Laboratory gas detection method and apparatus

Abstract

The application relates to a laboratory gas detection method and device. The method comprises the following steps: acquiring a plurality of original gas concentration data in a laboratory, performing anomaly detection on the plurality of original gas concentration data to obtain abnormal gas concentration data; determining interference gas concentration data that has cross interference with the abnormal gas concentration data, determining the probability of the abnormal gas concentration data deviating from a first dynamic baseline according to a first dynamic baseline and a first fluctuation range corresponding to the abnormal gas concentration data, and determining the change probability of the interference gas concentration data according to a second dynamic baseline and a second fluctuation range corresponding to the interference gas concentration data; determining a pollution probability according to the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the change probability of the interference gas concentration data, and determining whether the abnormal gas concentration data is real abnormal data according to the pollution probability. The method can eliminate cross gas interference, improve gas detection accuracy, and reduce the false alarm rate.

Description

Technical Field

[0001] This application relates to the field of laboratory environmental safety monitoring technology, and in particular to a laboratory gas detection method and equipment. Background Technology

[0002] With the development of scientific research and technology, laboratories are widely used in fields such as chemistry, biology, medicine, materials, and electronics. Various experiments often involve the use, release, or leakage of gases. For example, toxic and harmful gases, flammable gases, and volatile organic compounds frequently appear in experimental operations. Once these gases accumulate in the experimental environment, they not only pose a serious threat to the health of laboratory personnel but may also cause major safety accidents such as fires.

[0003] In traditional technologies, laboratories generally use gas detectors based on a single detection principle, which suffer from cross-gas interference, resulting in low gas detection accuracy, high false alarm rates, and an inability to meet the needs of precise monitoring. Summary of the Invention

[0004] Therefore, it is necessary to provide a laboratory gas detection method and equipment that can eliminate cross-gas interference, thereby improving gas detection accuracy and reducing false alarm rate, to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a laboratory gas detection method, comprising:

[0006] Acquire multiple raw gas concentration data in the laboratory, perform anomaly detection on the multiple raw gas concentration data, and obtain abnormal gas concentration data;

[0007] Identify interfering gas concentration data that cross-interferes with abnormal gas concentration data. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, determine the probability of change of the interfering gas concentration data. The first dynamic baseline is determined based on the ratio of abnormal gas concentration data to interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory.

[0008] The probability of contamination is determined based on the probability of abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in interfering gas concentration data. The probability of contamination is then used to determine whether the abnormal gas concentration data is genuine abnormal data.

[0009] In one embodiment, the various raw gas concentration data include infrared active gas concentration and volatile organic compound (VOC) concentration; the abnormal gas concentration data is VOC concentration.

[0010] Identify interfering gas concentration data that cross-interfere with the abnormal gas concentration data. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, determine the probability of change of the interfering gas concentration data, including:

[0011] The concentration data of interfering gases that cross-interfere with the concentration of volatile organic compounds are identified as infrared active gas concentrations;

[0012] Based on the first dynamic baseline and the first fluctuation range corresponding to the volatile organic compound concentration, determine the probability that the volatile organic compound concentration deviates from the first dynamic baseline;

[0013] The probability of change in infrared active gas concentration is determined based on the second dynamic baseline and the second fluctuation range corresponding to the infrared active gas concentration.

[0014] In one embodiment, determining the contamination probability based on the probability of abnormal gas concentration data deviating from a first dynamic baseline and the probability of changes in interfering gas concentration data includes:

[0015] Obtain the preset deviation coefficient and variation coefficient;

[0016] The probability of abnormal gas concentration data deviating from the first dynamic baseline is weighted according to the deviation coefficient to obtain the first weighting term;

[0017] The probability of change in the concentration of interfering gases is weighted according to the coefficient of change to obtain the second weighting term.

[0018] The pollution probability is obtained by summing the first and second weighted terms.

[0019] In one embodiment, multiple raw gas concentration data include infrared reactive gas concentration and toxic gas concentration; the method further includes:

[0020] When the abnormal gas concentration data is the concentration of toxic gas, the interfering gas concentration data that crosses with the toxic gas concentration is identified as the infrared active gas concentration.

[0021] The probability of change in infrared active gas concentration is determined based on the second dynamic baseline and the second fluctuation range corresponding to the infrared active gas concentration.

[0022] When the probability of change in infrared active gas concentration is within the range of change, the abnormal gas concentration data is determined to be false alarm data, and the range of change is determined according to the change threshold.

[0023] In one embodiment, anomaly detection is performed on multiple raw gas concentration data to obtain abnormal gas concentration data, including:

[0024] For each type of raw gas concentration data, determine the slope corresponding to the raw gas concentration data, and compare the slope corresponding to the raw gas concentration data with the slope threshold.

[0025] The raw gas concentration data is compared with the normal amplitude; the normal amplitude is determined based on the third dynamic baseline and the third fluctuation range corresponding to the raw gas concentration data.

[0026] If the slope of the original gas concentration data is greater than the slope threshold and the original gas concentration data exceeds the normal amplitude, the original gas concentration data is determined to be abnormal gas concentration data.

[0027] In one embodiment, the method further includes, prior to anomaly detection of multiple raw gas concentration data:

[0028] Obtain laboratory temperature data;

[0029] The zero-point offset compensation value is determined based on the preset temperature zero-point drift model and temperature data;

[0030] Temperature compensation is performed on each original gas concentration data based on the zero-point offset compensation value to obtain the compensated gas concentration data.

[0031] Correspondingly, anomaly detection is performed on various raw gas concentration data to obtain anomalous gas concentration data, including:

[0032] Anomaly detection is performed on the compensated gas concentration data to obtain abnormal gas concentration data.

[0033] In one embodiment, the method further includes:

[0034] If the abnormal gas concentration data is genuinely abnormal, activate the emergency ventilation system and send an early warning message.

[0035] Secondly, this application also provides a laboratory gas detection device, which includes:

[0036] The array sensor includes multiple types of gas sensors, which are used to detect gas concentrations in the laboratory, obtain multiple raw gas concentration data, and send the multiple raw gas concentration data to the cross-validation processor.

[0037] A cross-validation processor is used to detect anomalies in multiple raw gas concentration data to obtain abnormal gas concentration data; identify interfering gas concentration data that cross-interferes with the abnormal gas concentration data; determine the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data; and determine the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data. The first dynamic baseline is determined based on the ratio of the abnormal gas concentration data to the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory. The contamination probability is determined based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change of the interfering gas concentration data, and the contamination probability is used to determine whether the abnormal gas concentration data is genuine abnormal data.

[0038] In one embodiment, an array of sensors is arranged on the top of the laboratory, including infrared sensors, electrochemical sensors, and optical sensors arranged in layers from bottom to top. The infrared sensors, electrochemical sensors, and optical sensors are used to detect the concentration of infrared reactive gas, the concentration of toxic gas, and the concentration of volatile organic compounds, respectively.

[0039] The array sensor features a modular interface for detachable insertion of various types of gas sensors.

[0040] In one embodiment, the laboratory gas detection device further includes:

[0041] The self-cleaning device is located at the rear end of the array sensor, close to the optical sensor, and is used to activate a reverse airflow to purge the array sensor.

[0042] An air sampling port is located at the front end of the array sensor, close to the infrared sensor, and isolated from the self-cleaning device, for collecting laboratory gases;

[0043] A dustproof net is placed at the front end of the air sampling port to block particulate matter.

[0044] Thirdly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0045] Acquire multiple raw gas concentration data in the laboratory, perform anomaly detection on the multiple raw gas concentration data, and obtain abnormal gas concentration data;

[0046] Identify interfering gas concentration data that cross-interferes with abnormal gas concentration data. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, determine the probability of change of the interfering gas concentration data. The first dynamic baseline is determined based on the ratio of abnormal gas concentration data to interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory.

[0047] The probability of contamination is determined based on the probability of abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in interfering gas concentration data. The probability of contamination is then used to determine whether the abnormal gas concentration data is genuine abnormal data.

[0048] Fourthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0049] Acquire multiple raw gas concentration data in the laboratory, perform anomaly detection on the multiple raw gas concentration data, and obtain abnormal gas concentration data;

[0050] Identify interfering gas concentration data that cross-interferes with abnormal gas concentration data. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, determine the probability of change of the interfering gas concentration data. The first dynamic baseline is determined based on the ratio of abnormal gas concentration data to interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory.

[0051] The probability of contamination is determined based on the probability of abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in interfering gas concentration data. The probability of contamination is then used to determine whether the abnormal gas concentration data is genuine abnormal data.

[0052] The aforementioned laboratory gas detection methods, equipment, computer-readable storage media, and computer program products detect multiple gas concentration data using various types of sensors, enabling simultaneous and accurate detection of pollutants of different properties. They identify interfering gas concentration data that cross-interferes with abnormal gas concentration data. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, they determine the probability of the abnormal gas concentration data deviating from the first dynamic baseline. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, they determine the probability of change in the interfering gas concentration data. The pollution probability is determined based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in the interfering gas concentration data. The pollution probability is then used to determine whether the abnormal gas concentration data is truly abnormal. By performing probability fusion judgment on gases that cross-interfere in the laboratory, gas cross-verification is achieved to eliminate interference errors, improve gas detection accuracy, reduce the false alarm rate caused by interfering gases, and meet the needs of precise monitoring. Furthermore, the first dynamic baseline is determined based on the ratio of abnormal gas concentration data to interfering gas concentration data within a preset time window in the clean area of ​​the laboratory. This ratio can offset the common dependent variables of the abnormal gas concentration data and interfering gas concentration data, thus amplifying pollution characteristics. By collecting data within a preset sliding time window to determine a dynamic baseline, the system can dynamically adapt to the environment, provide real-time statistical references, and achieve more stable and accurate gas anomaly detection. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a schematic diagram of the structure of a laboratory gas detection device in one embodiment;

[0055] Figure 2 This is a flowchart illustrating a laboratory gas detection method in one embodiment;

[0056] Figure 3 This is a flowchart illustrating the cross-verification process of toxic gas concentration and infrared reactive gas concentration in one embodiment.

[0057] Figure 4 This is a flowchart illustrating a laboratory gas detection method in another embodiment. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0059] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0060] The laboratory gas detection method provided in this application embodiment can be applied to, for example... Figure 1 The laboratory gas detection equipment shown includes an array sensor 102 and a cross-validation processor 104. The array sensor 102 includes various types of gas sensors, each used to detect gas concentrations in the laboratory, obtaining multiple raw gas concentration data, which are then sent to the cross-validation processor 104. The cross-validation processor 104 performs anomaly detection on the raw gas concentration data, obtaining abnormal gas concentration data; identifies interfering gas concentration data that cross-interferes with the abnormal gas concentration data; determines the probability that the abnormal gas concentration data deviates from the first dynamic baseline based on the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data; and determines the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data. The first dynamic baseline is determined based on the ratio of the abnormal gas concentration data to the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory. The probability of contamination is determined based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change of the interfering gas concentration data; and the probability of contamination is used to determine whether the abnormal gas concentration data is genuine abnormal data. The cross-validation processor 104 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and servers.

[0061] In one exemplary embodiment, such as Figure 2 As shown, a laboratory gas detection method is provided, which can be applied to... Figure 1 The cross-validation processor in the example is used to illustrate the process, including steps 202 to 206. Wherein:

[0062] Step 202: Obtain various raw gas concentration data in the laboratory, perform anomaly detection on the various raw gas concentration data, and obtain abnormal gas concentration data.

[0063] Here, raw gas concentration data refers to the initial gas concentration data detected. Abnormal gas concentration data refers to the abnormal gas concentration data obtained by performing anomaly detection on each type of raw gas concentration data separately.

[0064] In some implementations, multiple types of gas sensors in the array sensor perform gas concentration detection in the laboratory, obtaining multiple raw gas concentration data, which are then sent to a cross-validation processor. The cross-validation processor performs anomaly detection on the multiple raw gas concentration data to obtain abnormal gas concentration data. Anomaly detection can be performed by comparing a fixed threshold; when the threshold is exceeded, the gas concentration data is determined to be abnormal.

[0065] Step 204: Identify interfering gas concentration data that cross-interferes with the abnormal gas concentration data; determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline based on the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data; determine the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data; the first dynamic baseline is determined based on the ratio of the abnormal gas concentration data to the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory.

[0066] Interfering gas concentration data refers to gas concentration data that can affect abnormal gas concentration data. Abnormal gas concentration data is usually accompanied by fluctuations in interfering gas concentration data, making it difficult to distinguish using a single sensor. The first dynamic baseline refers to the normal reference value of the abnormal gas concentration data that is continuously updated over time.

[0067] The first fluctuation range refers to the normal fluctuation range near the first dynamic baseline. The probability of abnormal gas concentration data deviating from the first dynamic baseline refers to the likelihood or confidence level of the currently detected abnormal gas concentration data showing an abnormal change compared to the first dynamic baseline. The second dynamic baseline refers to the normal reference value of the interfering gas concentration data, which is continuously updated over time. The second fluctuation range refers to the normal fluctuation range near the second dynamic baseline. The probability of change in interfering gas concentration data refers to the likelihood or confidence level of the currently detected interfering gas concentration data showing an abnormal change compared to the second dynamic baseline.

[0068] In some implementations, interfering gas concentration data that cross-interferes with the abnormal gas concentration data is identified. The ratio of abnormal gas concentration data to interfering gas concentration data within a preset sliding time window is obtained in a clean area of ​​the laboratory, and the mean of the ratio is calculated to obtain a first dynamic baseline. Here, a clean area refers to an area free from pollution or with the most stable air quality. For example, the width of the preset sliding time window can be a historical one-hour period. The standard deviation of the ratio is calculated to obtain a first fluctuation range. Based on the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data, the probability of the abnormal gas concentration data deviating from the first dynamic baseline is determined. The ratio of the current abnormal gas concentration data to the interfering gas concentration data is obtained, and a first standardized deviation value is determined based on this ratio, the first dynamic baseline, and the first fluctuation range. The first standardized deviation value can be calculated using the following formula:

[0069]

[0070] in, Indicates the first standardized deviation value. This represents the ratio of the current abnormal gas concentration data to the interfering gas concentration data. Indicates the first dynamic baseline. This indicates the first fluctuation range.

[0071] The first standardized deviation is converted into the probability of the anomalous gas concentration data deviating from the first dynamic baseline using methods such as statistical tail probability and confidence / outlier mapping. For example, the conversion formula for statistical tail probability can be:

[0072]

[0073]

[0074] in, This represents the two-tailed p-value corresponding to the first standardized deviation (assuming normality), where Φ is the cumulative distribution function of the standard normal distribution. This indicates the probability that the abnormal gas concentration data deviates from the first dynamic baseline.

[0075] The mean of the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory is calculated as the second dynamic baseline. The standard deviation of the interfering gas concentration data within the preset sliding time window in the clean area of ​​the laboratory is calculated as the second fluctuation range. A second standardized deviation is determined based on the current interfering gas concentration data, the second dynamic baseline, and the second fluctuation range. For example, the second standardized deviation can be calculated using the following formula:

[0076]

[0077] in, This represents the second standardized deviation. This indicates the current concentration data of interfering gases. Indicates the second dynamic baseline. This indicates the second fluctuation range.

[0078] The second standardized deviation is converted into the probability of change in the interfering gas concentration data using methods such as statistical tail probability and confidence / outlier mapping. For example, the conversion formula for statistical tail probability can be:

[0079]

[0080]

[0081] in, This represents the two-tailed p-value corresponding to the second standardized deviation (assuming normality), where Φ is the cumulative distribution function of the standard normal distribution. This indicates the probability of changes in the concentration data of interfering gases.

[0082] Step 206: Determine the pollution probability based on the probability that the abnormal gas concentration data deviates from the first dynamic baseline and the probability of change in the interfering gas concentration data, and determine whether the abnormal gas concentration data is real abnormal data based on the pollution probability.

[0083] The pollution probability refers to the probability of anomalies in the abnormal gas concentration data determined after cross-validation.

[0084] In some implementations, the probability of abnormal gas concentration data deviating from a first dynamic baseline and the probability of change in interfering gas concentration data are used to determine the pollution probability, which is then subjected to probability fusion evaluation to obtain the pollution probability. If the pollution probability is greater than a pollution threshold, the abnormal gas concentration data is determined to be genuine abnormal data. If the pollution probability is less than or equal to the pollution threshold, the abnormal gas concentration data is determined not to be genuine abnormal data. The abnormal gas concentration data is then stored. For example, the pollution threshold can be 0.8.

[0085] In traditional technologies, laboratories generally use gas detectors based on a single detection principle. A single detection principle makes it difficult to accurately detect pollutants of different properties, such as acidic gases and volatile organic compounds, at the same time. Furthermore, it lacks anti-cross-interference mechanisms. Common laboratory solvents (alcohol, acetone) can significantly affect the readings of volatile organic compounds and combustible gases, resulting in low air detection accuracy, high false alarm rate, and inability to meet the needs of precise monitoring.

[0086] The aforementioned laboratory gas detection method utilizes multiple types of sensors to detect various gas concentrations, enabling simultaneous and accurate detection of pollutants with different properties. It identifies interfering gas concentration data that cross-interferes with abnormal gas concentration data. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, it determines the probability of the abnormal gas concentration data deviating from the first dynamic baseline. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, it determines the probability of change in the interfering gas concentration data. The pollution probability is then determined based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in the interfering gas concentration data. Finally, the pollution probability is used to determine whether the abnormal gas concentration data is genuinely abnormal. By performing probability fusion judgment on gases with cross-interference in the laboratory, gas cross-verification is achieved to eliminate interference errors, improve gas detection accuracy, reduce false alarm rates caused by interfering gases, and meet the needs of precise monitoring. Furthermore, the first dynamic baseline is determined based on the ratio of abnormal gas concentration data to interfering gas concentration data within a preset time window in the clean area of ​​the laboratory. This ratio can offset the common dependent variables of the abnormal gas concentration data and interfering gas concentration data, thus mitigating the amplification of pollution characteristics. By collecting data within a preset sliding time window to determine a dynamic baseline, the system can dynamically adapt to the environment, provide real-time statistical references, and achieve more stable and accurate gas anomaly detection.

[0087] In one exemplary embodiment, the various raw gas concentration data include infrared reactive gas concentration and volatile organic compound (VOC) concentration; the abnormal gas concentration data is VOC concentration.

[0088] Identify interfering gas concentration data that cross-interfere with the abnormal gas concentration data. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, determine the probability of change of the interfering gas concentration data, including:

[0089] The concentration data of interfering gases that cross-interfere with the concentration of volatile organic compounds are identified as the concentration of infrared active gases; the probability of the concentration of volatile organic compounds deviating from the first dynamic baseline is determined based on the first dynamic baseline and the first fluctuation range corresponding to the concentration of volatile organic compounds; the probability of change of the concentration of infrared active gases is determined based on the second dynamic baseline and the second fluctuation range corresponding to the concentration of infrared active gases.

[0090] In some implementations, the concentration of infrared reactive gases detected by an infrared sensor is obtained, such as carbon dioxide or methane. The concentration of volatile organic compounds (VOCs) detected by an optical sensor is also obtained; the optical sensor detects the total concentration of VOCs.

[0091] The interfering gas concentration data that cross-interferes with the volatile organic compound (VOC) concentration is identified as the carbon dioxide concentration in the infrared reactive gas concentration. Based on the first dynamic baseline and first fluctuation range corresponding to the VOC concentration, the probability of the VOC concentration deviating from the first dynamic baseline is determined. Based on the second dynamic baseline and second fluctuation range corresponding to the carbon dioxide concentration in the infrared reactive gas concentration, the probability of change in the carbon dioxide concentration is determined.

[0092] In the embodiments of this application, when there is an abnormality in the concentration of volatile organic compounds, since the concentration of volatile organic compounds generated by experimental operations is usually accompanied by fluctuations in carbon dioxide, the two interfere with each other. By performing probability fusion judgment on the concentration of volatile organic compounds and the concentration of carbon dioxide, interference errors can be eliminated, the detection accuracy of volatile organic compound concentration can be improved, and the false alarm rate caused by carbon dioxide can be reduced.

[0093] In an exemplary embodiment, determining the pollution probability based on the probability of abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in interfering gas concentration data includes: obtaining preset deviation coefficients and change coefficients; weighting the probability of abnormal gas concentration data deviating from the first dynamic baseline based on the deviation coefficients to obtain a first weighted term; weighting the probability of change in interfering gas concentration data based on the change coefficients to obtain a second weighted term; and summing the first weighted term and the second weighted term to obtain the pollution probability.

[0094] In some implementations, preset deviation and variation coefficients are obtained. The deviation coefficient is a weighted coefficient corresponding to the probability that the abnormal gas concentration data deviates from the first dynamic baseline, and the variation coefficient is a weighted coefficient corresponding to the probability of change in the interfering gas concentration data. Based on the deviation and variation coefficients, the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in the interfering gas concentration data are weighted and summed to obtain the pollution probability. Wherein, pollution probability = W1 × P v +W2×P c Where W1 and W2 represent the deviation coefficient and the variation coefficient, respectively. P c A value close to 1 indicates that the interfering gas concentration data has deviated significantly from the second dynamic baseline, and that the interfering gas concentration data changes synchronously with the abnormal gas concentration data. The sensor selection, deviation coefficient, and variation coefficient depend on the specific application and the level of risk. For example, the deviation coefficient and variation coefficient can be set to 0.7 and 0.3, respectively.

[0095] In this embodiment, the probability of abnormal gas concentration data deviating from the first dynamic baseline is weighted and fused with the probability of change in interfering gas concentration data. By combining the deviation probability of abnormal gas concentration data with the synchronous change of interfering gases and quantifying them uniformly using a probabilistic model, false alarms can be reduced, the detection rate of real abnormal events can be improved, and the deviation coefficient and change coefficient can be flexibly adjusted according to the scenario to achieve flexible adjustment of anomaly detection sensitivity.

[0096] In one exemplary embodiment, such as Figure 3 As shown, the original gas concentration data includes infrared reactive gas concentration and toxic gas concentration; the method also includes a cross-validation step between the toxic gas concentration and the infrared reactive gas concentration, including:

[0097] Step 302: When the abnormal gas concentration data is a toxic gas concentration, determine the interfering gas concentration data that crosses with the toxic gas concentration as the infrared active gas concentration.

[0098] Step 304: Determine the probability of change in infrared active gas concentration based on the second dynamic baseline and the second fluctuation range corresponding to the infrared active gas concentration.

[0099] Step 306: When the probability of change in infrared active gas concentration is within the range of change, the abnormal gas concentration data is determined to be false alarm data, and the range of change is determined according to the change threshold.

[0100] In some embodiments, the concentration of toxic gases detected by an electrochemical sensor is obtained, such as concentrations of toxic gases including hydrogen sulfide or chlorine. The concentration of infrared-reactive gases detected by an infrared sensor is also obtained, such as concentrations of infrared-reactive gases including carbon dioxide or methane.

[0101] When the concentration of toxic gas becomes abnormal, the concentration of carbon dioxide in the infrared reactive gas concentration cross-interferes with the toxic gas concentration. The probability of change in carbon dioxide concentration is determined based on the second dynamic baseline and the second fluctuation range corresponding to the carbon dioxide concentration. The probability of change in carbon dioxide concentration is compared with the fluctuation range. If the probability of change in carbon dioxide concentration is within the fluctuation range, it indicates that the carbon dioxide concentration has deviated significantly from the second dynamic baseline, and the carbon dioxide concentration changes synchronously with the toxic gas concentration. This suggests that the carbon dioxide concentration may be causing the abnormal toxic gas concentration, and the abnormal toxic gas concentration is determined to be false alarm data. The fluctuation range is determined based on a fluctuation threshold. For example, if the fluctuation threshold is 1, the fluctuation range is 0.9 to 1. Furthermore, when the abnormal toxic gas concentration is determined to be false alarm data, a power-off protection mechanism for the electrochemical sensor can be implemented if necessary.

[0102] In this embodiment, when the abnormal gas concentration data is a toxic gas concentration, the probability of change in the infrared active gas concentration is determined based on the second dynamic baseline and the second fluctuation range corresponding to the infrared active gas concentration, which cross-interferes with the toxic gas concentration. When the probability of change in the infrared active gas concentration is within the fluctuation range, it indicates that the abnormal toxic gas concentration may be caused by the carbon dioxide concentration. Therefore, the abnormal toxic gas concentration is determined to be false alarm data, effectively avoiding false alarms of abnormal toxic gas concentrations.

[0103] In an exemplary embodiment, anomaly detection is performed on multiple raw gas concentration data to obtain abnormal gas concentration data, including: for each type of raw gas concentration data, determining the slope corresponding to the raw gas concentration data, comparing the slope corresponding to the raw gas concentration data with a slope threshold; comparing the raw gas concentration data with a normal amplitude, where the normal amplitude is determined based on the third dynamic baseline and the third fluctuation range corresponding to the raw gas concentration data; and determining the raw gas concentration data as abnormal gas concentration data when the slope corresponding to the raw gas concentration data is greater than the slope threshold and the raw gas concentration data exceeds the normal amplitude.

[0104] The slope corresponding to the original gas concentration data refers to the rate of change of the original gas concentration data per unit time. The third dynamic baseline is the normal reference value of the original gas concentration data that is continuously updated over time. The third fluctuation range refers to the normal fluctuation range near the third dynamic baseline.

[0105] In some implementations, a slope-amplitude dual-factor judgment can be introduced to determine the anomaly of raw gas concentration data detected by a single sensor. Specifically, for each type of raw gas concentration data, the slope corresponding to the raw gas concentration data is determined and compared with a slope threshold. For example, the slope threshold can be 200 ppb / min. The normal amplitude is determined based on the third dynamic baseline and the third fluctuation range corresponding to the raw gas concentration data. For example, the normal amplitude can be the third dynamic baseline + 3 of the third fluctuation range. If the slope corresponding to the raw gas concentration data is greater than the slope threshold and the raw gas concentration data exceeds the normal amplitude, the raw gas concentration data is determined to meet the anomaly condition, and the raw gas concentration data is considered abnormal gas concentration data.

[0106] For example, the outlier condition is (slope of the original gas concentration data > slope threshold) and (original gas concentration data > third dynamic baseline + 3 third fluctuation range), where the third dynamic baseline is updated every 10 minutes and the third fluctuation range is the recent standard deviation.

[0107] In this embodiment, a slope-amplitude dual-factor method is introduced to determine whether there are abnormalities in the raw gas concentration data detected by a single sensor. The slope factor can detect abnormalities that change rapidly but have not yet exceeded the normal amplitude, while the amplitude factor can detect abnormalities with extremely high absolute values. The combination of the two can simultaneously capture both chronic accumulation-type and instantaneous sudden-type abnormalities, reducing the probability of missed detection.

[0108] In an exemplary embodiment, before performing anomaly detection on multiple raw gas concentration data, the method further includes: a step of temperature compensation for the raw gas concentration data, including: acquiring laboratory temperature data; determining a zero-point offset compensation value based on a preset temperature zero-point drift model and the temperature data; performing temperature compensation on each type of raw gas concentration data based on the zero-point offset compensation value to obtain compensated gas concentration data; correspondingly, performing anomaly detection on multiple raw gas concentration data to obtain abnormal gas concentration data includes: performing anomaly detection on the compensated gas concentration data to obtain abnormal gas concentration data.

[0109] In some implementations, changes in ambient temperature can significantly affect sensor performance. For example, the output of an electrochemical sensor increases with rising temperature, which may cause the sensor output to shift, resulting in zero-point drift. Zero-point drift refers to the phenomenon where the sensor output is not zero when the concentration of the measured gas is zero (or the background air concentration). Temperature compensation is a crucial measure to ensure the accuracy of sensor detection data.

[0110] Acquire laboratory temperature data. Determine the zero-point offset compensation value based on a pre-defined temperature zero-point drift model and the temperature data. The temperature zero-point drift model is derived by placing the sensor under standard gas (zero concentration) conditions at different temperatures, recording the sensor's zero-point output value at each temperature, and then fitting this zero-point output value to a curve. For example, the temperature zero-point drift model can be a linear model, a quadratic polynomial model, or a piecewise linear interpolation model. Input the laboratory temperature data into the temperature zero-point drift model to obtain the zero-point offset compensation value. Summate the original gas concentration data and the zero-point offset compensation value to achieve temperature compensation, obtaining the compensated gas concentration data. Then, perform anomaly detection on the compensated gas concentration data to obtain abnormal gas concentration data.

[0111] In this embodiment, by considering the impact of temperature changes on the original gas concentration data, a zero-point offset compensation value is determined based on a preset temperature zero-point drift model and laboratory temperature data. Temperature compensation is then applied to the original gas concentration data based on the zero-point offset compensation value, which can avoid concentration detection errors caused by zero-point drift and improve sensor detection accuracy.

[0112] In one exemplary embodiment, the method further includes: activating an emergency ventilation system and sending an early warning message when the abnormal gas concentration data is genuinely abnormal.

[0113] In the event that the abnormal gas concentration data is genuinely abnormal, the emergency ventilation system is activated via a differential serial communication standard bus, and a warning message is simultaneously sent to the designated responsible personnel via the communication module. The communication module can support GSM (Global System for Mobile Communications) or Wi-Fi transmission. In the event of a genuine abnormal event, timely ventilation and warning messages can be sent, effectively shortening emergency response time and reducing the risk of poisoning among personnel in the laboratory area.

[0114] In another exemplary embodiment, such as Figure 4 As shown, a gas detection method for a testing instrument is provided, including:

[0115] Step 402: Obtain laboratory temperature data; determine zero-point offset compensation value based on preset temperature zero-point drift model and temperature data; obtain multiple original gas concentration data in the laboratory, and perform temperature compensation on each original gas concentration data according to the zero-point offset compensation value to obtain compensated gas concentration data.

[0116] Step 404: For each type of compensated gas concentration data, determine the slope corresponding to the compensated gas concentration data, compare the slope corresponding to the compensated gas concentration data with the slope threshold, compare the compensated gas concentration data with the normal amplitude, the normal amplitude is determined based on the third dynamic base line and the third fluctuation range corresponding to the original gas concentration data, and if the slope corresponding to the compensated gas concentration data is greater than the slope threshold and the compensated gas concentration data exceeds the normal amplitude, the compensated gas concentration data is determined to be abnormal gas concentration data.

[0117] Step 406: Identify interfering gas concentration data that cross-interferes with the abnormal gas concentration data; determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline based on the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data; determine the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data; the first dynamic baseline is determined based on the ratio of the abnormal gas concentration data to the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory.

[0118] Step 408: Obtain the preset deviation coefficient and change coefficient; weight the probability of the abnormal gas concentration data deviating from the first dynamic baseline according to the deviation coefficient to obtain the first weighted term; weight the probability of the change of the interfering gas concentration data according to the change coefficient to obtain the second weighted term; sum the first weighted term and the second weighted term to obtain the pollution probability; determine whether the abnormal gas concentration data is real abnormal data based on the pollution probability.

[0119] Step 410: If the abnormal gas concentration data is indeed abnormal, activate the emergency ventilation system and send an early warning message.

[0120] In this embodiment, multiple types of sensors are used to detect various gas concentration data, enabling simultaneous and accurate detection of pollutants with different properties. Interfering gas concentration data that cross-interferes with abnormal gas concentration data is identified. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, the probability of the abnormal gas concentration data deviating from the first dynamic baseline is determined. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, the probability of change of the interfering gas concentration data is determined. The pollution probability is determined based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change of the interfering gas concentration data. The pollution probability is then used to determine whether the abnormal gas concentration data is truly abnormal. By performing probability fusion judgment on gases that cross-interfere in the laboratory, gas cross-verification is achieved to eliminate interference errors, improve gas detection accuracy, reduce the false alarm rate caused by interfering gases, and meet the requirements of precise monitoring. Furthermore, the first dynamic baseline is determined based on the ratio of abnormal gas concentration data to interfering gas concentration data within a preset time window in the clean area of ​​the laboratory. This ratio can offset the common dependent variables of the abnormal gas concentration data and interfering gas concentration data, thus amplifying pollution characteristics. By collecting data within a preset sliding time window to determine a dynamic baseline, the system can dynamically adapt to the environment, provide real-time statistical references, and achieve more stable and accurate gas anomaly detection.

[0121] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0122] Based on the same inventive concept, this application also provides a laboratory gas detection device for implementing the laboratory gas detection method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more laboratory gas detection device embodiments provided below can be found in the limitations of the laboratory gas detection method described above, and will not be repeated here.

[0123] In one exemplary embodiment, such as Figure 1 As shown, a laboratory gas detection device is provided, comprising:

[0124] The array sensor includes multiple types of gas sensors, which are used to detect gas concentrations in the laboratory, obtain multiple raw gas concentration data, and send the multiple raw gas concentration data to the cross-validation processor.

[0125] A cross-validation processor is used to detect anomalies in multiple raw gas concentration data to obtain abnormal gas concentration data; identify interfering gas concentration data that cross-interferes with the abnormal gas concentration data; determine the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data; and determine the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data. The first dynamic baseline is determined based on the ratio of the abnormal gas concentration data to the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory. The contamination probability is determined based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change of the interfering gas concentration data, and the contamination probability is used to determine whether the abnormal gas concentration data is genuine abnormal data.

[0126] In this embodiment, multiple types of sensors are used to detect various gas concentration data, enabling simultaneous and accurate detection of pollutants with different properties. Interfering gas concentration data that cross-interferes with abnormal gas concentration data is identified. Based on the first dynamic baseline and first fluctuation range corresponding to the abnormal gas concentration data, the probability of the abnormal gas concentration data deviating from the first dynamic baseline is determined. Based on the second dynamic baseline and second fluctuation range corresponding to the interfering gas concentration data, the probability of change of the interfering gas concentration data is determined. The pollution probability is determined based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change of the interfering gas concentration data. The pollution probability is then used to determine whether the abnormal gas concentration data is truly abnormal. By performing probability fusion judgment on gases that cross-interfere in the laboratory, gas cross-verification is achieved to eliminate interference errors, improve gas detection accuracy, reduce the false alarm rate caused by interfering gases, and meet the requirements of precise monitoring. Furthermore, the first dynamic baseline is determined based on the ratio of abnormal gas concentration data to interfering gas concentration data within a preset time window in the clean area of ​​the laboratory. This ratio can offset the common dependent variables of the abnormal gas concentration data and interfering gas concentration data, thus amplifying pollution characteristics. By collecting data within a preset sliding time window to determine a dynamic baseline, the system can dynamically adapt to the environment, provide real-time statistical references, and achieve more stable and accurate gas anomaly detection.

[0127] In one exemplary embodiment, an array of sensors is arranged on the top of the laboratory, including infrared sensors, electrochemical sensors, and optical sensors arranged in layers from bottom to top. The infrared sensors, electrochemical sensors, and optical sensors are used to detect the concentration of infrared reactive gas, the concentration of toxic gas, and the concentration of volatile organic compounds, respectively.

[0128] The array sensor features a modular interface for detachable insertion of various types of gas sensors.

[0129] In some embodiments, an array of sensors, including infrared sensors, electrochemical sensors, and optical sensors arranged in layers from bottom to top, are used to detect the concentrations of infrared reactive gases, toxic gases, and volatile organic compounds, respectively.

[0130] The array sensor features a modular interface for detachably inserting various types of gas sensors, allowing for flexible sensor combinations. Furthermore, appropriate sensor combinations can be selected based on specific application scenarios. For example, in a biological laboratory, infrared and electrochemical sensors can be selected to detect carbon dioxide and ammonia concentrations, respectively. In a lithium battery laboratory, electrochemical and optical sensors can be selected to detect hydrogen fluoride and organophosphorus concentrations, respectively. In a nanomaterials laboratory, photoionization and optical particulate matter sensors, such as a PM0.1 optical counter, can be selected to detect ultrafine particulate matter and metallic fume concentrations, respectively. The scalable design meets the customized detection needs of different laboratories.

[0131] Furthermore, electrochemical sensors can be replaced with semiconductor sensors. Modular interfaces can be replaced with fixed integrated designs to reduce costs.

[0132] In this embodiment, by integrating infrared sensors, electrochemical sensors, and optical sensors, the concentrations of infrared reactive gases, toxic gases, and volatile organic compounds are detected, respectively, enabling the simultaneous detection of pollutants with multiple properties. The array sensor is equipped with a modular interface for detachably inserting various types of gas sensors, allowing for flexible sensor combinations.

[0133] In one exemplary embodiment, the laboratory gas detection device further includes: a self-cleaning device disposed at the rear end of the array sensor, close to the optical sensor, for initiating a reverse airflow to purge the array sensor; an air sampling port disposed at the front end of the array sensor, close to the infrared sensor, isolated from the self-cleaning device, for collecting laboratory gas; and a dustproof net disposed at the front end of the air sampling port to block particulate matter.

[0134] In some implementations, the front end of the array sensor is its lower end, and the rear end is its upper end. An air sampling port is located at the front end of the array sensor, close to the infrared sensor, for collecting laboratory gases. The collected gas then enters the array sensor for gas concentration detection. A self-cleaning device is located at the rear end of the array sensor. The reverse airflow channel of the self-cleaning device is located at the rear end of the array sensor, and the airflow direction is controlled by a solenoid valve to achieve timed operation of sampling, detection, and cleaning, preventing the purge airflow from contaminating the sampled air. A dust filter is also installed at the front end of the air sampling port to block particulate matter. For example, the self-cleaning device can be a miniature air pump that activates a reverse airflow (flow rate 0.5 L / min) every 24 hours to purge the array sensor cavity, working in conjunction with an electrostatic dust filter to block particulate matter.

[0135] Furthermore, a pretreatment unit is arranged after the air sampling port to standardize the sampled gas for the sensor, ensuring its cleanliness and consistent temperature and humidity to reduce sensor interference and extend sensor lifespan. This includes two stages of processing: a first-stage cyclone separator removes particles >10μm (centrifugal force 1200g), and a second-stage semiconductor cooler stabilizes the gas temperature to 25±0.5℃ and condenses and dehumidifies (dew point temperature -10℃). Its core function is to eliminate the interference of dust and moisture on the optical / electrochemical sensor, ensuring that the detection environment meets the sensor calibration conditions, which is the foundation for the comparability of multi-sensor data.

[0136] Furthermore, self-cleaning devices can be simplified to baffle protection to reduce costs.

[0137] In traditional technologies, sensors are susceptible to contamination, leading to decreased sensitivity and irreversible damage, requiring frequent manual calibration (typically once a week), resulting in high maintenance costs. However, in this embodiment, by integrating a self-cleaning device to extend the calibration cycle, the sensor's lifespan is effectively extended, and the frequency of manual calibration is reduced.

[0138] Each module in the aforementioned laboratory gas detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0139] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0140] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0141] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0142] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0143] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0144] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A laboratory gas detection method, characterized in that, The method includes: Acquire multiple raw gas concentration data in the laboratory, perform anomaly detection on the multiple raw gas concentration data, and obtain abnormal gas concentration data; Identify interfering gas concentration data that cross-interferes with the abnormal gas concentration data; determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline based on the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data; determine the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data; the first dynamic baseline is determined based on the ratio of the abnormal gas concentration data to the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory. The probability of contamination is determined based on the probability that the abnormal gas concentration data deviates from the first dynamic baseline and the probability of change of the interfering gas concentration data. The probability of contamination is then used to determine whether the abnormal gas concentration data is genuine abnormal data.

2. The method according to claim 1, characterized in that, The original gas concentration data includes infrared reactive gas concentration and volatile organic compound (VOC) concentration; the abnormal gas concentration data is the VOC concentration. The step of identifying interfering gas concentration data that crosses with the abnormal gas concentration data, determining the probability that the abnormal gas concentration data deviates from the first dynamic baseline based on the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data, and determining the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data includes: The concentration data of the interfering gas that cross-interferes with the concentration of the volatile organic compounds is determined as the concentration of the infrared active gas; Based on the first dynamic baseline and the first fluctuation range corresponding to the volatile organic compound concentration, the probability that the volatile organic compound concentration deviates from the first dynamic baseline is determined; The probability of change in the infrared active gas concentration is determined based on the second dynamic baseline and the second fluctuation range corresponding to the infrared active gas concentration.

3. The method according to claim 1, characterized in that, The step of determining the pollution probability based on the probability of the abnormal gas concentration data deviating from the first dynamic baseline and the probability of change in the interfering gas concentration data includes: Obtain the preset deviation coefficient and variation coefficient; The probability of the abnormal gas concentration data deviating from the first dynamic baseline is weighted according to the deviation coefficient to obtain the first weighting term; The probability of change in the concentration data of the interfering gas is weighted according to the change coefficient to obtain a second weighting term; The first weighted term and the second weighted term are summed to obtain the pollution probability.

4. The method according to claim 1, characterized in that, The raw gas concentration data includes infrared reactive gas concentration and toxic gas concentration; the method further includes: When the abnormal gas concentration data is the concentration of the toxic gas, the interfering gas concentration data that crosses with the concentration of the toxic gas is determined to be the infrared active gas concentration. The probability of change of the infrared active gas concentration is determined based on the second dynamic baseline and the second fluctuation range corresponding to the infrared active gas concentration. When the probability of change in the infrared active gas concentration is within the range of change, the abnormal gas concentration data is determined to be false alarm data, and the range of change is determined according to the change threshold.

5. The method according to claim 1, characterized in that, The abnormal gas concentration data obtained by performing anomaly detection on various original gas concentration data includes: For each type of raw gas concentration data, determine the slope corresponding to the raw gas concentration data, and compare the slope corresponding to the raw gas concentration data with a slope threshold. The original gas concentration data is compared with the normal amplitude; the normal amplitude is determined based on the third dynamic baseline and the third fluctuation range corresponding to the original gas concentration data. If the slope corresponding to the original gas concentration data is greater than the slope threshold and the original gas concentration data exceeds the normal amplitude, the original gas concentration data is determined to be abnormal gas concentration data.

6. The method according to claim 1, characterized in that, Prior to performing anomaly detection on the various raw gas concentration data, the method further includes: Acquire the temperature data of the laboratory; The zero-point offset compensation value is determined based on the preset temperature zero-point drift model and the temperature data. Temperature compensation is performed on each of the original gas concentration data based on the zero-point offset compensation value to obtain the compensated gas concentration data. Correspondingly, the anomaly detection of various original gas concentration data to obtain abnormal gas concentration data includes: Anomaly detection is performed on the compensated gas concentration data to obtain abnormal gas concentration data.

7. The method according to claim 1, characterized in that, The method further includes: If the abnormal gas concentration data is indeed abnormal, the emergency ventilation system will be activated and an early warning message will be sent.

8. A laboratory gas detection device, characterized in that, The laboratory gas detection equipment includes: An array sensor, including multiple types of gas sensors, is used to detect gas concentrations in the laboratory, obtain multiple raw gas concentration data, and send the multiple raw gas concentration data to a cross-validation processor; A cross-validation processor is used to perform anomaly detection on multiple raw gas concentration data to obtain abnormal gas concentration data; identify interfering gas concentration data that cross-interferes with the abnormal gas concentration data; determine the probability that the abnormal gas concentration data deviates from the first dynamic baseline based on the first dynamic baseline and the first fluctuation range corresponding to the abnormal gas concentration data; determine the probability of change of the interfering gas concentration data based on the second dynamic baseline and the second fluctuation range corresponding to the interfering gas concentration data; the first dynamic baseline is determined based on the ratio of the abnormal gas concentration data to the interfering gas concentration data within a preset sliding time window in the clean area of ​​the laboratory; determine the contamination probability based on the probability that the abnormal gas concentration data deviates from the first dynamic baseline and the probability of change of the interfering gas concentration data; and determine whether the abnormal gas concentration data is genuine abnormal data based on the contamination probability.

9. The laboratory gas detection device according to claim 8, characterized in that, The array sensor is arranged on the top of the laboratory and includes infrared sensors, electrochemical sensors and optical sensors arranged in layers from bottom to top. The infrared sensors, electrochemical sensors and optical sensors are used to detect the concentration of infrared active gas, toxic gas and volatile organic compounds, respectively. The array sensor is equipped with a modular interface for detachably inserting various types of gas sensors.

10. The laboratory gas detection device according to claim 9, characterized in that, The laboratory gas detection equipment also includes: A self-cleaning device is arranged at the rear end of the array sensor, close to the optical sensor, for activating a reverse airflow to purge the array sensor; An air sampling port is located at the front end of the array sensor, close to the infrared sensor, and isolated from the self-cleaning device, for collecting laboratory gases; A dustproof net is placed at the front end of the air sampling port to block particulate matter.

Images