Triboelectricity pressure sensor and sleep monitoring method and system

Abstract

The invention relates to the field of pressure sensors, in particular to a triboelectric pressure sensor and a sleep monitoring method and system. The triboelectric pressure sensor comprises a positive friction layer, a negative friction layer, a deformation electrode connected to the negative friction layer and a flexible electrode connected to the positive friction layer. The contact ends of the tips of the multiple micro-cone structures of the positive friction layer all face the negative friction layer, and gaps are formed between the contact ends and the negative friction layer. The Young modulus of the negative friction layer is smaller than that of the micro-cone structure, so that when the negative friction layer approaches the deformed electrode, the negative friction layer sinks and makes contact with the side face of the micro-cone structure. The change amplitude of the contact surface along with the pressure is large, so that the sensitivity of the sensor is improved. Respiration and pulse frequencies are respectively detected by a triboelectric pressure sensor to obtain electric signals, the electric signals are segmented and enhanced into waves, then the waves are classified by a big data method, a learning model for identifying sleep state categories corresponding to the waves is obtained, the categories are further rapidly and accurately identified, and an alarm is given according to the number of abnormal times. The method and the system have high reliability.

Description

Technical Field

[0001] The present invention relates to the field of pressure sensors, and in particular to a friction electric pressure sensor and a sleep monitoring method and system. Background Art

[0002] Detecting vital signs through pressure sensors with high sensitivity is a common means of monitoring vital signs. Since the changes in vital signs such as breathing and pulse are small, there are high requirements for the sensitivity of the sensor. The triboelectric pressure sensor is a device that uses the principles of friction electrification and electrostatic induction to detect pressure changes. The basic principle is that when the surfaces of two different materials come into contact and deform, positive and negative charges will be generated on the contact surfaces of the two materials due to the transfer of electrons, forming an electrostatic field. The change in this electrostatic field can be detected by measuring the potential difference on the electrodes, thereby indirectly measuring the pressure acting on the sensor. This sensor has the ability to be self-powered, has high sensitivity, a wide detection range, strong environmental adaptability, and is easy to manufacture. It is suitable for close-fitting vital sign monitoring equipment.

[0003] In existing triboelectric pressure sensors, the positive friction layer usually includes a flexible micro-cone structure, and the tip of the micro-cone structure is arranged adjacent to the negative friction layer. When the positive friction layer is deformed by force, the tip of the micro-cone structure contacts the negative friction layer and continues to deform after contact, increasing the contact area. This causes a change in the electrostatic field, which in turn causes a change in the output electrical signal. However, in this structure, the contact area between the tip of the micro-cone structure and the negative friction layer changes little, and the area advantage brought by the micro-cone structure is not fully utilized, thus limiting the further improvement of the sensor sensitivity by the microstructure. Summary of the invention

[0004] The object of the present invention is to provide a friction electric pressure sensor with higher sensitivity and a sleep monitoring method and system.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A triboelectric pressure sensor, comprising:

[0007] A positive friction layer, comprising a plurality of micro-cone structures, each of which has a contact end with a smaller cross-sectional area, and the contact ends of the plurality of micro-cone structures face the same direction, and the micro-cone structures are constructed of a material having a Young's modulus of a first value;

[0008] a negative friction layer connected to a side of the positive friction layer close to the contact end of the micro-cone structure, with a gap between the negative friction layer and the contact end of the micro-cone structure, and a Young's modulus of the negative friction layer being a second value, and the second value being less than the first value;

[0009] A deformation electrode connected to the side of the negative friction layer away from the positive friction layer;

[0010] The flexible electrode is connected to the side of the positive friction layer away from the negative friction layer.

[0011] Optionally, the thickness of the negative friction layer is smaller than the height of the micro-cone structure, the Young's modulus of the deformation electrode is a third value, and the third value is smaller than the second value.

[0012] Optionally, the negative friction layer is made of a silicone material, the deformable electrode is a hydrogel structure, and the deformable electrode is wrapped inside the negative friction layer.

[0013] Optionally, the mass ratio of polyvinyl alcohol, phytic acid and deionized water in the deformable electrode is any value of (1-4): (9.8-10.2): (9.8-10.2).

[0014] Optionally, the height of the micro-cone structure is any value between 0.5 mm and 1 mm.

[0015] Optionally, the positive friction layer also includes a support portion, which has a first end and a second end opposite to the first end, and one end of each micro-cone structure away from the contact end is supported on the first end of the support portion, and the second end of the support portion is supported on the negative friction layer.

[0016] Optionally, the first sensitivity of the friction electric pressure sensor is any value between 3V / kPa and 8V / kPa, the second sensitivity of the friction electric pressure sensor is any value between 0.4V / kPa and 1.6V / kPa, and the third sensitivity of the friction electric pressure sensor is any value between 0.04V / kPa and 0.15V / kPa.

[0017] In the second aspect, the present invention also provides a sleep system, comprising a first sensor, a second sensor and a controller electrically connected to the first sensor and the second sensor, the first sensor and the second sensor are both the above-mentioned friction electric pressure sensors, the first sensor is used to convert the user's pulse signal into a first electrical signal, the second sensor is used to convert the user's breathing signal into a second electrical signal, the controller is used to receive the first electrical signal and the second electrical signal, and in combination with a learning model obtained by training a deep learning network model, output the user's sleep condition data.

[0018] In a third aspect, the present invention further provides a sleep monitoring method using the above system, comprising:

[0019] Collecting the pulse signals and the breathing signals of multiple volunteers to form multiple groups of the first electrical signals and the second electrical signals;

[0020] Performing data segmentation and data enhancement processing on each group of the first electrical signal and the second electrical signal respectively to obtain a plurality of model pulse waves and a plurality of model respiratory waves;

[0021] The model pulse wave and the model respiratory wave are used as inputs and the respiratory type is used as output to learn and obtain the learning model, wherein the respiratory type includes three classification groups: a normal type, a hypopnea type and an apnea type;

[0022] Collecting the pulse signal and the breathing signal of the user, and performing the data segmentation and data enhancement processing to obtain a detected pulse wave and a detected breathing wave;

[0023] Inputting the detected pulse wave and the detected respiratory wave into the learning model to obtain the corresponding respiratory type, and performing abnormal counting on the total number of the hypopnea type and the apnea type;

[0024] When the value obtained by counting the number of abnormal times reaches a preset value, an early warning signal is issued.

[0025] Optionally, the data segmentation is performed according to time series, and the data enhancement is a time series interpolation method.

[0026] According to the first aspect of the present invention, when an external force acts on the sensor, the deformable electrode undergoes elastic deformation, driving the positive friction layer to deform. The contact end of the micro-cone structure contacts the negative friction layer to achieve indentation deformation. Since the Young's modulus of the micro-cone structure in the present invention is relatively high and the Young's modulus of the negative friction layer is relatively low, as the micro-cone structure gradually approaches the negative friction layer, the position where the negative friction layer contacts the micro-cone structure is recessed inward, and the contact end of the micro-cone structure is inserted into the negative friction layer. At this time, the side of the micro-cone structure close to the contact end contacts the negative friction layer. Due to the triboelectric effect, due to the difference in dielectric constant, the positive friction layer and the negative friction layer will generate equal amounts of positive and negative charges, respectively, thereby forming an electrostatic field. This change in the electrostatic field can be detected by measuring the potential difference between the deformable electrode and the flexible electrode, thereby indirectly measuring the magnitude of the pressure acting on the sensor. As the external force weakens, the distance between the positive friction layer and the negative friction layer gradually increases until the contact end of the micro-cone structure no longer contacts the negative friction layer, and the potential also changes. In the prior art, the contact surface between the micro-cone structure and the negative friction layer is the cross section of its contact end. The micro-cone structure with a small Young's modulus deforms with pressure, causing the cross section of the contact end to become larger, thereby increasing the contact area. In the present invention, since the contact surface between the micro-cone structure and the negative friction layer is the side surface, when the deformation degree of the deformed electrode is the same, the change amplitude of the contact area between the micro-cone structure and the negative friction layer is greater than the change amplitude in the prior art, thereby increasing the output of the sensor and improving the sensitivity. Since the Young's modulus of the micro-cone structure is small, when the sensor is no longer under force, the shape of the positive friction layer returns to its initial state, and the risk of dislocation of the contact end of the micro-cone structure is small, which helps to improve the stability of the sensor.

[0027] According to the second aspect of the present invention, respiration and pulse are detected respectively by the first sensor and the second sensor, and the detected electrical signals are transmitted to the controller, and the controller performs data processing and analysis to obtain abnormal counts, and outputs an alarm signal based on the abnormal counts, which helps to accurately monitor sleep conditions.

[0028] According to the third aspect of the present invention, the friction electric pressure sensor in the present invention detects breathing and pulse respectively, which helps to improve the sensitivity and stability of detection and improve the credibility of the detection results. The electrical signal is divided into multiple model pulse waves and multiple model respiratory waves, and a large number of model pulse waves and model respiratory waves are used to train the big data model, and the model pulse waves and model respiratory waves are classified into three categories: regular, hypopnea and apnea. The detected pulse waves and detected respiratory waves obtained by detection and segmentation are then classified to determine the current sleep state. When the abnormal counts corresponding to the hypopnea and apnea categories reach a preset value, an early warning signal is issued, thereby realizing the monitoring of sleep apnea hypopnea syndrome. By pre-constructing a learning model and judging the breathing situation during sleep according to the corresponding method of the learning model classification, it is possible to realize the rapid processing of a large amount of data, thereby cooperating with the use environment of long-term real-time monitoring, and having a high accuracy.

[0029] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a structural schematic diagram of the friction electric pressure sensor shown in the first embodiment of the present invention;

[0031] Figure 2 It is a light microscope image of the triboelectric pressure sensor shown in the first embodiment of the present invention;

[0032] Figure 3 The strain-stress curves of the electrode layers in the triboelectric pressure sensors with different electrode layer compositions;

[0033] Figure 4 A schematic diagram of the conductivity of the electrode layer in the triboelectric pressure sensor with different electrode layer components;

[0034] Figure 5 Schematic diagram of the sensitivity of triboelectric pressure sensors with different electrode layer components;

[0035] Figure 6 Schematic diagram of the sensitivity of triboelectric pressure sensors with different micro-cone structure heights;

[0036] Figure 7 This is a flowchart of the sleep monitoring method shown in Embodiment 1 of the present invention;

[0037] Figure 8 The waveform of a portion of the first electrical signal in a measurement shown in the first embodiment of the present invention;

[0038] Fig. 91 is the waveform of the second electrical signal in different sleep states shown in the first embodiment of the present invention.

[0039] Legend: 1-flexible substrate, 2-flexible electrode, 3-positive friction layer, 31-connecting part, 32-micro-cone structure, 321-contact end, 33-support part, 331-first end, 332-second end, 4-negative friction layer, 5-deformation electrode, 51-electrode part, 52-packaging part. DETAILED DESCRIPTION

[0040] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0042] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0043] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0044] See also Figure 1The triboelectric pressure sensor protected by the present invention includes a positive friction layer 3, a negative friction layer 4, a deformation electrode 5 and a flexible electrode 2. The positive friction layer 3 includes a plurality of micro-cone structures 32, the micro-cone structures 32 have a contact end 321 with a smaller cross-sectional area, and the contact ends 321 of the plurality of micro-cone structures 32 face the same direction. The micro-cone structures 32 are constructed of a material having a first value of Young's modulus. The negative friction layer 4 is connected to a side of the positive friction layer 3 close to the contact end 321 of the micro-cone structure 32, and there is a gap between the negative friction layer 4 and the contact end 321 of the micro-cone structure 32. The Young's modulus of the negative friction layer 4 is a second value, and the second value is less than the first value. The deformation electrode 5 is connected to the side of the negative friction layer 4 away from the positive friction layer 3. The flexible electrode 2 is connected to the side of the positive friction layer 3 away from the negative friction layer 4.

[0045] When an external force acts on the sensor, the deformation electrode 5 undergoes elastic deformation, driving the positive friction layer 3 to deform. The contact end 321 of the microcone structure 32 contacts the negative friction layer 4 to achieve indentation deformation. Since the Young's modulus of the microcone structure 32 in the present invention is low and the Young's modulus of the negative friction layer 4 is high, as the microcone structure 32 gradually approaches the negative friction layer 4, the position where the negative friction layer 4 contacts the microcone structure 32 is recessed inward, and the contact end 321 of the microcone structure 32 is inserted into the negative friction layer 4. At this time, the side of the microcone structure 32 close to the contact end 321 contacts the negative friction layer 4. Due to the triboelectric effect, due to the difference in dielectric constants, the positive friction layer 3 and the negative friction layer 4 will generate equal amounts of positive and negative charges, respectively, thereby forming an electrostatic field. This change in the electrostatic field can be detected by measuring the potential difference between the deformation electrode 5 and the flexible electrode 2, thereby indirectly measuring the pressure acting on the sensor. As the external force weakens, the distance between the positive friction layer 3 and the negative friction layer 4 gradually increases until the contact end 321 of the micro-cone structure 32 is no longer in contact with the negative friction layer 4, and the electric potential will also change. In the prior art, the contact surface between the micro-cone structure 32 and the negative friction layer 4 is the cross section of its contact end 321. The micro-cone structure 32 with a smaller Young's modulus deforms with pressure, resulting in a larger cross section of the contact end 321, thereby increasing the contact area. In the present invention, since the contact surface between the micro-cone structure 32 and the negative friction layer 4 is a side surface, when the deformation degree of the deformed electrode 5 is the same, the change amplitude of the contact area between the micro-cone structure 32 and the negative friction layer 4 is greater than the change amplitude in the prior art, thereby increasing the output of the sensor and improving the sensitivity. Since the Young's modulus of the micro-cone structure 32 is small, when the sensor is no longer subjected to force, the shape of the positive friction layer 3 returns to its initial state, and the risk of dislocation of the contact end 321 of the micro-cone structure 32 is small, which helps to improve the stability of the sensor.

[0046] In some embodiments, the thickness of the negative friction layer 4 is less than the height of the micro-cone structure 32, and the Young's modulus of the deformation electrode 5 is a third value, and the third value is less than the second value. When the flexible electrode 2 is deformed by a large force, the deformation amplitude of the negative friction layer 4 is large, so that the surface of the deformation electrode 5 connected to the negative friction layer 4 is concave and deformed, and the contact area between the deformation electrode 5 and the negative friction layer 4 changes, so that the sensitivity is improved.

[0047] In some embodiments, the negative friction layer 4 is made of silicone material, the deformable electrode 5 is a hydrogel structure, and the deformable electrode 5 is wrapped inside the negative friction layer 4. The hydrogel has a low Young's modulus, and wrapping the hydrogel by the negative friction layer 4 helps to protect the structure of the hydrogel.

[0048] In some embodiments, the mass ratio of polyvinyl alcohol, phytic acid and deionized water in the deformable electrode 5 is any value of 1-4:9.8-10.2:9.8-10.2, for example, any value of 1:10:10, 2:9.8:10.1, 3:10.2:10 and 4:10.1:9.9, which is helpful to obtain a deformable electrode 5 with a lower Young's modulus.

[0049] In some embodiments, the height of the micro-cone structure 32 is any value between 0.5 mm and 1 mm, for example, any value between 0.5 mm, 0.6 mm, 0.7 mm, 0.8 mm, 0.9 mm and 1 mm.

[0050] In some embodiments, the positive friction layer 3 further includes a support portion 33, the support portion 33 has a first end 331 and a second end 332 opposite to the first end 331, one end of each micro-cone structure 32 away from the contact end 321 is supported on the first end 331 of the support portion 33, and the second end 332 of the support portion 33 is supported on the negative friction layer 4. The micro-cone structure 32 is supported by the support portion 33, so that the micro-cone structure 32 does not contact the negative friction layer 4 when the sensor is not subjected to external force.

[0051] In some embodiments, the first sensitivity of the friction electric pressure sensor is any value in the range of 3V / kPa to 8V / kPa, for example, any value in the range of 3V / kPa, 4V / kPa, 5V / kPa, 6V / kPa, 7V / kPa and 8V / kPa. The second sensitivity of the friction electric pressure sensor is any value in the range of 0.4V / kPa to 1.6V / kPa, for example, any value in the range of 0.4V / kPa, 0.8V / kPa, 1V / kPa, 1.2V / kPa, 1.4V / kPa and 1.6V / kPa. The third sensitivity of the friction electric pressure sensor is any value in the range of 0.04V / kPa to 0.15V / kPa, for example, any value in the range of 0.04V / kPa, 0.06V / kPa, 0.09V / kPa, 0.12V / kPa and 0.15V / kPa.

[0052] In the second aspect, the present invention also provides a sleep system, including a first sensor, a second sensor and a controller electrically connected to the first sensor and the second sensor, the first sensor and the second sensor are both the above-mentioned friction electric pressure sensors, the first sensor is used to convert the user's pulse signal into a first electrical signal, the second sensor is used to convert the user's breathing signal into a second electrical signal, the controller is used to receive the first electrical signal and the second electrical signal, and combine with the learning model obtained by training the deep learning network model to output the user's sleep condition data.

[0053] Respiration and pulse are detected by the first sensor and the second sensor respectively, and the detected electrical signals are transmitted to the controller, which processes and analyzes the data to obtain abnormal counts, and outputs an alarm signal based on the abnormal counts, which helps to accurately monitor sleep conditions.

[0054] Third, see Figure 7 The present invention also provides a sleep monitoring method using the above system, comprising:

[0055] S1. Collect pulse signals and breathing signals of multiple volunteers to form multiple groups of first electrical signals and second electrical signals.

[0056] S2. Perform data segmentation and data enhancement processing on each group of first electrical signals and second electrical signals to obtain multiple model pulse waves and multiple model respiratory waves.

[0057] S3. The model pulse wave and the model respiratory wave are used as inputs and the respiratory type is used as output to learn and obtain a learning model. The respiratory type includes three classification groups: regular type, hypopnea type and apnea type.

[0058] S4. Collect the pulse signal and breathing signal of the user, and perform data segmentation and data enhancement processing to obtain a detected pulse wave and a detected breathing wave.

[0059] S5. Input the detected pulse wave and the detected respiratory wave into the learning model to obtain the corresponding respiratory type, and count the total number of the corresponding hypopnea and apnea types for abnormality.

[0060] S6. When the value obtained by abnormal counting reaches a preset value, a warning signal is issued.

[0061] The friction electric pressure sensor in the present invention detects breathing and pulse respectively, which helps to improve the sensitivity and stability of detection and improve the credibility of the detection results. The electrical signal is divided into multiple model pulse waves and multiple model respiratory waves, and a large number of model pulse waves and model respiratory waves are used to train the big data model, and the model pulse waves and model respiratory waves are classified into three categories: regular, hypopnea and apnea. The detected pulse waves and detected respiratory waves obtained by detection and segmentation are then classified to determine the current sleep state. When the abnormal counts corresponding to the hypopnea and apnea categories reach a preset value, an early warning signal is issued to achieve monitoring of sleep apnea hypopnea syndrome. By pre-constructing a learning model and judging the breathing situation during sleep according to the corresponding method of the learning model classification, a large amount of data can be quickly processed, thereby cooperating with the use environment of long-term real-time monitoring and having a high accuracy.

[0062] In some embodiments, data segmentation is segmentation based on time series, and data enhancement is a time series interpolation method.

[0063] Please refer to the following examples for details.

[0064] Embodiment 1:

[0065] The sleep monitoring system shown in a preferred embodiment of the present application includes a first sensor, a second sensor, a controller and an audible and visual alarm. The first sensor and the second sensor are friction electric pressure sensors with the same structure, wherein the first pressure sensor can be integrated into a bracelet, which is worn on the radial artery of the user's wrist to detect the user's pulse signal; the second sensor is integrated into the sleeping pad and placed at the back position at the same height as the armpit to detect the user's breathing signal. The controller, the first sensor, the second sensor, the display and the audible and visual alarm receive the electrical signals generated by the first sensor and the second sensor, and judge the user's physical signs as regular signs, low-flow gas signs or apnea signs through software processing, and output an alarm signal to the audible and visual alarm after the user is in the low-flow gas sign or apnea sign for many times, so that the audible and visual alarm sounds an alarm.

[0066] See also Figure 1 and Figure 2 The triboelectric pressure sensor in this embodiment includes a flexible substrate 1, a flexible electrode 2, a positive friction layer 3, a negative friction layer 4 and a deformation electrode 5. In this embodiment, the positive friction layer 3 and the negative friction layer 4 are both made by a template obtained by laser etching an acrylic resin plate.

[0067] In this embodiment, the flexible substrate 1 is made of polyvinyl alcohol (PVA) film. In other embodiments, other flexible substrate 1 materials commonly used in the sensor field may also be used.

[0068] A flexible electrode 2 for connecting to an external circuit and conducting electricity is formed on one side of the flexible substrate 1 . In this embodiment, the flexible electrode 2 is an aluminum electrode.

[0069] The positive friction layer 3 includes a connecting portion 31, a supporting portion 33 and a plurality of micro-cone structures 32 with the same structure. The connecting portion 31 is configured in a sheet shape, and the contour shape is matched with the flexible electrode 2 and the flexible substrate 1. The supporting portion 33 is configured as a ring column as a whole, having a first end 331 and a second end 332 opposite to the first end 331, and the first end 331 of the supporting portion 33 is connected to the edge of the surface of one side of the connecting portion 31. The micro-cone structure 32 is conical, and its tip is the contact end 321. The end of the micro-cone structure 32 away from the contact end 321 is connected to the surface of the connecting portion 31, and the micro-cone structure 32 and the supporting portion 33 are arranged on the same side of the connecting portion 31, and the height of the micro-cone structure 32 is slightly smaller than the height of the supporting portion 33. When preparing the positive friction layer 3, polyvinyl alcohol particles and deionized water are mixed in a mass ratio of 1:10, dispersed and swollen, and then heated to dissolve, and added to the corresponding template for static curing and peeling, so that the positive friction layer 3 has a specific shape and a higher Young's modulus. In this embodiment, the height of the micro-cone structure 32 is 0.5 mm.

[0070] The negative friction layer 4 is a silicone layer, which is constructed into a flat plate structure. The negative friction layer 4 contacts the second end 332 of the support portion 33 of the positive friction layer 3, and is used to rub and deform with the positive friction layer 3 to generate a potential change. When preparing the negative friction layer 4, the Ecoflex-0030 silicone component A and component B are mixed evenly in a volume ratio of 1:1, added to the corresponding template, completely cured and peeled off, so that the structure of the negative friction layer 4 is stable and the Young's modulus is lower than that of the positive friction layer 3.

[0071] The deformable electrode 5 includes an electrode portion 51 made of polyvinyl alcohol-phytic acid hydrogel material and a packaging portion 52 matched with the negative friction layer 4. The packaging portion 52 is constructed as a rectangular box-shaped structure with an open top, which accommodates the electrode portion 51 and supports the negative friction layer 4, thereby protecting the structure of the electrode portion 51. The packaging portion 52 is used for the same material as the negative friction layer 4, and the preparation method is similar to that of the negative friction layer 4, only the template shape is different. And the outer wall at the opening of the packaging portion 52 is flush with the edge of the negative friction layer 4. The electrode portion 51 is constructed as a rectangular block as a whole. When preparing the electrode portion 51, polyvinyl alcohol, phytic acid and deionized water are mixed evenly, heated and cross-linked, and poured into the prepared packaging portion 52, and frozen in a refrigerator at -20°C to form the electrode portion 51. The Young's modulus of the electrode portion 51 is lower than that of the negative friction layer 4, and the structure is matched with the negative friction layer 4. Since the electrode portion 51 is entirely wrapped inside the negative friction layer 4 and the packaging layer, its structure is protected and is not easily damaged or water-lost, thereby helping to improve the stability of the friction electric pressure sensor.

[0072] When an external force acts on the flexible substrate 1, the contact end 321 of the micro-cone structure 32 of the positive friction layer 3 contacts the negative friction layer 4 due to the external force. Due to the difference in Young's modulus, the contact end 321 of the micro-cone structure 32 is embedded in the negative friction layer 4 to form an indentation deformation. Due to the triboelectric effect and the difference in dielectric constant, the two triboelectric layers will generate equal amounts of positive and negative charges. The flexible electrode 2 and the deformable electrode 5 will also induce charges due to the electrostatic induction effect. When the external force gradually weakens, the positive friction layer 3 gradually moves away from the negative friction layer 4, and the distance between the upper and lower plates increases. At this time, the potential difference between the flexible electrode 2 and the deformable electrode 5 also changes. When the positive friction layer 3 separates from the negative friction layer 4, the negative charge on the flexible electrode 2 will be transferred to the deformable electrode 5 under the action of electrostatics, generating current in the external circuit until a state of equilibrium is reached. When the positive friction layer 3 contacts the negative friction layer 4 again and realizes an indentation deformation, the charge flows from the deformable electrode 5 to the flexible electrode 2, generating a current in the opposite direction, thereby generating an electrical signal. When the external force is weak, only the surface of the negative friction layer 4 is deformed, and the sensitivity of the triboelectric pressure sensor is the first sensitivity. As the applied pressure increases, the micro-cone structure 32 is further embedded in the negative friction layer 4 until it is completely embedded. At this time, the electrode portion 51 of the deformed electrode 5 with a lower Young's modulus is compressed until the microstructure is completely firm. At this time, the sensitivity of the triboelectric pressure sensor is the second sensitivity. As the pressure further increases, the electrode portion 51 is compressed to the limit, and the sensitivity of the triboelectric pressure sensor is the third sensitivity.

[0073] In this embodiment, multiple deformable electrodes 5 are prepared with different mass ratios, and multiple triboelectric pressure sensors are obtained. When the mass ratio of polyvinyl alcohol, phytic acid and deionized water is 1:10:10, the corresponding triboelectric pressure sensor is recorded as PV1; when the mass ratio of polyvinyl alcohol, phytic acid and deionized water is 2:10:10, the corresponding triboelectric pressure sensor is recorded as PV2; when the mass ratio of polyvinyl alcohol, phytic acid and deionized water is 3:10:10, the corresponding triboelectric pressure sensor is recorded as PV3; when the mass ratio of polyvinyl alcohol, phytic acid and deionized water is 4:10:10, the corresponding triboelectric pressure sensor is recorded as PV4.

[0074] See also Figure 3 , the strain of the electrode layer material in each triboelectric pressure sensor when subjected to different vertical pressures was detected, and the curve was drawn. It can be seen that with the increase of the polyvinyl alcohol content in the hydrogel, the Young's modulus of the hydrogel gradually increases. Figure 4 ,The conductivity of the electrode layer materials in each triboelectric pressure sensor was tested. It can be seen that the conductivity gradually decreases with the increase of the polyvinyl alcohol content, but each electrode layer has a high conductivity.

[0075] See also Figure 5, respectively detect the difference between the output voltage and the bias voltage of each friction electric pressure sensor under different pressures, draw a sensitivity curve, and calculate the sensitivity of each friction electric pressure sensor under different strain conditions. The calculation results are shown in Table 1 below.

[0076] Table 1:

[0077] PV1 PV2 PV3 PV4 First sensitivity V / kPa 7.424 6.734 4.257 3.049 Second sensitivity V / kPa 1.525 0.719 0.671 0.445 The third sensitivity V / kPa 0.144 0.074 0.044 0.054

[0078] For easy observation, Figure 5 Only the sensitivity curve of PV1 is shown in Table 1. Figure 5 , it can be seen that PV1 has higher sensitivity.

[0079] See also Figure 6 , adjust the height of the micro-cone structure 32 in PV1, and draw the sensitivity curve again. It can be seen that as the height of the micro-cone structure 32 increases, the sensitivity decreases. The reason is that the change in contact area gradually decreases with the increase in the height of the micro-cone structure 32, resulting in a decrease in sensing performance.

[0080] Therefore, in this embodiment, PV1 with better performance is used as the first sensor and the second sensor.

[0081] See also Figure 7 , the sleep monitoring method adopted by the sleep monitoring system in this embodiment includes:

[0082] S1. Collect pulse signals and breathing signals of multiple volunteers to form multiple groups of first electrical signals and second electrical signals.

[0083] S2. Perform data segmentation and data enhancement processing on each group of first electrical signals and second electrical signals to obtain multiple model pulse waves and multiple model respiratory waves.

[0084] S3. The model pulse wave and the model respiratory wave are used as inputs and the respiratory type is used as output to learn and obtain a learning model. The respiratory type includes three classification groups: regular type, hypopnea type and apnea type.

[0085] S4. Collect the pulse signal and breathing signal of the user, and perform data segmentation and data enhancement processing to obtain a detected pulse wave and a detected breathing wave.

[0086] S5. Input the detected pulse wave and the detected respiratory wave into the learning model to obtain the corresponding respiratory type, and count the total number of the corresponding hypopnea and apnea types for abnormality.

[0087] S6. When the value obtained by abnormal counting reaches a preset value, a warning signal is issued.

[0088] See also Figure 8 and Fig. 9It can be seen that the first electrical signal obtained by converting the pulse signal by the first sensor can be divided into multiple regular segments according to the time sequence, while the second electrical signal obtained by converting the pulse signal by the second sensor has a large difference in waveform when the sleep state is different.

[0089] In step S2, data segmentation is performed according to time series and is performed through a deep learning network model. The continuously generated first electrical signal and the second electrical signal are segmented into a model pulse wave and a model respiratory wave generated in the same time, thereby graphing the electrical signal. In this embodiment, data enhancement is a time series interpolation method, which helps to introduce more data changes and details, thereby enhancing the diversity of the model pulse wave and the model respiratory wave, improving the visibility of the features, and avoiding overfitting problems. In other embodiments, data enhancement can also be performed using other existing methods such as data amplification counting.

[0090] In step S3, a learning model is formed by training a deep learning network model. The data classification layer in the deep learning network model classifies each group of corresponding model pulse waves and model respiratory waves, thereby obtaining waveforms of regular, hypopnea, and apnea classes to form a learning model. The amount of data of the model pulse wave and model respiratory wave trained before obtaining the learning model is 600, so as to adjust the weight and bias parameters of the deep learning network model to minimize the prediction error, thereby improving the performance of the model.

[0091] In this embodiment, the deep learning network model adopts the existing Bi-GRU bidirectional gated recurrent unit model, performs a series of time series modeling operations on the input pulse wave and respiratory wave, uses the bidirectional recurrent mechanism of Bi-GRU to extract the time-dependent features of the data, and performs classification processing through the fully connected layer, and finally realizes the recognition and classification results of the pulse wave and respiratory wave output by the Bi-GRU model. In other embodiments, other deep learning models can also be used.

[0092] In this embodiment, part of the code corresponding to step S3 is as follows:

[0093] def_augment_data(data, augmentation_factor=2): 2 means generating twice as much new data for each row;

[0094] for_in range(augmentation_factor):

[0095] noise=np.random.normal(loc=0,scale=0.01,size=data.shape)

[0096] augmented_data.append(data+noise)

[0097] , loc=0, scale=0.01 means that the mean of the newly added data is 0 and the variance is 0.01, and the enhanced data is a fusion of the original data and the newly generated data;

[0098] data = df[['b','p']].values, b is the input data of the respiratory wave, p is the input data of the pulse wave;

[0099] Pill_drilling(200,0.0005,264,2,3), 200 is the number of training traversal data sets, 0.0005 is the learning rate, 2 is the input dimension of 2, 64 is the number of hidden units in each direction of Bi-GRU, 2 is the number of layers of Bi_GRU, and 3 is the number of sleep breathing state classifications

[0100] After obtaining the learning model through steps S1 to S3, the detection results of the first sensor and the second sensor can be analyzed through the learning model, that is, steps S4 to S6. Usually, steps S1 to S3 need to be performed before the product is put into use, while steps S4 to S6 are monitoring methods during product use.

[0101] In step S4, step S2 is repeated for the user to obtain a detected pulse wave and a detected respiratory wave.

[0102] In step S5, the detected pulse wave and the detected respiratory wave are classified according to the learning model, and abnormality count is performed when the detected pulse wave and the detected respiratory wave are classified into the hypopnea class or the apnea class.

[0103] In step S6, when the value obtained by the abnormal count changes, it is compared with the preset value. When the value obtained by the abnormal count is equal to the preset value, an early warning signal is issued. In the present embodiment, the preset value is set according to international standards, and a total of three preset values ​​are set. When the value obtained by the abnormal count is less than the first-level preset value, the sleep state is normal and no early warning occurs; when the value obtained by the abnormal count is greater than or equal to the first-level preset value and less than the second-level preset value, the sleep state is mild apnea syndrome, and a first-level early warning occurs; when the value obtained by the abnormal count is greater than or equal to the second-level preset value and less than the third-level preset value, the sleep state is moderate apnea syndrome, and a second-level early warning occurs; when the value obtained by the abnormal count is greater than or equal to the third-level preset value, the sleep state is severe apnea syndrome, and a third-level early warning occurs. In the present embodiment, the first-level early warning, the second-level early warning, and the third-level early warning are sound and light alarms with increasing intensity. In other embodiments, an alarm signal can also be sent to a mobile terminal via the Internet.

[0104] In this embodiment, after the construction of the learning model is completed, the learning model is also optimized by partially detecting the pulse wave and detecting the breathing wave. Specifically, the partially detected pulse wave and the detected breathing wave obtained when monitoring the user's sleep quality are input into the database of the deep learning network model to further improve the learning model.

[0105] The present invention provides a novel friction electric pressure sensor with high sensitivity and long service life. The friction electric pressure sensor monitors the pulse and breathing of the user, and classifies and counts the learning model obtained through deep learning network training, so as to monitor the sleep state of the user in real time and prevent the dangers brought by obstructive sleep apnea hypopnea syndrome.

[0106] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0107] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A triboelectric pressure sensor, characterized in that: include: A positive friction layer (3), comprising a plurality of micro-cone structures (32), wherein the micro-cone structures (32) have contact ends (321) with smaller cross-sectional areas, and the contact ends (321) of the plurality of micro-cone structures (32) face the same direction, and the micro-cone structures (32) are constructed of a material having a Young's modulus of a first value; a negative friction layer (4) connected to a side of the positive friction layer (3) close to the contact end (321) of the micro-cone structure (32), with a gap between the negative friction layer (4) and the contact end (321) of the micro-cone structure (32), and a Young's modulus of the negative friction layer (4) being a second value, and the second value being smaller than the first value; A deformation electrode (5) connected to a side of the negative friction layer (4) away from the positive friction layer (3); The flexible electrode (2) is connected to the side of the positive friction layer (3) away from the negative friction layer (4).

2. The friction electric pressure sensor according to claim 1, characterized in that: The thickness of the negative friction layer (4) is smaller than the height of the micro-cone structure (32), and the Young's modulus of the deformation electrode (5) is a third value, and the third value is smaller than the second value.

3. The friction electric pressure sensor according to claim 1, characterized in that: The negative friction layer (4) is made of a silica gel material, the deformable electrode (5) is a hydrogel structure, and the deformable electrode (5) is wrapped inside the negative friction layer (4).

4. The friction electric pressure sensor according to claim 3, characterized in that: The mass ratio of polyvinyl alcohol, phytic acid and deionized water in the deformable electrode (5) is any value in the range of (1-4): (9.8-10.2): (9.8-10.2).

5. The friction electric pressure sensor according to claim 1, characterized in that: The height of the micro-cone structure (32) is less than or equal to any value between 0.5 mm and 1 mm.

6. The friction electric pressure sensor according to claim 1, characterized in that: The positive friction layer (3) further comprises a support portion (33), wherein the support portion (33) has a first end (331) and a second end (332) opposite to the first end (331), an end of each of the micro-cone structures (32) away from the contact end (321) is supported on the first end (331) of the support portion (33), and the second end (332) of the support portion (33) is supported on the negative friction layer (4).

7. The triboelectric pressure sensor according to any one of claims 1 to 6, characterized in that: The first sensitivity of the friction electric pressure sensor is any value between 3V / kPa and 8V / kPa, the second sensitivity of the friction electric pressure sensor is any value between 0.4V / kPa and 1.6V / kPa, and the third sensitivity of the friction electric pressure sensor is any value between 0.04V / kPa and 0.15V / kPa.

8. A sleeping system, characterized in that: The invention comprises a first sensor, a second sensor and a controller electrically connected to the first sensor and the second sensor, wherein the first sensor and the second sensor are friction electric pressure sensors according to any one of claims 1 to 7, the first sensor is used to convert a pulse signal of a user into a first electrical signal, the second sensor is used to convert a breathing signal of the user into a second electrical signal, the controller is used to receive the first electrical signal and the second electrical signal, and output the sleep condition data of the user in combination with a learning model obtained by training a deep learning network model.

9. A sleep monitoring method using the system as claimed in claim 8, characterized in that: include: Collecting the pulse signals and the breathing signals of multiple volunteers to form multiple groups of the first electrical signals and the second electrical signals; Performing data segmentation and data enhancement processing on each group of the first electrical signal and the second electrical signal respectively to obtain a plurality of model pulse waves and a plurality of model respiratory waves; The model pulse wave and the model respiratory wave are used as inputs and the respiratory type is used as output to learn and obtain the learning model, wherein the respiratory type includes three classification groups: a normal type, a hypopnea type and an apnea type; Collecting the pulse signal and the breathing signal of the user, and performing the data segmentation and data enhancement processing to obtain a detected pulse wave and a detected breathing wave; Inputting the detected pulse wave and the detected respiratory wave into the learning model to obtain the corresponding respiratory type, and performing abnormal counting on the total number of the hypopnea type and the apnea type; When the value obtained by counting the number of abnormal times reaches a preset value, an early warning signal is issued.

10. The method according to claim 9, characterized in that The data segmentation is performed according to time series, and the data enhancement is a time series interpolation method.

Images