Utafiti huu umeunda jukwaa kompakt la kuhisi gesi kwa CMUT linalolenga kubaini muundo wa gesi kupitia mabadiliko ya muda wa kusafiri na amplitudo ya mawimbi ya ultrasoniki. Mfumo una transmitter CMUT na receiver CMUT za capacitive micromachined ultrasonic transducers zilizowekwa zikikabiliana. Short ultrasonic signal inayozalishwa na transmitter CMUT hupita kwenye controlled gas path yenye urefu wa 6 mm na kufika receiver CMUT. Muundo wa gesi unapobadilika, speed of sound na acoustic attenuation pia hubadilika; kwa hiyo time of flight ya received signal na amplitudo yake katika frequency spectrum hubadilika.

Ubunifu mkuu wa jukwaa lililopendekezwa ni kuunganisha sehemu muhimu za laboratory setup ambazo hapo awali zilihitaji function generator, high-voltage source, wideband power amplifier, oscilloscope na separate preamplifier katika portable electronic system moja. Mfumo unajumuisha high-voltage converter inayozalisha takriban 100 V kutoka 5 V source, wideband transmitter amplifier yenye GaN transistor, transimpedance receiver circuit na embedded Red Pitaya platform inayowezesha simultaneous waveform generation na acquisition.

Compact prototype ina dimensions za takriban 250 × 150 × 50 mm, volume ya 1,9 lita na mass ya 1 kg. Kwa kuwa laboratory setup ya awali ilikuwa takriban 63 lita na 30 kg, volume imepunguzwa takriban mara 33 na mass takriban mara 30. Power consumption ya transmitter-receiver front-end circuit ni chini ya 1 W, huku total system consumption pamoja na Red Pitaya ikiwa takriban 11 W. Kulingana na theoretical calculation ya watafiti kutoka energy capacity, portable battery ya 5 V na 10 Ah inaweza kutoa takriban masaa 4,5 ya operation. Duration hii haikupatikana kwa direct long-duration battery test, bali kwa kugawanya energy capacity ya takriban 50 Wh kwa power consumption.

Katika experiments, H₂, CO₂, CH₄ na He zilichanganywa na nitrogen; concentration steps za %4, %3, %2 na %1 zilitumika kwa kila gas. Measurements zilifanywa katika 1,5 MHz na total gas flow ya 100 cm³/min. Time-of-flight change na spectrum-amplitude change zilipotathminiwa pamoja, gases nne zilionekana kusogea katika directions tofauti kwenye ΔA–ΔToF plane. Finding hii inaonyesha kwamba mfumo ni sensitive kwa gas type na concentration changes katika controlled conditions. Hata hivyo, utafiti hautoi classification algorithm, detection limit, uncertainty calculation au independent blind test.

Mfumo hautegemei special chemical-sensing coating inayoreact na gas. Badala yake, unapima physical properties za gas kama speed of sound na attenuation. Approach hii ina potential ya kupunguza coating aging au chemical-selectivity issues katika applications fulani. Hata hivyo, kwa kuwa utafiti haukufanya direct long-term comparison na chemical sensors, lower drift, longer lifetime au higher reliability bado hazijathibitishwa experimentally.

Kwa mtazamo wa Uturuki: Mfumo unaweza kuwa msingi wa research kuhusu portable leak na mixture monitoring katika hydrogen production na storage infrastructures, natural-gas facilities, biogas applications, chemical industry, laboratory gas lines, energy-storage systems na technical facilities zinazotumia helium nchini Uturuki. Kwa application nchini Uturuki, multipoint calibration kwa local gas compositions, temperature na humidity compensation, testing ya gases ndani ya air, lower-concentration experiments, electromagnetic compatibility, vibration na aging tests, indoor na outdoor field trials, pamoja na explosive-atmosphere safety assessments vinapaswa kufanywa. Kutokana na utafiti huu haiwezi kuhitimishwa kwamba mfumo unatosha kama commercial leak detector, unatoa legal metrology accuracy, ni salama katika explosive atmospheres, una %0,01 detection limit au unaweza kutambua unknown multigas mixtures kiotomatiki.

CMUT ni nini na inapimaje gesi?

CMUT inamaanisha capacitive micromachined ultrasonic transducer. Kila cell ina membrane nyembamba sana na flexible iliyowekwa juu ya fixed bottom electrode, pamoja na gap kati ya structures hizi mbili. Membrane na bottom electrode kwa pamoja huunda electrical capacitor.

Kielelezo 1 kwenye ukurasa wa 3 wa utafiti kinaonyesha transmit na receive mechanism ya CMUT kwa sectional view. Main layers katika structure ni:

• Aluminum top electrode.
• Silicon nitride membrane.
• Vacuum gap.
• Conductive polycrystalline silicon bottom electrode.
• Silicon dioxide insulation layer.
• Silicon carrier substrate.

Wakati wa transmission, fixed DC voltage na time-varying AC voltage hutumika pamoja kwenye CMUT. Electrostatic force hutikisa membrane nyembamba na ultrasonic wave hutolewa kwenye surrounding gas medium. Wakati wa reception, incoming acoustic pressure husogeza membrane, capacitance ya cell hubadilika na change hii hupimwa kama electrical current.

DC biasing ya CMUT huongeza transmit na receive sensitivity. Lakini ikiwa voltage inakaribia sana collapse au pull-in limit ambapo membrane inavutwa kwa instability kuelekea bottom electrode, linear operating range na safety margin hupungua. Hii ndiyo sababu transmitter na receiver ziliendeshwa katika voltages tofauti katika utafiti.

Muundo wa CMUT arrays zilizotumika

Sensor head ina circular CMUT arrays mbili zinazofanana zilizowekwa face to face. Kila array ina square cells 1.941 zilizounganishwa parallel. Lateral dimension ya cells ni 32 µm, diameter ya circular active aperture ni 7 mm na active-area ratio ni %41.

Kielelezo 2(a) kwenye ukurasa wa 3 kinaonyesha close-up ya square CMUT cells zilizounganishwa kwa top metal interconnects; Kielelezo 2(b) kinaonyesha circular array yenye diameter ya 7 mm na electrical connection pads kuzunguka.

Katika Table 1, notation ya cell area imepangwa kwa namna inayomaanisha “32² µm²”, lakini maandishi yanasema wazi kwamba cell side length ni 32 µm. Kwa hiyo cell size inaeleweka kuwa 32 µm × 32 µm.

Kwa nini gas composition hubadilisha time of flight na amplitude?

Ultrasonic wave inayozalishwa na transmitter CMUT husafiri katika gas medium mpaka ifike receiver. Gas composition inapobadilika, speed of sound inaweza pia kubadilika. Change hii huathiri muda unaochukuliwa na signal kukamilisha path ya 6 mm. Katika utafiti, quantity hii inaitwa time of flight, yaani ToF.

Acoustic attenuation ya gas pia huathiri amplitude ya received signal. Kwa hiyo watafiti walitumia acoustic features mbili pamoja badala ya kutegemea measurement moja:

• ΔToF: Change ya measured time of flight relative to pure-nitrogen reference.
• ΔA: Change ya FFT amplitude at selected frequency relative to pure-nitrogen reference.

Kutumia features hizi mbili pamoja hutoa additional information ya kutenganisha gases zenye responses zinazokaribiana katika time-of-flight change kwa amplitude behavior, au kinyume chake.

Tatizo la laboratory setup ya awali lilikuwa nini?

Kielelezo 3 kwenye ukurasa wa 4 kinaonyesha previous laboratory measurement chain. Setup hii ilijumuisha:

• Function au arbitrary waveform generator.
• External wideband power amplifier.
• Multiple high-voltage power supplies.
• Separate transmitter na receiver bias networks.
• Receiver preamplifier.
• Oscilloscope.
• Multiple power, signal na ground connections.

Ingawa structure hii hutoa flexibility katika research laboratory, inafikia takriban 63 lita volume na 30 kg mass. Long cables na ground connections za separate instruments zinaweza kuunda noise na impedance mismatch katika weak airborne ultrasound signal. Synchronization ya measurements kati ya devices tofauti na kupata results katika field environment pia huwa ngumu.

General architecture ya compact platform

Kielelezo 4 kwenye ukurasa wa 4 kinaonyesha compact measurement chain inayochukua nafasi ya laboratory instruments. Mfumo una sehemu nne kuu:

1. Face-to-face CMUT transmitter na receiver pair.
2. Custom electronics board yenye transmitter na receiver amplifiers, high-voltage generation na biasing.
3. Red Pitaya board inayotoa waveform generation, synchronization na signal acquisition.
4. 5 V power supply kutoka USB au portable battery.

Low-voltage waveform inayozalishwa na Red Pitaya huamplifyiwa katika custom transmitter circuit na kutumika kwenye CMUT. Very small current kutoka receiver CMUT hubadilishwa kuwa voltage kwa transimpedance amplifier na kudigitizeiwa na ADC ya Red Pitaya. Kusimamia transmitter na receiver timing kutoka system hiyo hiyo huongeza repeatability ya time-of-flight calculation.

Kwa nini GaN-based transmitter amplifier ilitumika?

High-voltage na wideband amplifier inahitajika kuendesha CMUT transmitter kwa smooth sinusoidal pulses katika wide frequency range. High parasitic capacitance na larger conduction resistance ya conventional silicon power transistors inaweza kuwa limitation katika compact high-frequency circuit.

Katika utafiti, gallium nitride, yaani GaN transistor, ilichaguliwa kwa properties hizi:

• Low conduction resistance.
• Low Miller capacitance.
• Low gate charge na low gate-drive voltage.
• High-frequency na high-voltage compatibility.
• Reduced reverse-recovery charge kutokana na kutokuwa na body diode.

Jambo muhimu katika utafiti huu ni kwamba GaN element imetumika si kama conventional ultrasonic square-wave switch, bali kama pulsed linear amplifier inayozalisha controlled sinusoidal signal. Transistor huendeshwa katika linear region wakati wa ultrasonic pulse pekee na kuzimwa kati ya pulses. Hivyo heating na static power consumption zinazoweza kutokana na linear operation hupunguzwa.

Kwa active window ya takriban 131 µs na pulse interval ya 16,5 ms, duty cycle ni takriban %0,8. Watafiti hawakuhitaji dedicated heatsink kwenye prototype kutokana na short duty cycle hii. Hata hivyo, thermal behavior inapaswa kuthibitishwa tofauti katika continuous au higher-repetition-rate operation.

Kwa nini transmitter waveform ililainishwa?

Standard ultrasonic transmitters mara nyingi hutumia square voltage pulses zenye sharp edges. Utafiti huu unaendesha CMUT chini ya resonance yake ya takriban 8 MHz katika wideband region ya 0,5–5 MHz. Sharp-edged pulse inaweza kuchochea undesired resonance na transient vibrations.

Kwa hiyo transmitted sinusoidal pulse ililainishwa kwa Hanning window. Mwanzo na mwisho wa pulse si wa ghafla. Kielelezo A3 katika Appendix A kinaonyesha amplified waveforms katika control voltages tofauti:

• Katika 165 mV circuit huendeshwa karibu na cutoff na output hupotoshwa.
• Katika 215 mV circuit hukaribia rail limitation na clipping hutokea.
• Katika 190 mV output huwa linear zaidi na distortion kidogo.

Appendix A inaripoti total gain kutoka input hadi output kuwa takriban 60 dB. Gain hii inajumuisha pia contribution ya takriban 20 dB kutoka operational amplifier.

Frequency-dependent load effect

CMUT hutengeneza capacitive load electrically. Kielelezo A4 katika Appendix A kinaonyesha kwamba open circuit na different CMUT loads zenye 3,1 nF, 0,95 nF na 0,15 nF hubadilisha transmitter gain kwa kiasi kikubwa. Output inaweza kupungua frequency inapoongezeka wakati input amplitude ile ile inatumika.

Katika Kielelezo A5, watafiti walirekebisha input amplitude kulingana na frequency na kwa kiasi kikubwa kuflatten reduction hii. Gain iliyopungua kwa wazi pamoja na frequency katika fixed drive ilibaki takriban constant katika frequency-adjusted drive. Property hii ni muhimu kwa future multi-frequency gas measurement; hata hivyo, multi-frequency method hii haikutumika katika gas experiments.

Receiver transimpedance amplifier

Receiver CMUT hubadilisha incoming acoustic wave kuwa very small electrical current. Katika utafiti, transimpedance amplifier ilitumika kubadilisha current hii kuwa measurable voltage.

Kielelezo 6 kwenye ukurasa wa 6 kinaonyesha receiver circuit. Main feedback components zilizochaguliwa ni:

• 1 MΩ feedback resistor.
• 0,2 pF nominal feedback capacitor.
• Takriban 0,4 pF effective feedback capacitance pamoja na PCB parasitics.

22 kΩ, 56 kΩ, 110 kΩ na 1 MΩ resistors zililinganishwa. Large feedback resistor hutoa higher signal gain katika low-megahertz region lakini hupunguza usable bandwidth. Small resistors hutoa wider bandwidth lakini hupunguza received signal amplitude.

Kwa takriban 1,5 MHz ambako gas experiment ilifanywa, 1 MΩ resistor ilichaguliwa kama balance inayofaa kati ya gain na bandwidth. Katika multi-frequency application, switchable feedback resistors au additional gain stages zinaweza kuhitajika.

Takriban 100 V ilitolewaje kutoka 5 V?

Ingawa takriban 70–90 V zinahitajika kwa CMUT biasing, portable system inalishwa tu na 5 V USB source. Kwa hiyo, boost DC-DC converter yenye discrete components ilitengenezwa kwenye board.

Converter iliyoonyeshwa kwenye Kielelezo 7 cha ukurasa wa 6 ina:

• Inductor.
• Low-side switching transistor.
• Rectifier diode.
• Parallel output capacitors.
• Voltage-limiting Zener diode.
• Self-oscillating PNP/NPN control stage

components. Current inayohitajika katika output ya takriban 100 V ni karibu 2 mA pekee.

Converter huswitch katika takriban 16 kHz. Switching hii na harmonics zake zinaweza kuzalisha noise inayochanganyika na weak receiver signal. Kielelezo B1 katika Appendix B kinaonyesha 16 kHz fundamental component pamoja na higher harmonics kwenye switching node.

Switching noise ilipunguzwaje?

Boost converter husimamishwa kwa muda transmitter pulse inapoanzishwa. Kielelezo B2 katika Appendix B kinaonyesha kwamba converter switching hupotea katika transmitter na receiver measurement window na huanza tena baada ya measurement kukamilika.

Converter ikiwa imezimwa, high voltage ya CMUT hutolewa kutoka energy iliyohifadhiwa katika output capacitors. Baada ya pulse voltage hupungua na converter huchaji capacitors tena. Kulingana na Kielelezo B3, karibu 16,5 ms zinahitajika ili high-voltage rail irudi karibu 100 V. Kwa hiyo minimum pulse interval ya system ilichaguliwa kuwa 16,5 ms.

Solution hii hutoa low noise lakini huzuia acquisition speed. Higher-current au more-efficient converter inaweza kupunguza recharge time.

Embedded measurement system inayotegemea Red Pitaya

Digital core ya system ni STEMlab 125-14 Gen 2 Red Pitaya board. DAC na ADC units za board zinaweza kusample hadi 125 MS/s. Speed hii hutoa temporal resolution ya kutosha kwa generation na acquisition ya 0,5–5 MHz waveforms.

Kielelezo 8 kwenye ukurasa wa 6 kinaonyesha hardware na software flow. Main tasks ni:

• Kuandaa mapema sinusoidal na Hanning-windowed transmitter pulse katika memory.
• Kuzalisha kwenye DAC output kupitia digital direct synthesis.
• Kusynchronize ADC acquisition na transmit pulse.
• Kuaverage multiple acquisitions on-board ikihitajika.
• Kutuma measurement data na setting information kwa host computer kupitia UDP.
• Signal processing na real-time visualization kwa Python 3 software.

UDP ilichaguliwa kwa low protocol overhead na short latency. Measurement index na time information huongezwa kwenye kila data frame ili kutoa traceability katika continuous measurements.

Hatua za signal processing

Received signal ilichambuliwa kwa mpangilio huu:

1. Kupata first received ultrasonic signal ndani ya predefined time window.
2. Kufanya cross-correlation kati ya received signal na programmed transmit waveform.
3. Kupunguza frequency components na noise zilizo nje ya correlation.
4. Kutoa signal envelope kwa Hilbert transform.
5. Kubaini time of flight kutoka peak position ya envelope.
6. Kutumia FFT kwenye valid signal region.
7. Kutoa spectrum amplitude katika selected excitation frequency.
8. Kuhesabu result kama difference relative to pure-nitrogen reference.

Time-of-flight change:

\[ \Delta \mathrm{ToF} = \mathrm{ToF}(n) - \mathrm{ToF}(N_2) \]

Spectrum-amplitude change:

\[ \Delta A = H_{\mathrm{FFT}}(n) - H_{\mathrm{FFT}}(N_2) \]

Hapa n inawakilisha current gas measurement na \(N_2\) pure-nitrogen reference. ΔA si absolute absorption coefficient; ni difference ya received FFT amplitude katika selected frequency relative to nitrogen measurement.

Gas-flow experiment

Gas mixtures ziliandaliwa kwa Brooks 5850S mass-flow controllers. Kila target gas ilipelekwa kwa sensor kama binary mixture na nitrogen.

Pure-nitrogen purge na reference measurement zilifanywa kati ya gas conditions. Hii ilitumika kupunguza athari za components zilizobaki kutoka gas ya awali na kufidia kwa sehemu slow temperature au electronic drifts.

Frequency performance

Kielelezo 10 kwenye ukurasa wa 9 kinaonyesha frequency response ya entire CMUT-transmitter-receiver chain kwa different receiver feedback resistors. Response inategemea combination ya effects hizi, si circuit block moja:

• CMUT transmitter sensitivity.
• CMUT receiver sensitivity.
• Bandwidth ya receiver na transmitter electronics.
• Acoustic diffraction katika low frequency.
• Airborne-ultrasound attenuation katika high frequency.
• Electrical capacitive loading.

System ilizalisha measurable signal katika range ya 0,5–5 MHz. Katika basic airborne bandwidth experiments, signal hadi 5 MHz iliweza kupokelewa bila additional interpolation au averaging. Katika gas experiment, acquisitions 20 ziliaveragewa kwa stable feature extraction.

Kielelezo 11 kinaonyesha received time signals na spectra katika different center frequencies. Main text inaita last panel 4 MHz, wakati figure caption inasema 3 MHz. Inconsistency hii inapaswa kusahihishwa.

Matokeo ya compactness na power consumption

Measurement electronics ya compact system hutumia chini ya 1 W. Red Pitaya hutumia hadi takriban 10 W wakati Wi-Fi inatumika. Kwa total consumption ya takriban 11 W, kugawanya nominal battery capacity ya 50 Wh hutoa theoretical runtime ya masaa 4,5. Real runtime inaweza kuwa fupi zaidi kwa sababu ya converter losses, battery temperature na capacity degradation.

Processed signals zinaonyesha nini?

Kielelezo 12 kwenye ukurasa wa 10 kinalinganisha signals zilizopatikana baada ya cross-correlation kwa pure N₂, %4 H₂ + %96 N₂ na %4 CO₂ + %96 N₂. Kila graph inaonyesha:

• Processed ultrasonic waveform katika time domain.
• Hilbert envelope.
• Envelope peak inayotoa time of flight.
• FFT amplitude katika 1,5 MHz

Inaonekana kwamba H₂ na CO₂ mixtures zilitofautiana na nitrogen reference katika peak position na spectrum amplitude.

Response kwa concentration steps

Kielelezo 13(a) kwenye ukurasa wa 11 kinaonyesha time-dependent changes za ΔToF na ΔA katika experiment; Kielelezo 13(b) kinaonyesha applied gas na concentration program. Takriban data points 540 zilikusanywa katika kila three-minute step.

Short transition deviations zilitokea wakati wa gas switching. Watafiti walihusisha hizi na delay za flow-control valves na flow disturbances wakati wa kubadilisha gas. Representative values zilichukuliwa kutoka middle section ya kila step ili kuepuka transient regions.

First-order linear trends zilitumika kwa concentration series zinazoshuka kutoka %4 hadi %1. Lines hizi zinaonyesha response direction katika working range; kwa kuwa independent validation haikufanywa, hazipaswi kuchukuliwa kuwa full calibration models zinazotumika kwa concentrations zote.

ΔA–ΔToF map

Kielelezo 14 kwenye ukurasa wa 11 kinaonyesha measured amplitude change dhidi ya time-of-flight change kwa H₂, CO₂, CH₄ na He. Gas families nne ziliunda trajectories tofauti:

• Points za H₂, CO₂, CH₄ na He hazioverlap kabisa kwenye line moja.
• Gas concentration inapobadilika, kila gas husogea kwenye trend yake.
• Mixed-gas point yenye %1 ya kila gas iko katika intermediate region.

Result hii inaonyesha kwamba kutumia acoustic features mbili pamoja hubeba information kuhusu gas type. Hata hivyo, figure inatoa visual separability tu; kwa kuwa hakuna trained classifier, blind test, confidence region au error matrix, quantitative “gas recognition accuracy” haiwezi kuhesabiwa.

Stability na drift

Watafiti wanaeleza kwamba temperature drift ilibaki limited wakati wa short experiment na nitrogen references zilipunguza slow changes. Maandishi pia yanasema kwamba katika longer observations residual ΔToF baseline drift ilikuwa takriban %0,5 ndani ya takriban masaa 11.

Hata hivyo, detailed time series ya long-term test, number of repeats na equivalent under different environmental conditions hazijawasilishwa kwenye main figures. Kwa hiyo value ya %0,5 haipaswi kutumika kama long-term product stability au field-calibration interval.

Nguvu za utafiti

• Unaunganisha CMUT, high-voltage generation, transmitter, receiver na fast acquisition katika portable chain moja.
• Unatumia GaN transistor kama pulsed linear amplifier kuzalisha controlled sinusoidal high voltage.
• Bias voltages za transmitter na receiver CMUT zinaweza kurekebishwa independently.
• High-voltage converter husimamishwa wakati wa measurement ili kupunguza switching noise.
• Receiver feedback resistor inaweza kubadilishwa ili kurekebisha gain na bandwidth.
• Measurable airborne ultrasonic transmission imeonyeshwa katika range ya 0,5–5 MHz.
• Real controlled gas-flow experiments zimefanywa kwa H₂, CO₂, CH₄ na He.
• Time of flight na spectrum amplitude zimetolewa pamoja kutoka measurement hiyo hiyo.
• Volume na mass zimepunguzwa takriban mara 30 dhidi ya previous laboratory system.
• Battery na Wi-Fi zimetumika kuonyesha measurement independent of mains power.

Mapungufu ya utafiti

• Utafiti ni preprint ambayo haijapitia peer review.
• Gas-sensing experiment ilifanywa katika 1,5 MHz pekee.
• Acoustic path ni 6 mm pekee.
• Lowest tested single-gas concentration ni %1.
• Detection limit, quantitative accuracy na uncertainty budget hazijatolewa.
• Gas-classification accuracy au false-alarm rate hazijahesabiwa.
• Independent repeats na error bars kwa kila gas hazijaripotiwa wazi.
• Mixed-gas experiment ina composition point moja pekee.
• Gas behavior katika air, oxygen au real industrial background haijajaribiwa.
• Humidity haikubadilishwa kwa controlled manner.
• Pressure variation haikuchunguzwa.
• Gas transition na purge times hazijatathminiwa kwa quantitative response-time model.
• Gas experiment ilitumia averaging ya signals 20.
• Signal processing bado inafanywa kwenye host computer na haijahamishwa kikamilifu on-board.
• Prototype ina 1 kg na si final mini-sensor size.
• Real battery operating time haijaonyeshwa experimentally.
• Temperature ya GaN element na long-term linear-operation reliability hazijapimwa.
• High-voltage safety na enclosure design hazijatathminiwa.
• EMC, vibration na mechanical-shock tests hazijafanywa.
• Explosive-atmosphere certification au spark safety haijaonyeshwa.
• Manufacturing cost na mass-production analysis hazipo.
• Kuna internal inconsistencies kuhusu GaN element name, measurement frequency na DAC operating point.

Utafiti unaunga mkono nini?

Utafiti unaunga mkono kwamba time of flight na amplitude ya ultrasonic signal inayosafiri kati ya CMUT mbili hujibu concentration changes katika controlled mixtures za H₂, CO₂, CH₄ na He. Pia unaonyesha kwamba basic transmit, receive, high-voltage na data-acquisition functions ambazo hapo awali zilihitaji laboratory instruments nyingi zinaweza kuunganishwa katika battery-powered platform ya takriban lita 1,9.

Experiments zinaunga mkono kwamba pulsed linear GaN amplifier inaweza kutumika kwa wideband sinusoidal CMUT drive, transimpedance receiver inaweza kurekebishwa kwa feedback elements, na kuzima boost converter wakati wa measurement husaidia weak-signal measurement.

Utafiti hauthibitishi nini?

Utafiti hauthibitishi kwamba mfumo unatambua unknown gases kiotomatiki na bila makosa, unadetect gas katika level ya %0,01, unajibu kwa kasi ya kutosha kwa real industrial leaks au hauathiriwi na temperature na humidity changes. Separation ya ΔA–ΔToF curves peke yake haimaanishi certified selectivity au quantitative classification performance.

Utafiti pia hauonyeshi moja kwa moja kwamba CMUT approach ina longer lifetime, lower cost au higher reliability kuliko chemical gas sensors. Benefits hizi ni research potential inayotokana na kutotumia reactive sensing coating na zinahitaji long-term comparative experiments kwa uthibitisho.

Mbinu na Matokeo ya Utafiti

Basic properties za sensor head

Components za electronic system

Embedded platform na communication

Receiver feedback comparison

Gas-experiment sequence

1. Pure N₂ ilipelekwa kwenye sensor head na reference measurement ikachukuliwa.
2. Target gas na N₂ zilichanganywa kwa desired ratio kwa mass-flow controllers.
3. Total gas flow ilidumishwa katika 100 cm³/min.
4. Pulse ya 1,5 MHz yenye duration ya 20 µs ilitumika kwenye transmitter CMUT.
5. Signal kutoka receiver CMUT iliamplyfiwa kwa TIA.
6. Acquisitions 20 ziliaveragewa kwa kila measurement point.
7. Processed data ilitumwa kwa host computer katika takriban sekunde 0,32.
8. ToF ilibainishwa kutoka cross-correlation na Hilbert envelope.
9. Amplitude ya 1,5 MHz ilitolewa kutoka FFT.
10. ΔToF na ΔA zilihesabiwa relative to pure N₂ reference.
11. Stable values kutoka middle section ya kila concentration step zilitumika.

Main quantitative results

Technical inconsistencies ndani ya maandishi

Validations zinazohitajika kabla ya application

• Gas experiments zinapaswa kurudiwa kama multi-frequency katika range ya 0,5–5 MHz.
• Multiple calibration points chini ya %1 zitumike kwa kila gas.
• Detection na quantification limits zihesabiwe.
• Classification accuracy ipimwe kwa independent blind gas samples.
• Wide mixture dataset yenye gases nyingi katika ratios tofauti iundwe.
• Temperature, humidity na pressure variations zijumuishwe katika calibration.
• Realistic background gases zenye air na oxygen zitumike.
• Response, purge na valve-delay times zipimwe tofauti.
• Sensor-to-sensor variability kati ya different CMUT fabrication batches ibainishwe.
• Long-term baseline, repeatability na aging experiments zifanywe.
• Temperature na linear-operation reliability ya GaN element ifuatiliwe.
• Smaller na lower-power embedded hardware itathminiwe badala ya Red Pitaya.
• Signal processing ihamishwe kikamilifu on-board.
• High-voltage section iwekwe katika electrical-safety enclosure.
• EMC, vibration, shock na portable-battery tests zifanywe.
• Explosive-atmosphere safety ithibitishwe kwa hydrogen na methane applications.
• Field trials zifanywe katika pilot facilities chini ya real leak na ventilation conditions.

Dokezo la Chanzo na Mbinu

Jina kamili la awali la utafiti:Compact CMUT Gas Sensor Integrating GaN-Based Amplifier

Waandishi na mpangilio: Yilihamu Abudujiasuer; Rida Yachouti; Etienne Lemaire; Flavien Barcella; Thierry Potier; Fabrice Mathieu; Isabelle Dufour; Dominique Certon.

Equal-contribution author information: Hakuna equal-contribution au co-first-authorship statement katika utafiti.

Corresponding author: Yilihamu Abudujiasuer.

Corresponding-author email: yilihamu.abudujiasuer@univ-tours.fr

Institutional affiliations:

1. GREMAN Laboratory UMR-CNRS 7347, University of Tours, 37071 Tours, France.
2. IMS Laboratory UMR-CNRS 5218, University of Bordeaux, 33405 Talence, France.
3. LAAS Laboratory, LAAS-CNRS UPR8001, University of Toulouse, 31031 Toulouse, France.

Author-institution mapping:

• Yilihamu Abudujiasuer, Rida Yachouti, Etienne Lemaire, Flavien Barcella, Thierry Potier na Dominique Certon: GREMAN Laboratory, University of Tours.
• Isabelle Dufour: IMS Laboratory, University of Bordeaux.
• Fabrice Mathieu: LAAS Laboratory, University of Toulouse.

Source type: Experimental preprint research article yenye CMUT sensor fabrication, custom electronic-circuit development, frequency characterization na controlled gas-flow experiments.

Subject area: MEMS gas sensors, CMUT, airborne ultrasound, GaN amplifiers, embedded electronics, time-of-flight measurement na acoustic gas discrimination.

Peer-review status: Utafiti haujapitia peer review. Kila ukurasa wa uploaded version una preprint na peer-review warning.

Publication platform: SSRN.

Platform provider: Elsevier SSRN Preprint Services.

SSRN record number: 7201800.

DOI:10.2139/ssrn.7201800

Official link:SSRN record page

Publication year: 2026.

Journal: Haijathibitishwa kwamba imechapishwa katika peer-reviewed journal.

Original journal publisher: Hakuna verified peer-reviewed journal au final journal publisher. SSRN ni preprint platform na haipaswi kuonyeshwa kama peer-reviewed journal.

Funding: Utafiti ulifadhiliwa ndani ya MIGNON project inayoungwa mkono na French National Research Agency. Funding number katika article ni ANR-22-CE42-0012-04. Utafiti pia ulitumia CERTeM technological platform katika University of Tours.

MIGNON main project record:ANR-22-CE42-0012

Conflict of interest: Hakuna separate na explicit conflict-of-interest statement katika uploaded main text.

Author contributions: Hakuna CRediT au detailed author-contribution statement inayofanana.

Data na code access: Hakuna open repository link iliyotolewa kwa raw gas-measurement data, PCB fabrication files, circuit-design files, Red Pitaya source code au Python analysis code.

Supporting-information status: Main circuit na characterization details zimetolewa katika Appendix A na Appendix B za main text. Hakuna separate supporting-information file iliyorejelewa.

Bibliographic na technical warning: Main text ina statements tofauti kuhusu model ya GaN element, final frequency ya Kielelezo 11 na nominal DAC DC control level. Maeneo haya yanapaswa kusahihishwa kabla ya final publication.

Maelezo haya ya Kiswahili yameandaliwa baada ya kuchunguza main text yote, tables, circuit diagrams, frequency curves, time signals, gas-response graphs na technical analyses katika Appendix A–B za uploaded study. Scientific content imewekewa kikomo kwenye experimental results na interpretations za watafiti zilizowasilishwa katika utafiti. External sources zilitumika tu kwa bibliographic verification ya title, authors, DOI, platform, institutions na project information.

Results zinapaswa kutathminiwa kwa kuzingatia kwamba utafiti haujapitia peer review, gas experiments zimewekewa kikomo kwa single frequency na short acoustic path, lowest tested concentration ni %1, multigas experiment ina composition moja pekee, classification na detection-limit metrics hazijatolewa, environmental variables hazijajaribiwa na prototype bado imeunganishwa na host computer.