A method for qualitative identification and quantitative measurement of gases based on a broad-spectrum sensor array

Abstract

The present invention relates to the field of gas detection technology, and specifically to a method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array, comprising the following steps: S1: acquiring sensor array measurement data; S2: converting the measurement data into a gas spectrum and a gas intensity value; S3: qualitative identification, determining the category and name of the gas; S4: quantitative measurement, determining the gas concentration. The present invention can replace Fourier transform infrared spectrometers (FTIR) and mass spectrometers (MS) in most application scenarios, greatly reducing the cost of use, and can be widely used in emergency disasters, environmental monitoring, garbage disposal, landfill monitoring, food processing, livestock and poultry breeding monitoring, medical and health care and other fields.

Description

Technical Field

[0001] The present invention relates to the technical field of gas detection, and in particular to a method for qualitatively identifying and quantitatively measuring gas based on a broad-spectrum sensor array. Background Art

[0002] At present, the following methods are mainly used for gas identification:

[0003] 1. Gas sensor: Use gas sensors to detect gas composition and concentration. This method requires knowing the composition of the gas to be measured in advance and selecting sensors according to the gas composition. If the composition of the gas to be measured is not known, a large number of sensors need to be selected, which is unrealistic in actual use.

[0004] 2. Mass spectrometer (hereinafter referred to as MS): In MS, the gas sample is first ionized to produce ions of gas molecules. Then, these ions are accelerated by an accelerator and separated by a mass analyzer, and detected according to the mass-to-charge ratio of different gas molecules. Finally, the detected signal is converted into a mass spectrum. By analyzing and comparing the mass spectrum, the types and contents of various components in the gas can be determined. However, the cost of MS is relatively high (hundreds of thousands to millions), and it has high requirements for the working environment and operators.

[0005] 3. Fourier transform infrared spectrometer (hereinafter referred to as FTIR): FTIR is an instrument used to measure spectra. It obtains the spectral information of the sample by measuring the sample's absorption or reflection of light of different wavelengths. For gas identification, FTIR can be used to analyze the absorption spectrum of gas samples within a specific wavelength range to determine the presence and concentration of different components in the gas. The gas can be identified by comparing the sample spectrum with the spectral data of known gases in the standard library. However, FTIR's identification of gas may be affected by factors such as gas concentration and gas reaction, so certain calibration and verification are required in practical applications. FTIR has the same problems as MS, high cost (hundreds of thousands) and high requirements (high requirements for working environment and operators).

[0006] In the actual application of gas analysis, MS and FTIR are more inclined to be used in laboratory scenarios. In the application scenarios of handheld portable analysis at the gas leakage site or online real-time analysis, the use of MS and FTIR will lead to high costs and stringent requirements on the use environment. At the same time, there is still a lack of instruments that meet the requirements of low cost, are not picky about the use environment, and can qualitatively identify and quantitatively measure gases. Summary of the invention

[0007] In view of the shortcomings of the prior art, the present invention provides a method for qualitative identification and quantitative measurement of gases based on a broad-spectrum sensor array, with the aim of enabling rapid qualitative identification (identification of gas components and gas categories) and quantitative measurement (gas concentration) of unknown gases.

[0008] To achieve the above objectives, the present invention is implemented by the following technical solution: A method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array, comprising the following steps:

[0009] S1: Obtain sensor array measurement data;

[0010] S2: converting the measurement data into a gas spectrum and a gas intensity value;

[0011] S3: Qualitative identification, determine the type and name of the gas;

[0012] S4: Quantitative measurement to determine gas concentration.

[0013] Preferably, the sensor circuit needs to be designed before S1, and at least 8 types of MOS sensors are selected to form a MOS sensor array. The original data measured by the sensor array is obtained through the sensor acquisition circuit, and the sensor resistance Rs is calculated by Formula 1;

[0014] Formula 1

[0015] Where Rs is the sensor resistance, RL1 is the digital potentiometer, RL2 is the fixed resistor, (RL1+RL2) is the load resistance, Vc is the power supply voltage, and Vout is the voltage on the load resistor.

[0016] Preferably, S1 specifically adopts a trend judgment algorithm to identify the trend of sensor measurement values, and defines the sensor trend as: rising, falling, fluctuating and smoothing. When the sensor measurement value trend is an rising or falling trend, the average of the latest 8 groups of data in the data buffer is taken as the real-time value data; when the sensor measurement value is a fluctuating trend, the average of the latest 60 groups of data in the data buffer is taken as the real-time data; when the sensor measurement value is a smoothing trend, the average of all values ​​in the buffer is taken as the real-time data.

[0017] Preferably, S2 specifically extracts the sensor array measurement data and converts the data into concentration values ​​and gas spectra, so that the gas spectra are independent of the gas concentration. The raw data measured by the sensor array measured from S1 is as follows:

[0018] {Rs1,Rs2,Rs3,Rs4,Rs5,Rs6,Rs7,Rs8}

[0019] Among them, Rs1 ~ Rs8 represent the resistance of sensor 1 ~ sensor 8 respectively;

[0020] The calculation method of the gas spectrum is as follows: consider each sensor as a dimension, find the greatest common divisor X according to the actual measured data of the sensor, and replace Rs 1~ Rs 8 with A×X~H×X in sequence, as shown below:

[0021] {A×X , B×X , C×X , D×X , E×X , F×X , G×X , H×X}

[0022] Remove the greatest common divisor and get the gas spectrum:

[0023] Gas spectrum = {A, B, C, D, E, F, G, H};

[0024] The gas spectrum library contains various gas spectra.

[0025] Preferably, in S2, the calculation method of the gas intensity is as follows: after setting RL1 in zero-level air, the value of the sensor resistance Rs of each sensor is detected respectively, and the value of the sensor resistance Rs measured by each sensor in zero-level air is defined as Ro, corresponding to Ro1~Ro8 respectively, the gas intensity value of each sensor is Ro / Rs, and the gas intensity value Vs of the sensor array is calculated as follows:

[0026] Vs=Ro1 / Rs1+Ro2 / Rs2+Ro3 / Rs3+Ro4 / Rs4+Ro5 / Rs5+Ro6 / Rs6+Ro7 / Rs7+Ro8 / Rs8 Formula 2.

[0027] Preferably, the specific steps of S3 are: first determine whether the gas intensity is greater than the detection threshold. Only when the gas intensity is greater than the detection threshold, the gas spectrum can be relatively unique, and the gas can be qualitatively identified. According to the gas spectrum and gas spectrum library calculated by S2, the closest spectrum is found through the qualitative identification algorithm to determine the gas category, and complete the qualitative identification of the gas. The most critical part of the qualitative identification algorithm is to determine the spectrum difference. The calculation formula of the spectrum difference is as follows:

[0028] Gas Spectrum Library = {A L1 , B L1 , C L1 , D L1 , E L1 , F L1 , G L1 , H L1}

[0029] {A L2 , B L2 , C L2 , D L2 , EL2 , F L2 , G L2 , H L2}

[0030] {A L3 , B L3 , C L3 , D L3 , E L3 , F L3 , G L3 , H L3}

[0031] {A Ln , B Ln , C Ln , D Ln , E Ln , F Ln , G Ln , H Ln}

[0032] Among them, the suffix numbers L1, L2, L3…Ln respectively represent the order of the gas spectra in the spectrum library;

[0033] Current Gas Spectrum = {A, B, C, D, E, F, G, H}

[0034] Formula 3

[0035] Wherein, n in Formula 3 means that the spectrum difference between the current gas spectrum and the n spectra in the spectrum library needs to be calculated in sequence;

[0036] By calculating the spectrum difference, the spectrum with the smallest difference is found in the gas spectrum library, and the gas name and the category are obtained.

[0037] Preferably, the specific steps of S4 are: the calculation formula of the gas concentration Val is as follows:

[0038] Formula 4

[0039] Wherein, Vs represents the gas intensity, and the calculation method is shown in Formula 2, k represents the calibration slope, and b represents the calibration intercept;

[0040] After the quantitative measurement is successful, the qualitative identification and quantitative measurement of the gas are completed as a whole, and the sensor outputs the detection results.

[0041] Preferably, the values ​​of the calibration slope k and the calibration intercept b need to be obtained by calibrating the sensor in advance, and the sensor is calibrated by a multi-point calibration method.

[0042] Val零点 = k0× Vs + b0

[0043] Val 浓度1 = k1× Vs + b1

[0044] Val 浓度2 = k2× Vs + b2

[0045] …

[0046] Val 浓度n = k n × Vs + b n

[0047] The suffixes 1 to n of the calibration slope k and the calibration intercept b represent the order of the multi-point calibration.

[0048] Preferably, after the sensor is designed, the sensor array needs to be calibrated. The sensor array calibration first needs to complete the zero point calibration. The sensor array zero point calibration needs to complete two tasks:

[0049] 1) Obtain the default configuration parameters of the digital potentiometer in each sensor detection circuit;

[0050] 2) The detection value of the sensor in zero-level air is implemented as follows: First, by calculating and adjusting the digital potentiometer RL1 in the sensor detection circuit, Vout approaches 2.2v in a zero-level air environment. This step is repeated 8 times. When the ratio of the difference between the maximum and minimum values ​​of the digital potentiometer RL1 to the average of 8 times is less than 5%, the digital potentiometer RL1 is configured as the average, and the calibration of the digital potentiometer RL1 is completed. If the calculation result is ≥5%, return to the previous step and start the calculation again; secondly, after the digital potentiometer is configured, the value of the sensor resistance Rs is calculated. The calculation is also repeated 8 times to obtain the values ​​of 8 sensor resistance Rs. If the ratio of the difference between the maximum and minimum values ​​to the average of 8 times is less than 5%, the sensor zero point calibration is completed, and the average of the 8 sensor resistance Rs is written into Ro of each sensor. Ro is the value of the sensor resistance Rs in the zero-level air state.

[0051] Preferably, after completing the zero point calibration of the sensor, it is also necessary to perform concentration calibration on the sensor array, which specifically includes the following steps:

[0052] Step 1: Connect the calibration gas and record the calibration concentration value

[0053] Step 2: Enter the calibration state, read the sensor array data, and store it in the buffer

[0054] Step 3: Compare the latest 8 data to see if the difference between the maximum and minimum values ​​is less than 5% of the mean.

[0055] Step 4: When the difference between the maximum and minimum values ​​is less than 5% of the mean, the calibration is completed and the sensor array value is stored as the concentration calibration value.

[0056] The beneficial effects of the present invention are as follows: compared with the traditional MS and FTIR gas identification methods, the present method can realize rapid identification of unknown gases at a relatively low cost (currently, gas identification of more than 100 organic compounds can be realized), and at the same time, the concentration information of the gas can be identified, and the operation and maintenance are simple. The method can be widely used in emergency disasters, environmental monitoring, garbage disposal, landfill monitoring, food processing, livestock and poultry breeding monitoring and medical and health fields, and has high economic value. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a circuit diagram of the sensor array in the present invention;

[0058] Figure 2 It is a MOS sensor detection circuit diagram in the prior art;

[0059] Figure 3 This is a MOS sensor detection circuit diagram of the present invention;

[0060] Figure 4 This is a zero point calibration flow chart of the present invention;

[0061] Figure 5 It is the concentration calibration flow chart of the present invention;

[0062] Figure 6 It is a schematic diagram of the gas spectrum in the present invention;

[0063] Figure 7 A flow chart of qualitative identification and quantitative measurement of gas in the present invention;

[0064] Figure 8 A flowchart for obtaining sensor measurement data in the present invention;

[0065] Fig. 9 A flow chart of gas qualitative identification in the present invention;

[0066] Fig.10 This is a flow chart of the quantitative measurement of gas in the present invention. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0068] A method for qualitative identification and quantitative measurement of gases based on a broad-spectrum sensor array is described from four parts: sensor circuit design, sensor array calibration, gas library format standard, and implementation steps of qualitative identification and quantitative measurement. Sensor circuit design, sensor array calibration, and gas library are prerequisites for achieving qualitative identification and quantitative measurement. Sensor circuit design ensures the accuracy of sensor raw data acquisition; sensor array calibration includes zero point calibration and concentration calibration, and calibration is used to achieve the parallelism and detection accuracy of the sensor's raw detection data to meet the design standards; the gas library is the key to qualitative identification, and the gas library contains all gas spectra that the sensor can qualitatively analyze, and the sensor can perform qualitative analysis by identifying the spectra.

[0069] Part 1: Sensor Circuit Design

[0070] MOS sensors are broad-spectrum sensors that can detect multiple gases, but the response to each gas is different. Using the response of MOS sensors to different gases, the present invention selects at least 8 types of MOS sensors to form a MOS sensor array (8 types of MOS sensors are selected in this embodiment, or more than 8 types are also possible), and obtains the raw data measured by the sensor array through the acquisition circuit. The sensor array circuit block diagram is as follows Figure 1 As shown. The basis for accurate gas identification is the accuracy of sensor test data. The measurement deviation of the same type of sensor detecting the same gas determines the measurement accuracy of mass-produced products. The application of MOS sensors is based on the change of sensor internal resistance. However, due to the limitations of manufacturing process and sensor principle, the internal resistance of MOS sensors of the same type in the same environment varies by dozens of times and is discretely distributed. The generated electrical signals also differ by dozens of times. Therefore, it is difficult to achieve accurate measurement in the basic analog-to-digital detection circuit. Therefore, the first problem to be solved is to solve the measurement problem caused by the discreteness of MOS sensors.

[0071] The key to the sensor array circuit design is the sensor acquisition circuit (the principles of sensors 1 to 8 are the same). The acquisition circuit design of the present invention solves the shortcomings of the MOS universal acquisition circuit, such as poor mass production consistency and high calibration difficulty.

[0072] The basic test circuit of MOS sensor is as follows Figure 2 As shown in the figure, V C Represents the power supply voltage, V H represents the voltage applied across the sensor, R H Represents the sensor heating resistor, RL represents the load resistor, Rs represents the sensor resistance, Vout represents the voltage on the load resistor, and the sensor resistance Rs is calculated by the formula shown below.

[0073] Formula 0

[0074] In the case of different individuals of the same model sensor, the Vout value varies greatly, and the calculated Rs value also varies greatly. The present invention modifies the basic test circuit into Figure 3 As shown in the figure, V C Represents the power supply voltage, V H represents the voltage applied across the sensor, R H Represents the sensor heating resistor, RL1 represents the digital potentiometer, RL2 represents the fixed resistor, Rs represents the sensor resistance, Vout is the voltage on the load resistor, RL resistor is replaced by the digital potentiometer RL1 and the fixed resistor RL2, when the sensor is calibrated in zero-level air, the RL1 resistance is adjusted to make Vout approach the same value. At this time, the Rs value is calculated by formula 1.

[0075] Formula 1

[0076] By adding the digital potentiometer RL1, the load circuit can be adjusted dynamically to compensate for the discreteness of the MOS sensor design, improve the test consistency, solve the batch production problem of the present invention and the requirement of the gas library in Part 3 on sensor consistency, and realize the standard gas library.

[0077] Part 2: Sensor Array Calibration

[0078] The sensor array calibration first needs to complete the zero point calibration. The sensor array zero point calibration of the present invention needs to complete two tasks: obtain the default configuration parameters of the digital potentiometer in each sensor detection circuit (see Part 1 Sensor Circuit Design for details) and the detection value of the sensor in zero-level air, and implement the steps as follows: Figure 4 As shown: First, by calculating and adjusting the digital potentiometer RL1 in the sensor detection circuit, Vout approaches 2.2v in a zero-level air environment. This step is repeated 8 times. When the ratio of the difference between the maximum and minimum values ​​of RL1 and the average of 8 times is less than 5%, the digital potentiometer RL1 is configured as the average value, and the calibration of RL1 is completed. If the calculation result is ≥5%, return to the previous step and start the calculation again. Secondly, after the digital potentiometer is configured, the value of the sensor resistance Rs is calculated. The calculation is repeated 8 times to obtain the values ​​of the 8 sensor resistance Rs. If the ratio of the difference between the maximum and minimum values ​​and the average of 8 times is less than 5%, the sensor zero point calibration is completed, and the average of the 8 sensor resistance Rs is written into Ro of each sensor. Ro is the value of the sensor resistance Rs in the zero-level air state.

[0079] After completing the zero point calibration of the sensor, the concentration calibration of the sensor array is also required. Concentration calibration is the key to achieve quantitative measurement. Concentration calibration requires recording the gas name, gas concentration, and sensor array Rs value. The calibration steps are as follows: Figure 5As shown. Specifically including the following steps:

[0080] Step 1: Connect the calibration gas and record the calibration concentration value

[0081] Step 2: Enter the calibration state, read the sensor array data, and store it in the buffer

[0082] Step 3: Compare the latest 8 data to see if the difference between the maximum and minimum values ​​is less than 5% of the mean.

[0083] Step 4: When the difference between the maximum and minimum values ​​is less than 5% of the mean, the calibration is completed and the sensor array value is stored as the concentration calibration value.

[0084] Part 3: Gas library format standard

[0085] The MOS sensor array measurement data is extracted and converted into concentration values ​​and gas spectra, so that the gas spectra are independent of gas concentration. The raw data measured by the sensor array is as follows:

[0086] {Rs1,Rs2,Rs3,Rs4,Rs5,Rs6,Rs7,Rs8}

[0087] Among them, Rs1~Rs8 represent the resistance values ​​of sensor 1~sensor 8 respectively.

[0088] Gas spectrum: Consider each sensor as a dimension, find the greatest common divisor X based on the actual sensor data, and replace Rs1~Rs8 with A×X~H×X in sequence, as shown below:

[0089] {A×X , B×X , C×X , D×X , E×X , F×X , G×X , H×X}

[0090] Remove the greatest common divisor and express it as follows:

[0091] Gas spectrum = {A, B, C, D, E, F, G, H}

[0092] The gas library contains spectra of various gases.

[0093] In order to make gas identification more vivid, it can be represented by a diagram, for example Figure 6 As shown, the vertical axis value in the figure is the value of the sensor resistance Rs1~Rs8 divided by the greatest common divisor.

[0094] The gas spectrum after the above normalization process removes the influence of gas intensity on the spectrum. When the gas intensity reaches the detection threshold, the generated gas spectrum will be independent of the gas intensity. In this way, the spectrum of the same gas generated is unique and can be used for gas identification.

[0095] Part 4: Steps to implement qualitative identification and quantitative measurement

[0096] The steps for qualitative identification and quantitative measurement are as follows: Figure 7 shown.

[0097] Step 1: Obtain sensor array measurement data.

[0098] In order to quickly and accurately obtain sensor measurement data, this article adds a trend judgment algorithm to identify the trend of sensor measurement values, and defines the sensor trend as: rising, falling, fluctuating, and smoothing. When the sensor measurement value trend is an rising or falling trend, the average of the latest 8 groups of data in the data buffer is taken as the real-time value data. This design can quickly respond to data trends and improve the sensor response speed; when the sensor measurement value is a fluctuating trend, the average of the latest 60 groups of data in the data buffer is taken as the real-time data, which can effectively suppress data fluctuations and make the real-time monitoring data smooth; when the sensor measurement value is a smooth trend, the average of all values ​​in the buffer is taken as the real-time data. The process of obtaining sensor array measurement data is as follows: Figure 8 As shown:

[0099] Step 2: The detected quantity data is converted into gas spectrum and gas intensity.

[0100] Extract the MOS sensor array measurement data and convert the data into concentration values ​​and gas spectra, so that the gas spectra are independent of gas concentration. From step 1, we can see that the raw data measured by the sensor array is as follows:

[0101] {Rs1,Rs2,Rs3,Rs4,Rs5,Rs6,Rs7,Rs8}

[0102] Among them, Rs1~Rs8 represent the resistance values ​​of sensor 1~sensor 8 respectively.

[0103] Gas spectrum: Consider each sensor as a dimension, find the greatest common divisor X based on the actual sensor data, and replace Rs 1~ Rs 8 with A×X~H×X in sequence, as shown below:

[0104] {A×X , B×X , C×X , D×X , E×X , F×X , G×X , H×X}

[0105] Remove the greatest common divisor and get the gas spectrum:

[0106] Gas spectrum = {A, B, C, D, E, F, G, H}

[0107] Gas intensity: The detection of gas intensity is related to system calibration. After setting RL1 in zero-level air in step 1, the sensor resistance Rs of each sensor needs to be detected respectively. At the same time, the sensor resistance Rs measured by each sensor in zero-level air is defined as Ro, corresponding to Ro1~Ro8 respectively. The gas intensity value of each sensor is Ro / Rs. The calculation formula of the gas intensity value Vs of the sensor array is as follows:

[0108] Vs=Ro1 / Rs1+Ro2 / Rs2+Ro3 / Rs3+Ro4 / Rs4+Ro5 / Rs5+Ro6 / Rs6+Ro7 / Rs7+Ro8 / Rs8 Formula 2

[0109] The gas intensity detected by the sensor array can be calculated using Formula 2 and reduced to a value to facilitate subsequent gas concentration calculations.

[0110] Step 3: Qualitative identification, determine the gas category and name.

[0111] The process of gas qualitative identification is as follows Fig. 9 As shown, first determine whether the gas intensity is greater than the detection threshold. Only when the gas intensity is greater than the detection threshold can the gas spectrum be relatively unique and qualitative gas identification can be performed.

[0112] According to the gas spectrum and gas spectrum library calculated in step 2, the qualitative identification algorithm is used to find the closest spectrum to determine the gas category and complete the qualitative identification of the gas. The most critical part of the qualitative identification algorithm is to determine the spectrum difference. The following is the calculation formula for the spectrum difference.

[0113] Gas library spectrum = {A L , B L , C L , D L , E L , F L , G L , H L}

[0114] Current Gas Spectrum = {A, B, C, D, E, F, G, H}

[0115] Formula 3

[0116] By calculating the spectrum difference, the spectrum with the smallest difference is found in the gas library, and the gas name and the category are obtained.

[0117] In order to realize accurate and rapid gas identification using limited resources on an embedded system, the gas identification algorithm of the present invention is designed in three aspects: gas identification library, sensor correction coefficient, and gas type identification as follows:

[0118] 1) Gas identification library data structure

[0119] The design principles are:

[0120] a) Add detection limit to the spectrum data structure, which is associated with the initial formula of gas concentration in the above text to reduce the detection error;

[0121] b) The gas number is represented by the gas spectrum data storage address;

[0122] c) Spectral data is stored discretely, and the 10k storage space is divided into 8 areas, each area stores 128 types of gases.

[0123] 2) The current sensor spectrum data structure is the same as the gas identification library, but based on the measured data of the sensor array, a correction factor uint8_tk_factor[8] is added to each sensor according to the calibration data at the factory. The measured spectrum of the sensor is adjusted with the correction factor and then compared with the data in the spectrum library for identification.

[0124] 3) According to the spectrum difference calculation formula in formula 3, traverse each spectrum in the spectrum database for comparison, select the gas closest to the current spectrum, and calculate the storage address of the current spectrum as the gas number.

[0125] Step 4: Quantitative measurement to determine gas concentration.

[0126] The calculation formula of gas concentration Val is as follows:

[0127] Formula 4

[0128] Wherein, Vs represents the gas intensity, and the calculation method is shown in Formula 2, k represents the calibration slope, and b represents the calibration intercept.

[0129] The values ​​of the calibration slope k and the calibration intercept b need to be obtained by calibrating the sensor in advance. In the present invention, a multi-point calibration method is adopted to calibrate the sensor.

[0130] Val 零点 = k0× Vs + b0

[0131] Val 浓度1 = k1× Vs + b1

[0132] Val 浓度2 = k2× Vs + b2

[0133] …

[0134] Val 浓度n = k n × Vs + b n

[0135] The suffixes 1 to n of the calibration slope k and the calibration intercept b represent the order of the multi-point calibration.

[0136] The process of gas quantitative measurement is as follows Fig.10 shown.

[0137] After the quantitative measurement is successful, the qualitative identification and quantitative measurement of the gas are completed as a whole, and the sensor outputs the detection results.

[0138] The present invention can identify unknown types of polluted gases, determine the type of polluted gases and the specific type of polluted gases, and can measure the concentration of gases, with a measurement accuracy of ±10% within the designed range. The present invention classifies and identifies gases based on embedded, weak signal acquisition, zero-point calibration, linear calibration, multi-point calibration, dynamic calibration, normalized spectrogram and pattern recognition algorithm, and obtains gas concentrations based on the sensor's response to different types of gases. The gas analyzer designed based on the present invention can replace FTIR and MS in most application scenarios, greatly reducing the cost of use, and can be widely used in emergency disasters, environmental monitoring, garbage disposal, landfill monitoring, food processing, livestock and poultry breeding monitoring, and medical and health care.

[0139] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array, characterized in that: The following steps are involved: S1: Obtain sensor array measurement data; S2: converting the measurement data into a gas spectrum and a gas intensity value; S3: Qualitative identification, determine the type and name of the gas; S4: quantitative measurement to determine gas concentration; Before S1, the sensor circuit needs to be designed. Eight types of MOS sensors are selected to form a MOS sensor array. The original data measured by the sensor array is obtained through the sensor acquisition circuit. The sensor resistance Rs is calculated by formula 1. Formula 1 Where Rs is the sensor resistance, RL1 is the digital potentiometer, RL2 is the fixed resistor, (RL1+RL2) is the load resistance, Vc is the power supply voltage, and Vout is the voltage on the load resistor; Specifically, S2 extracts the sensor array measurement data and converts the data into concentration values ​​and gas spectra, so that the gas spectra are independent of the gas concentration. The raw data measured by the sensor array from S1 are as follows: {Rs1,Rs2,Rs3,Rs4,Rs5,Rs6,Rs7,Rs8} Among them, Rs1 ~ Rs8 represent the resistance of sensor 1 ~ sensor 8 respectively; The calculation method of the gas spectrum is as follows: consider each sensor as a dimension, find the greatest common divisor X according to the actual measured data of the sensor, and replace Rs 1~ Rs 8 with A×X~H×X in sequence, as shown below: {A×X , B×X , C×X , D×X , E×X , F×X , G×X , H×X} Remove the greatest common divisor and get the gas spectrum: Gas spectrum = {A, B, C, D, E, F, G, H}; The gas spectrum library contains various gas spectra; In S2, the calculation method of gas intensity is as follows: after setting RL1 in zero-level air, the value of the sensor resistance Rs of each sensor is detected respectively, and the value of the sensor resistance Rs measured by each sensor in zero-level air is defined as Ro, corresponding to Ro1~Ro8 respectively, the gas intensity value of each sensor is Ro / Rs, and the gas intensity value Vs of the sensor array is calculated as follows: Vs=Ro1 / Rs1+Ro2 / Rs2+Ro3 / Rs3+Ro4 / Rs4+Ro5 / Rs5+Ro6 / Rs6+Ro7 / Rs7+Ro8 / Rs8 Formula 2; The specific steps of S3 are: first determine whether the gas intensity is greater than the detection threshold. Only when the gas intensity is greater than the detection threshold, the gas spectrum can be relatively unique, and the gas can be qualitatively identified. According to the gas spectrum and gas spectrum library calculated by S2, the closest spectrum is found through the qualitative identification algorithm to determine the gas category and complete the qualitative identification of the gas. The most critical part of the qualitative identification algorithm is to determine the spectrum difference. The calculation formula of the spectrum difference is as follows: Gas Spectrum Library = {A L1 , B L1 , C L1 , D L1 , E L1 , F L1 , G L1 , H L1 } {A L2 , B L2 , C L2 , D L2 , E L2 , F L2 , G L2 , H L2 } {A L3 , B L3 , C L3 , D L3 , E L3 , F L3 , G L3 , H L3 } ... {A Ln , B Ln , C Ln , D Ln , E Ln , F Ln , G Ln , H Ln } Among them, the suffix numbers L1, L2, L3…Ln respectively represent the order of gas spectra in the spectrum library; Current Gas Spectrum = {A, B, C, D, E, F, G, H} Formula 3 Wherein, n in Formula 3 means that the spectrum difference between the current gas spectrum and the n spectra in the spectrum library needs to be calculated in sequence; By calculating the spectrum difference, the spectrum with the smallest difference is found in the gas spectrum library, and the gas name and the category are obtained.

2. The method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array according to claim 1, characterized in that: Specifically, S1 adopts a trend judgment algorithm to identify the trend of sensor measurement values, and defines the sensor trend as: rising, falling, fluctuating and smoothing. When the sensor measurement value trend is an rising or falling trend, the average of the latest 8 groups of data in the data buffer is taken as the real-time value data; when the sensor measurement value is a fluctuating trend, the average of the latest 60 groups of data in the data buffer is taken as the real-time data; when the sensor measurement value is a smoothing trend, the average of all values ​​in the buffer is taken as the real-time data.

3. The method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array according to claim 1, characterized in that: The specific steps of S4 are as follows: the calculation formula of gas concentration Val is as follows: Equation 4 Wherein, Vs represents the gas intensity, and the calculation method is shown in Formula 2, k represents the calibration slope, and b represents the calibration intercept; After the quantitative measurement is successful, the qualitative identification and quantitative measurement of the gas are completed as a whole, and the sensor outputs the detection results.

4. The method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array according to claim 3, characterized in that: The values ​​of the calibration slope k and the calibration intercept b need to be obtained by calibrating the sensor in advance, and the sensor is calibrated by a multi-point calibration method. Wall 零点 = k0× Vs + b0 Wall 浓度1 = k1× Vs + b1 Val 浓度2 = k2× Vs + b2 … Val 浓度n = k n × Vs + b n The suffixes 1 to n of the calibration slope k and the calibration intercept b represent the order of the multi-point calibration.

5. The method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array according to claim 1, characterized in that: After the sensor is designed, the sensor array needs to be calibrated. The sensor array calibration first needs to complete the zero point calibration. The sensor array zero point calibration requires completing two tasks: 1) Obtain the default configuration parameters of the digital potentiometer in each sensor detection circuit; 2) The detection value of the sensor in zero-level air is implemented as follows: First, by calculating and adjusting the digital potentiometer RL1 in the sensor detection circuit, Vout approaches 2.2v in a zero-level air environment. This step is repeated 8 times. When the ratio of the difference between the maximum and minimum values ​​of the digital potentiometer RL1 to the average of 8 times is less than 5%, the digital potentiometer RL1 is configured as the average, and the calibration of the digital potentiometer RL1 is completed. If the calculation result is ≥5%, return to the previous step and start the calculation again; secondly, after the digital potentiometer is configured, the value of the sensor resistance Rs is calculated. The calculation is also repeated 8 times to obtain the values ​​of 8 sensor resistance Rs. If the ratio of the difference between the maximum and minimum values ​​to the average of 8 times is less than 5%, the sensor zero point calibration is completed, and the average of the 8 sensor resistance Rs is written into Ro of each sensor. Ro is the value of the sensor resistance Rs in the zero-level air state.

6. The method for qualitative identification and quantitative measurement of gas based on a broad spectrum sensor array according to claim 5, characterized in that: After completing the zero point calibration of the sensor, the concentration calibration of the sensor array is also required, which includes the following steps: Step 1: Connect the calibration gas and record the calibration concentration value; Step 2: Enter the calibration state, read the sensor array data, and store it in the buffer; Step 3: Compare the latest 8 data to see if the difference between the maximum and minimum values ​​is less than 5% of the mean; Step 4: When the difference between the maximum and minimum values ​​is less than 5% of the mean, the calibration is completed and the sensor array value is stored as the concentration calibration value.

Images