Electronic nose device based on microwave resonator array and gas analysis method

Abstract

The invention discloses an electronic nose device based on a microwave resonator array and a gas analysis method. The device comprises an upper sensing array layer and a lower feed layer located below the upper sensing array layer. The lower feed layer comprises a grounding plane, a lower dielectric substrate arranged on the grounding plane, and a microwave coupling feeder line arranged on the lower dielectric substrate; the upper sensing array layer comprises an upper dielectric substrate and a resonator array arranged on the upper dielectric substrate, the resonator array comprises a plurality of resonator units, and the plurality of resonator units are coated with different gas sensitive material coatings and are used for detecting multi-element gas. According to the device, the characteristic that different gas sensitive materials have different adsorption capacities on different gases is utilized, selective detection and multi-mode combined analysis of the gases are realized, multi-element gas detection can be realized at the same time, the influence of environment temperature and humidity on gas detection precision is effectively resisted, and the device has a wide application prospect in the fields of gas monitoring and sensor device integration.

Description

TECHNICAL FIELD

[0001] The present application relates to a microwave resonator array-based electronic nose device and gas analysis method, and belongs to the technical field of microwave sensors. BACKGROUND

[0002] Microwave gas sensing technology has become an important research direction in the fields of environmental monitoring, industrial safety protection, food safety monitoring, etc. due to its non-invasive, high penetration ability and sensitive response to dielectric properties of the medium.

[0003] However, the prior art has significant limitations: traditional single-layer microwave sensors have a fixed integration of sensing units and feed networks, resulting in fixed device functions and limited service life. Once the sensitive material fails or the detection target changes, the entire device needs to be scrapped. On the other hand, the mainstream single-resonant cavity structure can only coat one type of sensitive material in the sensitive area. When facing complex mixed gas environments, multiple devices need to be replaced, several devices need to be deployed, or serial detection is required, which is inefficient, costly, and difficult to capture the interaction effects between gases. In addition, electrochemical sensors that rely on surface redox reactions usually require high temperatures above 200℃ to activate the metal oxide semiconductor material as the sensitive material, which not only brings high energy consumption and thermal control challenges, but also easily causes material structure degradation, restricting the long-term stability and application scenarios of the equipment. SUMMARY

[0004] To solve the above problems, the present application provides a microwave resonator array-based electronic nose device and gas analysis method. The electronic nose device adopts a double-layer detachable microwave sensing array architecture, i.e., a modular design with physical separation, so that the upper sensing array layer can be flexibly replaced for different detection scenarios, significantly reducing maintenance costs and breaking through the shackles of single-use devices. On the surface of the upper sensing array layer, nine independently designed interdigital resonator units are arranged in a 3x3 matrix. Each resonator unit is optimized by unique geometric parameters to excite a highly localized electromagnetic field distribution, forming a frequency-separated resonant mode. The surface of each resonator unit can be functionalized to coat a gas-sensitive material coating specific to a particular gas. The dielectric constant change caused by gas molecule adsorption disturbs the resonant frequency and amplitude, enabling real-time detection of nine gases in parallel. The electronic nose device can achieve high sensitivity at ppm level at room temperature, completely avoiding the need for high-temperature operation, and significantly improving the energy efficiency ratio and safety of the equipment.

[0005] The technical solutions of the present application are as follows:

[0006] The application provides a microwave resonator array-based electronic nose device, which comprises an upper sensor array layer and a lower feeding layer located below the upper sensor array layer; the lower feeding layer comprises a ground plane, a lower dielectric substrate arranged on the ground plane and a microwave coupling feed line arranged on the lower dielectric substrate; the upper sensor array layer comprises an upper dielectric substrate and a resonator array arranged on the upper dielectric substrate, and the resonator array comprises a plurality of resonator units, and different gas-sensitive material coatings are coated on the plurality of resonator units for detecting multi-element gas.

[0007] In an embodiment of the application, the equivalent circuit of each resonator unit of the resonator array is represented as an RLC series circuit, and the resonant frequency is:

[0008]

[0009] wherein L sensor is the equivalent inductance of the resonator, C D is the capacitance change of the gas-sensitive material after combining with gas molecules, and C is the equivalent capacitance of the resonator.

[0010] In an embodiment of the application, different types of gas-sensitive material coatings are coated on the resonator array, including P-type semiconductor, N-type semiconductor, polymer and strong dielectric ceramic material.

[0011] Each of the gas-sensitive material coatings is sensitive to one or more gases, and when the gas is adsorbed onto the gas-sensitive material coating, the change of the dielectric property of the material is determined by the Debye model:

[0012]

[0013] wherein ω = 2πf, τ = 1 / (2πf R ), f R is the relaxation frequency, ε′ ∞ is the real part when f→∞, ε′ s is the real part when f→0, the higher the working frequency, the greater the dielectric constant change, and the greater the capacitance change C D of the resonator caused by gas adsorption, and the more significant the resonant frequency change and S21 parameter.

[0014] In an embodiment of the application, the upper sensor array layer and the lower feeding layer are detachably connected through screws.

[0015] In an embodiment of the application, the resonator array, the microwave coupling feed line and the ground plane are made of conductive material.

[0016] In an embodiment of the present application, the upper sensing array layer and the lower feeding layer are prepared on a polytetrafluoroethylene medium substrate through a wet etching process.

[0017] In an embodiment of the present application, the resonator array includes nine independently designed interdigital resonator units arranged in a 3x3 matrix, which can generate resonance modes at different frequencies according to the size of the resonator.

[0018] In an embodiment of the present application, the resonator array is coated with nine gas-sensitive materials on the exposed surface, including three P-type semiconductors, three N-type semiconductors, one polymer, one two-dimensional material, and one ceramic material, to selectively adsorb target gas molecules; the N-type semiconductor includes SnO2, ZnO, and TiO2, the anatase phase of TiO2 enhances the NO2 capture ability of the surface oxygen vacancy, and the SnO2 and ZnO double N-type semiconductors cross-verify the NO2 signal, with SnO2 high sensitivity and ZnO fast recovery characteristics complementary; the P-type semiconductor includes NiO, Co3O4, and CuO, the hole-dominant conduction mechanism of NiO has intrinsic selectivity to reducing gases, NH3 molecules react with pre-adsorbed oxygen to release electrons and neutralize holes; the spinel structure of Co2O4 provides multiple oxidation states, enhancing the catalytic oxidation activity of NH3, with sensitivity higher than NiO, and the P-type characteristics of CuO assist NH3 detection, with the surface Cu + active sites preferentially adsorbing ammonia molecules; the two-dimensional material selects Ti3C2T x xene, the surface termination group forms a directional hydrogen bond network with the ethanol hydroxyl group, and the dielectric constant imaginary part changes significantly, and the layered structure provides a specific surface area of >800 m 2 / g, achieving high sensitivity for ethanol detection; polyaniline as a polymer material, the acetone carbonyl induces the conformational change of the aniline unit molecular chain, changing the polaron migration path, and can specifically distinguish between ketone and alcohol interference; the strong dielectric ceramic material barium titanate responds to polar molecules with a ferroelectric domain flip, with a dielectric constant real part change amplitude of 40%, widening the temperature and humidity adaptability, and used to exclude the influence of environmental humidity on the accuracy of the electronic nose device.

[0019] The present application also provides a gas analysis method based on a microwave resonator array, which adopts the electronic nose device based on a microwave resonator array, and is based on a graph neural network model, and specifically includes the following steps:

[0020] Step 1, introducing the gas to be measured into the resonator array, collecting the S21 parameters of each resonator unit, including the resonance frequency shift Δf, amplitude attenuation Δ|S21|, phase change and calculating the change amount of the above features;

[0021] Step 2, construct a graph structure by regarding each resonator unit as an independent graph node, and the node features include Delta f, Delta |S21|, Generate dynamic edges based on electromagnetic coupling characteristics: establish frequency coupling edges when the resonant frequency difference is less than 50MHz, and establish response correlation edges when the response change correlation coefficient is greater than 0.7;

[0022] Step 3, analyze the mixed gas information using a hierarchical attention mechanism, including a gas level attention mechanism and a concentration level attention mechanism, calculate the gas-specific weight according to the pre-set learnable query table of each gas, generate the concentration-sensitive weight through a lightweight multi-layer perception, filter the resonator nodes using a multi-head attention, and finally focus on the key feature area of the response of different gases;

[0023] Step 4, output three kinds of collaborative tasks in parallel: one is gas existence detection, using a sigmoid classifier to output the gas existence probability; one is gas type classification, concatenating the node level features and using a Softmax classifier to give the probability of different combinations; and the last one is concentration probability modeling, outputting the 95% confidence interval according to the concentration expectation.

[0024] The application also provides an application of the microwave resonator array-based electronic nose device in multi-gas specificity identification.

[0025] The application has the following beneficial effects:

[0026] (1) The microwave resonator array-based electronic nose device provided by the application can realize multiplexing of the lower feeding layer by designing a double-layer detachable structure of the upper sensing array layer and the lower feeding layer, and can freely replace the upper sensing array layer according to actual use requirements in different gas detection environments, and only needs to re-coat or attach the customized gas-sensitive material coating. The electronic nose device adopts an array resonator mode, which can realize mixed detection of multiple gases compared with traditional single resonator detection, and can realize specific detection and quantitative analysis of multiple gases by using the changes of different resonant modes under different gases and different concentrations.

[0027] (2) The application can be directly functionalized and coated with nine different gas-sensitive materials, including three types of P-type semiconductors, three types of N-type semiconductors, and one type of high polymer, two-dimensional material and strong dielectric ceramic, which can significantly improve the selective capture ability of complex mixed gases through the specific interaction between material characteristics and target gases. The multi-mode electromagnetic field excited by the nine independent resonator units can synchronously perceive the multi-microwave parameter changes caused by gas adsorption, and construct an information density far exceeding that of a single-mode sensor. This multi-mode joint analysis mechanism effectively overcomes the misjudgment problem caused by cross-sensitivity of traditional sensors.

[0028] (3) The application is based on a graph neural network model, which realizes the paradigm shift of gas detection from single-point response analysis to systematic relationship reasoning. Nine resonator units are constructed as a dynamic graph model, each resonator unit is modeled as a graph node, the response characteristics of the unit are taken as node attributes, and the electromagnetic coupling effect and spatial correlation between the units are taken as dynamic edge weights. Through hierarchical graph convolution and attention mechanism, the network can adaptively learn the microwave parameter changes caused by different gas combinations, and first complete the qualitative classification of multi-component gas, and then accurately quantify the concentration of each gas through the regression branch. This hardware and algorithm co-innovation not only breaks through the detection bottleneck of traditional sensors "single target per time", but also lays a technical paradigm for building a high-throughput, low-power intelligent gas sensing platform. The model architecture has greatly improved the recognition accuracy of four-component mixed gas (NH3 / NO2 / ethanol / acetone) compared with traditional methods, and completely solves the false alarm problem caused by cross-sensitivity.

[0029] (4) The application utilizes the different adsorption abilities of different gas-sensitive materials for different gases to realize selective detection of gases and multi-mode combined analysis. Since the dielectric properties of the resonator near field change after the gas-sensitive material adsorbs gas molecules, the corresponding resonant mode also changes. The electronic nose device can simultaneously realize multi-component gas detection and effectively resist the influence of environmental temperature and humidity on gas detection accuracy by utilizing this multi-mode change characteristic. The electronic nose device has simple manufacturing process and high sensitivity, and is suitable for industrial environment and food quality monitoring, etc., and has wide application prospects in the fields of gas monitoring and sensor device integration. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can also obtain the remaining drawings according to these drawings without creative labor.

[0031] Figure 1 The structure schematic diagram of the electronic nose device based on the microwave resonator array provided by the present application is shown in the figure.

[0032] Figure 2 The structure schematic diagram of the nine resonator units of different sizes provided by the present application is shown in the figure.

[0033] Figure 3 The surface electric field distribution diagram of the nine resonator units of different sizes provided by the present application is shown in the figure.

[0034] Figure 4 The simulation S21 parameter diagram of the microwave resonator array provided by the present application is shown in the figure.

[0035] Figure 5 S21 response curve of the microwave resonator array provided by the present application under different concentrations of gas;

[0036] Figure 6 Fitting curve diagram of the present application using a neural network to predict gas concentration and actual target gas concentration;

[0037] Figure 7 Structure block diagram of the graph neural network model provided by the present application.

[0038] In the figure: 1, upper sensing array layer; 2, lower feeding layer; 3, ground plane; 4, lower dielectric substrate; 5, microwave coupling feed line; 6, upper dielectric substrate; 7, resonator array; 8, gas sensitive material coating. DETAILED DESCRIPTION

[0039] The technical solutions of the present application will be described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0040] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance. Among them, the terms "first position" and "second position" are two different positions.

[0041] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixed connection, but also can be detachable connection; can be mechanical connection, but also can be electrical connection; can be directly connected, but also can be indirectly connected through intermediate medium, can be the communication between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0042] Example 1

[0043] As Figure 1 , Figure 2 and Figure 3As shown, the embodiment provides a microwave resonator array-based electronic nose device for gas category detection and concentration identification, which comprises an upper sensor array layer 1 and a lower feed layer 2 located below the upper sensor array layer 1; the lower feed layer 2 comprises a ground plane 3, a lower dielectric substrate 4 arranged on the ground plane 3, and a microwave coupling feed line 5 arranged on the lower dielectric substrate 4; the upper sensor array layer 1 comprises an upper dielectric substrate 6 and a resonator array 7 arranged on the upper dielectric substrate 6, the resonator array 7 comprises a plurality of resonator units, and different gas-sensitive material coatings 8 are coated on the plurality of resonator units for detecting multi-component gas.

[0044] Optionally, the resonator array 7, the microwave coupling feed line 5 and the ground plane 3 are made of conductive material, and copper can be optionally used.

[0045] Optionally, the upper sensor array layer 1 and the lower feed layer 2 are detachably connected by screws.

[0046] Optionally, the upper sensor array layer 1 and the lower feed layer 2 are prepared on a polytetrafluoroethylene dielectric substrate (Teflon®0.54mm, Cu@30um) through a wet etching process, and a hole puncher is used to punch a positioning screw hole with a diameter of 4mm at the top corner of the lower dielectric substrate 4 and the upper dielectric substrate 6, respectively.

[0047] Optionally, the resonator array 7 comprises nine interdigital resonator units independently designed and arranged in a 3×3 matrix, which can produce resonance modes at different frequencies according to different resonator sizes. The number of resonator units can be adjusted according to actual use requirements, such as 2×2, 4×4, and the embodiment selects a 3×3 array size for detecting four-component mixed gas.

[0048] The equivalent circuit of each resonator unit of the resonator array 7 can be represented as an RLC series circuit, and the resonant frequency thereof is:

[0049]

[0050] wherein L sensor is the equivalent inductance of the resonator, C D is the capacitance change of the gas-sensitive material combined with gas molecules, and C is the equivalent capacitance of the resonator.

[0051] The exposed metal copper on the resonator array 7 is coated with different types of gas-sensitive material coating 8, including P-type semiconductor, N-type semiconductor, polymer, strong dielectric ceramic material, etc. Different gas-sensitive materials have different response characteristics to different gases, thereby forming detection units with different gas sensitivity with the microwave resonator units. The response difference between each unit provides data support for the identification of gas species and the prediction of concentration in a complex gas atmosphere, thereby realizing multi-mode joint analysis using the changes of nine different resonance modes.

[0052] In this embodiment, nine different gas-sensitive materials are coated on the microwave resonator array, including three P-type semiconductors, three N-type semiconductors, one polymer, one two-dimensional material, and one ceramic material. In actual use, the gas-sensitive material coating 8 can also be replaced according to the needs, i.e. replacing the upper sensing array layer 1 to realize the multiplexing of the lower feed layer 2.

[0053] Further, each gas-sensitive material coating 8 is sensitive to one or more gases. When the gas is adsorbed onto the gas-sensitive material coating 8, the change in material dielectric properties is determined by the Debye model:

[0054]

[0055] where ω = 2πf, τ = 1 / (2πf R ), f R is the relaxation frequency, ε′ ∞ is the real part when the resonance frequency f→∞, ε′ s is the real part when f→0, the higher the working frequency, the greater the change in dielectric constant, the greater the change in resonator capacitance C D caused by gas adsorption, the more significant the resonance frequency change and the S21 parameter.

[0056] There are significant differences in the response characteristics of different gas-sensitive materials to different gases. This difference is reflected in the specific reaction of the materials (such as P-type semiconductors, N-type semiconductors, high molecular polymers or strong dielectric ceramics) coated on the microwave resonator unit when exposed to the gas. Each resonator unit therefore constitutes a detection unit with unique gas sensitivity. The response differences between the units are mainly manifested as changes in radio frequency multi-derivative parameters, including resonant frequency offset, amplitude attenuation, phase angle changes, and quality factor (Q factor) fluctuations. The differences in these multi-derivative parameters are due to the electromagnetic property disturbances caused by the interaction between gas molecules and gas-sensitive materials, providing a rich data matrix for the identification of gas types in complex gas atmospheres. By analyzing the patterns of these parameters (such as using machine learning algorithms to process frequency-amplitude correlations), the system can achieve highly selective gas identification and concentration prediction, thereby supporting multi-mode joint analysis and precise detection, showing olfactory perception capabilities that surpass humans. Therefore, the electronic nose device based on microwave resonator arrays provided by the present invention can be used in the specific identification of multiple gases.

[0057] Example 2

[0058] This embodiment provides an electronic nose device based on a microwave resonator array for gas type detection and concentration identification. The electronic nose device comprises an upper sensor array layer 1 and a lower feed layer 2. The upper sensor array layer 1 comprises a 0.54 mm thick polytetrafluoroethylene dielectric substrate 6, an interdigital resonator array 7 formed by etching copper, a gas-sensitive material coating 8 applied to the surface of the resonator array 7, and four corner positioning screw holes. The lower feed layer 2 comprises a ground plane 3, a 0.54 mm thick polytetrafluoroethylene dielectric substrate 4, an S-shaped microwave coupling feed line 5, and positioning screw holes corresponding to the corner positioning screw holes of the upper sensor array layer 1. The dielectric substrate is fabricated on a copper-clad substrate (Cu@30μm) using a wet etching process, and mechanical alignment is achieved using 4 mm diameter positioning screw holes. The lower feed layer 2 is connected to a vector network analyzer via a coaxial interface, responsible for exciting the resonator array 7 and collecting S21 transmission parameters in real time.

[0059] As attached Figure 2 As shown, the interdigital resonator array 7 consists of nine independent copper resonator units arranged in a 3×3 matrix. Each resonator unit is suspended above the S-shaped microwave coupling feeder 5 through the bottom coupling region. By differentiating the resonator size and changing the unit electrical length, the array produces nine spectrally separated resonance peaks in the 2.5–6 GHz frequency band, as shown in Figure 1. Figure 4 As shown in Figure 1, resonator units 1 to 9 correspond to 2.77 GHz, 3.11 GHz, 3.44 GHz, 3.68 GHz, 4.00 GHz, 4.36 GHz, 4.73 GHz, 5.13 GHz, and 5.41 GHz, respectively. The surface electric field of each resonator unit is highly localized.Figure 4 As shown, the gas sensitive region is formed. By coating nine gas sensitive materials, including three P-type semiconductors, three N-type semiconductors, one polymer, one two-dimensional material, and one ceramic material, on the exposed copper surface of the resonator array 7, the target gas molecules can be selectively adsorbed.

[0060] The resonator array 7 of the preferred 3x3 array size in this embodiment balances the demand for quaternary mixed gas detection and spectrum resource occupation, but the resonator unit topology can be expanded to 2x2 or 4x4 configurations according to actual application. The gas detection mechanism is based on: gas molecule adsorption induces dielectric constant change of gas sensitive material → disturbs electromagnetic field distribution of corresponding resonator unit → causes resonance frequency shift, amplitude attenuation, and phase jump of S21 parameter → through multi-resonance mode collaborative analysis, gas type and concentration are analyzed. This design verifies the technical feasibility of the microwave resonator array-based electronic nose device with a double-layer detachable structure in realizing high multiplexing, multi-gas parallel detection.

[0061] Embodiment 3

[0062] This embodiment provides a gas sensitive material selection and gas analysis method for a microwave resonator array-based electronic nose device. The microwave resonator array-based electronic nose device has the advantages of multiple independent resonant modes, more parameters that can be analyzed, including resonance frequency shift, amplitude attenuation, and phase change, and can realize multi-parameter inversion of gas category and gas concentration by utilizing the advantages of microwave multi-parameters.

[0063] This embodiment details the gas sensitive material configuration strategy and gas analysis mechanism of the microwave resonator array-based electronic nose device. The core advantage of the electronic nose device is: through the multi-modal electromagnetic response (resonance frequency shift Δf, amplitude attenuation Δ|S21|, phase change ) excited by nine independent resonator units, a three-dimensional parameter space is constructed to realize holographic analysis of gas fingerprints. In view of the urgent needs of industrial toxic gas leakage (such as NO2) and food safety monitoring (such as ethanol / acetone residue), ethanol, acetone, NH3, and NO2 are selected as target gases.

[0064] The gas sensitive materials can be divided into N-type semiconductors, P-type semiconductors, polymers, two-dimensional materials, and ceramic materials according to their categories. The following is the material selection of this embodiment:

[0065] N-type semiconductor includes tin dioxide (SnO2), zinc oxide (ZnO), titanium dioxide (TiO2). Among them, SnO2 is the most commonly used gas sensitive material, which is sensitive to various oxidizing and reducing gases such as ethanol, NO2, CO, etc., and ZnO is sensitive to ethanol, propanol and other gases. According to the sensitivity of the material to different gases, SnO2, ZnO and TiO2 can be selected as N-type semiconductor gas sensitive materials.

[0066] Further, the anatase phase surface oxygen vacancy of TiO2 enhances the NO2 capture ability, and the SnO2 and ZnO double N-type semiconductor cross-verify the NO2 signal, and the high sensitivity of SnO2 and the rapid recovery characteristics of ZnO are complementary.

[0067] P-type semiconductor includes nickel oxide (NiO), cobalt oxide (Co3O4), and copper oxide (CuO), among which Co3O4 is sensitive to VOCs, ethanol and other gases, NiO is sensitive to reducing gases, and CuO is relatively sensitive to CO, NO2 and other gases. Therefore, these three materials are selected as P-type semiconductor gas sensitive materials.

[0068] Further, the hole-dominant conduction mechanism of NiO has intrinsic selectivity to reducing gases. NH3 molecules react with pre-adsorbed oxygen (4NH3+5O2→4NO+6H2O) to release electrons and neutralize holes; the spinel structure of Co3O4 provides multiple oxidation states (Co 2+ / Co 3+ ), enhancing the catalytic oxidation activity of NH3, and the sensitivity is higher than that of NiO, and the P-type characteristics of CuO assist NH3 detection, and the surface Cu + active sites preferentially adsorb ammonia molecules.

[0069] MXene (Ti3C2T x ) is selected as a gas sensitive material in two-dimensional materials, and the surface termination group (-OH / -F) forms a directional hydrogen bond network with the hydroxyl group of ethanol, and the imaginary part of the dielectric constant (ε") changes significantly. Its layered structure provides a specific surface area of >800 m 2 / g, realizing high sensitivity of ethanol detection.

[0070] Polyaniline (PANI) as a high polymer material, acetone carbonyl (C=O) induces the conformational change of aniline unit molecular chain, changes the polaron migration path, and can specifically distinguish ketone and alcohol interference.

[0071] Strong dielectric ceramic material barium titanate (BaTiO3), its ferroelectric domain flip response polarity molecule, the real part of dielectric constant (ε') changes by 40%, and the temperature and humidity adaptability is widened (-20℃~85℃). It can be used to exclude the influence of environmental humidity on the accuracy of electronic nose device.

[0072] The differentiated responses of the nine materials when the mixed gas contacts the resonator array are manifested in three aspects: physical adsorption, chemical adsorption, and node disturbance: MXene / PANI captures ethanol / acetone through van der Waals force / hydrogen bond; NiO / Co3O4 catalyzes the oxidation of NH3, and TiO2 / SnO2 bonds NO2; BaTiO3 ferroelectric polarization responds to the change of gas dipole moment. Each material may respond to the remaining gases, therefore, the neural network is used to jointly analyze the Δf, Δ|S21| of the nine resonator units, and can specifically detect ethanol, acetone, NH3, and NO2.

[0073] As shown in Figure 5 , the S21 parameter changes caused by the introduction of 10-500 ppm acetone gas, and the S21 responses generated by the nine resonators can be seen. Due to the different gas-sensitive materials coated thereon, they cause different S21 changes when different concentrations of acetone are introduced. Based on this, multiple groups of data are tested, and the S21 parameters are input into the neural network for classification and regression. The fitting curve of the predicted concentration and the target concentration is shown in Figure 6 , R=0.99, which shows extremely high detection accuracy, indicating that the electronic nose device based on the microwave resonator array can be suitable for gas discrimination and detection, and provides a new monitoring scheme for the industrial and food safety fields.

[0074] The above examples are only preferred embodiments for fully illustrating the present application, and the size of the 3x3 resonator array is only introduced. The size of the resonator array can be adjusted according to actual needs, such as 2x2, 2x3, 4x4, etc. The working mode of the resonator array introduced in the above examples is not limited to a single mode, and multiple working modes can also be performed simultaneously, for example, resonant mode 1 and resonant mode 2 are performed simultaneously, which can realize single gas detection; resonant mode 1 to resonant mode 6 are performed simultaneously, which can realize the discrimination and concentration detection of binary gas; resonant mode 1 to resonant mode 9 are combined, which can realize the specific discrimination and specific concentration detection of four-component gas, and effectively eliminate the influence of environmental temperature and humidity on the detection accuracy.

[0075] In addition, according to the different gases to be detected, the electronic nose device can be combined with different sensitive materials, including but not limited to the gas-sensitive materials proposed above. By combining the advantages of the detachable upper sensing array layer and microwave detection, the electronic nose can be reconstructed at room temperature. The specific material selection of the gas-sensitive material also includes:

[0076] Indium oxide, tungsten oxide, iron oxide, etc. in N-type semiconductors are sensitive to hydrogen, ozone, oxygen, etc. and are often doped with Pt, Pd, Au, etc. noble metals or other oxides to improve selectivity and sensitivity;

[0077] Chromium trioxide, lanthanum oxide in P-type semiconductor, which are also sensitive to ethanol, NO2, acetone and other gases, but due to the problem of selectivity, can be modified and optimized by doping, compounding and nanostructure regulation.

[0078] The gas-sensitive materials that can be used in the conductive polymer aspect include polypyrrole (sensitive to NO2, NH3, H2S, etc., good film-forming property), polythiophene and its derivatives (sensitive to VOCs, NH3, NO2, etc., good performance adjustability), polystyrene sulfonate, etc.

[0079] The carbon-based materials include graphene, carbon nanotubes, carbon black, etc., which have high specific surface area, conductivity and chemical stability.

[0080] Metal-organic frameworks (MOF) are a new type of porous crystal material, which are self-assembled from metal ions / clusters and organic ligands, and typical representatives include ZIF series, MIL series, UiO series, etc., which have super-high specific surface area, pore size and highly adjustable chemical environment, strong adsorption capacity for specific gas molecules and great potential for selectivity.

[0081] The remaining materials include but are not limited to the above categories. In general, the selection of a gas-sensitive material depends on the target gas, application scenario (such as whether high-temperature operation is required, power consumption limitation, cost requirement, stability requirement, whether flexibility is required) and required performance indicators (sensitivity, selectivity, response speed, etc.), and the selection of the gas-sensitive material can be flexibly adjusted according to the actual use requirement and the size of the resonator array, and finally combined with a neural network for classification and regression.

[0082] In summary, the electronic nose device based on a microwave resonator array provided by the application adopts a double-layer detachable microwave sensing array architecture, that is, a modular design with physical separation, the lower feeding layer integrates a ground plane, a lower dielectric substrate and a microwave coupling feed line, the upper sensing array layer is designed to be pluggable, and the upper sensing array layer and the lower feeding layer are tightly attached through the fixing screw holes and fixing screws of the upper and lower layers. This design enables the upper sensing array layer to be flexibly replaced for different detection scenarios, greatly reduces the maintenance cost and breaks through the shackles of single use of the device. On the surface of the upper sensing array layer, nine independently designed interdigital resonator units are arranged in a 3x3 matrix, each resonator unit excites a highly localized electromagnetic field distribution through unique geometric parameter optimization, forming a frequency-spectrally separated resonant mode. The surface of each resonator unit can be functionalized to coat a gas-sensitive material coating for specific gases (such as NH3, VOCs), and the dielectric constant change caused by gas molecule adsorption is used to disturb the resonant frequency and amplitude, realizing parallel real-time detection of nine kinds of gases. Most importantly, the electronic nose device can achieve high-sensitivity response at ppm level at room temperature, completely avoiding the need for high-temperature operation, and significantly improving the energy efficiency ratio and safety of the equipment.

[0083] As Figure 7 shown, in order to fully analyze the response signals such as frequency offset, amplitude attenuation and phase change generated by the sensing array, and solve the problem of mixed gas cross interference, the application provides a gas analysis method based on a microwave resonator array, which adopts an electronic nose device based on a microwave resonator array. The gas analysis method is based on a graph neural network (GNN) model, each resonator unit is modeled as a graph node, the response characteristics are taken as node attributes, and the electromagnetic coupling effect and spatial correlation between units are taken as dynamic edge weights. Through hierarchical graph convolution and attention mechanism, the network can adaptively learn the microwave parameter changes caused by different gas combinations, and first complete the qualitative classification of multi-component gas, and then accurately quantify the concentration of each gas through the regression branch; the gas analysis method specifically includes the following steps:

[0084] Step 1, the gas to be measured is introduced into the resonator array 7, and the S21 parameters of each resonator unit are collected, including resonant frequency offset Δf, amplitude attenuation Δ|S21|, phase change , etc. (quality factor, Smith chart, etc. can be additionally used), and the change amount of the above characteristics is calculated;

[0085] Step 2, each resonator unit is regarded as an independent graph node to construct a graph structure, and the node features include Δf, Δ|S21|, , etc. Dynamic edges are generated based on electromagnetic coupling characteristics: when the resonant frequency difference is <50MHz, frequency coupling edges are established, and when the response change correlation coefficient is >0.7, response correlation edges are established;

[0086] Step 3, use hierarchical attention mechanism to analyze mixed gas information, gas level attention mechanism and concentration level attention mechanism, calculate gas-specific weights according to the pre-set learnable query table of each gas, generate concentration-sensitive weights through a lightweight multilayer perceptron, use multi-head attention to filter key resonator nodes, and finally focus on the key feature area of different gas responses;

[0087] Step 4, output three kinds of cooperative tasks in parallel: one is gas existence detection, using a sigmoid classifier to output the probability of gas existence; one is gas type classification, concatenating node-level features through a Sofrmax classifier to give the probability of different combinations; and the last one is concentration probability modeling, according to the concentration expectation to output the 95% confidence interval.

[0088] This hardware and algorithm co-innovation not only breaks through the detection bottleneck of traditional sensors "single time and single target", but also provides a method basis for building a high-throughput, low-power intelligent gas sensing platform.

[0089] The application utilizes the different adsorption ability of different gas-sensitive materials to different gases to realize selective detection and multi-mode combined analysis of the gases. After the gas-sensitive material adsorbs the gas molecules, the dielectric properties of the near field of the resonator will change, and thus the corresponding resonance mode will also change. The electronic nose device utilizes the multi-mode change characteristics to realize multi-element gas detection at the same time, and effectively resists the influence of environmental temperature and humidity on the gas detection accuracy. The electronic nose device has simple manufacturing process and high sensitivity, is suitable for industrial environment and food quality monitoring, and has wide application prospect in the fields of gas monitoring and sensor device integration.

[0090] The principles and implementation manners of the present application are described by using specific examples in the present article, and the above examples are only used to help understand the method and core idea of the present application. It should be noted that, for ordinary skilled in the art, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application. Therefore, the protection scope of the present application patent should be subject to the appended claims.

Claims

1. A microwave resonator array based electronic nose device, characterized in that, The device comprises an upper sensor array layer (1) and a lower feeding layer (2) below the upper sensor array layer (1); the lower feeding layer (2) comprises a ground plane (3), a lower dielectric substrate (4) arranged on the ground plane (3), and a microwave coupling feed line (5) arranged on the lower dielectric substrate (4); the upper sensor array layer (1) comprises an upper dielectric substrate (6) and a resonator array (7) arranged on the upper dielectric substrate (6), the resonator array (7) comprises a plurality of resonator units, and different gas-sensitive material coatings (8) are coated on the plurality of resonator units for detecting multiple gases.

2. A microwave resonator array based electronic nose device according to claim 1, characterized in that, The equivalent circuit of each resonator unit of the resonator array (7) is represented as an RLC series circuit, and the resonant frequency is: where L sensor is the equivalent inductance of the resonator, C D is the change in capacitance of the gas sensitive material after binding a gas molecule, C is the equivalent capacitance of the resonator.

3. A microwave resonator array based electronic nose device according to claim 2, wherein, The resonator array (7) is coated with different types of gas-sensitive material coatings (8), including P-type semiconductor, N-type semiconductor, polymer, and strong dielectric ceramic material; Each gas-sensitive material coating (8) is sensitive to one or more gases, and when the gas is adsorbed onto the gas-sensitive material coating (8), the change in material dielectric properties is determined by the Debye model: where ω = 2πf, τ = 1 / (2πf) R ), f R is the relaxation frequency, ε' ∞ is the real part for f→∞, ε' s is the real part for f→0, the higher the operating frequency, the greater the change in dielectric constant, the greater the change in capacitance C D caused by gas adsorption in the resonator, the more significant the change in resonance frequency and the S21 parameter.

4. The microwave resonator array based electronic nose device according to claim 1, wherein, The upper sensor array layer (1) and the lower feeding layer (2) are detachably connected by screws.

5. The microwave resonator array based electronic nose device according to claim 1, wherein, The resonator array (7), the microwave coupling feed line (5), and the ground plane (3) are made of conductive material.

6. The microwave resonator array based electronic nose device according to claim 1, wherein, The upper sensor array layer (1) and the lower feeding layer (2) are prepared on a polytetrafluoroethylene dielectric substrate through a wet etching process.

7. The microwave resonator array based electronic nose device according to claim 1, wherein, The resonator array (7) comprises nine independently designed interdigital resonator units arranged in a 3x3 matrix, which can produce resonant modes at different frequencies according to different resonator sizes.

8. A microwave resonator array based electronic nose device according to claim 7, wherein, The resonator array (7) is coated with nine gas-sensitive materials on the exposed surface, including three P-type semiconductors, three N-type semiconductors, one polymer, one two-dimensional material, and one ceramic material, to selectively adsorb target gas molecules; the N-type semiconductors include SnO2, ZnO, and TiO2, the anatase phase of TiO2 enhances the NO2 capture ability of the surface oxygen vacancies, and the SnO2 and ZnO double N-type semiconductors cross-verify the NO2 signal, with SnO2 high sensitivity and ZnO fast recovery characteristics complementary; the P-type semiconductors include NiO, Co3O4, and CuO, the hole-dominant conduction mechanism of NiO has intrinsic selectivity for reducing gases, NH3 molecules react with pre-adsorbed oxygen to release electrons and neutralize holes; the spinel structure of Co3O4 provides multiple oxidation states, enhancing the catalytic oxidation activity of NH3 and improving the sensitivity compared to NiO, and the P-type characteristics of CuO assist NH3 detection, with its surface Cu + Active sites preferentially adsorb ammonia molecules; the two-dimensional material selects Ti3C2T x xene, the surface termination groups form a directional hydrogen bond network with the ethanol hydroxyl group, the imaginary part of the dielectric constant changes significantly, and the layered structure provides a specific surface area of >800 m 2 / g, achieving high sensitivity for ethanol detection; polyaniline as a high polymer material, the acetone carbonyl induces changes in the conformation of aniline unit molecular chains, changing the polaron migration path, and can specifically distinguish between ketone and alcohol interference; strong dielectric ceramic material barium titanate, whose ferroelectric domain flips in response to polar molecules, with a real part of the dielectric constant change amplitude of up to 40%, widening the temperature and humidity adaptability, used to exclude the influence of environmental humidity on the accuracy of the electronic nose device.

9. A method of gas analysis based on an array of microwave resonators, characterized by, The gas analysis method based on the microwave resonator array electronic nose device of any one of claims 1-8 is based on a graph neural network model, and specifically comprises the following steps: Step 1, the test gas is introduced into the resonator array (7), and the S21 parameters of each resonator unit are collected, including the resonant frequency shift Δf, the amplitude attenuation Δ|S21|, and the phase change and the change amount of the above features is calculated; Step 2: Treat each resonator unit as an independent graph node to build a graph structure. The node features include Δf, Δ|S21|, Generate dynamic edges based on electromagnetic coupling characteristics: establish a frequency coupling edge when the resonant frequency difference is less than 50MHz, and establish a response correlation edge when the response change correlation coefficient is greater than 0.7; Step 3: Use the hierarchical attention mechanism to analyze the mixed gas information, gas level attention mechanism, and concentration level attention mechanism, calculate the gas-specific weight according to the pre-set learnable query table of each gas, generate the concentration-sensitive weight through a lightweight multi-layer perception, use the multi-head attention to filter the resonator nodes, and finally focus on the key feature area of different gas responses; Step 4: Parallel output three cooperative tasks: one is gas existence detection, using a sigmoid classifier to output the gas existence probability; one is gas type classification, concatenating the node-level features and using a Sofrmax classifier to give the probability of different combinations; and the last one is concentration probability modeling, outputting the 95% confidence interval according to the concentration expectation.

10. Application of the microwave resonator array-based electronic nose device of any one of claims 1-8 in multi-gas-specific identification.

Images