Waterproof battery-free flexible electronic nose capable of being used for gas recognition

Abstract

The invention relates to the technical field of gas sensors and intelligent wearing, in particular to a waterproof battery-free flexible electronic nose capable of being used for gas recognition, which comprises an upper layer waterproof breathable film, a lower layer waterproof breathable film and a porous flexible circuit board arranged between the upper layer waterproof breathable film and the lower layer waterproof breathable film, a processor and a gas sensor array are integrated on the porous flexible circuit board, and the processor is used for collecting analog signals of the gas sensor array and carrying out gas identification based on the analog signals. According to the invention, high-sensitivity gas detection, omni-directional waterproof and low-power-consumption operation can be realized, and the device is suitable for wearable medical monitoring, industrial environment detection, emergency rescue and other scenes.

Description

Technical Field

[0001] The present invention relates to the field of gas sensors and smart wearable technologies, and in particular to a waterproof, battery-free, flexible electronic nose that can be used for gas identification. Background Art

[0002] With the increasing problem of harmful gas pollution in indoor environments and vehicle spaces, gas sensing technology based on metal oxide semiconductors (MOS) and carbon-based materials has become a research hotspot in the field of environmental monitoring due to its high sensitivity and fast response characteristics. Traditional electronic nose systems generally have the problems of bloated size and excessive power consumption. For example, although the array structure composed of 2-100 sensors in the existing technology can realize the classification detection of volatile organic compounds (VOCs) such as benzene, the complex hardware architecture makes it difficult to miniaturize the equipment, and the cumulative power consumption generated by the collaborative work of multiple sensors seriously restricts the endurance in wearable scenarios.

[0003] Furthermore, traditional electronic noses often utilize rigid circuit boards and wired power supply designs, which present significant deficiencies in flexibility, environmental protection, and wearer comfort. These devices are particularly difficult to meet the needs of long-term monitoring in confined spaces within vehicles or on dynamic human surfaces. Although MEMS technology has enabled sensor miniaturization through micro-nanofabrication in recent years, the cross-sensitivity issue of electronic noses in complex mixed gas environments remains a challenge. Existing research has primarily focused on material modification or array optimization, failing to fundamentally address the conflict between hardware redundancy and insufficient selectivity.

[0004] In the field of wearable gas detection devices, existing technologies face two core challenges: power supply and packaging. The current mainstream solution relies on lithium batteries for power, but battery modules significantly increase the size and weight of the device, and frequent charging affects the user experience. Furthermore, the energy supply problem for flexible electronic devices has not yet been effectively solved, and wireless energy transmission technology faces bottlenecks such as low efficiency and poor stability in practical applications. Furthermore, traditional packaging materials struggle to balance breathability and protection: rigid packaging affects device flexibility, while ordinary polymer films are susceptible to water vapor penetration, interfering with sensing performance.

[0005] Gas identification algorithms and embedded system deployment capabilities are core bottlenecks hindering the practical application of electronic nose technology. At the algorithmic level, traditional machine learning models face two major technical challenges: first, the processing of multi-dimensional time-series sensor data suffers from high computational complexity and delayed real-time response; second, there are few successful cases of existing gas identification algorithms being deployed on embedded platforms. Especially when the size of the sensor array is limited, the limited response feature dimensions severely restrict the feature extraction efficiency of traditional algorithms, making it difficult to capture discriminative gas features from low-dimensional response curves. This technical dilemma poses a key challenge to wearable electronic nose systems. How to build an intelligent gas identification model with controllable computational complexity and support real-time processing while ensuring classification accuracy has become the key to breaking through the current technical bottleneck. Summary of the Invention

[0006] The purpose of this invention is to solve the problems of traditional gas detection equipment such as bulky size, reliance on wired power supply, insufficient waterproof performance, and high cost and complexity brought by multi-sensor arrays. It provides a waterproof and battery-free flexible electronic nose that can be used for gas identification. Through the collaborative innovation of flexible packaging technology, wireless power supply architecture and embedded intelligent algorithms, it achieves highly sensitive gas detection, omnidirectional waterproofness and low power consumption operation, and is suitable for scenarios such as wearable medical monitoring, industrial environment detection and emergency rescue.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A waterproof, battery-free, flexible electronic nose that can be used for gas identification comprises an upper waterproof, breathable membrane, a lower waterproof, breathable membrane, and a porous flexible circuit board disposed between the upper and lower waterproof, breathable membranes. The porous flexible circuit board integrates a processor and a gas sensor array. The processor is configured to collect analog signals from the gas sensor array and perform gas identification based on the analog signals.

[0009] Preferably, the upper waterproof breathable membrane and the lower waterproof breathable membrane are both porous fiber structures with a pore size of 0.001-10 μm, a porosity of 60-90%, a water contact angle ≥90°, and a gas permeability ≥30%.

[0010] Preferably, the bending radius of the porous flexible circuit board is ≤50 mm, and the overall thickness is ≤10 mm.

[0011] Preferably, the gas sensor array includes at least one MEMS gas sensor, a MEMS micro-hotplate and a gas-sensitive material are provided on the MEMS gas sensor, and the MEMS micro-hotplate is used to detect a resistance change of the gas-sensitive material.

[0012] Preferably, the MEMS micro-hotplate includes an insulating layer, a metal heater, and interdigital electrodes. The insulating layer is a silicon dioxide insulating layer or a silicon nitride insulating layer. The insulating layer is arranged between the metal heater and the interdigital electrodes. The metal heater is used to provide a preset operating temperature, and the interdigital electrodes are used to detect the resistance change of the gas-sensitive material.

[0013] Preferably, the metal heater provides a preset working temperature through pulse width modulation control, and the metal heater performs periodic dual temperature switching, alternating between a high temperature phase and a low temperature phase to modulate the response characteristics of the gas sensitive material, wherein the high temperature phase temperature is 300-500°C, and the low temperature phase temperature is 200-400°C.

[0014] Preferably, the gas sensitive material comprises at least one of a metal oxide semiconductor, a conductive polymer, a carbon-based material, a metal sulfide, a perovskite oxide, a metal organic framework and a composite material.

[0015] Preferably, the processor uses a pre-trained MiniRocket model to perform gas identification based on the analog signal, and the MiniRocket model is used to pre-process and extract features of the analog signal to identify the gas type and concentration.

[0016] Preferably, the electronic nose also includes a power supply module, which is used to provide power to the processor and gas sensor array through electromagnetic induction. The power supply module includes a receiving coil, a resonant circuit, a rectifier circuit, a filter circuit, a voltage stabilizing circuit, and an energy storage unit connected in sequence. Taking the receiving coil as the starting point, the resonant circuit realizes efficient coupling and frequency matching of electromagnetic waves, and then the alternating signal is converted into direct current by the rectifier circuit. The filter circuit eliminates noise interference, the voltage stabilizing circuit ensures the stability of the output voltage, and the energy storage unit acts as a buffer to finally realize power supply.

[0017] Preferably, the electronic nose further comprises a communication module, which supports data transmission of the electronic nose by wireless communication, and the wireless communication comprises at least one of Bluetooth, WiFi, near field communication, satellite communication, 5G / 4G, UWB, Zigbee, and LoRa.

[0018] The beneficial effects of the present invention are:

[0019] (1) Multi-gas identification can be achieved: through the synergistic effect of temperature modulation strategy and heterojunction gas-sensitive materials, a single or a small number of sensor arrays can output multi-dimensional response signals, combined with a lightweight model to achieve multi-gas classification, which is lower in cost than traditional multi-sensor solutions.

[0020] (2) Wireless power supply compatibility: Supports mainstream wireless charging protocols to ensure that the device is reverse-powered by the smartphone when there is no battery, solving the bottleneck of traditional electronic nose endurance and improving portability and ease of use.

[0021] (3) Omnidirectional waterproof flexible packaging: The parametric design of the waterproof and breathable membrane balances protection and breathability. The FPC substrate ensures that the equipment can operate stably in a bent state and is suitable for complex environments such as industrial inspections and motion monitoring.

[0022] (4) Dynamic temperature modulation optimization: The periodic dual-temperature switching strategy enhances the dynamic response characteristics of the sensor, and combined with the real-time processing of the embedded model, the accuracy of gas recognition is improved.

[0023] (5) Non-sensing human-computer interaction: Remote monitoring and data management can be achieved through mobile phone APP, meeting the dual needs of concealment and ease of use in medical, industrial and other fields.

[0024] (6) Rapid emergency response: The entire process from gas detection to result output takes ≤100 seconds, which is suitable for emergency scenarios such as toxic gas leakage and fire warning, and has a faster response speed than traditional electronic noses. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1 This is a macroscopic structural diagram of a waterproof, battery-free, flexible electronic nose that can be used for gas identification according to an embodiment of the present invention;

[0027] Figure 2 Schematic diagram of the relationship between modules and units in a waterproof, battery-free, flexible electronic nose that can be used for gas identification according to an embodiment of the present invention;

[0028] Figure 3 This is a circuit diagram of a wireless function of a waterproof, battery-free, flexible electronic nose that can be used for gas identification according to an embodiment of the present invention;

[0029] Figure 4 This is a flow chart of the MiniRocket algorithm embedded in a microprocessor for a waterproof, battery-free, flexible electronic nose that can be used for gas identification according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following will clearly and completely describe the technical solutions in the embodiments of the present invention 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 making creative efforts are within the scope of protection of the present invention.

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] This embodiment provides a waterproof, battery-free, flexible electronic nose that can be used for gas identification. Figure 1 As shown, it includes: an upper waterproof breathable membrane, a lower waterproof breathable membrane, and a porous flexible circuit board arranged between the upper waterproof breathable membrane and the lower waterproof breathable membrane, wherein the porous flexible circuit board integrates a processor and a gas sensor array, and the processor is used to collect analog signals from the gas sensor array and perform gas identification based on the analog signals.

[0033] Furthermore, both the upper and lower waterproof and breathable membranes are porous fiber structures with pore sizes ranging from 0.001 to 10 μm, a porosity of 60-90%, a water contact angle ≥ 90°, and a gas permeability ≥ 30%. This parameter range effectively blocks liquid water penetration (such as rainwater and sweat) while allowing gas molecules to quickly diffuse to the sensor surface, achieving a balanced balance between waterproofness and detection response speed.

[0034] Furthermore, the bending radius of the porous flexible circuit board is ≤50mm, and the overall thickness is ≤10mm. This design allows the electronic nose to conform to curved surfaces of the human body (such as the wrist and nose) or complex equipment surfaces, meeting the flexible deployment requirements of wearable devices and industrial scenarios.

[0035] Furthermore, the gas sensor array includes at least one MEMS gas sensor, a MEMS micro-hotplate and a gas sensitive material are provided on the MEMS gas sensor, and the MEMS micro-hotplate is used to detect a resistance change of the gas sensitive material.

[0036] Among them, the MEMS micro-hotplate includes an insulating layer, a metal heater, and interdigital electrodes. The insulating layer is a silicon dioxide insulating layer or a silicon nitride insulating layer. The insulating layer is arranged between the metal heater and the interdigital electrodes. The metal heater is used to provide a preset operating temperature, and the interdigital electrodes are used to detect the resistance change of the gas-sensitive material.

[0037] The metal heater provides a preset operating temperature through pulse width modulation control. The metal heater performs periodic dual temperature switching, alternating between a high temperature phase and a low temperature phase to modulate the response characteristics of the gas sensitive material, wherein the high temperature phase temperature is 300-500°C and the low temperature phase temperature is 200-400°C.

[0038] The gas sensitive material includes at least one of a metal oxide semiconductor, a conductive polymer, a carbon-based material, a metal sulfide, a perovskite oxide, a metal organic framework and a composite material.

[0039] Furthermore, the processor uses a pre-trained MiniRocket model to perform gas identification based on the analog signal, and the MiniRocket model is used to pre-process and extract features of the analog signal to identify the gas type and concentration.

[0040] Furthermore, the electronic nose further comprises a power supply module, which is used to provide electrical energy to the processor and the gas sensor array through electromagnetic induction.

[0041] Among them, the power supply module includes a receiving coil, a resonant circuit, a rectifier circuit, a filter circuit, a voltage stabilizing circuit, and an energy storage unit connected in sequence. Taking the receiving coil as the starting point, the resonant circuit realizes efficient coupling and frequency matching of electromagnetic waves, and then the alternating signal is converted into direct current through the rectifier circuit. The filter circuit eliminates noise interference, the voltage stabilizing circuit ensures the stability of the output voltage, and the energy storage unit plays a buffering role, finally realizing power supply.

[0042] Furthermore, the electronic nose also includes a communication module, which supports data transmission of the electronic nose by wireless communication, and the wireless communication includes at least one of Bluetooth, WiFi, near-field communication, satellite communication, 5G / 4G, UWB, Zigbee, and LoRa.

[0043] Specifically, Figure 1 This is a macroscopic cross-sectional view of the electronic nose of this embodiment. The top and bottom layers are encapsulated with a waterproof and breathable membrane. The upper and lower layers form a sandwich protection system, which not only ensures the air permeability of the system but also achieves omnidirectional waterproofing. The middle layer is a porous flexible circuit board (FPC) that integrates Figure 2 The ESP32S3 SoC processor, button module, indicator light, power module, single MOS-MEMS gas sensor, and connected application are embedded in a smartphone terminal.

[0044] In this embodiment, the waterproof and breathable membrane is embodied as a waterproof and breathable PU membrane, which is prepared by electrospinning. 1.3g of solid PU particles are added to 10mL of a tetrahydrofuran and DMF solution with a volume ratio of 3:2, and stirred at 60°C until completely dissolved to obtain a precursor solution. 3mL of the prepared solution is placed in a syringe and placed in an electrospinning machine. The electrospinning distance between the syringe and the collector is set to 20cm, the electrostatic field voltage of the electrospinning machine is set to 12kV, and the peristaltic pump propulsion rate is 0.8mL / h. After the electrospinning is completed, the collected microporous fiber structure PU membrane is placed in a drying oven to dry to remove residual solvent.

[0045] The fiber diameter of the waterproof, breathable PU membrane in this embodiment is approximately 200 nm. This micro-nanostructure imparts excellent air permeability to the PU membrane, allowing air and target gases to pass smoothly through the membrane surface, providing the structural foundation for achieving breathability in the system. The membrane also exhibits significant hydrophobicity, effectively repelling water droplets. This property ensures that even in the event of accidental water splashes or high humidity, water droplets remain confined to the membrane surface and cannot penetrate the sensors and circuitry beneath the membrane, thus protecting the system.

[0046] In this embodiment, the flexible substrate is a flexible printed circuit board (FPC), with a bending radius of ≤50mm and an overall thickness of ≤10mm. This design allows the electronic nose to conform to curved surfaces (such as the wrist and nose) or complex equipment surfaces, meeting the flexible deployment requirements of wearable devices and industrial scenarios.

[0047] In this embodiment, the gas sensor array includes at least one MEMS gas sensor, which includes a MEMS micro-hotplate and a gas-sensitive material. The MEMS micro-hotplate serves as a heating tool for the MOS gas-sensitive material. The MEMS micro-hotplate includes: a silicon dioxide or silicon nitride insulating layer; a metal heater with a heating area of 0.2-2mm 2 , operating temperature range is 100-500℃; interdigital electrodes are used to detect resistance changes of gas sensitive materials.

[0048] Among them, the metal heater adopts platinum metal heater, and the heating area of platinum heater is 0.5-2mm 2 The operating temperature range is 100-500°C. The interdigitated electrodes are spaced at the micron level and are used to detect changes in the resistance of the gas-sensitive material. This structure achieves efficient heating and precise detection at the microscale, reducing power consumption by over 60% compared to traditional sensors.

[0049] The MEMS micro-hotplate integrates tiny interdigitated electrodes and a microheater made of metal platinum, separated by an insulating layer of silicon dioxide and silicon nitride. The microheater provides the appropriate operating temperature for the gas sensor, while the interdigitated electrodes detect changes in the resistance of the gas-sensitive material on the plate. The microheater has a heating area of 1×1 mm, distributing heat energy over a tiny region. While the local temperature may be high, the overall heat released is limited and cannot significantly heat the surrounding environment or objects. Therefore, people will not feel the heat when touching it, ensuring safety.

[0050] In this embodiment, the gas-sensitive material is at least one of a metal oxide semiconductor, a conductive polymer, a carbon-based material, and a metal sulfide. The gas-sensitive material is selected from a single or composite system of nanostructured metal oxide semiconductors (such as SnO2, ZnO, WO3), conductive polymers (polyaniline, polypyrrole), carbon-based materials (graphene, carbon nanotubes), metal sulfides (MoS2), perovskite oxides, metal-organic frameworks, and composite materials. A 5-50 μm thick sensitive film layer is formed on the interdigitated electrodes to achieve selective response to gases such as VOCs, CO, and NOx.

[0051] The gas-sensitive material is a tin oxide-nickel oxide heterojunction (SnO2@NiO): 995.4mg of Ni(CH3COO)2·4H2O and 379.2mg of SnCl2·2H2O are added to 10mL of a 1:1 volume ratio of ethanol and DMF solution. The mixture is stirred at room temperature until completely dissolved. 1g of PVP is then added and sonicated to dissolve the mixture completely. The prepared solution is then placed in a syringe and placed in an electrospinning machine. The distance between the collector and the syringe is set to 20cm to receive the electrospinning precursor fibers. The electrospinning machine parameters are then set to apply an 18kV electric field and generate an electrojet at a peristaltic pump propulsion rate of 0.8mL / h. After electrospinning, the collected precursor fiber film is dried in a drying oven and then calcined in a tube furnace at 600°C for 5h to produce the SnO2@NiO nanomaterial.

[0052] The button module and the display light in the porous flexible circuit board (FPC) of this embodiment are common spring buttons and red LED lights, which are used to reset the device and indicate the working status.

[0053] The processor used is the ESP32-S3, an MCU chip with integrated 2.4GHz Wi-Fi and Bluetooth 5 (LE), supporting Long Range mode. The ESP32-S3 is equipped with a 32-bit LX7 dual-core processor with a clock speed of up to 240MHz, 512KB of internal SRAM (TCM), 45 programmable GPIO pins, and a rich set of communication interfaces. The ESP32-S3 MCU also adds vector instructions for accelerating neural network calculations and signal processing, enabling the deployment of machine learning and even deep learning models at the edge.

[0054] The chip integrates the pre-trained MiniRocket model, and the model algorithm flow is as follows: Figure 4 As shown, the sensor first collects response data for eight VOCs (benzene, toluene, ethylbenzene, xylene, styrene, formaldehyde, acetone, and ethanol) at varying concentrations and temperatures to create a dataset. The raw data is then preprocessed with standardization and feature flattening, converting the multidimensional time series data into a tensor structure suitable for MiniRocket processing. Feature extraction is then performed using an automatically generated set of dilated convolution kernels, and a multidimensional feature space is constructed using random dilation coefficients and kernel parameters. The extracted features are then optimized through channel combination and fed into a classification head, where pattern recognition is performed using a multi-layer perceptron with batch normalization and dropout regularization. During the training phase, the model parameters are dynamically updated using the backpropagation algorithm, while validation set performance is monitored to prevent overfitting. Finally, the trained model is quantized and formatted for embedded hardware. A lightweight TensorFlow Lite file is generated and embedded into the ESP32-S3 firmware, where the inference engine is deployed and executed in real time.

[0055] The power module in this embodiment receives energy through wireless energy reception and includes a resonant circuit, a rectifier circuit, and a voltage regulator circuit. It supports at least one of the Qi, AirFuel, Rezence, or NFC wireless charging protocols, with an output power range of 0.1-10W. Through multi-stage processing of the resonant circuit, rectifier circuit, and voltage regulator circuit, it is compatible with mainstream wireless charging devices (such as reverse charging of smartphones) and ensures stable energy acquisition within a distance of 0.5m.

[0056] The power module controls the temperature of the microheater via pulse-width modulation (PWM), performing periodic dual-temperature switching, alternating between high-temperature and low-temperature phases to modulate the response characteristics of the gas-sensitive material. The high-temperature phase is set at 300-500°C, and the low-temperature phase is set at 200-400°C. The temperature modulation frequency is 0.1-10Hz. By alternately activating different surface reaction mechanisms of the gas-sensitive material, the electronic nose's ability to distinguish between multiple gas types is enhanced, allowing detection of more than eight gases.

[0057] Power modules such as Figure 3 As shown in the figure, a top-down energy transmission path design is adopted. Starting from the receiving coil, a resonant circuit achieves efficient coupling and frequency matching of electromagnetic waves. The alternating signal is then converted into direct current by a rectifier circuit, and then a filter circuit eliminates noise interference. The voltage regulator module ensures stable output voltage, and the energy storage unit (such as a supercapacitor or lithium battery) acts as a buffer, ultimately providing power to external devices. The protection circuit added at the bottom includes overvoltage, overcurrent, and temperature protection functions to ensure safe and reliable operation. The microprocessor also outputs a square wave signal with adjustable duty cycle to drive the microheater through PWM regulation of the power module. A high temperature phase (400°C) and a low temperature phase (300°C) are set to temperature modulate the MEMS gas sensor.

[0058] The power module uses a Qi chip as the wireless power receiving circuit chip, achieving an output power of up to 5W, far exceeding the system's required 1W power consumption. When a wireless charging transmitter (such as a smartphone) generates a changing magnetic field, the receiving coil senses this field and converts it into electrical energy. After rectification and voltage regulation by the Qi wireless chip, a stable power supply is provided. This technology is highly compatible with a wide range of wireless charging transmitters and seamlessly integrates with smartphones equipped with wireless reverse charging, greatly enhancing the system's applicability.

[0059] The electronic nose provided in this embodiment also includes an application, which is connected to the processor via wireless communication. The application has functions such as device scanning, data reading, image display, and viewing connection status. It is used to receive sensor data and display gas type, concentration and safety alarm information, and supports wireless data synchronization.

[0060] Wireless communication utilizes at least one of the following transmission methods: Bluetooth, WiFi, Near Field Communication (NFC), satellite communication, 5G / 4G, UWB, Zigbee, LoRa, etc. The electronic nose communicates with the device via Bluetooth, WiFi, Near Field Communication (NFC), satellite communication, 5G / 4G, UWB, Zigbee, LoRa, etc., providing device scanning, real-time data display (waveform / concentration curve), safety threshold alarms, and historical data backtracking. The app features a built-in gas database that supports user-defined detection targets (such as formaldehyde and benzene series). Detection results simultaneously display the gas type, concentration, and health risk level (e.g., green / yellow / red alert).

[0061] Specifically, this embodiment is based on the development of a mobile app. This mobile app is developed for the Android system and is named "Wearable Gas Sensor Bracelet." The Wearable Gas Sensor Bracelet app features a clear interface layout and multifunctional design. Specifically, it provides functions such as device scanning, data reading, image jump display, and connection status viewing. The development of this app effectively enhances the user interactivity and practicality of the electronic nose, providing a simple and efficient way to transmit, display, and manage gas sensor data, providing users with a friendly and intuitive user experience.

[0062] The device is activated through wireless energy; gas molecules penetrate the waterproof and breathable membrane and react with sensitive materials, and the sensor resistance value changes with the gas concentration; the processor collects the time-series resistance signal and inputs it into the machine learning model; the model outputs the gas type and concentration to the terminal display.

[0063] The gas identification method involves: activating the device with wireless energy, allowing the gas to penetrate a permeable membrane and react with the gas-sensitive material, collecting dynamic resistance signals, inputting them into a machine learning model, and outputting the gas information to the mobile phone. The model utilizes a lightweight sequential network architecture to process multi-temperature modulation data from the sensor, achieving an identification accuracy rate exceeding 85% and a memory footprint of less than 2000KB.

[0064] In this embodiment, the user wears a waterproof, battery-free, flexible electronic nose that can be used for gas identification, turns on the reverse wireless charging function of the smartphone to start the device, the indicator light comes on, and the device is brought close to the gas to be measured. The gas passes through the waterproof and breathable membrane to reach the MEMS gas sensor and reacts with the synthetic gas-sensitive material. The resistance of the MEMS sensor changes, and the MEMS micro-hotplate alternately heats the material at two temperatures. The resistance data is converted into a voltage signal through a matching resistor and is read by the AD conversion module of the MCU and displayed on the mobile phone APP. When a certain amount of data is collected, the data is analyzed by the machine learning model deployed by the processor to determine the gas type and concentration information, which is displayed on the mobile phone APP.

[0065] After the user wears the waterproof battery-free electronic nose device based on flexible electronic technology, he turns on the reverse wireless charging function through the smartphone to power the device. At this time, the onboard LED indicator light turns on and the device enters working state.

[0066] When the device is placed close to the gas source to be measured, ambient gas molecules penetrate the selectively permeable waterproof and breathable membrane, enter the MEMS gas sensor, and react with the SnO2@NiO heterostructure nanomaterial on its surface. At this point, the micro-hotplate within the MEMS sensor, integrated on the flexible substrate, performs a dual-temperature modulation program (alternating heating at 300°C / 400°C), driving a periodic redox reaction in the sensitive material, resulting in specific changes in the sensor array's resistance.

[0067] The resistance signal is converted into a voltage waveform through a precision matching circuit and captured in real time by the high-precision ADC module built into the low-power MCU at a sampling rate of 100Hz. The original waveform data is then synchronously transmitted to the mobile phone APP interface via the Bluetooth 5.0 protocol to display the real-time waveform.

[0068] When 5 seconds of valid data are collected, the embedded processor automatically triggers the deployed lightweight MiniRocket machine learning model, uses pre-trained feature convolution kernels to perform temporal pattern recognition, completes gas type determination and concentration inversion calculation within 300ms, and finally synchronously displays structured detection results such as "Benzene series -0.8ppm" on the mobile phone app. At the same time, it generates dynamic concentration curves and safety threshold alarm prompts, realizing full-process closed-loop monitoring from physical signal perception to chemical information decoupling.

[0069] This embodiment provides a waterproof, battery-free, flexible electronic nose for gas identification. Based on a fabricated MEMS gas sensor, it is controlled by an ESP32 and equipped with a series of test circuits. This electronic nose provides a more convenient, rapid, inexpensive, and accurate solution for gas detection and identification, offering advantages such as portability, real-time performance, low cost, ease of operation, and high sensitivity.

[0070] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A waterproof, battery-free, flexible electronic nose for gas identification, characterized in that: include: An upper waterproof breathable membrane, a lower waterproof breathable membrane, and a porous flexible circuit board arranged between the upper waterproof breathable membrane and the lower waterproof breathable membrane, wherein the porous flexible circuit board integrates a processor and a gas sensor array, and the processor is used to collect analog signals from the gas sensor array and perform gas identification based on the analog signals.

2. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 1, characterized in that: The upper waterproof breathable membrane and the lower waterproof breathable membrane are both porous fiber structures with a pore size of 0.001-10 μm, a porosity of 60-90%, a water contact angle of ≥90°, and a gas permeability of ≥30%.

3. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 1, characterized in that: The bending radius of the porous flexible circuit board is ≤50 mm, and the overall thickness is ≤10 mm.

4. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 1, characterized in that: The gas sensor array includes at least one MEMS gas sensor. A MEMS micro-hotplate and a gas-sensitive material are arranged on the MEMS gas sensor. The MEMS micro-hotplate is used to detect the resistance change of the gas-sensitive material.

5. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 4, characterized in that: The MEMS micro-hotplate includes an insulating layer, a metal heater, and interdigital electrodes. The insulating layer is a silicon dioxide insulating layer or a silicon nitride insulating layer. The insulating layer is arranged between the metal heater and the interdigital electrodes. The metal heater is used to provide a preset operating temperature, and the interdigital electrodes are used to detect the resistance change of the gas-sensitive material.

6. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 5, characterized in that: The metal heater provides a preset operating temperature through pulse width modulation control. The metal heater performs periodic dual temperature switching, alternating between a high temperature phase and a low temperature phase to modulate the response characteristics of the gas sensitive material, wherein the high temperature phase temperature is 300-500°C and the low temperature phase temperature is 200-400°C.

7. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 4, characterized in that: The gas sensitive material includes at least one of a metal oxide semiconductor, a conductive polymer, a carbon-based material, a metal sulfide, a perovskite oxide, a metal organic framework and a composite material.

8. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 1, characterized in that: The processor uses a pre-trained MiniRocket model to perform gas identification based on the analog signal. The MiniRocket model is used to pre-process and extract features of the analog signal to identify the gas type and concentration.

9. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 1, characterized in that: The electronic nose also includes a power supply module, which is used to provide power to the processor and gas sensor array through electromagnetic induction. The power supply module includes a receiving coil, a resonant circuit, a rectifier circuit, a filter circuit, a voltage stabilization circuit, and an energy storage unit connected in sequence. Starting from the receiving coil, the resonant circuit achieves efficient coupling and frequency matching of electromagnetic waves. The alternating signal is then converted into direct current through the rectifier circuit. The filter circuit eliminates noise interference, the voltage stabilization circuit ensures output voltage stability, and the energy storage unit acts as a buffer to ultimately achieve power supply.

10. The waterproof, battery-free, flexible electronic nose for gas identification according to claim 1, characterized in that: The electronic nose also includes a communication module, which supports data transmission of the electronic nose by wireless communication, and the wireless communication includes at least one of Bluetooth, WiFi, near-field communication, satellite communication, 5G / 4G, UWB, Zigbee, and LoRa.

Images