Two-dimensional pressure sensing array based on pressure change fibers, intelligent cushion and system

Abstract

The invention belongs to the field of smart home, and particularly relates to a two-dimensional pressure sensing array based on pressure change fibers, an intelligent cushion and a system. The two-dimensional pressure sensing array comprises a base material made of a textile material, and a plurality of warp fibers and weft fibers which are embedded into the base material by adopting a textile process. The pressure change fibers can generate corresponding electric signals according to the pressure borne by the pressure change fibers; the sheath layer is composed of a tubular sheath layer and a linear core layer in the sheath layer; the core layer adopts a wire of which the surface is coated with a dielectric layer; and the sheath layer is prepared from a material which has electronegativity difference with the dielectric layer. A two-dimensional pressure sensing array and a driving circuit of the two-dimensional pressure sensing array are integrated into a conventional cushion, the intelligent cushion with a pressure monitoring function is obtained, and various physiological monitoring and early warning functions can be further achieved based on data collected by the intelligent cushion. The problems that the resolution ratio of a pressure sensor in an existing intelligent cushion product is low, and diversified physiological feature perception cannot be achieved are solved.

Description

Technical Field

[0001] This invention belongs to the field of smart home, specifically relating to a two-dimensional pressure sensing array based on pressure-varying fibers, a corresponding smart cushion, and a physiological sensing and early warning system based on the smart cushion. Background Technology

[0002] In existing technologies, a few seat cushions integrate pressure sensing functions, but their sensors are usually based on traditional capacitive or resistive principles. These sensors are often sparsely distributed and have low resolution, only able to achieve coarse pressure monitoring of large areas to determine "occupant / unoccupied" status or obtain the overall average pressure value, unable to accurately capture subtle differences in sitting posture (such as slight shifts in the body's center of gravity). Therefore, based on this low-precision data, existing products can only achieve a simple "sedentary reminder" function, unable to accurately identify, classify, and warn of specific poor sitting postures, let alone provide high-resolution visual feedback with practical guidance for posture correction and mental state monitoring. In addition, the aforementioned pressure sensors based on capacitive or resistive principles require continuous external power to maintain the electric field or detect changes in resistance, resulting in high system power consumption. This forces the product to either charge frequently or remain connected to a power cord, severely limiting its convenience and long-term user experience in mobile scenarios such as driving, making it difficult to achieve truly seamless, all-day sitting health management. Summary of the Invention

[0003] To address the issue that existing smart cushion products have low-resolution pressure sensors that cannot detect diverse physiological characteristics, this invention provides a two-dimensional pressure sensing array based on pressure-varying fibers, a corresponding smart cushion, and a physiological sensing and early warning system based on the smart cushion.

[0004] This invention is achieved using the following technical solution: A two-dimensional pressure sensing array based on piezoelectric fibers includes a substrate made of textile material and several warp and weft fibers embedded in the substrate using a textile process. The warp and weft fibers are piezoelectric fibers capable of outputting an electrical signal related to the pressure value when pressure is applied at any point on the fiber. The piezoelectric fiber consists of a tubular sheath and an internal linear core layer; the core layer is a conductive wire with a dielectric layer on its surface; the sheath layer is made of a material with a different electronegativity than the dielectric layer. The sheath layer serves both as an encapsulation layer to provide the tactile feel of the fabric and as a friction layer, forming a contact-separated triboelectric nanogenerator with the dielectric layer.

[0005] Each warp or weft fiber determines the pressure value in the corresponding direction based on the output electrical signal. By combining the electrical signals output by all warp and weft fibers in the array, the pressure value of the sensing node corresponding to any intersection of warp and weft fibers is determined, thereby realizing two-dimensional pressure sensing.

[0006] As a further improvement of the present invention, any intersection of a set of warp and weft fibers is used as a sensing node; the number of warp and weft fibers embedded in the substrate are N and M, respectively; the resolution of the two-dimensional pressure sensing array is N×M.

[0007] As a further improvement of the present invention, the conductor of the core layer of the compression-converter fiber is made of metal wire or filament formed of non-metallic conductive fiber.

[0008] As a further improvement of the present invention, the sheath is made of cotton yarn, nylon, polyester or any one or more other textile fibers, including natural fibers and synthetic fibers.

[0009] As a further improvement of the present invention, the dielectric layer is made of organic polymer material or inorganic material.

[0010] As a further improvement of the present invention, the positions of the warp and weft fibers are optimized to achieve patterning of the spatial distribution of the sensing nodes.

[0011] As a further improvement of the present invention, the method for generating the pressure values ​​of each sensing node in the two-dimensional pressure sensing array is as follows: Based on the pre-calibrated "pressure-electric signal" mapping of the pressure-transformer fibers, the electrical signals output by each warp and weft fiber are converted into corresponding pressure values.

[0012] The pressure value of each pressure-transformer fiber is the sum of the pressure values ​​of all sensing nodes in the corresponding row or column.

[0013] When the resolution of the two-dimensional pressure sensing array is less than the preset resolution threshold, the pressure value of each sensing node is solved by point-by-point calibration, combining the pressure values ​​of all warp and weft fibers; otherwise, a pre-trained neural network-based deep learning model is used to generate the pressure value of each sensing node based on the pressure values ​​of each warp and weft fiber.

[0014] The present invention also includes a smart cushion, comprising a cushion body, and further comprising: a two-dimensional pressure sensing array based on piezoelectric fibers as described above, a signal acquisition circuit, and a data processing circuit. The two-dimensional pressure sensing array based on piezoelectric fibers is embedded in the cushion body. The signal acquisition circuit is electrically connected to the conductors in each warp and weft fiber of the two-dimensional pressure sensing array based on piezoelectric fibers; and acquires the electrical signals output by each warp and weft fiber.

[0015] The data processing module first converts the electrical signals output by each warp and weft fiber into corresponding pressure values ​​according to the pre-calibrated "pressure-electrical signal" mapping of the pressure-transformer fibers. Then, it combines the pressure values ​​of each warp and weft fiber to solve for the pressure value of each sensing node, thereby generating a pressure distribution map.

[0016] As a further improvement of the present invention, the smart cushion also includes a wireless transmission module, which is electrically connected to the data processing module and is used to upload the collected data, and / or process data, and / or processing results of the data processing module to a cloud platform or smart terminal.

[0017] The cloud platform or user terminal is used to analyze the user's physiological state based on the time-series data of the collected pressure distribution map, thereby realizing multiple life sensing and early warning functions, including pressure monitoring, pressure sore warning, sitting posture monitoring, poor sitting posture warning, fall warning, and mental state monitoring; and to visualize the various data.

[0018] As a further improvement of the present invention, the method for realizing the life sensing and early warning function includes: (1) Determine the sensing node corresponding to the maximum pressure at each moment based on the pressure distribution map, and issue a pressure ulcer warning when the duration of the pressure state of the corresponding sensing node exceeds the preset time threshold.

[0019] (2) Analyze the overall pressure distribution characteristics, including the pressure center and contact area, and use machine learning algorithms to identify and warn of hunching, tilting or other bad sitting postures.

[0020] (3) Monitor sudden and drastic changes in pressure, including a sudden drop in total pressure and a rapid shift in the center of pressure, and then determine whether the user has suddenly fallen and issue an alarm.

[0021] (4) Analyze the dynamic change pattern of the user's long-term sitting posture, including the slight shaking of the pressure center and the frequency of sitting posture switching; and combine the breathing and heart rate data extracted from the pressure signal to assess the user's fatigue and mental stress status.

[0022] As a further improvement of the present invention, the signal acquisition circuit includes a signal sampling circuit, a signal amplification circuit, a filtering circuit, and an analog-to-digital converter; As a further improvement of the present invention, the wireless transmission module includes any one or more of the following: IoT communication module, Wi-Fi module, Bluetooth module, LoRa module, and ZigBee module.

[0023] The present invention also includes a physiological sensing and early warning system based on a smart cushion, which includes: the aforementioned smart cushion, cloud platform and user terminal.

[0024] The smart cushion generates a pressure distribution map at a preset sampling frequency when the user uses the product and uploads it to a cloud platform. The cloud platform analyzes the user's physiological state based on the time-series data of the collected pressure distribution map, thereby enabling multiple life-sensing and early warning functions, including pressure monitoring, pressure sore warning, posture monitoring, poor posture warning, fall warning, and mental state monitoring. The user terminal communicates with the cloud platform. Users can log in to their personal accounts to interact with the cloud platform, gain access to various data, use the life-sensing and early warning functions, and visualize the monitoring data.

[0025] The technical solution provided by this invention has the following beneficial effects: This invention provides a novel two-dimensional pressure sensing array for seat cushions, woven from pressure-varying fibers based on the TENG (triboelectric nanogenerator) principle, enabling the monitoring of vertical pressure. This two-dimensional pressure sensing array is a flexible sensor with a fabric-like shape and feel, and can achieve high-resolution detection of pressure distribution in a plane at low cost, thus laying the foundation for realizing various intelligent sensing functions based on seat cushions.

[0026] The smart seat cushion and its integrated system provided in this application utilize user pressure to deform the fibrous pressure sensor, generating an electrical signal. The signal acquisition circuit and data processing module collect and analyze the data in real time, enabling functions such as pressure monitoring, pressure ulcer warning, posture monitoring, poor posture warning, fall warning, and mental state monitoring. The accompanying mobile terminal allows for data storage, real-time display, and human-computer interaction. The technology used in this application requires no power supply to the sensor, offering extremely high safety. The interface displays rich information, and the human-computer interaction is simple, making it suitable for use in homes, communities, hospitals, driving, offices, and other scenarios, with broad application prospects. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the structure of the compression-modulated fiber provided in Embodiment 1 of the present invention.

[0028] Figure 2 This is a schematic diagram illustrating the principle of pressure sensing achieved by pressure-sensitive fiber in Embodiment 1 of the present invention.

[0029] Figure 3 This is a schematic diagram of the structure of the two-dimensional pressure sensing array provided in Embodiment 1 of the present invention.

[0030] Figure 4 This is a schematic diagram of the intelligent seat cushion in Embodiment 2 of the present invention.

[0031] Figure 5A schematic diagram of a smart pillow with wireless data transmission and alarm functions is provided for Embodiment 2 of the present invention.

[0032] Figure 6 This is an architecture diagram of the physiological sensing and early warning system based on a smart cushion provided in Embodiment 3 of the present invention.

[0033] The diagram is marked as follows: 1. Substrate; 2. Warp fiber; 3. Weft fiber; 4. Sheath; 5. Dielectric layer; 6. Conductor. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0035] Example 1 To address the limitations of existing smart cushion products on the market, such as limited functionality, reliance on electricity for operation (posing safety risks), high power consumption, and restricted application scenarios, this embodiment provides a two-dimensional pressure sensing array based on pressure-varying fibers. This two-dimensional pressure sensing array is fabricated from fibrous, self-driven, passive pressure-varying fibers based on the principle of triboelectric nanogenerators. When applied to a smart cushion, this two-dimensional pressure sensing array can provide a tactile experience consistent with traditional fabrics, and, using existing textile manufacturing processes, achieve high-resolution, dot-matrix acquisition of positive pressure, thereby generating a pressure distribution map. This lays the foundation for functions such as posture recognition, fall warning, and multi-dimensional health risk monitoring when users use the smart cushion.

[0036] In this embodiment, the pressure-sensitive fiber is an innovative "pressure sensor." This type of sensor is a thread that resembles conventional textile yarn in appearance and feel. The difference lies in the fact that any point within the pressure-sensitive fiber undergoes localized deformation when subjected to pressure, generating an electrical signal related to the magnitude of the localized pressure at the end of the conductor 6 in its core layer. This achieves the pressure sensing function.

[0037] Specifically, such as Figure 1As shown, the piezoelectric fiber consists of a tubular sheath 4 and an internal linear core layer. The core layer is made of a conductor 6 with a dielectric layer 5 covering its surface. The sheath 4 is made of a material with a different electronegativity than the dielectric layer 5. In practical applications, the conductor 6 in the core layer of the piezoelectric fiber can be made of metal wire, such as highly conductive and ductile elemental metals like gold, silver, copper, iron, and aluminum, or alloy materials containing the aforementioned metal elements. Alternatively, it can be made of non-metallic conductive fibers, such as carbon fibers with conductive additives, doped carbon nanotubes, or coated with graphene. The dielectric layer 5 covering the surface of the conductor 6 in the core layer can be made of organic polymer materials or inorganic materials, such as polyethylene, polyvinyl chloride, polypropylene, and various fluorinated resins (including FEP, PVDF, PVDF-TrFE, PVDF-HFP, PFA, PFPE, etc.). The sheath layer 4 can be made of any one or more textile fibers. In this embodiment, the textile fibers used in the sheath layer 4 can include various natural fibers, such as cotton, linen, and wool, or various synthetic fibers, such as nylon, acrylic, polyester, polyurethane fibers, etc. The sheath layer 4 in the pressure fiber serves a dual function: firstly, it acts as an encapsulation layer, providing a fabric-like feel; secondly, it acts as a friction layer, forming a contact-separated triboelectric nanogenerator with the dielectric layer 5. This allows for triboelectric charging with the dielectric layer 5 under pressure, triggering the output of an electrical signal representing the pressure value from the conductors 6 in the core layer.

[0038] In detail, the pressure sensing principle of the pressure-sensitive fiber in this embodiment is as follows: Figure 2 As shown, in the initial state, there is a certain gap between the core layer and the sheath layer 4 of the pressure-sensitive fiber. Under electrostatic induction, the dielectric layer 5 and the sheath layer 4 will carry opposite charges. When pressure is applied to any part of the pressure-sensitive fiber, the sheath layer 4 at that location will deform and gradually move closer to the core layer. When the sheath layer 4 contacts the core layer, the sheath layer 4 will rub against the dielectric layer 5 on the surface of the core layer, triggering charge transfer. According to the principle of triboelectric power generation, the wire 6, which is in direct electrical contact with the dielectric layer 5, will output a corresponding electrical signal. When the external force is removed, the deformation of the sheath layer 4 will recover, and it will re-detach from the core layer. The above description uses the example of a gap between the sheath layer 4 and the core layer. In practical applications, the sheath layer 4 and the core layer can also be in direct physical contact in the initial state, and the friction generated by the relative sliding between the sheath layer 4 and the core layer after deformation under pressure will generate electricity and output an electrical signal corresponding to the pressure. Specifically, in this embodiment, the electrical signal output by the pressure-transformer fiber is related to the magnitude of the pressure and the size of the physical contact area caused by the pressure. Under single-point contact conditions, the greater the applied pressure, the greater the output electrical signal; the smaller the pressure, the smaller the output electrical signal. When the sheath layer 4 and the core layer make physical contact due to pressure at various points, the larger the contact area, the greater the electrical signal; the smaller the contact area, the smaller the electrical signal.

[0039] Based on the pressure-sensitive fiber described above, the two-dimensional pressure sensing array provided in this embodiment is as follows: Figure 3 As shown, it includes a substrate 1 made of textile material, and several warp fibers 2 and weft fibers 3 embedded into the substrate 1 using a textile process. The warp fibers 2 and weft fibers 3 are pressure-varying fibers that can output an electrical signal related to the pressure value when pressure is applied at any point on the fiber. In practical applications, when using the substrate 1 for fabric weaving, the pressure-varying fibers can be incorporated into the substrate 1 and embedded into the processed fabric along the warp and weft directions, respectively, forming a mesh-like distribution of warp fibers 2 and weft fibers 3.

[0040] exist Figure 3 In the structure shown, each warp fiber 2 or weft fiber 3 in the two-dimensional pressure sensing array provided in this embodiment can be used to detect pressure in the corresponding direction, and then determine the pressure value in the corresponding direction based on the output electrical signal. This embodiment uses the intersection of any set of warp fibers 2 and weft fibers 3 in the two-dimensional pressure sensing array as a sensing node; based on this, combined with the electrical signals output by all warp fibers 2 and weft fibers 3 in the array, the pressure value of the sensing node corresponding to any intersection of warp fibers 2 and weft fibers 3 can be further determined; thus realizing two-dimensional pressure sensing. Assuming that the number of warp fibers 2 and weft fibers 3 embedded in the substrate 1 are N and M respectively, the resolution of the two-dimensional pressure sensing array is N×M. That is, the pressure values ​​at N×M points can be detected respectively. The number of sensing nodes is not limited in the scheme provided in this embodiment. In practical applications, the number of sensing nodes depends on the number of pressure-varying fibers used in the warp and weft directions. The more pressure fibers there are, the denser the distribution of sensing nodes, and the higher the resolution of the pressure distribution imaging. Given the low cost of pressure-varying fibers in this embodiment, it is easy to achieve higher resolution pressure distribution monitoring, which further lays the foundation for achieving higher accuracy in seat posture classification, etc.

[0041] In practical applications, this embodiment uses the following steps to generate the pressure values ​​of each sensing node based on the electrical signals output by the warp fibers 2 and weft fibers 3 in the two-dimensional pressure sensing array: (1) Based on the pre-calibrated pressure-electric signal mapping of the pressure-transformer fiber, the electrical signals output by each warp fiber 2 and weft fiber 3 are converted into corresponding pressure values.

[0042] As previously mentioned, the strength of the electrical signal output by the pressure-transformer fiber is related to the pressure value. Therefore, through a large number of tests, the value of the electrical signal output by the pressure-transformer fiber under different pressure states can be obtained, and then the "pressure-electrical signal" mapping can be established.

[0043] (2) The pressure value of each pressure-transformer fiber is the sum of the pressure values ​​of all sensing nodes in the corresponding row or column.

[0044] As mentioned earlier, applying pressure at any point on the PBT fiber will generate an electrical signal output. Therefore, the magnitude of the final electrical signal output by the PBT fiber is actually the sum of the electrical signals generated by all points along the entire line under pressure. According to the principle of vector superposition, the pressure value measured on each PBT fiber should actually be the sum of the pressure values ​​of all sensing nodes in the corresponding row or column.

[0045] (3) Finally, by combining the pressure values ​​of all warp fibers 2 and weft fibers 3, the pressure value of each sensing node is calculated: This embodiment provides different methods for calculating pressure values ​​for two-dimensional pressure sensing arrays with different resolutions. When the resolution is low, i.e., the number of sensing nodes is small, the pressure value of each sensing node can be solved using a point-by-point calibration method. However, when the resolution of the two-dimensional pressure sensing array is high, the difficulty of solving by point-by-point calibration increases, and the space of feasible solutions becomes larger. In this case, this embodiment can design and train a deep learning model based on a neural network. The pre-trained deep learning model generates the pressure value of each sensing node based on the pressure values ​​of each warp fiber 2 and weft fiber 3.

[0046] In this embodiment, during the deep learning model training phase, a two-dimensional pressure sensing array can be placed on a traditional pressure sensor array with a dotted distribution. Pressures with different spatial distributions are then applied to it, and the electrical signals D output by each warp fiber 2 and weft fiber 3 in the two-dimensional pressure sensing array are acquired: D = {d1, d2, ..., D...}. N+M}; and the pressure values ​​F at each point in a traditional pressure sensor array: F ​​= {f1, f2, ..., f N×M Then, the associated binary array DM is used as sample data to obtain a large number of measured binary arrays representing the outputs of two-dimensional pressure sensor arrays under different pressure states, which are then used as sample datasets for training the aforementioned deep learning model.

[0047] The previously described scheme uses a grid-like array of sensing nodes formed by the cross-shaped intersection of warp and weft fibers, with each fiber extending in a straight line. In practical applications, to meet different needs, the warp fibers in each direction can also extend along a curved trajectory. This allows the sensing nodes formed by the intersection of warp fiber 2 and weft fiber 3 to be distributed in a planar space according to a specified pattern, thus meeting the differentiated pressure detection requirements in different scenarios. With this non-uniform patterned distribution of sensing nodes, the number of sensing nodes can be increased in key areas to enhance local resolution, while in non-key areas, the number of sensing nodes can be decreased to reduce local resolution.

[0048] Example 2 Based on the two-dimensional pressure sensor array with a flexible fabric feel and high-resolution pressure detection function provided in Embodiment 1, this embodiment further provides a smart cushion. For example... Figure 4 As shown, the smart cushion includes a cushion body, a two-dimensional pressure sensing array based on pressure-varying fibers as in Example 1, a signal acquisition circuit, and a data processing circuit.

[0049] The seat cushion body refers to a traditional seat cushion without detection functions, which is generally composed of a surface layer and internal filling material. In the smart seat cushion of this embodiment, a two-dimensional pressure sensor array based on pressure-varying fibers is embedded in the seat cushion body. For example, in practical applications, the two-dimensional pressure sensor array can be glued or sewn to the inside of the surface layer of the seat cushion body, or it can be embedded in the center of the filling layer.

[0050] The signal acquisition circuit is electrically connected to the conductors 6 in each warp fiber 2 and weft fiber 3 of the two-dimensional pressure sensing array based on piezoelectric fibers; and acquires the electrical signals output by each warp fiber 2 and weft fiber 3. In practical applications, the signal acquisition circuit further includes a signal sampling circuit, a signal amplification circuit, a filtering circuit, and an analog-to-digital converter. The signal sampling circuit acquires the electrical signals output by each piezoelectric fiber at a preset sampling frequency; the signal amplification circuit amplifies the weak electrical signals acquired to a threshold range suitable for subsequent processing; the filtering circuit filters out noise contained in the acquired signals; and the analog-to-digital converter converts the acquired analog signals using an analog-to-digital converter (AD converter) to obtain the corresponding digital signals.

[0051] The data processing module is electrically connected to the signal acquisition circuit. It acquires the digital signals representing the strength of the electrical signals output by each pressure-sensitive fiber from the signal acquisition circuit. First, according to a pre-calibrated "pressure-electrical signal" mapping for the pressure-sensitive fibers, the electrical signals output by each warp fiber 2 and weft fiber 3 are converted into corresponding pressure values. Then, combining the pressure values ​​of each warp fiber 2 and weft fiber 3, the pressure value of each sensing node is calculated, thereby generating a pressure distribution map. The pressure distribution map contains the position signal of each sensing node in the two-dimensional pressure sensing array, as well as the pressure magnitude of that node. In practical applications, the pressure distribution map can be represented and transmitted in the form of a matrix or vector.

[0052] In practical applications, the spatial distribution information of pressure collected by the smart cushion in different areas can be transmitted to external systems or devices to support diverse functions and applications. In this embodiment, no limitations are placed on the data transmission method or protocol. Data transmission can be performed via wired connection or wireless local area network (LAN) setup. In a further optimized embodiment, the smart cushion may include a wireless transmission module electrically connected to the data processing module, used to upload the collected data, and / or process data, and / or processing results from the data processing module to a cloud platform or smart terminal. The wireless transmission module in this embodiment can employ any one or more of the following: a 4G or 5G IoT-based IoT communication module, a Wi-Fi module, a Bluetooth module, a LoRa module, or a ZigBee module. In actual use, the module can be flexibly configured according to the actual data transmission volume and network environment.

[0053] Based on the pressure distribution map collected by the smart cushion during user use, the data processing module in this embodiment can further analyze the user's physiological state using the time-series data of the collected pressure distribution map, thereby realizing multiple life sensing and early warning functions. Of course, in practical applications, the data processing load for these functions is relatively large, so it can also be implemented through a cloud platform or user terminal that receives the corresponding data. The life sensing and early warning functions that can be implemented in this embodiment include pressure monitoring, pressure sore warning, posture monitoring, poor posture warning, fall warning, and mental state monitoring; and the data is visualized. Specifically, the implementation methods of the above-mentioned life sensing and early warning functions include: (1) Determine the sensing node corresponding to the maximum pressure at each moment based on the pressure distribution map, and issue a pressure ulcer warning when the duration of the pressure state of the corresponding sensing node exceeds the preset time threshold.

[0054] (2) Analyze the overall pressure distribution characteristics, including the pressure center and contact area, and use machine learning algorithms to identify and warn of hunching, tilting or other bad sitting postures.

[0055] For example, features such as total pressure, average pressure, pressure center coordinates, and effective contact area applied to the seat cushion during user use can be extracted from the time-series data of the pressure distribution map. Then, machine learning classification algorithms (such as Bayesian optimized decision tree models) are used to classify sitting postures and determine poor sitting postures such as tilted sitting posture and 1 / 3 sitting posture. In addition, warnings are triggered for poor sitting postures that have been present for a long time.

[0056] (3) Monitor sudden and drastic changes in pressure, including a sudden drop in total pressure and a rapid shift in the center of pressure, and then determine whether the user has suddenly fallen and issue an alarm.

[0057] (4) Analyze the dynamic change pattern of the user's long-term sitting posture, including the slight shaking of the pressure center and the frequency of sitting posture switching; and combine the breathing and heart rate data extracted from the pressure signal to assess the user's fatigue and mental stress status.

[0058] For example, when a user uses a smart cushion, the trajectory of the pressure center is calculated based on changes in posture over a long period. This trajectory data is then subjected to high-frequency filtering to detect subtle tremors caused by excessive mental stress. The percentage of time the user maintains their primary sitting posture is calculated, and combined with changes in the pressure center's movement trajectory and effective contact area, a cumulative entropy value for posture switching is defined, which is correlated with fatigue-induced inattention.

[0059] Alternatively, by filtering and performing wavelet transform on the pressure signal, respiratory and cardiac impact signals can be separated, and then respiratory and heart rates can be calculated. The calculated respiratory and heart rates, along with their variability and other characteristics, are then fused with dynamic posture data for multimodal data analysis. A temporal deep learning model (such as an attention-based transformer model) is then used to analyze fatigue or mental stress, issuing warnings for excessive mental stress or fatigue. The algorithms for detecting mental stress and fatigue can be calibrated using heart rate variability data.

[0060] To achieve the alarm function, such as Figure 5 As shown, the data processing module is also connected to an alarm module, which issues an audible and visual alarm upon receiving an alarm command to alert the user. In practical applications, the thresholds for triggering alarms based on factors such as pressure ulcer risk, poor posture, mental stress level, and fatigue can be manually set by the user.

[0061] Example 3 Based on the two schemes in Example 2, such as Figure 6 As shown, this embodiment further provides a physiological sensing and early warning system based on a smart cushion, which includes: a smart cushion, a cloud platform, and a user terminal as in embodiment 2.

[0062] The smart cushion generates a pressure distribution map at a preset sampling frequency when the user uses the product and uploads it to a cloud platform. The cloud platform's backend server analyzes the user's physiological state based on the time-series data of the collected pressure distribution map, thereby enabling multiple life-sensing and early warning functions, including pressure monitoring, pressure sore warning, posture monitoring, poor posture warning, fall warning, and mental state monitoring. The user terminal communicates with the cloud platform. Users can log in to their personal accounts to interact with the cloud platform, gain access to various data, and use the various life-sensing and early warning functions. In practical applications, the user terminal can also use a display screen to visualize the monitoring data during the process, such as the heat map corresponding to the pressure distribution, or display various physiological monitoring data extracted based on the collected pressure signals, and statistically display various alarm events generated by overweighting.

[0063] For example, in a typical application scenario, a smart cushion can communicate with the cloud platform's backend server via a wireless communication module (such as Wi-Fi or 4G). After the user uses the cushion, the smart cushion periodically collects and uploads pressure detection data. The backend service processes the collected data and generates relevant data for each user's life-sensing and early warning functions, storing this data in a database. The monitoring functions and services provided by the cloud platform can be offered via an app. Users or their guardians can log in to the app using their mobile phones, tablets, smartwatches, or other user terminals to access the cloud platform's database, view the relevant data, and receive alerts directly from the cloud platform when alarm events occur.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A two-dimensional pressure sensing array based on piezoelectric fibers, characterized in that, It includes a substrate made of textile material, and several warp and weft fibers embedded in the substrate using a textile process; the warp and weft fibers are piezoelectric fibers that can output an electrical signal related to the pressure value when pressure is applied at any point on the fiber; the piezoelectric fibers consist of a tubular sheath and an internal linear core layer; the core layer is a conductor with a dielectric layer on its surface. The sheath is made of a material with a different electronegativity than the dielectric layer; the sheath serves both as an encapsulation layer to provide the feel of a fabric and as a friction layer to form a contact-separated triboelectric nanogenerator with the dielectric layer. Each warp or weft fiber determines the pressure value in the corresponding direction based on the output electrical signal. By combining the electrical signals output by all warp and weft fibers in the array, the pressure value of the sensing node corresponding to any intersection of warp and weft fibers is determined, thereby realizing two-dimensional pressure sensing.

2. The two-dimensional pressure sensing array based on piezoelectric fibers as described in claim 1, characterized in that: The intersection of any set of warp and weft fibers serves as a sensing node; the number of warp and weft fibers embedded in the substrate are N and M, respectively; the resolution of the two-dimensional pressure sensing array is N×M.

3. The two-dimensional pressure sensing array based on piezoelectric fibers as described in claim 1, characterized in that: The conductors of the core layer of the compression-transformer fiber are made of metal wires or filaments formed of non-metallic conductive fibers. And / or, the sheath is made of cotton yarn, nylon, polyester or any one or more other textile fibers, including natural fibers and synthetic fibers; And / or, the dielectric layer is made of organic polymer material or inorganic material.

4. The two-dimensional pressure sensing array based on piezoelectric fibers as described in claim 3, characterized in that: The positions of warp and weft fibers are optimized to achieve patterning of the spatial distribution of sensing nodes.

5. The two-dimensional pressure sensing array based on piezoelectric fibers as described in claim 1, characterized in that: The method for generating the pressure values ​​of each sensing node in a two-dimensional pressure sensor array is as follows: Based on the pre-calibrated "pressure-electric signal" mapping of the pressure-transformer fiber, the electrical signal output by each warp and weft fiber is converted into the corresponding pressure value; The pressure value of each pressure-transformer fiber is used as the sum of the pressure values ​​of all sensing nodes in the corresponding row or column; When the resolution of the two-dimensional pressure sensing array is less than the preset resolution threshold, the pressure value of each sensing node is solved by combining the pressure values ​​of all warp and weft fibers through a point-by-point calibration method. Conversely, a pre-trained neural network-based deep learning model is used to generate the pressure values ​​of each sensing node based on the pressure values ​​of each warp and weft fiber.

6. A smart seat cushion, comprising a seat cushion body, characterized in that, It also includes: The two-dimensional pressure sensing array based on pressure-varying fibers as described in any one of claims 1-5 is embedded in the seat cushion body; The signal acquisition circuit is electrically connected to the conductors in each warp and weft fiber of the two-dimensional pressure sensing array based on the pressure-varying fiber; and acquires the electrical signals output by each warp and weft fiber. The data processing module first converts the electrical signals output by each warp and weft fiber into corresponding pressure values ​​according to the pre-calibrated "pressure-electrical signal" mapping of the pressure-transformer fiber; then, it combines the pressure values ​​of each warp and weft fiber to solve for the pressure value of each sensing node; and finally generates a pressure distribution map.

7. The smart seat cushion as described in claim 6, characterized in that: It also includes a wireless transmission module, which is electrically connected to the data processing module and is used to upload the collected data, and / or process data, and / or processing results of the data processing module to a cloud platform or smart terminal. The cloud platform or user terminal is used to analyze the user's physiological state based on the time-series data of the collected pressure distribution map, thereby realizing multiple life sensing and early warning functions, including pressure monitoring, pressure sore warning, sitting posture monitoring, poor sitting posture warning, fall warning, and mental state monitoring; and to visualize the various data.

8. The smart seat cushion as described in claim 7, characterized in that: The methods for the life sensing and early warning functions include: (1) Determine the sensing node corresponding to the maximum pressure at each moment based on the pressure distribution map, and issue a pressure ulcer warning when the duration of the pressure state at the corresponding sensing node exceeds the preset duration threshold. (2) Analyze the overall pressure distribution characteristics, including the pressure center and contact area, and use machine learning algorithms to identify and warn of hunching, leaning to the side or other bad sitting postures; (3) Monitor sudden and drastic changes in pressure, including a sudden drop in total pressure and a rapid shift in the center of pressure, and then determine whether the user has suddenly fallen and issue an alarm; (4) Analyze the dynamic change pattern of the user's long-term sitting posture, including the slight shaking of the pressure center and the frequency of sitting posture switching; and combine the breathing and heart rate data extracted from the pressure signal to assess the user's fatigue and mental stress status.

9. The smart seat cushion as described in claim 7, characterized in that: The signal acquisition circuit includes a signal sampling circuit, a signal amplification circuit, a filtering circuit, and an analog-to-digital converter; And / or, the wireless transmission module includes any one or more of the following: IoT communication module, Wi-Fi module, Bluetooth module, LoRa module, and ZigBee module.

10. A physiological sensing and early warning system based on a smart seat cushion, comprising: The smart cushion as described in claim 8 is used to generate the pressure distribution map according to a preset sampling frequency when the user uses it, and upload it to a cloud platform; The cloud platform is used to analyze the user's physiological state based on the time-series data of the collected pressure distribution map, thereby realizing multiple life sensing and early warning functions, including pressure monitoring, pressure sore warning, sitting posture monitoring, poor sitting posture warning, fall warning, and mental state monitoring. as well as The user terminal communicates with the cloud platform; the user interacts with the cloud platform by logging into their personal account, thereby obtaining access to various data, using various life sensing and early warning functions, and realizing the visualization of monitoring data.

Images