneously, and deformation occurs in the entire pressure loading area between the upper and lower electrodes. In this case, further deformation becomes very difficult, and the contact area increases very slowly. Therefore, at this stage, the sensitivity becomes low.

[0062] Example 2

[0063] In an embodiment of the present disclosure, a method for preparing a flexible pressure sensor is provided, as Figure 1 shown, including:

[0064] Preparation of P(VDF-TrFE) fibers: Dissolve P(VDF-TrFE) powder in 60:40 (DMF:THF) to prepare a 16 wt% P(VDF-TrFE) solution, and then ultrasonically treat the solution until the P(VDF-TrFE) powder is completely dissolved. P(VDF-TrFE) nanofibers are prepared by electrospinning at a working voltage of 14 kV, a stock solution flow rate of 1.2 mL h -1 and a receiving distance of 11 cm.

[0065] Preparation of PDA-coated P(VDF-TrFE) fibers: Prepare a dopamine (DA) solution by dissolving dopamine hydrochloride (100 mg, 98%, Aladdin, China) in Tris-HCl buffer solution. Immerse the fibers in the DA solution for 6 h and rinse with deionized water.

[0066] Preparation of PDA-coated P(VDF-TrFE) / TiO₂ nanocolumn fibers: TiO₂ nanocolumns were deposited on PDA-coated P(VDF-TrFE) fibers by a two-step low-temperature hydrothermal method. First, 0.2 mL of tetrabutyl titanate was added to a mixed solution of 0.1 mL of deionized water, 5 mL of acetic acid, and 15 mL of ethanol, and stirred thoroughly for 60 min to synthesize the precursor solution of the TiO₂ seed layer. Then, the PDA-coated P(VDF-TrFE) fibers and the precursor solution were loaded into a 50 mL Teflon-lined stainless steel autoclave and subjected to a hydrothermal reaction at a low temperature of 110 °C for 4 h to obtain TiO₂ seeds fixed on the P(VDF-TrFE) optical fibers. After that, the fibers were thoroughly washed with deionized water and then dried in an oven at 60 °C. Subsequently, a mixed solution of 0.33 mL of tetrabutyl titanate, 10 mL of deionized water, and 10 mL of 12 M HCl was used to perform a second low-temperature hydrothermal treatment on the fibers with TiO₂ seeds at 110 °C for 12 h to achieve the growth of TiO₂ nanocolumns on the P(VDF-TrFE) fibers. The fibers were rinsed with deionized water to remove residues and then dried in an oven at 60 °C for 6 h.

[0067] Preparation of P(VDF-TrFE) / TiO₂ / MWCNT thin films: 0.5 mL of MWCNT solution (0.15%, China) was loaded into a test tube, diluted with deionized water to a 0.015% MWCNT aqueous solution and poured into a petri dish. P(VDF-TrFE) / TiO₂ nanocolumn fibers with an area of 1 cm * 1 cm were immersed in the diluted MWCNT solution for 10 s, taken out, and dried in an oven at 60 °C for 18 min, and repeated 3 times.

[0068] Sensor assembly: The prepared P(VDF-TrFE) / TiO₂ / MWCNT three-layer film was cut into 3 pieces with a size of 1 cm * 1 cm. The A4 paper was cut into 2 pieces with a size of 1 cm * 1 cm and the middle part of 0.8 cm * 0.8 cm was hollowed out. The P(VDF-TrFE) / TiO₂ / MWCNT film was bonded to the trimmed paper with double-sided tape. The cut silver tape electrodes were placed on the upper and lower surfaces of the piezoresistive layer, and finally encapsulated face to face with PI tape. In this way, a multilayer high-sensitivity, ultra-stable, wear-resistant, and waterproof piezoresistive pressure sensor was successfully fabricated.

[0069] Example 3

[0070] In one embodiment of the present disclosure, a method for preparing a flexible pressure sensor is provided, as Figure 1 shown, including:

[0071] Preparation of P(VDF-TrFE) fibers: Dissolve P(VDF-TrFE) powder in 50:50 (DMF:THF) to prepare a 20 wt% P(VDF-TrFE) solution, and then ultrasonically treat the solution until the P(VDF-TrFE) powder is completely dissolved. P(VDF-TrFE) nanofibers are prepared by electrospinning at a working voltage of 16 kV, a stock solution flow rate of 1 mL h-1, and a receiving distance of 12 cm.

[0072] Preparation of PDA-coated P(VDF-TrFE) fibers: Prepare a dopamine (DA) solution by dissolving dopamine hydrochloride (100 mg, 98%, Aladdin, China) in Tris-HCl buffer solution. Immerse the fibers in the DA solution for 6 h, take them out, and rinse with deionized water.

[0073] Preparation of PDA-coated P(VDF-TrFE) / TiO2 nanorod fibers: Deposit TiO2 nanorods on PDA-coated P(VDF-TrFE) fibers by two-step low-temperature hydrothermal method. First, add 0.2 mL of tetrabutyl titanate to a mixed solution of 0.1 mL of deionized water, 5 mL of acetic acid, and 15 mL of ethanol, and stir well for 60 min to synthesize the precursor solution of the TiO2 seed layer. Then, put the PDA-coated P(VDF-TrFE) fibers and the precursor solution into a 50 mL Teflon-lined stainless steel autoclave and carry out a hydrothermal reaction at a low temperature of 110 °C for 4 h to obtain TiO2 seeds fixed on the P(VDF-TrFE) optical fiber. After that, thoroughly wash the fibers with deionized water and then dry them in an oven at 60 °C. Subsequently, use a mixed solution of 0.33 mL of tetrabutyl titanate, 10 mL of deionized water, and 10 mL of 12 M HCl to carry out the second low-temperature hydrothermal treatment of the fibers with TiO2 seeds at 100 °C for 12 h to achieve the growth of TiO2 nanorods on the P(VDF-TrFE) fibers. Rinse the fibers with deionized water to remove the residues, and then dry them in an oven at 60 °C for 6 h.

[0074] Preparation of P(VDF-TrFE) / TiO2 / MWCNT films: Take 0.5 mL of MWCNT solution (0.15%, China) and put it into a test tube, add deionized water to dilute it to a 0.015% MWCNT aqueous solution and pour it into a petri dish. Immerse the P(VDF-TrFE) / TiO2 nanorod fibers with an area of 1 cm * 1 cm into the diluted MWCNT solution for 10 s, take them out, and dry them in an oven at 60 °C for 15 min. Repeat this process 3 times.

[0075] Sensor assembly: The prepared P(VDF-TrFE) / TiO2 / MWCNT three-layer film was cut into three pieces with a size of 1 cm * 1 cm. The A4 paper was cut into two pieces with a size of 1 cm * 1 cm, and the middle part of 0.8 cm * 0.8 cm was hollowed out. The P(VDF-TrFE) / TiO2 / MWCNT film was bonded to the trimmed paper with double-sided tape. The cut silver tape electrodes were placed on the upper and lower surfaces of the piezoresistive layer, and finally, it was encapsulated face to face with PI tape. In this way, a multi-layer highly sensitive, ultra-stable, wear-resistant, and waterproof piezoresistive pressure sensor was successfully fabricated.

[0076] Example 4

[0077] In an embodiment of the present disclosure, a flexible pressure sensor is provided, which includes a plurality of acquisition circuits. The plurality of acquisition circuits include a single-channel acquisition circuit and a multi-channel acquisition circuit. The single-channel acquisition circuit is used for signal acquisition of handwritten pressure recognition, and the multi-channel acquisition circuit is used for signal acquisition of a password lock system.

[0078] As an embodiment, the construction of the acquisition circuit includes:

[0079] A resistive tactile sensor means that when an external force acts on the device, the resistance value of the sensor will change. We convert the change in resistance into a change in voltage through a voltage division circuit. We designed a single-channel acquisition circuit for signal acquisition of handwritten person recognition and a 9-channel acquisition circuit for signal acquisition of a password lock system. The power supply uses a voltage of 5V, as Figure 4 shown. A handwritten person recognition and identity authentication password unlocking system was constructed. Single-channel and multi-channel acquisition circuits were built, and the handwritten signals of four experimental subjects and the key signals of the input password were collected. The data set was processed (removing noise, replacing missing values with the average of adjacent two values, normalizing, etc.), and machine learning techniques were combined to identify the human identity, with the recognition accuracy reaching more than 90%. And a front-end page was developed for real-time page display.

[0080] The recognition technology based on the combination of pressure sensing technology and machine learning mainly includes five parts: data acquisition, data preprocessing, feature extraction, data classification, and decision-making.

[0081] Data acquisition is an operation to obtain original dynamic data. The obtained original data is a set of repeated (multiple) input samples obtained within a specified time period. There may be situations such as partial data sample loss or data acquisition information omission in the collected original data. Through data preprocessing, the incomplete data is made complete, the wrong data is corrected, and the redundant data is removed, and then the required data is selected to improve the data quality and accuracy.

[0082] The main task of feature extraction is to identify and extract the common characteristics of users from the acquired raw data. These features are used for model generation. The features extracted from human touch dynamics data are divided into three major categories, namely, time, space, and motion features. The effectiveness of features affects the design and performance of classifiers. If we send all the raw features to the classifier as classification features, it will make the classifier complex and the probability of classification errors may not be small.

[0083] Data classification is mainly to predict the classification label. By analyzing the input data and learning the characteristics of the training set data, an accurate description or model is found for each class. Although the labels of the test data are unknown, we can still use the generated class descriptions to predict the classes of new data. To improve the accuracy and effectiveness of classification, data is usually preprocessed before classification, including: data cleaning, correlation analysis, and data transformation.

[0084] After the features are successfully extracted, the data will be stored in a system constructed by complex neural networks, decision trees, or other machine learning techniques. Handwritten person recognition system:

[0085] (1) Feedforward neural network:

[0086] A multi-layer feedforward neural network, where each neuron in the network receives the input from the previous-level neuron and outputs to the next level. The information processing ability of the multi-layer feedforward neural network comes from the multiple compositions of simple non-linear functions, which can realize the mapping transformation from the input space to the output space.

[0087] (2) Handwritten person recognition process:

[0088] In this disclosure, the handwritten signatures of 4 people are collected through Arduion, and each signature action is collected for 10 seconds. As Figure 5 shown, each sample collects 600 points, and a total of 1028 handwritten signature data are collected. Among them, 900 are used for training, 128 are used for verification, and each point is used as a feature and input into a fully connected neural network, which contains two hidden layers, an input layer, and an output layer. The cross-entropy loss function is adopted, and after 15 epochs, the accuracy rates on the training set, validation set, and test set reach over 90% respectively. Three front-end pages are constructed, namely, the login interface, the login success interface, and the login failure interface. Four users each write an English sentence on the flexible pressure sensor, and a model is trained through the fully connected neural network for prediction. Only the user predicted as 0 can successfully log in to the system, while the users predicted as 1, 2, or 3 will log in fail.

[0089] Password and identity authentication unlocking system:

[0090] The current unlocking verification method is usually password-based, that is, the user identity is judged by whether the entered password is correct. However, this knowledge-based authentication method is prone to password forgetting or password stealing. When typing the password, each user has a unique behavior pattern, such as the typing duration, strength, and interval time, which are unique. This typing behavior pattern can be obtained through a flexible pressure sensor array and extracted together with the touch events related to the entered password. Machine learning technology is used to identify legitimate users and enhance the security of unlocking.

[0091] A 3*3 pressure sensor array was made, and a 9-channel acquisition circuit was built based on Arduino. We collected a dataset of 3 users, with 177 samples for each user, including 150 for user training and 27 for testing. Each channel collected 400 points. And the dataset was preprocessed through PyCharm. Regular matching was used to make the data format more standardized. Missing values were replaced with the average of adjacent two sampling points, and the data was normalized. The data was processed into a tensor of [9,20,20] and input into a convolutional neural network to further extract features, with the output being a tensor of [9,3,3]. After flattening, it was connected to a fully connected layer, and finally, a 3-classification was output, with the recognition accuracy reaching 93%.

[0092] This disclosure is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the disclosure. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a machine for implementing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 a device for the functions specified in one or more blocks.

[0093] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, so that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable devices provide steps for implementing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 a step for the functions specified in one or more blocks.

[0094] Although the specific embodiments of the present disclosure have been described above in conjunction with the accompanying drawings, they are not intended to limit the scope of protection of the present disclosure. Those skilled in the art should understand that various modifications or variations that can be made without creative efforts on the basis of the technical solutions of the present disclosure are still within the scope of protection of the present disclosure.

Claims

1. A preparation method of a flexible pressure sensor, characterized in that, It includes the following steps: Electrospinning the poly(vinylidene fluoride-trifluoroethylene) copolymer as the matrix material to obtain a poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer, uniformly coating a layer of p-phenylenediamine on the poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer, and growing a TiO2 nanocolumn structure on the p-phenylenediamine-coated poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer through two-step hydrothermal treatment to obtain a fiber membrane; then infiltrating it into an MWCNT solution, taking it out, and drying it to obtain a conductive nanofiber membrane as the piezoresistive layer; Repeat 3 times, and assemble a piezoresistive sensor after preparing and obtaining 3 layers of conductive nanofiber membranes; Among them, during assembly, A4 paper cuttings are added between each layer of conductive nanofiber membranes, and the A4 paper cuttings are double-sidedly adhered to the conductive nanofiber membranes; Specifically, the assembly process of the piezoresistive sensor is as follows: Cut the prepared multi-layer conductive nanofiber membranes into the same size, cut the A4 paper cuttings into multiple pieces of the same size and with the middle part hollowed out according to a set size, place the paper pieces between the conductive nanofiber membranes, and use the double-sided adhesion method to place the cut conductive silver tape as the electrode layer on the upper and lower surfaces of the piezoresistive layer, and finally encapsulate it face to face with a polyimide tape.

2. The preparation method of a flexible pressure sensor according to claim 1, wherein, The preparation method of the poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer is: Dissolve the poly(vinylidene fluoride-trifluoroethylene) copolymer powder in a mixed solution of N,N-dimethylformamide and tetrahydrofuran to prepare a poly(vinylidene fluoride-trifluoroethylene) copolymer solution, then perform ultrasonic treatment on the poly(vinylidene fluoride-trifluoroethylene) copolymer solution until the poly(vinylidene fluoride-trifluoroethylene) copolymer powder is completely dissolved, and then perform electrospinning under certain conditions to obtain a poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer.

3. The preparation method of a flexible pressure sensor according to claim 1, characterized in that, The method of growing the TiO2 nanocolumn structure on the p-phenylenediamine-coated poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer is: First, dissolve dopamine hydrochloride in a tris(hydroxymethyl)aminomethane hydrochloride buffer solution to prepare a dopamine solution, then immerse the poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer in the dopamine solution for a period of time, rinse it with deionized water, and deposit TiO2 nanocolumns on the p-phenylenediamine-coated poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer through two-step low-temperature hydrothermal treatment.

4. The preparation method of a flexible pressure sensor according to claim 1, characterized in that, The preparation method of the conductive nanofiber membrane is: Take a certain amount of multi-walled carbon nanotube solution and put it into a test tube, add multi-walled carbon nanotube aqueous solution diluted to a certain extent with deionized water and pour it into a petri dish, immerse a certain area of the poly(vinylidene fluoride-trifluoroethylene) copolymer fiber layer deposited with TiO2 nanocolumns and coated with p-phenylenediamine into the diluted multi-walled carbon nanotube aqueous solution for a period of time, take it out, and put it in an oven to dry for a period of time; Repeat several times to prepare multi-layer conductive nanofiber membranes.

5. The method for preparing a flexible pressure sensor according to claim 1, wherein, The double-sided adhesion method is to use double-sided tape for adhesion.

6. A flexible pressure sensor prepared by the method for preparing a flexible pressure sensor according to any one of claims 1-5, characterized in that, It includes multiple acquisition circuits, and the multiple acquisition circuits include single-channel acquisition circuits and multi-channel acquisition circuits.

7. A flexible pressure sensor prepared by the method for preparing a flexible pressure sensor according to claim 6, characterized in that, The single-channel acquisition circuit is used for signal acquisition of handwriting pressure recognition, and the multi-channel acquisition circuit is used for signal acquisition of a password lock system.

8. A flexible pressure sensor prepared by the method for preparing a flexible pressure sensor according to claim 6, characterized in that, Both the single-channel acquisition circuit and the multi-channel acquisition circuit use a voltage of 5V.

Images