Electrochemical sensor calibration method and device

Abstract

The invention discloses an electrochemical sensor calibration method and equipment. The method comprises the following steps: acquiring probe data, temperature and humidity data and gas characteristic data of a sensor to be calibrated; according to the probe data, constructing a dynamic updating model for dynamically updating the zero data of the sensor to be calibrated; according to the temperature and humidity data, constructing a temperature and humidity compensation model for performing temperature and humidity compensation on the to-be-calibrated sensor; according to the gas characteristic data, a multi-gas correction model is constructed, and the multi-gas correction model is used for correcting output data when the sensor to be calibrated is in different gas types; the dynamic updating model, the temperature and humidity compensation model and the multi-gas correction model form a calibration model of the sensor to be calibrated; and calibrating the to-be-calibrated sensor according to the calibration model to obtain a calibration result which is used for indicating the to-be-calibrated sensor to be calibrated and then output. The output data of the sensor to be calibrated in different gas types can be corrected, and the problem that accurate measurement cannot be achieved in a multi-gas interference scene is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrochemical sensors, and in particular to an electrochemical sensor calibration method and equipment. Background Art

[0002] In fields such as environmental monitoring and industrial safety, accurate measurement of electrochemical sensors is crucial. Calibration, a key step in ensuring measurement accuracy, directly impacts sensor performance. Calibration establishes a correspondence between a known concentration of a standard gas and the sensor's output signal, enabling accurate measurement of unknown gas concentrations.

[0003] Most of the existing electrochemical sensor calibration methods use the following two methods:

[0004] 1. Traditional static calibration collects multiple sets of data in a constant temperature and humidity environment. Its essence is to establish a model based on the response law of a single gas or a simple mixed gas under ideal conditions. It cannot accurately measure in sudden pollution scenarios with multiple gas interferences;

[0005] 2. Deep learning methods rely on a large amount of training data to obtain the mapping relationship between input and output. It is difficult to learn the interference mechanism between different gases from the theoretical level of intermolecular interaction. There is a problem of insufficient compensation for multi-gas cross-interference, resulting in inaccurate measurement in scenarios with multi-gas interference.

[0006] Therefore, the existing electrochemical sensor calibration method cannot effectively solve the problem of inaccurate measurement in multi-gas interference scenarios. Summary of the Invention

[0007] In order to overcome the shortcomings of the existing technology, the purpose of the present invention is to provide an electrochemical sensor calibration method and equipment, which can correct the output data of the sensor to be calibrated when it is exposed to different gas types based on the sensitivity of the sensor to be calibrated to different gases, thereby solving the problem of inaccurate measurement in multi-gas interference scenarios.

[0008] In order to solve the above problems, the present invention is implemented according to the following scheme:

[0009] A method for calibrating an electrochemical sensor is provided, comprising:

[0010] Obtain probe data, temperature and humidity data, and gas characteristic data of the sensor to be calibrated;

[0011] Constructing a dynamic update model for dynamically updating the zero point data of the sensor to be calibrated according to the probe data;

[0012] Constructing a temperature and humidity compensation model for performing temperature and humidity compensation on the sensor to be calibrated based on the temperature and humidity data;

[0013] Constructing a multi-gas correction model based on the gas characteristic data, wherein the multi-gas correction model is used to correct the output data of the sensor to be calibrated when it is exposed to different gas types;

[0014] The dynamic update model, the temperature and humidity compensation model, and the multi-gas correction model constitute a calibration model of the sensor to be calibrated;

[0015] The sensor to be calibrated is calibrated according to the calibration model to obtain a calibration result for indicating the output of the sensor to be calibrated after the calibration.

[0016] Compared with the prior art, the beneficial effects of the electrochemical sensor calibration method of the present invention are as follows: by obtaining gas characteristic data reflecting the sensitivity of the sensor to be calibrated to different gases, a multi-gas correction model is constructed based on the gas characteristic data to correct the output data of the sensor to be calibrated when it is in different gas types, thereby solving the problem of inaccurate measurement in multi-gas interference scenarios.

[0017] Optionally, the probe data includes output data sets and initial zero point data sets of the sensor to be calibrated in various gas environments;

[0018] Obtain the probe data of the sensor to be calibrated, including:

[0019] placing the sensor to be calibrated in a test environment with different electrochemical gas concentrations under standard conditions at different times, respectively, to obtain the output data set of the sensor to be calibrated including the output data at different times;

[0020] The sensor to be calibrated is placed in a test environment without electrochemical gas under standard conditions at different times, and the initial zero point data set including the initial zero point data output by the sensor to be calibrated at different times is obtained.

[0021] Optionally, constructing a dynamic update model for dynamically updating the zero point data of the sensor to be calibrated based on the probe data includes:

[0022] Determine the time interval for dynamically updating the zero point data of the sensor to be calibrated;

[0023] Determine, according to the time interval, the real-time output data at the current moment and the historical output data with the time interval between the current moment and the real-time output data in the output data set;

[0024] Determining probe change parameters of the sensor to be calibrated based on the real-time output data and the historical output data;

[0025] The dynamic update model is constructed according to the probe change parameters and the initial zero-point data set.

[0026] Optionally, constructing the dynamic update model according to the probe change parameter and the initial zero point data set includes:

[0027] determining, according to the probe change parameter, an updated zero point data set including updated zero point data at different moments;

[0028] determining a zero-point concentration change parameter according to the updated zero-point data set and the initial zero-point data set;

[0029] The dynamic update model is constructed according to the zero-point concentration change parameter.

[0030] Optionally, the temperature and humidity data include temperature compensation data and humidity compensation data;

[0031] Get the temperature and humidity data of the sensor to be calibrated, including:

[0032] Determine the temperature change data for the temperature change test and the humidity change data for the humidity change test of the sensor to be calibrated;

[0033] Placing the sensor to be calibrated in a gas environment consisting of standard humidity and the temperature change data to obtain the temperature compensation data;

[0034] The sensor to be calibrated is placed in a gas environment consisting of a standard temperature and the temperature change data to obtain the humidity compensation data.

[0035] Optionally, constructing a temperature and humidity compensation model for determining the temperature and humidity compensation coefficients of the sensor to be calibrated based on the temperature and humidity data includes:

[0036] Determining a temperature compensation coefficient of the sensor to be calibrated according to the temperature compensation data;

[0037] Determining a humidity compensation coefficient of the sensor to be calibrated according to the humidity compensation data;

[0038] The temperature and humidity compensation model is constructed according to the temperature compensation coefficient and the humidity compensation coefficient.

[0039] Optionally, the gas characteristic data includes gas sensitivity of the sensor to be calibrated to different electrochemical gases;

[0040] Obtain gas characteristic data of the sensor to be calibrated, including:

[0041] The sensor to be calibrated is placed in a gas environment including different electrochemical gases under standard conditions, and the sensor resistance values ​​of the sensor to be calibrated in the different electrochemical gas environments are obtained;

[0042] Place the sensor to be calibrated in a gas environment without electrochemical gas under standard conditions, and obtain the sensor resistance value of the sensor to be calibrated in the clean environment;

[0043] The gas sensitivity is determined according to the sensor resistance values ​​of the sensor to be calibrated in the different electrochemical gas environments and the sensor resistance value in the clean environment.

[0044] Optionally, calibrating the sensor to be calibrated according to the calibration model to obtain a calibration result indicating an output of the sensor to be calibrated after the calibration, includes:

[0045] Get the current probe data output by the sensor to be calibrated;

[0046] According to the current probe data, obtaining compensated probe data after the current probe data is compensated by the temperature and humidity compensation model;

[0047] According to the compensation probe data, actual zero point data output after the zero point value is updated by the dynamic update model is obtained;

[0048] The multi-gas correction model determines the gas type of the gas environment in which the sensor to be calibrated is located based on the compensation probe data, corrects the compensation probe data based on the gas type, and outputs the corrected probe data;

[0049] determining a concentration change value according to the corrected probe data and the actual zero point data;

[0050] The concentration change value is subjected to curve processing and fitting to obtain the calibration result, wherein the calibration result includes actual concentration data output after the sensor to be calibrated is calibrated.

[0051] Optionally, the multi-gas correction model determines the gas type of the gas environment in which the sensor to be calibrated is located based on the compensation probe data, including:

[0052] determining gas sensitivity based on the compensation probe data;

[0053] The type of electrochemical gas included in the gas environment where the sensor to be calibrated is located is determined according to the gas sensitivity and the gas characteristic data.

[0054] A computer device is also provided, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the electrochemical sensor calibration method. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flow chart of the calibration method of the present invention;

[0056] Figure 2 The graphs are the probe data output without compensation by the temperature and humidity compensation model, the probe data output after compensation by the temperature and humidity compensation model, and the actual output probe data.

[0057] Figure 3 This is a calibration flow chart of the sensor to be calibrated by the calibration model of the present invention. DETAILED DESCRIPTION

[0058] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0059] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0060] See also Figure 1 As shown, a method for calibrating an electrochemical sensor of the present invention includes:

[0061] S1: Obtain the probe data, temperature and humidity data, and gas characteristic data of the sensor to be calibrated; wherein, the probe data is the output data of the sensor to be calibrated under standard conditions at different times and in different gas environments, which is used to reflect the response characteristics of the sensor to be calibrated to multiple electrochemical gases and different concentrations; the temperature and humidity data is the output data of the sensor to be calibrated under different temperature and humidity conditions in the same gas environment, which is used to reflect the impact of temperature and humidity changes on the output data of the sensor to be calibrated; the gas characteristic data is the sensitivity of the sensor to be calibrated to the electrochemical gas present in the gas environment when it is in different gas environments under standard conditions.

[0062] In one embodiment of the present invention, the probe data includes output data sets and an initial zero point data set of the sensor to be calibrated in various gas environments. Acquiring the probe data of the sensor to be calibrated includes:

[0063] At different times, the sensor to be calibrated is placed in a test environment under standard conditions including different electrochemical gas concentrations, and an output data set of the sensor to be calibrated including output data at different times is obtained; wherein the standard conditions are an environment with a temperature of 25°C and a humidity of 50%. By recording the output data of the sensor to be calibrated in a gas environment with the same electrochemical gas at different times, the response law of the sensor to be calibrated to the electrochemical gas at different times can be reflected. By recording the output data of the sensor to be calibrated in a gas environment with different electrochemical gases, the response law of the sensor to be calibrated to different electrochemical gases can be reflected. For example, in an environment with different electrochemical gases such as formaldehyde and ethanol and with different concentrations, the output data of the sensor to be calibrated will be different, and these data can show its response law.

[0064] At different times, the sensor to be calibrated is placed in a test environment without electrochemical gas under standard conditions, and an initial zero point data set of the sensor to be calibrated is obtained, including the initial zero point data output at different times; wherein the standard conditions are an environment with a temperature of 25°C and a humidity of 50%. By recording the output data of the sensor to be calibrated in a clean environment without electrochemical gas at different times, the drift of the sensor to be calibrated in the absence of electrochemical gas can be reflected. As time goes by, the zero point of the sensor to be calibrated may change. By recording the initial zero point data at different times, the zero point drift of the sensor to be calibrated can be reflected, and the gas concentration detected by the sensor to be calibrated can be accurately output.

[0065] In one embodiment of the present invention, the temperature and humidity data include temperature compensation data and humidity compensation data; obtaining the temperature and humidity data of the sensor to be calibrated includes:

[0066] First, determine the temperature change data for the temperature change test and the humidity change data for the humidity change test of the sensor to be calibrated. The temperature change data refers to the data formed by setting different temperature values ​​while maintaining the same humidity, here the standard humidity is 50%, and the temperature change gradient between different temperature values ​​is the same. In this embodiment, 5°C, 15°C, 25°C, 35°C, 45°C, and 55°C with a temperature change gradient of 10°C are used as temperature change points to form the temperature change data. The probe data changes of the sensor to be calibrated when the sensor to be calibrated is in these temperature environments are simulated. The humidity change data refers to the data formed by setting different humidity values ​​while maintaining the same temperature, here the standard temperature is 25°C. The humidity change gradient between different humidity values ​​is the same. In this embodiment, 5%, 20%, 35%, 50%, 65%, and 80% with a humidity change gradient of 15% are used as humidity change points to form the humidity change data. The probe data changes of the sensor to be calibrated when the sensor to be calibrated is in these humidity environments are simulated.

[0067] Finally, the sensor to be calibrated is placed in a gas environment consisting of standard humidity and temperature change data to obtain temperature compensation data; specifically, the sensor to be calibrated is placed in a standard gas environment consisting of standard humidity 50%, standard temperature 25℃, the same electrochemical gas and the same concentration, and the probe data output by the sensor to be calibrated is obtained; then the temperature is set to 5℃, 15℃, 25℃, 35℃, 45℃, and 55℃ in sequence, and the sensor to be calibrated is obtained at [5℃ / 50%], [15℃ / 50%], [25℃ / 50%], [ The probe data output under the six conditions of [35°C / 50%], [45°C / 50%], and [55°C / 50%] in the same gas and gas environment constitute the temperature compensation data of the sensor to be calibrated. The probe data output by the sensor to be calibrated at different temperatures in the same electrochemical gas and gas environment with the same concentration can reflect the influence of temperature on the sensor to be calibrated. Based on this influence, the probe data of the sensor to be calibrated is temperature compensated to eliminate the influence of temperature on the probe data output by the sensor to be calibrated.

[0068] The sensor to be calibrated is placed in a gas environment consisting of standard temperature and temperature change data to obtain humidity compensation data. Specifically, the sensor to be calibrated is placed in a standard gas environment consisting of standard temperature 25°C, standard humidity 50%, the same electrochemical gas and the same concentration to obtain the probe data output by the sensor to be calibrated. Then, the humidity is set to 5%, 20%, 35%, 50%, 65%, and 80% in sequence to obtain the sensor to be calibrated at [25°C / 5%], [25°C / 20%], [25°C / 35%], [2 The probe data output under the following six conditions: [5℃ / 50%], [25℃ / 65%], and [25℃ / 80%], in an environment with the same gas and the same concentration, constitute the humidity compensation data of the sensor to be calibrated. The probe data output by the sensor to be calibrated under different humidity conditions in an environment with the same electrochemical gas and the same concentration can reflect the influence of humidity on the sensor to be calibrated. Based on this influence, humidity compensation is performed on the probe data of the sensor to be calibrated to eliminate the influence of humidity on the probe data output by the sensor to be calibrated.

[0069] Inside the sensor to be calibrated, which is used to detect electrochemical gas concentrations, there are sensitive materials that can react specifically with the gas to be measured. When different electrochemical gases come into contact with these sensitive materials, electrochemical reactions will be triggered, thereby changing the electrical properties of the sensitive materials. For example, the carrier concentration inside the material changes, which ultimately manifests as a change in the sensor resistance value.

[0070] In one embodiment of the present invention, the gas characteristic data includes the gas sensitivity of the sensor to be calibrated to different electrochemical gases; obtaining the gas characteristic data of the sensor to be calibrated includes:

[0071] First, the sensor to be calibrated is placed in a gas environment including different electrochemical gases under standard conditions, and the sensor resistance value of the sensor to be calibrated in different electrochemical gas environments is obtained; the standard conditions are an environment with a temperature of 25°C and a humidity of 50%. By obtaining the sensor resistance value of the sensor to be calibrated in different electrochemical gases, the resistance value can be used to reflect the impact of different electrochemical gases on the sensitive materials inside the sensor that react specifically with the gas.

[0072] Next, the sensor to be calibrated is placed in a gas environment without electrochemical gas under standard conditions, and the sensor resistance value of the sensor to be calibrated in the clean environment is obtained; the standard conditions are an environment with a temperature of 25°C and a humidity of 50%. The sensor resistance value is not only affected by the gas to be measured, but also by multiple factors. Therefore, the sensor resistance value of the sensor to be calibrated in different electrochemical gas environments obtained above cannot determine whether the change in the resistance value is caused by the electrochemical gas or other factors, and it cannot directly reflect the sensitivity of the sensor to be calibrated to different electrochemical gases. Therefore, by obtaining the sensor resistance value of the sensor to be calibrated in a clean environment without electrochemical gas, other factors that may affect the sensor resistance value are excluded, and the degree of resistance change of the sensor to be calibrated caused by different electrochemical gases is accurately calculated to obtain the sensitivity of the sensor to be calibrated to different electrochemical gases.

[0073] Finally, the gas sensitivity is determined based on the sensor resistance values ​​of the sensor to be calibrated in different electrochemical gas environments and the sensor resistance values ​​in a clean environment. The calculation formula is as follows:

[0074]

[0075] Among them, S gas is the gas sensitivity of the sensor to be calibrated in the target electrochemical gas environment, R g is the sensor resistance value of the sensor to be calibrated in the target electrochemical gas environment, and R0 is the sensor resistance value of the sensor to be calibrated in a clean environment.

[0076] In one embodiment of the present invention, the sensor to be calibrated includes a dynamic temperature control management program for controlling temperature, and a Kalman filter for suppressing noise. For electrochemical sensors, for the same electrochemical gas, such as ethanol, the sensor has different sensitivities to different concentrations of ethanol. The dynamic temperature control management program is used to heat or suppress heating of the sensor material, so that the material is at a temperature with better response characteristics, thereby optimizing the response characteristics of the gas-sensitive material and improving the output signal quality. At the same time, the temperature fluctuations introduced by the dynamic temperature control management program will increase signal noise. Such noise is suppressed by the Kalman filter to ensure the stability of the output probe data.

[0077] S2: Based on the probe data, a dynamic update model is constructed for dynamically updating the zero-point data of the sensor to be calibrated. The probe data includes the output data sets of the sensor to be calibrated in various gas environments and the initial zero-point data set. The initial zero-point data set is used to reflect the zero-point drift of the sensor to be calibrated in a clean environment without electrochemical gas over time. Therefore, the zero-point drift of the sensor to be calibrated over time is learned through the probe data, and a dynamic update model for dynamically updating the zero-point data of the sensor to be calibrated is constructed.

[0078] In one embodiment of the present invention, a dynamic update model for dynamically updating zero point data of a sensor to be calibrated is constructed based on probe data, including:

[0079] First, a time interval for dynamically updating the zero-point data of the sensor to be calibrated is determined. In this embodiment, the time interval is 24 hours. As for sensor zero-point drift, there will be no significant drift in a short period of time. However, if the time interval is too long and the zero-point data of the sensor to be calibrated is not updated for a long time, the detection accuracy of the sensor to be calibrated will be affected. Therefore, 24 hours is selected as the time interval in this embodiment. Next, based on the time interval, the output data set is determined to include the current real-time output data and the historical output data that are time-intervals apart from the current time. Based on the real-time output data and the historical output data, the probe change parameter of the sensor to be calibrated is determined. Specifically, a fitting calculation is performed on the real-time output data and the historical output data. By comparing the real-time output data at the current time with the historical output data 24 hours (time interval) ago, the slope of the probe data change rate of the sensor to be calibrated is obtained. This slope is used as the probe change parameter of the sensor to be calibrated. The probe change parameter can reflect the change of the sensor to be calibrated during this time interval.

[0080] Finally, a dynamic update model is constructed based on the probe change parameters and the initial zero-point data set.

[0081] In one embodiment of the present invention, a dynamic update model is constructed based on the probe change parameters and the initial zero point data set, including:

[0082] First, according to the probe change parameters, the updated zero point data set including the updated zero point data at different times is determined. The calculation formula of the updated zero point data is as follows:

[0083] Z new =α·Z old +0.1min(Z t [t-24:t])

[0084] Among them, Z new is the updated zero point data of the sensor to be calibrated after the zero point value is dynamically updated, α is the probe change parameter, Z old The historical zero point data of the sensor to be calibrated before the zero point value is dynamically updated. When the probe change parameter approaches 0, the updated zero point value is preferably selected as the updated zero point data. When the probe change parameter approaches 1, the zero point value before the update is preferably selected as the updated zero point data.

[0085] Then, the zero-point concentration variation parameters are determined based on the updated zero-point data set and the initial zero-point data set; finally, a dynamic update model is constructed based on the zero-point concentration variation parameters.

[0086] The expression for updating the zero-point data set is as follows:

[0087] Z c ={C1, C2, ..., C n}

[0088] Among them, Z c To update the zero-point dataset, C n Update zero point data for the nth moment.

[0089] The expression of the initial zero-point data set is as follows:

[0090] VOC e ={E1, E2, ..., E n}

[0091] Among them, VOC e is the initial zero-point data set, E n is the initial zero point data at the nth moment.

[0092] In one embodiment of the present invention, the zero-point concentration change parameter is determined by fitting a curve to the updated zero-point data set and the initial zero-point data set to obtain a zero-point concentration change curve, which is expressed as follows:

[0093]

[0094] Among them, VOC0 is the zero-point value calculated from the zero-point concentration change curve, k is the zero-point concentration change parameter, which is obtained from the experimental fitting test, and Z new is the updated zero point data at the current moment, C i is the updated zero point data at the i-th moment, E i is the initial zero point data at the i-th moment.

[0095] S3: Based on the temperature and humidity data, a temperature and humidity compensation model is constructed for performing temperature and humidity compensation on the sensor to be calibrated; wherein the temperature and humidity data include temperature compensation data for reflecting the influence of temperature on the probe data output by the sensor to be calibrated, and humidity compensation data for reflecting the influence of humidity on the probe data output by the sensor to be calibrated. Therefore, the offset of the probe data output by the sensor to be calibrated under the influence of temperature and humidity is learned through the temperature and humidity data, and a temperature compensation and humidity compensation model for the probe data output by the sensor to be calibrated is constructed to eliminate the influence of temperature and humidity on the probe data output by the sensor to be calibrated.

[0096] In one embodiment of the present invention, a temperature and humidity compensation model for determining the temperature and humidity compensation coefficients of the sensor to be calibrated is constructed based on the temperature and humidity data and Poisson's law and the Hall effect, including:

[0097] First, determine the temperature compensation coefficient of the sensor to be calibrated based on the temperature compensation data. The calculation formula for the temperature compensation coefficient is as follows:

[0098]

[0099] ΔZ i =Z k -Z k-1

[0100] Where β is the temperature compensation coefficient, N T The number of probe data included in the temperature compensation data, ΔT i Z is the temperature change between the standard temperature of 25℃ and the temperature value of group i. k-1 is the probe data output by the sensor to be calibrated at a standard temperature of 25°C and a standard humidity of 50%, Z k ΔZ is the probe data output by the sensor to be calibrated after the temperature changes from the standard temperature of 25°C to the temperature value of the i-th group, i The change in probe data output by the sensor to be calibrated after the temperature changes.

[0101] According to the humidity compensation data, determine the humidity compensation coefficient of the sensor to be calibrated. The calculation formula of the humidity compensation coefficient is as follows:

[0102]

[0103] ΔZ i =Z k -Z k-1

[0104] Among them, γ is the humidity compensation coefficient, N T The number of probe data included in the humidity compensation data, ΔH i Z is the humidity change between the standard humidity of 50% and the humidity value of group i. k-1 is the probe data output by the sensor to be calibrated at a standard temperature of 25°C and a standard humidity of 50%, Z k ΔZ is the probe data output by the sensor to be calibrated after the temperature changes from the standard humidity of 50% to the humidity value of the i-th group, i The change in probe data output by the sensor to be calibrated after the humidity changes.

[0105] According to the temperature compensation coefficient and humidity compensation coefficient, a temperature and humidity compensation model is constructed, and its expression is as follows:

[0106] Z k|k-1 =Z k-1 +βΔT+γΔH+w k

[0107] Among them, Z k|k-1 is the estimated value of the probe data at time k based on the probe data at time k-1, Z k-1 is the probe data at the k-1th moment, β is the temperature compensation coefficient, ΔT is the temperature change between the current temperature value and the standard temperature of 25°C, γ is the humidity compensation coefficient, ΔH is the humidity change between the current humidity value and the standard humidity of 50%, w k is the process noise, which obeys the Gaussian distribution N(0,Q).

[0108] See also Figure 2 The figure shows a comparison between the sensor data before and after compensation using the temperature and humidity compensation model, the sensor data after compensation using the temperature and humidity compensation model, and the actual sensor data. The figure shows that the sensor data after compensation using the temperature and humidity compensation model is more consistent with the actual sensor data. S4: Based on the gas characteristic data, a multi-gas correction model is constructed. The multi-gas correction model is used to correct the output data of the sensor to be calibrated when exposed to different gas types. The expression of the multi-gas correction model is as follows:

[0109] Z gas =S gas ·Z k

[0110] Among them, Z gas The corrected probe data is obtained by correcting the probe data output by the sensor to be calibrated, Sgas is the gas sensitivity of the sensor to be calibrated to the current detection gas, Z k This is the probe data output before correction based on gas sensitivity of the sensor to be calibrated.

[0111] S5: Dynamically update the model, temperature and humidity compensation model, and multi-gas correction model to form the calibration model of the sensor to be calibrated.

[0112] In one embodiment of the present invention, a dynamic update model is constructed based on probe data and is used to dynamically update the zero-point data of the sensor being calibrated. Specifically, by determining an update interval, comparing the current real-time output data with historical output data, and deriving probe change parameters, the model is then combined with the initial zero-point data set to construct a dynamic update model that continuously tracks zero-point changes, ensuring the accuracy and stability of measurement results.

[0113] The temperature and humidity compensation model is built based on temperature and humidity data and is used to compensate for the temperature and humidity of the sensor being calibrated. Temperature and humidity variations can significantly affect the measurement accuracy of the sensor being calibrated. The temperature and humidity compensation model obtains temperature and humidity compensation data, determines the temperature and humidity compensation coefficients, and then constructs the model. During actual measurements, the temperature and humidity compensation model uses these coefficients to correct for deviations in the sensor being calibrated due to temperature and humidity variations. This reduces the interference of temperature and humidity factors on the output results and improves the measurement accuracy of the sensor being calibrated in different temperature and humidity environments.

[0114] The multi-gas correction model, constructed based on gas characteristic data, is used to correct the output data of the sensor being calibrated when exposed to different gas types. The sensor being calibrated responds differently to different electrochemical gases, which can easily cause cross-interference and affect measurement accuracy. The multi-gas correction model obtains the sensor's sensitivity to different electrochemical gases, combines it with compensation probe data to determine the gas type, and then corrects the compensation probe data. This effectively resolves the issue of inaccurate measurements in multi-gas interference scenarios, enabling the sensor to accurately identify and measure the concentrations of different electrochemical gases.

[0115] By dynamically updating the model to ensure the accuracy of the zero point of the sensor to be calibrated, the temperature and humidity compensation model eliminates the impact of environmental factors on the sensor to be calibrated, and the multi-gas correction model solves problems in multi-gas interference scenarios. The calibration model composed of these three models effectively improves the measurement accuracy and reliability of the sensor to be calibrated, enabling the calibrated sensor to accurately measure the electrochemical gas concentration in complex environments.

[0116] See also Figure 3 As shown, S6: calibrate the sensor to be calibrated according to the calibration model to obtain a calibration result for indicating the output of the sensor to be calibrated after calibration.

[0117] In one embodiment of the present invention, calibrating a sensor to be calibrated according to a calibration model to obtain a calibration result indicating an output of the sensor to be calibrated after calibration includes:

[0118] First, the current probe data output by the sensor to be calibrated is obtained. Based on the current probe data, the compensated probe data after the current probe data is compensated by the temperature and humidity compensation model is obtained to eliminate the influence of the two environmental factors of temperature and humidity on the model to be calibrated.

[0119] Next, based on the compensated probe data, the actual zero point data output after the zero point value is updated by the dynamic update model is obtained to avoid the zero point drift of the sensor to be calibrated over time, which may cause inaccurate output probe data.

[0120] After updating the zero point value of the sensor to be calibrated, the multi-gas correction model determines the gas type of the gas environment in which the sensor to be calibrated is located based on the compensation probe data, corrects the compensation probe data according to the gas type, and outputs the corrected probe data to correct the compensation probe data according to the gas sensitivity corresponding to different gas types.

[0121] Finally, the concentration change value is determined based on the corrected probe data and the actual zero point data; the concentration change value is subjected to curve processing and fitting to obtain the calibration result, which includes the actual concentration data output after the sensor to be calibrated is calibrated.

[0122] In one embodiment of the present invention, the concentration change value is the difference between the corrected probe data and the actual zero point data. The concentration change value is processed and fitted using a quadratic polynomial curve to obtain a fitted quadratic polynomial of the concentration change value, which is expressed as follows:

[0123] VOC y =a o +a1·X+a2·X 2

[0124] Among them, VOC y is the fitted concentration change value after fitting, X is the concentration change value before fitting, and a0, a1, and a2 are the constant coefficients obtained after fitting. If the zero point of the sensor to be calibrated has not drifted, the fitted concentration change value is the final calibration result (actual concentration data). If the zero point of the sensor to be calibrated has drifted, the difference between the fitted concentration change value and the actual zero point data is the final calibration result (actual concentration data).

[0125] In one embodiment of the present invention, the multi-gas correction model determines the gas type of the gas environment in which the sensor to be calibrated is located based on the compensation probe data, including:

[0126] First, the gas sensitivity is determined based on the compensated probe data. Finally, the type of electrochemical gas in the gas environment of the sensor to be calibrated is determined based on the gas sensitivity and gas characteristic data. In multi-gas scenarios, different electrochemical gases can cause cross-interference in the sensor response, and the sensor's sensitivity to different gases varies greatly. Failure to confirm the electrochemical gas type will result in an inability to accurately convert the response signal into the corresponding gas concentration value. After confirming the gas type, the output probe data can be corrected based on the gas characteristics and the sensor's sensitivity to the gas, reducing multi-gas interference errors and improving the accuracy of the final output calibration result.

[0127] At the same time, after determining the gas type, the dynamic temperature control management program can optimize the temperature of the gas-sensitive material of the sensor to be calibrated according to the gas type to optimize the response characteristics of the sensor to be calibrated in the gas environment.

[0128] In the actual calibration process, in order to further improve the accuracy of calibration, the spatiotemporal convolutional network (STCNN) will be used to analyze the collaborative drift rules of multiple sensor nodes. The spatiotemporal convolutional network (STCNN) can comprehensively consider the changing characteristics of sensor data in the time and space dimensions, automatically extract key information from the data through convolution operations, and then accurately analyze the collaborative drift rules between multiple sensor nodes.

[0129] Based on these analysis results, optimized compensation parameters are generated, including temperature and humidity compensation coefficients β and γ, gain and offset, etc. These optimized compensation parameters are used to optimize the subsequent calibration process. Specifically, these parameters are applied to the calibration model of the sensor to be calibrated, such as using new β and γ values ​​in the temperature and humidity compensation model, and using the updated gain and offset in subsequent steps such as calculating the concentration change value, thereby improving the accuracy of the calibration results.

[0130] Among them, in the multi-gas scenario, a gain compensation model is required to compensate the calibration results, which plays an important role in the entire calibration process. The expression of the difficult-to-compensate model is a nonlinear regression model:

[0131]

[0132] Among them, VOC Level is the calibration result after gain compensation, Z k To compensate the probe data, Z new is the actual zero point data, θ k is the dynamic weight coefficient output automatically, is the compensation concentration of the kth interfering gas. Gain and offset are the key parameters obtained after the above nonlinear regression optimization.

[0133] When long-term background changes are detected, the gain is also optimized jointly with the Levenberg-Marquardt algorithm and gradient descent, and adjusted according to exponential decay. Its expression is as follows:

[0134] Gain new =Gain old ·e -λt

[0135] Among them, Gain new Gain is the adjusted gain. old is the gain before adjustment, λ=0.05, which is the experimental calibration attenuation coefficient.

[0136] In this way, the gain compensation model can better adapt to complex environments, and combined with optimized parameters such as temperature and humidity compensation coefficients, the accuracy of sensor calibration and detection can be comprehensively improved.

[0137] The calibration method provided by the present invention shortens the calibration time of the sensor to be calibrated from 24 hours to 90 minutes, and the data volume is reduced by 70%. At the same time, the RMSE in the low-concentration section is reduced by 57%, and the cross-interference error is reduced by 50%. In addition, a temperature and humidity compensation model constructed based on Poisson's law and the Hall effect reduces the error increase when the temperature change value ΔT>5°C from ±25% to ±8%. The drift error within 30 days is reduced by 68.5% (±35%→±11%), and the maintenance cycle is extended to 3 months.

[0138] The present invention also provides a computer device comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the above-mentioned calibration method.

[0139] The processor can be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0140] The memory can be used to store the computer program or module, and the processor implements the various functions of the calibration method by running or executing the computer program or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created based on the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0141] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for calibrating an electrochemical sensor, characterized in that: include: Obtain probe data, temperature and humidity data, and gas characteristic data of the sensor to be calibrated; Constructing a dynamic update model for dynamically updating the zero point data of the sensor to be calibrated according to the probe data; Constructing a temperature and humidity compensation model for performing temperature and humidity compensation on the sensor to be calibrated based on the temperature and humidity data; Constructing a multi-gas correction model based on the gas characteristic data, wherein the multi-gas correction model is used to correct the output data of the sensor to be calibrated when it is exposed to different gas types; The dynamic update model, the temperature and humidity compensation model, and the multi-gas correction model constitute a calibration model of the sensor to be calibrated; The sensor to be calibrated is calibrated according to the calibration model to obtain a calibration result for indicating the output of the sensor to be calibrated after the calibration.

2. The electrochemical sensor calibration method according to claim 1, characterized in that: The probe data includes output data sets and initial zero point data sets of the sensor to be calibrated in various gas environments; Obtain the probe data of the sensor to be calibrated, including: placing the sensor to be calibrated in a test environment with different electrochemical gas concentrations under standard conditions at different times, respectively, to obtain the output data set of the sensor to be calibrated including the output data at different times; The sensor to be calibrated is placed in a test environment without electrochemical gas under standard conditions at different times, and the initial zero point data set including the initial zero point data output by the sensor to be calibrated at different times is obtained.

3. The electrochemical sensor calibration method according to claim 2, characterized in that: Constructing a dynamic update model for dynamically updating the zero point data of the sensor to be calibrated according to the probe data, including: Determine the time interval for dynamically updating the zero point data of the sensor to be calibrated; Determine, according to the time interval, the real-time output data at the current moment and the historical output data with the time interval between the current moment and the real-time output data in the output data set; Determining probe change parameters of the sensor to be calibrated based on the real-time output data and the historical output data; The dynamic update model is constructed according to the probe change parameters and the initial zero-point data set.

4. The electrochemical sensor calibration method according to claim 3, characterized in that: Constructing the dynamic update model according to the probe change parameter and the initial zero point data set includes: determining, according to the probe change parameter, an updated zero point data set including updated zero point data at different moments; determining a zero-point concentration change parameter according to the updated zero-point data set and the initial zero-point data set; The dynamic update model is constructed according to the zero-point concentration change parameter.

5. The electrochemical sensor calibration method according to claim 1, characterized in that: The temperature and humidity data include temperature compensation data and humidity compensation data; Get the temperature and humidity data of the sensor to be calibrated, including: Determine the temperature change data for the temperature change test and the humidity change data for the humidity change test of the sensor to be calibrated; Placing the sensor to be calibrated in a gas environment consisting of standard humidity and the temperature change data to obtain the temperature compensation data; The sensor to be calibrated is placed in a gas environment consisting of a standard temperature and the temperature change data to obtain the humidity compensation data.

6. The electrochemical sensor calibration method according to claim 5, characterized in that: Based on the temperature and humidity data, a temperature and humidity compensation model for determining the temperature and humidity compensation coefficients of the sensor to be calibrated is constructed, including: Determining a temperature compensation coefficient of the sensor to be calibrated according to the temperature compensation data; Determining a humidity compensation coefficient of the sensor to be calibrated according to the humidity compensation data; The temperature and humidity compensation model is constructed according to the temperature compensation coefficient and the humidity compensation coefficient.

7. The electrochemical sensor calibration method according to claim 1, characterized in that: The gas characteristic data includes the gas sensitivity of the sensor to be calibrated to different electrochemical gases; Obtain gas characteristic data of the sensor to be calibrated, including: The sensor to be calibrated is placed in a gas environment including different electrochemical gases under standard conditions, and the sensor resistance values ​​of the sensor to be calibrated in the different electrochemical gas environments are obtained; Place the sensor to be calibrated in a gas environment without electrochemical gas under standard conditions, and obtain the sensor resistance value of the sensor to be calibrated in the clean environment; The gas sensitivity is determined according to the sensor resistance values ​​of the sensor to be calibrated in the different electrochemical gas environments and the sensor resistance value in the clean environment respectively.

8. The electrochemical sensor calibration method according to claim 1, characterized in that: Calibrate the sensor to be calibrated according to the calibration model to obtain a calibration result indicating the output of the sensor to be calibrated after calibration, including: Get the current probe data output by the sensor to be calibrated; According to the current probe data, obtaining compensated probe data after the current probe data is compensated by the temperature and humidity compensation model; According to the compensation probe data, actual zero point data output after the zero point value is updated by the dynamic update model is obtained; The multi-gas correction model determines the gas type of the gas environment in which the sensor to be calibrated is located based on the compensation probe data, corrects the compensation probe data based on the gas type, and outputs the corrected probe data; determining a concentration change value according to the corrected probe data and the actual zero point data; The concentration change value is subjected to curve processing and fitting to obtain the calibration result, wherein the calibration result includes actual concentration data output after the sensor to be calibrated is calibrated.

9. The electrochemical sensor calibration method according to claim 8, characterized in that: The multi-gas correction model determines the gas type of the gas environment in which the sensor to be calibrated is located based on the compensation probe data, including: determining gas sensitivity based on the compensation probe data; The type of electrochemical gas included in the gas environment where the sensor to be calibrated is located is determined according to the gas sensitivity and the gas characteristic data.

10. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement an electrochemical sensor calibration method according to any one of claims 1 to 9.

Images