Semiconductor gas sensor calibration compensation method

By collecting and fitting the AD values ​​of the sensor in different environments, generating unique compensation values, the problems of individual differences and environmental impact of semiconductor gas sensors are solved, the detection accuracy and stability are improved, and the sensor is high in different environments Performance and reliability.

CN119086659APending Publication Date: 2024-12-06CHENGDU XINHAOSI ELECTRONICS DETECTING TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411422776.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-12
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In actual applications, semiconductor gas sensors are not linear between the output voltage value and the gas concentration to be measured, and their working performance is easily affected by the environment, resulting in inaccurate detection results. The existing calibration methods cannot fully cover various scenarios and cannot be fundamentally solved. Problems of individual sensor differences and environmental changes.

Method used

By collecting the zero point and alarm action value AD values ​​of the sensor under different environmental conditions, using a linear relationship fitting algorithm, a unique compensation value is generated for each sensor, eliminating the impact of individual differences and environmental changes, and improving detection accuracy and stability .

Benefits of technology

Accurate compensation for each sensor is achieved, which reduces the interference of environmental changes on the sensor output response, improves the accuracy and stability of the detection results, and ensures the high performance and reliability of the sensor in different environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119086659A_ABST
    Figure CN119086659A_ABST
Patent Text Reader

Abstract

The invention discloses a semiconductor gas sensor calibration compensation method, and relates to the technical field of sensors, and the compensation method comprises the following steps: S100, sampling in a standard environment: in the standard environment, sampling an AD value of a sensor at a zero point and an AD value of an alarm point as reference values, according to the method, the zero reference AD value x0 and the alarm action value reference AD value y0 are uniquely set for each semiconductor gas sensor, so that the individual difference of the sensors is accurately considered, and the highly personalized compensation strategy effectively avoids the accumulation of measurement errors caused by neglecting the small difference between the sensors in the traditional method, so that the measurement accuracy is improved. Through a compensation scheme customized for each sensor, not only is the initial accuracy of gas detection improved, but also the measurement precision of the sensors can be continuously kept at a high level along with the prolonging of the service time, so that the performance of the whole gas detection system is remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of sensor technology, and in particular to a calibration and compensation method for a semiconductor gas sensor. Background Art

[0002] With the advancement of science and technology and the development of industry, semiconductor gas sensors, as an important detection element, are playing an increasingly important role in environmental monitoring, industrial safety, and home security. Semiconductor gas sensors use semiconductor gas sensors as sensitive elements to detect changes in conductivity caused by the adsorption of gas molecules on the surface of semiconductor electrodes to identify and quantify gas concentrations. This type of sensor is widely used in the detection of methane, liquefied gas, and hydrogen combustible gases due to its low cost, fast response, and high sensitivity.

[0003] However, despite the many advantages of semiconductor gas sensors, they still face many challenges in practical applications. Traditionally, there is no strict linear relationship between the output voltage value of semiconductor gas sensors and the concentration of the gas to be measured, which increases the difficulty of sensor calibration. In addition, the working performance of the sensor is easily affected by the environment, resulting in fluctuations in the output response, which in turn affects the accuracy of the detection results. Most existing sensor calibration methods are based on calibration under a preset environment. Although this method can improve the accuracy of the sensor to a certain extent, it is labor-intensive and difficult to fully cover various practical application scenarios. At the same time, sensor manufacturers usually classify the sensors according to the voltage differences under standard environments to screen qualified sensors, but this method cannot fundamentally solve the problems caused by individual differences in sensors and environmental influences.

[0004] Therefore, a calibration and compensation method for semiconductor gas sensors was developed. Through a compensation mechanism based on zero-point response, individual compensation of each semiconductor gas sensor based on its own zero-point response was achieved. This not only eliminated the influence of individual differences of sensors on the detection results, but also significantly reduced the interference of changes in ambient temperature and humidity on the sensor output response, thereby improving the accuracy and reliability of semiconductor gas sensors in practical applications. Summary of the Invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a calibration and compensation method for semiconductor gas sensors. It can accurately collect and analyze the zero point and alarm action value AD value of the sensor under different environmental conditions, and use a linear relationship fitting algorithm to generate a unique compensation value for each sensor, effectively eliminating the influence of individual differences of sensors and changes in ambient temperature and humidity on the detection results, and improving the accuracy and stability of sensor detection.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a calibration and compensation method for a semiconductor gas sensor, the compensation method comprising the following steps:

[0007] S100, sampling under standard environment: Under standard environment, the AD value of the sensor at zero point and the AD value of the alarm point are sampled as the reference value;

[0008] S200, sampling in a non-standard environment: repeat the above operation in a non-standard environment to obtain the zero point AD value and the alarm point AD value as calibration values;

[0009] S300, calculate the linear correspondence: based on the four values ​​x0, x1, y0, and y1, calculate the linear correspondence between the zero-point AD value change and the alarm point AD value change, and obtain the constant slope a and intercept b;

[0010] S400, actual application environment data collection: In the actual application environment, when there is no gas concentration, the average AD value after the sensor is powered on is collected as the actual zero-point AD value, and when there is gas concentration, the average AD value is collected at a certain time interval as the AD value corresponding to the actual concentration value;

[0011] S500, fitting and alarm determination: Substitute x0, x2, y2 and linear relationship constants a and b into the fitting formula, calculate the AD value y3 corresponding to the fitting concentration value, compare y3 with the alarm action value reference value y0, and make an alarm determination.

[0012] Furthermore, the S100 ages the sensor for 48 hours under a standard environment with the gas concentration around the sensor being 0. Within the last 24 hours after aging, the sensor is collected every second for 60 minutes, and the collected AD values ​​are averaged using the windowed averaging method to obtain a zero-point reference AD ​​value x0. A preset alarm gas concentration is introduced, and after the gas concentration stabilizes, the sensor's AD value is collected every second for n minutes. The collected AD values ​​are averaged using the windowed averaging method to obtain an alarm action value reference AD ​​value y0.

[0013] Furthermore, the standard environment in S100 is an atmospheric condition with a temperature of 20° C., a relative humidity of 50%, and an atmospheric pressure of 96 kPa to 106 kPa.

[0014] Furthermore, in the step of using the windowed averaging method to average the collected AD values ​​in S100, when the AD value is recorded 60 times per second, the 60 AD values ​​are averaged, and the averaged AD value is used as the representative value of the minute to generate 1-minute granular data. After the 1-minute granular data is generated, the data of the second-level granular dimension is cleared, and the counting starts again in seconds, and the AD values ​​of the next minute are continued to be collected, and the data collection and averaging are continued according to the above steps to generate the average of hours and days.

[0015] Furthermore, the S200 maintains the gas concentration at 0 in a non-standard environment, continuously collects AD values ​​within n minutes at a frequency of collecting AD values ​​once per second, and processes the collected AD values ​​using the window averaging method to obtain the zero-point calibration AD value x1 in the non-standard environment. After completing the acquisition of the zero-point calibration AD value x1, without changing the conditions of the non-standard environment, a certain amount of gas to be tested is injected to make the gas concentration in the environment reach the alarm setting value. At the same sampling frequency as when obtaining x1, the AD values ​​within n minutes are continuously collected, and the collected AD values ​​are also processed using the window averaging method to obtain the alarm action value calibration AD value y1 in the non-standard environment when the gas concentration is the alarm setting value.

[0016] Furthermore, the S300 calculates the zero point AD difference based on the zero point reference AD ​​value x0, the alarm action value reference AD ​​value y0, the alarm action value reference AD ​​value x1, and the alarm action value calibration AD value y1 collected by sampling under the standard environment and sampling under the non-standard environment. The formula is: △x = x1-x0, and the alarm action value difference is calculated. The formula is: △y = y1-y0. According to the two obtained changes △x and △y, the slope a and intercept b of the linear correspondence are calculated. The slope a calculation formula is: The calculation formula for the intercept b is: b=y0-a·x0.

[0017] Furthermore, in S400, the semiconductor gas sensor is placed in an actual application environment, and the data acquisition frequency is set to once per second. In an environment with no gas concentration, the sensor is started and timed to collect AD values ​​for 5-20 minutes. During this time period, the AD value per second is continuously recorded and processed using a windowed averaging method to obtain the actual zero-point AD value x2. While keeping the sensor operating state unchanged, the target gas is injected to place the sensor in an environment with gas concentration. Starting from 20 minutes later, AD values ​​are collected in real time at 5-second intervals, and the AD values ​​in each interval are windowed and averaged to obtain the corresponding AD value y2.

[0018] Further, the S500 substitutes x0, x2, y2, and the slope a and intercept b of the linear relationship into the fitting formula. The fitting formula fits the AD value y3 of the current concentration value under the standard environment according to the current actual measured values (x2 and y2) and reference values (x0, a, b). The fitting formula is: y3 = y2 - a(x2 - x0) - b. Compare the calculated y3 with the alarm action value reference y0 to determine whether to alarm.

[0019] Further, the alarm determination in the S500: If y3 ≥ y0, it is determined as an alarm, indicating that the gas concentration in the current environment has reached the preset alarm concentration; if y3 < y0, there is no alarm, indicating that the gas concentration in the current environment is within the safe range.

[0020] Further, the sampling frequency of the semiconductor gas sensor is set to 1 Hz, and the sampling AD accuracy is set to 12 bits.

[0021] Compared with the prior art, the calibration compensation method of this semiconductor gas sensor has the following beneficial effects:

[0022] First, by uniquely setting the zero-point reference AD value x0 and the alarm action value reference AD value y0 for each semiconductor gas sensor, the present invention realizes the precise consideration of the individual differences of the sensors. This highly personalized compensation strategy effectively avoids the accumulation of measurement errors caused by ignoring the small differences between sensors in the traditional method. Through the compensation scheme tailored for each sensor, not only the initial accuracy of gas detection is improved, but also it is ensured that as the usage time extends, the measurement accuracy of the sensor can continuously remain at a high level, thereby significantly improving the performance of the overall gas detection system.

[0023] Second, by constructing a linear relationship model between the zero-point change and the alarm action value change, the present invention demonstrates strong environmental adaptability. The reference values set under the standard environment, combined with the calibration data under the non-standard environment, enable the sensor to dynamically sense and adapt to the changes in environmental factors. This intelligent adjustment mechanism not only reduces the direct impact of environmental fluctuations on the sensor performance, but also significantly improves the stability and reliability of the sensor in different environments. Whether in extreme temperature and humidity conditions or in an environment with rapid pressure changes, the sensor can maintain accurate measurement and a stable alarm threshold, further reducing the risk of false alarms and missed alarms caused by environmental changes.

[0024] Other advantages, objectives, and features of the present invention will be described in part in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0026] Figure 1 A flow chart of a semiconductor gas sensor calibration compensation method;

[0027] Figure 2 Schematic diagram of AD value under the same concentration in different atmospheric environments;

[0028] Figure 3 Schematic diagram of the concentration fitting curve. DETAILED DESCRIPTION

[0029] The following is a clear and complete description of the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0030] Example 1:

[0031] A semiconductor gas sensor calibration compensation method includes the following steps: sampling under a standard environment, sampling under a non-standard environment, calculating a linear correspondence, collecting data in an actual application environment, fitting, and alarm determination.

[0032] First, enter the sampling stage under the standard environment (S100). Under the standard environment of temperature 20°C, relative humidity 50%, and atmospheric pressure 96kPa-106kPa, when the gas concentration around the sensor is 0, the sensor is aged for 48 hours. In the last 24 hours after aging, the AD value of the sensor is collected every second for 60 minutes, and the collected AD value is averaged using the windowed averaging method: after the AD value is recorded 60 times per second, these 60 AD values ​​are averaged, and the averaged AD value is used as the representative value of the minute to generate 1-minute particle data. After generating the 1-minute particle data, the data of the second-level particle dimension is cleared, and the counting is restarted by seconds. The AD value of the next minute is continued to be collected and averaged according to the above steps to generate the average of hours and days to obtain the zero-point reference AD ​​value x0. The preset alarm gas concentration is introduced. After the gas concentration stabilizes, the AD value of the sensor is collected every second for n minutes. The collected AD values ​​are averaged using the windowed averaging method to obtain the alarm action value reference AD ​​value y0.

[0033] Next, enter the sampling phase under non-standard conditions (S200). Under non-standard conditions, keep the gas concentration at 0, and collect the AD values within n minutes continuously at a frequency of collecting the AD value once per second. Use the window averaging method to process the collected AD values to obtain the zero-point calibration AD value x1 under this non-standard condition. After obtaining the zero-point calibration AD value x1, without changing the conditions of the non-standard environment, inject a certain amount of the gas to be measured to make the gas concentration in the environment reach the alarm setting value, and collect the AD values within n minutes continuously at the same sampling frequency as when obtaining x1. Also use the window averaging method to process the collected AD values to obtain the alarm action value calibration AD value y1 when the gas concentration is at the alarm setting value under this non-standard condition.

[0034] Subsequently, enter the stage of calculating the linear correspondence relationship (S300). Based on the zero-point reference AD value x0, alarm action value reference AD value y0, alarm action value reference AD value x1, and alarm action value calibration AD value y1 collected during sampling under standard conditions and non-standard conditions, calculate the zero-point AD difference: Δx = x1 - x0, calculate the alarm action value difference: Δy = y1 - y0. According to the two obtained change amounts Δx and Δy, calculate the slope a and intercept b of the linear correspondence relationship: The calculation formula for the slope a is: The calculation formula for the intercept b is: b = y0 - a·x0.

[0035] Then, enter the stage of collecting data in the actual application environment (S400). Place the semiconductor gas sensor in the actual application environment, set the data collection frequency to collect once per second. In an environment without gas concentration, start the sensor and time it, collect the AD values for 5 - 20 minutes. During this period, continuously record the AD values per second and use the window averaging method to process them to obtain the actual zero-point AD value x2. Without changing the operating state of the sensor, inject the target gas to make the sensor in an environment with gas concentration, and start collecting the AD values in real time at intervals of 5 seconds from 20 minutes later. Perform window averaging processing on the AD values within each interval to obtain the corresponding AD value y2.

[0036] Finally, enter the fitting and alarm determination stage (S500). Substitute x0, x2, y2, and the slope a and intercept b of the linear relationship into the fitting formula: y3 = y2 - a(x2 - x0) - b, fit the AD value y3 of the current concentration value under the standard environment, compare the calculated y3 with the alarm action value reference value y0 to determine whether to alarm. When y3 ≥ y0, it is determined to alarm, indicating that the gas concentration in the current environment has reached the preset alarm concentration. When y3 < y0, it does not alarm, indicating that the gas concentration in the current environment is within the safe range.

[0037] Embodiment 2:

[0038] Specific steps of a calibration compensation method for a semiconductor gas sensor:

[0039] S100. In a standard environment, when the gas concentration around the sensor is 0, let the sensor age for 48 hours. In the last 24 hours of aging, collect the AD value once every second for 60 minutes, and use the windowed averaging method to process the collected AD values to obtain the zero-point reference AD value x0. Introduce a preset alarm gas concentration. After the gas concentration stabilizes, collect the AD value once every second for n minutes, and also use the windowed averaging method to process the AD values to obtain the alarm action value reference AD value y0.

[0040] S200. Sampling in a non-standard environment. In a non-standard environment such as different temperatures and humidities, keep the gas concentration at 0, collect the AD value once every second for n minutes, and use the windowed averaging method to process it to obtain the zero-point calibration AD value x1 in this environment. Without changing the environmental conditions, inject a certain amount of the待测 gas to the alarm set concentration, collect the AD value at the same frequency, and process it to obtain the alarm action value calibration AD value y1.

[0041] S300. Calculate the linear correspondence relationship. Calculate the zero-point AD difference △x = x1 - x0 and the alarm action value difference △y = y1 - y0, and calculate the slope a and intercept b of the linear correspondence relationship according to △x and △y.

[0042] S400. Collect actual application environment data. Place the sensor in the actual application environment, start and collect the AD values for 5 - 20 minutes when there is no gas concentration to obtain the actual zero-point AD value x2. Inject the target gas, and start collecting the AD value once every 5 seconds from 20 minutes later, and process it to obtain y2.

[0043] S500. Fitting and alarm determination. Use the fitting formula y3 = y2 - a(x2 - x0) - b, substitute x0, x2, y2, slope a, and intercept b into it, calculate to obtain y3, and compare y3 with the alarm action value reference y0: If y3 ≥ y0, it is determined as an alarm, indicating that the gas concentration reaches the preset alarm concentration; if y3 < y0, there is no alarm, indicating that the gas concentration is within the safe range.

[0044] Example 3:

[0045] Specific process steps of a calibration compensation method for a semiconductor gas sensor;

[0046] (1) Through experiments, obtain x0, x1, y0, y1;

[0047] (2) Calculate the zero-point difference △x = x1 - x0;

[0048] (3) Calculate the alarm action value difference △y = y1 - y0;

[0049] (4) Calculate the slope and intercept b = y0-a·x0;

[0050] (5) Place the sensor in an environment with no gas concentration and time the sensor power-on time;

[0051] (6) Collect the AD value 5-20 minutes after power-on and calculate the actual zero point value x2;

[0052] (7) 20 minutes after power-on, the AD values ​​collected in real time are averaged over 5 seconds to calculate the actual concentration value corresponding to the AD value y2;

[0053] (8) Substitute x0, x2, y2, y3, a, and b into the fitting curve formula of the relationship between the alarm point and the zero point: y3 = y2-a(x2-x0)-b, and calculate the AD value y3 corresponding to the fitting concentration value;

[0054] (9) Compare y3 and y0. When y3 ≥ y0, an alarm is issued.

[0055] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A calibration compensation method for semiconductor gas sensors. It is characterized in that The compensation method includes the following steps: S100, sampling under standard environment: under standard environment, the AD value of the sensor at zero point and the AD value of the alarm point are sampled as the reference value; S200, sampling in a non-standard environment: repeat the above operation in a non-standard environment to obtain the zero point AD value and the alarm point AD value as calibration values; S300, calculating the linear correspondence: according to the four values ​​of x0, x1, y0, and y1, calculating the linear correspondence between the change in the zero-point AD value and the change in the alarm-point AD value, and obtaining the constant slope a and intercept b; S400, actual application environment data collection: in the actual application environment, when there is no gas concentration, the average value of the AD value after the sensor is powered on is collected as the actual zero-point AD value, and when there is gas concentration, the average value of the AD value is collected at a certain time interval as the AD value corresponding to the actual concentration value; S500, fitting and alarm determination: Substitute x0, x2, y2 and linear relationship constants a and b into the fitting formula, calculate the AD value y3 corresponding to the fitting concentration value, compare y3 with the alarm action value reference value y0, and make an alarm determination.

2. A semiconductor gas sensor calibration compensation method according to claim 1, It is characterized in that The S100 is aged for 48 hours under a standard environment with the gas concentration around the sensor being 0. Within the last 24 hours after aging, data are collected every one second for 60 minutes, and the collected AD values ​​are averaged using the windowed averaging method to obtain a zero-point reference AD ​​value x0. A preset alarm gas concentration is introduced, and after the gas concentration stabilizes, the sensor AD values ​​are collected every one second for n minutes, and the collected AD values ​​are averaged using the windowed averaging method to obtain an alarm action value reference AD ​​value y0.

3. A semiconductor gas sensor calibration compensation method according to claim 2, It is characterized in that The standard environment in S100 is an atmospheric condition with a temperature of 20° C., a relative humidity of 50%, and an atmospheric pressure of 96 kPa to 106 kPa.

4. A semiconductor gas sensor calibration compensation method according to claim 2, It is characterized in that In the step S100, the windowed averaging method is used to average the collected AD values. When the AD value is recorded 60 times per second, the 60 AD values ​​are averaged, and the averaged AD value is used as the representative value of the minute to generate 1-minute granular data. After the 1-minute granular data is generated, the data of the second-level granular dimension is cleared, and the counting starts again by seconds, and the AD value of the next minute is continuously collected. The data collection and averaging are continued according to the above steps to generate the average of hours and days.

5. A semiconductor gas sensor calibration compensation method according to claim 1, It is characterized in that In a non-standard environment, the S200 keeps the gas concentration at 0, and continuously collects the AD values within n minutes at a frequency of collecting the AD value once per second. The collected AD values are processed using the window averaging method to obtain the zero-point calibration AD value x1 in this non-standard environment. After obtaining the zero-point calibration AD value x1, without changing the conditions of the non-standard environment, a certain amount of the gas to be measured is injected to make the gas concentration in the environment reach the alarm set value. Then, at the same sampling frequency as when obtaining x1, the AD values within n minutes are continuously collected, and the collected AD values are also processed using the window averaging method to obtain the alarm action value calibration AD value y1 when the gas concentration is at the alarm set value in this non-standard environment.

6. The method for calibrating and compensating a semiconductor gas sensor according to claim 1, characterized in that The S300 calculates the zero point AD difference based on the zero point reference AD ​​value x0, the alarm action value reference AD ​​value y0, the alarm action value reference AD ​​value x1 and the alarm action value calibration AD value y1 collected by sampling under the standard environment and sampling under the non-standard environment. The formula is: △x=x1-x0, and the alarm action value difference is calculated. The formula is: △y=y1-y0. According to the two obtained changes △x and △y, the slope a and intercept b of the linear correspondence are calculated. The slope a calculation formula is: The calculation formula for intercept b is: b=y0-a·x0.

7. The method for calibrating and compensating a semiconductor gas sensor according to claim 1, characterized in that In the S400, the semiconductor gas sensor is placed in the actual application environment, and the data collection frequency is set to collect once per second. In an environment without gas concentration, the sensor is started and timed, and the AD values are collected for 5 - 20 minutes. During this time period, the AD value per second is continuously recorded and processed using the window averaging method to obtain the actual zero-point AD value x2. Without changing the operating state of the sensor, the target gas is injected to make the sensor in an environment with gas concentration. Starting from 20 minutes later, the AD values are collected in real time at 5-second intervals, and the AD values within each interval are processed by window averaging to obtain the corresponding AD value y2.

8. The method for calibrating and compensating a semiconductor gas sensor according to claim 1, characterized in that S500 substitutes x0, x2, y2 and the slope a and intercept b of the linear relationship into the fitting formula. The fitting formula fits the AD value y3 of the current concentration value under the standard environment according to the current actual measured value (x2 and y2) and the reference value (x0, a, b). The fitting formula is: y3 = y2-a(x2-x0)-b. 3 and the alarm action value reference value y 0 Compare and determine whether to issue an alarm.

9. The method for calibrating and compensating a semiconductor gas sensor according to claim 8, characterized in that The alarm determination in the S500: If y3 ≥ y0, it is determined as an alarm, indicating that the gas concentration in the current environment has reached the preset alarm concentration; if y3 < y0, no alarm is given, indicating that the gas concentration in the current environment is within the safe range.

10. The method for calibrating and compensating a semiconductor gas sensor according to any one of claims 1 - 9, characterized in that The sampling frequency of the semiconductor gas sensor is set to 1 Hz, and the sampling AD accuracy is set to 12 bits.