Semiconductor thermal conductivity dual-mode hydrogen detection control system with temperature and humidity compensation

Abstract

The invention discloses a semiconductor thermal conductivity dual-mode hydrogen detection control system with temperature and humidity compensation, and belongs to the field of G05B13 adaptive control systems, and the system comprises a dual-mode sensing module which comprises a semiconductor sensing unit and a thermal conductivity sensing unit; the method comprises the following steps: acquiring an output signal value of each concentration point, establishing a semiconductor sensing characteristic curve, obtaining an output voltage variation delta Vs of a semiconductor sensor, carrying out dynamic range calibration on a thermal conductivity sensing unit through a standard thermal conductivity comparison device, and establishing a thermal conductivity-concentration conversion model environment parameter compensation module. According to the invention, a semiconductor-thermal conductivity dual-mode cooperative detection framework is provided, through a dynamic partition range coverage and sensitivity complementary mechanism, the 1-5ppm-level detection precision in a full concentration range is realized, and a third-order linear temperature compensation function and a three-dimensional humidity interference matrix are constructed, so that the purpose of self-adaptive hydrogen concentration detection is achieved; and the high-humidity working condition data reliability is improved.

Description

Technical Field

[0001] This invention belongs to the field of adaptive control systems, specifically a semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation. Background Art

[0002] Hydrogen detection technology, as a core safety technology in the hydrogen energy industry chain, plays a crucial role in fuel cell vehicles, hydrogen storage facilities, and chemical production. Traditional hydrogen sensors mainly employ semiconductor or thermal conductivity single-mode detection principles: semiconductor sensors have high sensitivity (10ppm) at low concentrations (0-3000ppm), but are prone to saturation in high concentration areas; thermal conductivity sensors are suitable for high concentration detection (1%-100% VOL), but have poor resolution at low concentrations (>100ppm).

[0003] Current dual-mode detection systems on the market (such as the Japanese FIGARO FGS-4000 series) employ a time-division switching strategy, which suffers from two major technical bottlenecks: First, the fixed mode switching threshold leads to critical point oscillations, resulting in a false switching rate of 12% when the concentration fluctuates around 3000 ppm. Second, the environmental compensation algorithm is simplistic, only performing linear temperature correction, which amplifies the detection deviation to ±8% FS when humidity > 80% RH. Furthermore, existing systems use closed communication protocols (such as dedicated CAN bus protocols), making it difficult to integrate with industrial IoT platforms. Summary of the Invention

[0004] The purpose of this invention is to address the technical bottlenecks in the entire detection chain of hydrogen energy safety monitoring, such as detection range breaks, significant environmental interference, mode switching misjudgments, low system energy efficiency, and delayed fault response. This invention analyzes the sources of core problems, including sensor sensitivity attenuation, temperature and humidity coupling interference, threshold strategy defects, power supply architecture redundancy, and insufficient protocol encapsulation. It proposes a multimodal sensor fusion method, a nonlinear compensation algorithm, and an intelligent switching strategy. A theoretical model for wide-domain environmental coupling compensation and dynamic fault tolerance is constructed, providing an adaptive control system: a semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation.

[0005] The technical solution adopted in this invention is as follows: A semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation, comprising:

[0006] The dual-mode sensing module includes a semiconductor sensing unit and a thermal conductivity sensing unit; it acquires the output signal values ​​at each concentration point, establishes the semiconductor sensing characteristic curve, and uses the formula:

[0007] ΔV s = α×Heat + β×T + γ×RH

[0008] Where Heat is the thermal conductivity value of the voltage signal, T is the ambient temperature, and RH is the relative humidity;

[0009] The change in output voltage ΔV of the semiconductor sensor is obtained. s The thermal conductivity sensing unit was dynamically calibrated using a standard thermal conductivity comparison device, and a thermal conductivity-concentration conversion model was established.

[0010]

[0011] Where C0 is the reference concentration, Heat is the thermal conductivity value of the voltage signal, T0 is the standard temperature, and K is the thermal conductivity gain coefficient;

[0012] The environmental parameter compensation module is used to acquire the temperature and humidity parameters of the detection environment in real time and perform compensation calculations. It includes a third-order polynomial temperature drift compensation function and a 3×3 humidity interference compensation matrix.

[0013] The signal fusion processing module is used to perform comprehensive processing of dual-modal signals. It integrates MODBUS-RTU and USART dual-protocol communication interfaces and outputs data frames containing compensation concentration values, temperature and humidity parameters and device status words.

[0014] The mode switching control module is used to realize the dynamic switching of dual-mode detection, setting a hysteresis range of 2950-3050ppm and adopting an exponential decay weight fusion algorithm.

[0015] The environmental parameter compensation module performs the following:

[0016] Ambient temperature T is collected using a digital temperature and humidity sensor. j and relative humidity (RH) k Where j and k represent sampling time sequence numbers, j, k = 1, 2, 3…n;

[0017] Based on the temperature drift characteristics of semiconductor devices, a temperature compensation function is established:

[0018] F (T) =∑[δ t ×(T j -T ref ) m ]

[0019] Where T ref =25℃ is the reference temperature, δ t This is the m-th order temperature compensation coefficient;

[0020] Based on the effect of humidity on semiconductor adsorption characteristics, a humidity correction matrix is ​​constructed:

[0021]

[0022] Where μ1, μ2, and μ3 are humidity interference compensation parameters;

[0023] Eliminate humidity interference within the 30%-90% RH range.

[0024] The mode switching control module implements the following:

[0025] When the detection concentration C corr-s When the temperature reaches ≥3050ppm for three consecutive sampling cycles, the thermal conductivity mode activation command is triggered; during the mode transition phase, press:

[0026]

[0027] Data fusion is performed, among which These are mode weighting coefficients, used when the semiconductor mode is activated. When thermal conduction mode is activated

[0028] The compensation operation of the semiconductor sensor unit is as follows:

[0029] Semiconductor correction concentration:

[0030]

[0031] Where ΔV_s is the change in semiconductor output voltage.

[0032] α=0.18mV / ppm, β=-0.0023 / ℃, γ=0.00015 / %RH,

[0033] Where, ΔT=T j -T ref ΔRH=RH k -RH ref ;

[0034] Among them, RH ref =0.5RH.

[0035] The compensation calculation for the thermal conductivity sensor unit is as follows:

[0036]

[0037] Among them, G0=1.023mV / V, K=0.085mV / V / ppm, θ=-0.0017 / ℃, ζ=0.00023 / kPa.

[0038] The device status word includes:

[0039] BIT0 is the status flag for the semiconductor module, BIT1 is the status flag for the thermal conductivity module, and BIT2-BIT7 are reserved as environmental parameter exception codes.

[0040] One method for hydrogen detection and control, including temperature and humidity compensation, comprises the following steps:

[0041] Step 1: System initialization, dual-mode sensing module power-on self-test, environmental parameter compensation module starts loading reference parameters;

[0042] Step 2: Perform initial semiconductor mode detection to obtain the raw signal V. s And simultaneously collect T j RH k ;

[0043] Step 3: The compensation calculation unit calculates according to the formula. Calculate the corrected concentration value;

[0044] Step 4: Threshold comparator determines C corr-s With C th The relationship is such that when the set threshold is exceeded, a mode switch is triggered;

[0045] Step 5: After activating the thermal conductivity mode, press... The formula is used to compensate for thermal conductivity.

[0046] Step Six: The data fusion unit selects the output final concentration value C according to the working mode. final ;

[0047] Step 7: The output module synchronously updates the display interface and uploads data packets through the communication interface;

[0048] Step 8: The system monitors the status of each module in real time, and executes the fail-safe protocol when a sensor malfunction is detected.

[0049] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0050] 1. This invention provides a semiconductor-thermal conductivity dual-mode collaborative detection architecture. Through dynamic partitioned range coverage (0-3000ppm semiconductor mode / 3000-100000ppm thermal conductivity mode) and a sensitivity complementarity mechanism, it achieves a detection accuracy of 1-5ppm across the entire concentration range, achieving an adaptive control effect. Compared with traditional single-mode sensors, the range is extended by 300 times and the critical point detection error is reduced by 82%.

[0051] 2. In this invention, a third-order linear temperature compensation function and a three-dimensional humidity interference matrix are constructed. The temperature drift from -20℃ to 60℃ and the humidity coupling effect from 30-90%RH are eliminated by an orthogonal decomposition algorithm, which reduces the detection deviation in a wide temperature range environment from ±3%FS to ±0.5%FS and improves the reliability of data under high humidity conditions (>80%RH).

[0052] 3. In this invention, a dynamic switching strategy for the hysteresis interval (2950-3050ppm) is designed in combination with a three-cycle continuous judgment mechanism to solve the problem of erroneous switching at the critical concentration point of the traditional fixed threshold, reduce the mode jump rate from 12% to 0.02%, and achieve a smooth transition of 0.5 seconds through an exponential decay fusion algorithm to eliminate data step. Attached Figure Description

[0053] Figure 1 This is a simplified flowchart illustrating Embodiment 1 of the present invention;

[0054] Figure 2 This is a simplified schematic diagram illustrating the steps of Embodiment 2 of the present invention. Detailed Implementation

[0055] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] Example 1, as Figure 1 As shown, a semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation includes a dual-mode sensing module, an environmental parameter compensation module, a signal fusion processing module, a mode switching control module, and a data output terminal.

[0057] The dual-mode sensing module is used for dual-mode detection of hydrogen concentration in the target gas. It includes a semiconductor sensing unit and a thermal conductivity sensing unit. The semiconductor sensing unit operates in a concentration range of 0-3100 ppm with a detection sensitivity calibrated to 0.01 ppm. The thermal conductivity sensing unit operates in a concentration range of 3100-100000 ppm with a detection sensitivity calibrated to 0.05 ppm. The specific calibration process for the detection parameters is as follows:

[0058] The linearity of the semiconductor sensing unit was calibrated using a calibration gas within the range of 0-3100 ppm. Output signal values ​​at each concentration point were obtained, and a semiconductor sensing characteristic curve was established based on the formula:

[0059] ΔV s = α×Heat + β×T + γ×RH

[0060] Where, ΔV s It is set as the change in output voltage of the semiconductor sensor;

[0061] Heat voltage signal thermal conductivity value, T is ambient temperature, RH is relative humidity, α, β, γ represent the concentration sensitivity coefficient, temperature drift coefficient and humidity interference coefficient of semiconductor element, respectively;

[0062] The thermal conductivity sensing unit was dynamically calibrated using a standard thermal conductivity comparison device, and a thermal conductivity-concentration conversion model was established.

[0063]

[0064] Where C0 is the reference concentration of 3100ppm, T0 is the standard temperature of 25℃, K is the thermal conductivity gain coefficient of 0.085mV / V / ppm, and θ is the temperature compensation factor of -0.0017 / ℃;

[0065] The environmental parameter compensation module is used to acquire the temperature and humidity parameters of the detection environment in real time. The specific compensation algorithm is implemented as follows:

[0066] Ambient temperature (T) and relative humidity (RH) are collected using a digital temperature and humidity sensor, and the collected values ​​are calibrated as T. j and RH k Where j and k represent sampling time sequence numbers, j, k = 1, 2, 3…n;

[0067] Based on the temperature drift characteristics of semiconductor devices, a temperature compensation function is established:

[0068] F (T) =∑[δ t ×(T j -T ref ) m ]

[0069] Where T ref =25℃ is the reference temperature, δ t This is the m-th order temperature compensation coefficient;

[0070] Based on the effect of humidity on semiconductor adsorption characteristics, a humidity correction matrix is ​​constructed:

[0071]

[0072] Where μ1, μ2, and μ3 are humidity interference compensation parameters;

[0073] The signal fusion processing module is used to perform comprehensive processing of dual-mode signals. The specific processing flow includes:

[0074] When the detection concentration C d When the value is ≤3100ppm, the semiconductor signal processing channel is activated to perform compensation calculations on the original signal.

[0075]

[0076] Where ΔT=T j -T ref ΔRH=RH k -RH ref ;

[0077] Among them, RH ref =0.5RH;

[0078] When the detection concentration C d When the value is >3100ppm, switch to the thermal conductivity signal processing channel and perform thermal conductivity compensation calculation:

[0079]

[0080] Where P is the pressure correction term and ζ is the pressure influence factor 0.00023 / kPa;

[0081] Set the mode switching threshold C th =3100ppm±50ppm hysteresis range to prevent the detection value from oscillating frequently at the critical point;

[0082] The mode switching control module is used to realize intelligent switching between dual-mode detection. Its control logic implementation is as follows:

[0083] During the initialization phase, the semiconductor sensing unit is activated by default, while the thermal conductivity sensing unit is in standby mode.

[0084] Real-time comparison of the concentration value C after processing corr With threshold C th When three consecutive sampling periods satisfy C corr ≥C th When +Δh (Δh = 50ppm is the anti-shake margin), the mode switching command is triggered, the power supply to the semiconductor sensing unit is turned off, and the thermal conductivity sensing unit is started.

[0085] In thermal conductivity operating mode, if C is detected corr ≤C th When -ΔL (ΔL=50ppm), a reverse switch is performed, and the semiconductor sensing unit is restarted;

[0086] The data output terminal integrates a display unit and a communication interface, and the display interface presents the compensated concentration value in real time.

[0087]

[0088] in, These are mode weighting coefficients, used when the semiconductor mode is activated. When thermal conduction mode is activated

[0089] Among them, semiconductor correction concentration:

[0090]

[0091] The thermal conductivity sensor unit compensation calculation is as follows:

[0092]

[0093] Among them, G raw The original output of the thermal conductivity bridge is P0, which is the standard atmospheric pressure, or 101.325 kPa.

[0094] The communication protocol supports dual-mode output of MODBUS-RTU and USART. The data frame contains concentration value, temperature and humidity parameters and device status word. The device status word includes: BIT0 is the status flag of the semiconductor module, BIT1 is the status flag of the thermal conductivity module, and BIT2-BIT7 are reserved as environmental parameter abnormal codes.

[0095] It should be noted that the temperature drift coefficient β was determined by an experiment in a temperature chamber ranging from -20℃ to 60℃, with the output drift of the semiconductor device recorded at 5℃ intervals, and obtained by least squares fitting; the humidity interference coefficient γ was calibrated in 10% increments within the range of 30%-90%RH.

[0096] Specifically, α = 0.18 mV / ppm (at 25°C); β = -0.0023 / °C; γ = 0.00015 / %RH;

[0097] Example 2, as Figure 2 As shown, the system workflow includes the following steps:

[0098] Step 1: System initialization, dual-mode sensing module power-on self-test, environmental parameter compensation module starts loading reference parameters;

[0099] Step 2: Perform initial semiconductor mode detection to obtain the raw signal V. s And simultaneously collect T j RH k ;

[0100] Step 3: The compensation calculation unit calculates according to the formula. Calculate the corrected concentration value;

[0101] Step 4: Threshold comparator determines C corr-s With C th The relationship is such that when the set threshold is exceeded, a mode switch is triggered;

[0102] Step 5: When C orr-s When the value continuously exceeds 3050 ppm, switch to thermal conductivity mode and calculate. The formula is used to compensate for thermal conductivity.

[0103] Step Six: The data fusion unit fuses the dual-mode data using an exponential decay algorithm according to the operating mode, outputs the final concentration value, and selects the final concentration value C to output. final ;

[0104] Step 7: The output module synchronously updates the display interface and uploads data packets through the communication interface;

[0105] Step 8: The system monitors the status of each module in real time, and executes the fail-safe protocol when a sensor malfunction is detected.

[0106] Experimental example

[0107] The hydrogen detection data used in the experiment were collected from the National Hydrogen Energy Safety Laboratory from October 2023 to March 2024. The equipment used included:

[0108] Hydrogen concentration calibrator (GASERAP4100, accuracy ±0.1%)

[0109] Wide-range temperature and humidity logger (Testo 635, temperature range -40℃ 100% RH)

[0110] Dynamic pressure sensor (Honeywell TSC series, measuring range 0–200 kPa)

[0111] The experimental dataset consists of the following three types of operating conditions:

[0112] Low-concentration leakage data (0-3000ppm, semiconductor mode): 1500 sets

[0113] High concentration release data (3000-100000ppm, thermal conductivity mode): 1200 sets

[0114] Environmental disturbance data (including sudden changes in temperature and humidity, and pressure fluctuations): 800 sets

[0115] A total of 3,500 sets of data were collected. Table 1 shows the composition of the dataset. After collection, the data were normalized (0-5V voltage range) and environmental parameters were labeled (temperature, humidity, air pressure). The data were stored in CSV and JSON formats and divided into training set (2,450 sets), validation set (350 sets), and test set (700 sets) in a ratio of 7:1:2.

[0116] Table 1. Composition of the hydrogen detection dataset

[0117] Serial number Data Category quantity 0 Low concentration leakage data 1500 1 High concentration release data 1200 2 Environmental interference data 800

[0118] The study employed a fourth-order nonlinear temperature compensation function and a three-dimensional humidity interference matrix, and conducted comparative experiments in a CNAS-certified laboratory (temperature 25±0.5℃, humidity 50±5%RH). Table 2 shows the comparison of detection accuracy, Table 3 shows the comparison of environmental adaptability, and Table 4 shows the stability of mode switching.

[0119] Table 2 Comparison of Detection Accuracy (Unit: ppm)

[0120] Concentration range Error in this solution Traditional semiconductor sensors Conventional thermal conductivity meter 0-3000 (Semiconductors) ±0.8 ±5.2 - 3000-100000 (thermal conductivity) ±15 - ±50

[0121] Table 3 Comparison of Environmental Adaptability

[0122]

[0123] Table 4 Stability of Mode Switching

[0124] index This program Traditional fixed threshold scheme Critical point mis-switching rate 0.02％ 12％ Standard deviation of transitional data 0.8ppm 15ppm

[0125] The performance comparison of the models was obtained by deploying and testing on edge devices, as shown in Table 5.

[0126] Table 5 Model Performance Comparison

[0127] Model Recognition rate (%) Weight (MB) FPS (frames per second) Traditional single-mode detection model 92.3 108 1.35 This solution features a dual-mode model. 99.1 38.5 3.07

[0128] Experimental data show that:

[0129] Extended measurement range: Achieves full coverage from 0-100,000 ppm, a 300-fold increase in measurement range compared to traditional semiconductor sensors;

[0130] Improved accuracy: Detection error at the critical point (3000ppm) is reduced by 82% (from ±5.2ppm to ±0.8ppm);

[0131] Environmental robustness: Temperature drift error from -20℃ to 60℃ is reduced to ±0.5%FS, and the effect of humidity is reduced by 96%;

[0132] Energy efficiency optimization: power consumption 148mW ​​(traditional solution 520mW), standby power consumption 4.7mW.

[0133] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation, characterized in that: include: A dual-mode sensing module, including a semiconductor sensing unit and a thermal conductivity sensing unit; Obtain the output signal value of each concentration point and establish the semiconductor sensor characteristic curve according to the formula: ΔV s =α×Heat+β×T+γ×RH Where, Heat is the thermal conductivity value of the voltage signal, T is the ambient temperature, and RH is the relative humidity; Get the output voltage change of the semiconductor sensor ΔV s , the dynamic range of the thermal conductivity sensing unit is calibrated by a standard thermal conductivity comparison device, and a thermal conductivity-concentration conversion model is established: Where C0 is the reference concentration, T0 is the standard temperature, and K is the thermal conductivity gain coefficient; Environmental parameter compensation module, used to obtain the temperature and humidity parameters of the detection environment in real time and perform compensation operations, including a third-order polynomial temperature drift compensation function and a 3×3 humidity interference compensation matrix; The signal fusion processing module is used to perform comprehensive processing of dual-mode signals, integrates MODBUS-RTU and USART dual-protocol communication interfaces, and outputs data frames containing compensation concentration values, temperature and humidity parameters, and device status words; The mode switching control module is used to realize the dynamic switching of dual-mode detection, set the hysteresis range of 2950-3050ppm and adopt the exponential decay weight fusion algorithm.

2. A semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation according to claim 1, characterized in that: The environmental parameter compensation module performs: The ambient temperature T is collected by a digital temperature and humidity sensor j and relative humidity RH k , where j, k represent the sampling sequence numbers, j, k = 1, 2, 3…n; According to the temperature drift characteristics of semiconductor components, a temperature compensation function is established: F (T) =∑[δ t ×(T j -T ref ) m ] Where T ref =25℃ is the reference temperature, δ t is the mth-order temperature compensation coefficient; According to the influence of humidity on the adsorption characteristics of semiconductors, a humidity correction matrix is ​​constructed: Among them, μ1, μ2, and μ3 are humidity interference compensation parameters; Eliminate humidity interference in the range of 30%-90%RH.

3. A semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation according to claim 2, characterized in that: The mode switching control module implements: When the detection concentration C corr-s When the value is ≥3050ppm for three consecutive sampling cycles, the thermal conductivity mode activation command is triggered; during the mode transition stage, press: Data fusion is performed, where is the mode weighting coefficient, when the semiconductor mode is activated When thermal conductivity mode is activated 4. A semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation according to claim 3, characterized in that: The semiconductor sensor unit compensation operation is: Semiconductor Corrected Concentration: Where ΔV_s is the change in semiconductor output voltage, α=0.18mV / ppm, β=-0.0023 / ℃, γ=0.00015 / %RH, Where ΔT = T j -T ref , ΔRH=RH k -RH ref ; Among them, RH ref =0.5RH.

5. A semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation according to claim 4, characterized in that: The thermal conductivity sensor unit compensation operation is: Among them, G0=1.023mV / V, K=0.085mV / V / ppm, θ=-0.0017 / ℃, ζ=0.00023 / kPa.

6. A semiconductor thermal conductivity dual-mode hydrogen detection and control system with temperature and humidity compensation according to claim 5, characterized in that: The device status word includes: BIT0 is the semiconductor module status flag, BIT1 is the thermal conductivity module status flag, and BIT2-BIT7 are reserved for environmental parameter abnormality codes.

7. A hydrogen detection control method with temperature and humidity compensation, comprising a detection control system as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: Step 1: System initialization, dual-mode sensor module power-on self-test, environmental parameter compensation module starts benchmark parameter loading; Step 2: Perform semiconductor mode initial inspection to obtain the original signal V s And synchronously collect T j RH k ; Step 3: Compensation calculation unit according to the formula Calculate the corrected concentration value; Step 4: Threshold comparator determines C corr-s With C th When the set threshold is exceeded, the mode switching is triggered; Step 5: After the thermal mode is activated, press G corr =G raw +θ×(T j -T0)+ζ×P formula to compensate thermal conductivity; Step 6: The data fusion unit selects the output of the final concentration value C according to the working mode final ; Step 7: The output module synchronously updates the display interface and uploads the data packet through the communication interface; Step 8: The system monitors the status of each module in real time and executes the fail-safe protocol when a sensor abnormality is detected.

Images